net: do not sense pfmemalloc status in skb_append_pagefrags()

In the Linux kernel, the following vulnerability has been resolved:

net: do not sense pfmemalloc status in skb_append_pagefrags()

skb_append_pagefrags() is used by af_unix and udp sendpage()

implementation so far.

In commit 326140063946 ("tcp: TX zerocopy should not sense

pfmemalloc status") we explained why we should not sense

pfmemalloc status for pages owned by user space.

We should also use skb_fill_page_desc_noacc()

in skb_append_pagefrags() to avoid following KCSAN report:

BUG: KCSAN: data-race in lru_add_fn / skb_append_pagefrags

write to 0xffffea00058fc1c8 of 8 bytes by task 17319 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:323 [inline]

lru_add_fn+0x327/0x410 mm/swap.c:228

folio_batch_move_lru+0x1e1/0x2a0 mm/swap.c:246

lru_add_drain_cpu+0x73/0x250 mm/swap.c:669

lru_add_drain+0x21/0x60 mm/swap.c:773

free_pages_and_swap_cache+0x16/0x70 mm/swap_state.c:311

tlb_batch_pages_flush mm/mmu_gather.c:59 [inline]

tlb_flush_mmu_free mm/mmu_gather.c:256 [inline]

tlb_flush_mmu+0x5b2/0x640 mm/mmu_gather.c:263

tlb_finish_mmu+0x86/0x100 mm/mmu_gather.c:363

exit_mmap+0x190/0x4d0 mm/mmap.c:3098

__mmput+0x27/0x1b0 kernel/fork.c:1185

mmput+0x3d/0x50 kernel/fork.c:1207

copy_process+0x19fc/0x2100 kernel/fork.c:2518

kernel_clone+0x166/0x550 kernel/fork.c:2671

__do_sys_clone kernel/fork.c:2812 [inline]

__se_sys_clone kernel/fork.c:2796 [inline]

__x64_sys_clone+0xc3/0xf0 kernel/fork.c:2796

do_syscall_x64 arch/x86/entry/common.c:50 [inline]

do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80

entry_SYSCALL_64_after_hwframe+0x63/0xcd

read to 0xffffea00058fc1c8 of 8 bytes by task 17325 on cpu 1:

page_is_pfmemalloc include/linux/mm.h:1817 [inline]

__skb_fill_page_desc include/linux/skbuff.h:2432 [inline]

skb_fill_page_desc include/linux/skbuff.h:2453 [inline]

skb_append_pagefrags+0x210/0x600 net/core/skbuff.c:3974

unix_stream_sendpage+0x45e/0x990 net/unix/af_unix.c:2338

kernel_sendpage+0x184/0x300 net/socket.c:3561

sock_sendpage+0x5a/0x70 net/socket.c:1054

pipe_to_sendpage+0x128/0x160 fs/splice.c:361

splice_from_pipe_feed fs/splice.c:415 [inline]

__splice_from_pipe+0x222/0x4d0 fs/splice.c:559

splice_from_pipe fs/splice.c:594 [inline]

generic_splice_sendpage+0x89/0xc0 fs/splice.c:743

do_splice_from fs/splice.c:764 [inline]

direct_splice_actor+0x80/0xa0 fs/splice.c:931

splice_direct_to_actor+0x305/0x620 fs/splice.c:886

do_splice_direct+0xfb/0x180 fs/splice.c:974

do_sendfile+0x3bf/0x910 fs/read_write.c:1255

__do_sys_sendfile64 fs/read_write.c:1323 [inline]

__se_sys_sendfile64 fs/read_write.c:1309 [inline]

__x64_sys_sendfile64+0x10c/0x150 fs/read_write.c:1309

do_syscall_x64 arch/x86/entry/common.c:50 [inline]

do_syscall_64+0x2b/0x70 arch/x86/entry/common.c:80

entry_SYSCALL_64_after_hwframe+0x63/0xcd

value changed: 0x0000000000000000 -> 0xffffea00058fc188

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 17325 Comm: syz-executor.0 Not tainted 6.1.0-rc1-syzkaller-00158-g440b7895c990-dirty #0

Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/11/2022