In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
| linux | | https://git.kernel.org/linus/3261400639463a853ba2b3be8bd009c2a8089775 | https://git.kernel.org/linus/c07aea3ef4d4076f18f567b98ed01e082e02ed51 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
IntheLinuxkernel,thefollowingvulnerabilityhasbeenresolved:tcp:TXzerocopyshouldnotsense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
| linux | | https://git.kernel.org/linus/3261400639463a853ba2b3be8bd009c2a8089775 | https://git.kernel.org/linus/c07aea3ef4d4076f18f567b98ed01e082e02ed51 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got arecent syzbot report [1] showing apossible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 (net: introduce__skb_fill_page_desc_noacc )as aprereq.Race is more probable after commit c07aea3ef4d4( mm: add asignature in struct page )because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly afalse positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn /tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
| linux | | https://git.kernel.org/linus/3261400639463a853ba2b3be8bd009c2a8089775 | https://git.kernel.org/linus/c07aea3ef4d4076f18f567b98ed01e082e02ed51 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe gota recent syzbot report [1] showinga possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6( net: introduce__skb_fill_page_desc_noacc) asa prereq.Race is more probable after commit c07aea3ef4d4( mm: adda signature in struct page) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostlya false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn/ tcp_build_fragwrite to 0xffffea0004a1d2c8 of8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU:1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
| linux | | https://git.kernel.org/linus/3261400639463a853ba2b3be8bd009c2a8089775 | https://git.kernel.org/linus/c07aea3ef4d4076f18f567b98ed01e082e02ed51 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got arecent syzbot report [1] showing apossible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 (net: introduce__skb_fill_page_desc_noacc )as aprereq.Race is more probable after commit c07aea3ef4d4( mm: add asignature in struct page )because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly afalse positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn /tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe got a recent syzbot report [1] showing a possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6 ( net: introduce__skb_fill_page_desc_noacc ) as a prereq.Race is more probable after commit c07aea3ef4d4( mm: add a signature in struct page ) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostly a false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn / tcp_build_fragwrite to 0xffffea0004a1d2c8 of 8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of 8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU: 1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
| linux | | https://git.kernel.org/linus/3261400639463a853ba2b3be8bd009c2a8089775 | https://git.kernel.org/linus/c07aea3ef4d4076f18f567b98ed01e082e02ed51 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:tcp: TX zerocopy should not sense pfmemalloc statusWe gota recent syzbot report [1] showinga possible misuseof pfmemalloc page status in TCP zerocopy paths.Indeed, for pages coming from user space or other layers,using page_is_pfmemalloc() is moot, and possibly could givefalse positives.There has been attempts to make page_is_pfmemalloc() more robust,but not using it in the first place in this context is probably better,removing cpu cycles.Note to stable teams :You need to backport 84ce071e38a6( net: introduce__skb_fill_page_desc_noacc) asa prereq.Race is more probable after commit c07aea3ef4d4( mm: adda signature in struct page) because page_is_pfmemalloc()is now using low order bit from page->lru.next, which can changemore often than page->index.Low order bit should never be set for lru.next (when used as an anchorin LRU list), so KCSAN report is mostlya false positive.Backporting to older kernel versions seems not necessary.[1]BUG: KCSAN: data-race in lru_add_fn/ tcp_build_fragwrite to 0xffffea0004a1d2c8 of8 bytes by task 18600 on cpu 0:__list_add include/linux/list.h:73 [inline]list_add include/linux/list.h:88 [inline]lruvec_add_folio include/linux/mm_inline.h:105 [inline]lru_add_fn+0x440/0x520 mm/swap.c:228folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246folio_batch_add_and_move mm/swap.c:263 [inline]folio_add_lru+0xf1/0x140 mm/swap.c:490filemap_add_folio+0xf8/0x150 mm/filemap.c:948__filemap_get_folio+0x510/0x6d0 mm/filemap.c:1981pagecache_get_page+0x26/0x190 mm/folio-compat.c:104grab_cache_page_write_begin+0x2a/0x30 mm/folio-compat.c:116ext4_da_write_begin+0x2dd/0x5f0 fs/ext4/inode.c:2988generic_perform_write+0x1d4/0x3f0 mm/filemap.c:3738ext4_buffered_write_iter+0x235/0x3e0 fs/ext4/file.c:270ext4_file_write_iter+0x2e3/0x1210call_write_iter include/linux/fs.h:2187 [inline]new_sync_write fs/read_write.c:491 [inline]vfs_write+0x468/0x760 fs/read_write.c:578ksys_write+0xe8/0x1a0 fs/read_write.c:631__do_sys_write fs/read_write.c:643 [inline]__se_sys_write fs/read_write.c:640 [inline]__x64_sys_write+0x3e/0x50 fs/read_write.c:640do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdread to 0xffffea0004a1d2c8 of8 bytes by task 18611 on cpu 1:page_is_pfmemalloc include/linux/mm.h:1740 [inline]__skb_fill_page_desc include/linux/skbuff.h:2422 [inline]skb_fill_page_desc include/linux/skbuff.h:2443 [inline]tcp_build_frag+0x613/0xb20 net/ipv4/tcp.c:1018do_tcp_sendpages+0x3e8/0xaf0 net/ipv4/tcp.c:1075tcp_sendpage_locked net/ipv4/tcp.c:1140 [inline]tcp_sendpage+0x89/0xb0 net/ipv4/tcp.c:1150inet_sendpage+0x7f/0xc0 net/ipv4/af_inet.c:833kernel_sendpage+0x184/0x300 net/socket.c:3561sock_sendpage+0x5a/0x70 net/socket.c:1054pipe_to_sendpage+0x128/0x160 fs/splice.c:361splice_from_pipe_feed fs/splice.c:415 [inline]__splice_from_pipe+0x222/0x4d0 fs/splice.c:559splice_from_pipe fs/splice.c:594 [inline]generic_splice_sendpage+0x89/0xc0 fs/splice.c:743do_splice_from fs/splice.c:764 [inline]direct_splice_actor+0x80/0xa0 fs/splice.c:931splice_direct_to_actor+0x305/0x620 fs/splice.c:886do_splice_direct+0xfb/0x180 fs/splice.c:974do_sendfile+0x3bf/0x910 fs/read_write.c:1249__do_sys_sendfile64 fs/read_write.c:1317 [inline]__se_sys_sendfile64 fs/read_write.c:1303 [inline]__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1303do_syscall_x64 arch/x86/entry/common.c:50 [inline]do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80entry_SYSCALL_64_after_hwframe+0x63/0xcdvalue changed: 0x0000000000000000 -> 0xffffea0004a1d288Reported by Kernel Concurrency Sanitizer on:CPU:1 PID: 18611 Comm: syz-executor.4 Not tainted 6.0.0-rc2-syzkaller-00248-ge022620b5d05-dirty #0Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022
