Andrei Vagin 58b7064773 fs: sendfile handles O_NONBLOCK of out_fd ... commit bdeb77bc2c upstream. sendfile has to return EAGAIN if out_fd is nonblocking and the write into it would block. Here is a small reproducer for the problem: #define _GNU_SOURCE /* See feature_test_macros(7) */ #include <fcntl.h> #include <stdio.h> #include <unistd.h> #include <errno.h> #include <sys/stat.h> #include <sys/types.h> #include <sys/sendfile.h> #define FILE_SIZE (1UL << 30) int main(int argc, char **argv) { int p[2], fd; if (pipe2(p, O_NONBLOCK)) return 1; fd = open(argv[1], O_RDWR | O_TMPFILE, 0666); if (fd < 0) return 1; ftruncate(fd, FILE_SIZE); if (sendfile(p[1], fd, 0, FILE_SIZE) == -1) { fprintf(stderr, "FAIL\n"); } if (sendfile(p[1], fd, 0, FILE_SIZE) != -1 || errno != EAGAIN) { fprintf(stderr, "FAIL\n"); } return 0; } It worked before b964bf53e5, it is stuck after b964bf53e5, and it works again with this fix. This regression occurred because do_splice_direct() calls pipe_write that handles O_NONBLOCK. Here is a trace log from the reproducer: 1) | __x64_sys_sendfile64() { 1) | do_sendfile() { 1) | __fdget() 1) | rw_verify_area() 1) | __fdget() 1) | rw_verify_area() 1) | do_splice_direct() { 1) | rw_verify_area() 1) | splice_direct_to_actor() { 1) | do_splice_to() { 1) | rw_verify_area() 1) | generic_file_splice_read() 1) + 74.153 us | } 1) | direct_splice_actor() { 1) | iter_file_splice_write() { 1) | __kmalloc() 1) 0.148 us | pipe_lock(); 1) 0.153 us | splice_from_pipe_next.part.0(); 1) 0.162 us | page_cache_pipe_buf_confirm(); ... 16 times 1) 0.159 us | page_cache_pipe_buf_confirm(); 1) | vfs_iter_write() { 1) | do_iter_write() { 1) | rw_verify_area() 1) | do_iter_readv_writev() { 1) | pipe_write() { 1) | mutex_lock() 1) 0.153 us | mutex_unlock(); 1) 1.368 us | } 1) 1.686 us | } 1) 5.798 us | } 1) 6.084 us | } 1) 0.174 us | kfree(); 1) 0.152 us | pipe_unlock(); 1) + 14.461 us | } 1) + 14.783 us | } 1) 0.164 us | page_cache_pipe_buf_release(); ... 16 times 1) 0.161 us | page_cache_pipe_buf_release(); 1) | touch_atime() 1) + 95.854 us | } 1) + 99.784 us | } 1) ! 107.393 us | } 1) ! 107.699 us | } Link: https://lkml.kernel.org/r/20220415005015.525191-1-avagin@gmail.com Fixes: b964bf53e5 ("teach sendfile(2) to handle send-to-pipe directly") Signed-off-by: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>		2022-08-03 12:05:16 +02:00
..
9p	9p: fix EBADF errors in cached mode	2022-06-29 09:04:26 +02:00
adfs
affs
afs	netfs: do not unlock and put the folio twice	2022-07-22 10:21:42 +02:00
autofs
befs
bfs
btrfs	btrfs: zoned: fix a leaked bioc in read_zone_info	2022-07-22 10:21:21 +02:00
cachefiles
ceph	netfs: do not unlock and put the folio twice	2022-07-22 10:21:42 +02:00
cifs	smb3: workaround negprot bug in some Samba servers	2022-07-22 10:21:42 +02:00
coda
configfs
cramfs
crypto
debugfs
devpts
dlm	dlm: fix pending remove if msg allocation fails	2022-07-29 17:28:15 +02:00
ecryptfs
efivarfs
efs
erofs	erofs: fix 'backmost' member of z_erofs_decompress_frontend	2022-06-14 18:44:57 +02:00
exfat	exfat: use updated exfat_chain directly during renaming	2022-07-29 17:28:16 +02:00
exportfs	exportfs: support idmapped mounts	2022-06-09 10:30:56 +02:00
ext2
ext4	ext4: add reserved GDT blocks check	2022-06-22 14:28:11 +02:00
f2fs	f2fs: do not count ENOENT for error case	2022-06-29 09:04:38 +02:00
fat	fat: add ratelimit to fat*_ent_bread()	2022-06-09 10:29:49 +02:00
freevxfs
fscache	fscache: Fix if condition in fscache_wait_on_volume_collision()	2022-07-12 16:42:17 +02:00
fuse
gfs2	gfs2: use i_lock spin_lock for inode qadata	2022-06-09 10:29:47 +02:00
hfs
hfsplus
hostfs
hpfs
hugetlbfs	hugetlbfs: fix hugetlbfs_statfs() locking	2022-06-09 10:30:31 +02:00
iomap	iomap: iomap_write_failed fix	2022-06-09 10:30:08 +02:00
isofs
jbd2
jffs2	jffs2: fix memory leak in jffs2_do_fill_super	2022-06-14 18:44:55 +02:00
jfs	fs: jfs: fix possible NULL pointer dereference in dbFree()	2022-06-09 10:29:49 +02:00
kernfs	kernfs: Separate kernfs_pr_cont_buf and rename_lock.	2022-06-14 18:45:11 +02:00
ksmbd	ksmbd: use SOCK_NONBLOCK type for kernel_accept()	2022-07-22 10:21:46 +02:00
lockd	lockd: fix nlm_close_files	2022-07-22 10:21:35 +02:00
minix
netfs	netfs: do not unlock and put the folio twice	2022-07-22 10:21:42 +02:00
nfs	NFSv4: Add an fattr allocation to _nfs4_discover_trunking()	2022-07-07 17:54:52 +02:00
nfs_common
nfsd	NFSD: Decode NFSv4 birth time attribute	2022-07-22 10:21:35 +02:00
nilfs2	nilfs2: fix incorrect masking of permission flags for symlinks	2022-07-22 10:21:21 +02:00
nls
notify	fanotify: refine the validation checks on non-dir inode mask	2022-07-07 17:54:57 +02:00
ntfs	ntfs: fix use-after-free in ntfs_ucsncmp()	2022-08-03 12:05:16 +02:00
ntfs3	fs/ntfs3: Fix invalid free in log_replay	2022-06-09 10:30:56 +02:00
ocfs2	Revert "ocfs2: mount shared volume without ha stack"	2022-08-03 12:05:16 +02:00
omfs
openpromfs
orangefs
overlayfs
proc	vmcore: convert copy_oldmem_page() to take an iov_iter	2022-06-29 09:04:36 +02:00
pstore
qnx4
qnx6
quota	quota: Prevent memory allocation recursion while holding dq_lock	2022-06-22 14:27:51 +02:00
ramfs
reiserfs
romfs
smbfs_common
squashfs
sysfs
sysv
tracefs
ubifs
udf	udf: Avoid using stale lengthOfImpUse	2022-05-10 13:30:32 +02:00
ufs
unicode
vboxsf
verity
xfs
zonefs	zonefs: fix zonefs_iomap_begin() for reads	2022-06-25 15:29:46 +02:00
aio.c
anon_inodes.c
attr.c	fs: account for group membership	2022-06-22 14:28:10 +02:00
bad_inode.c
binfmt_aout.c
binfmt_elf.c
binfmt_elf_fdpic.c
binfmt_elf_test.c
binfmt_flat.c	binfmt_flat: do not stop relocating GOT entries prematurely on riscv	2022-06-09 10:29:26 +02:00
binfmt_misc.c
binfmt_script.c
buffer.c
char_dev.c
compat_binfmt_elf.c
coredump.c
d_path.c
dax.c	dax: fix cache flush on PMD-mapped pages	2022-06-09 10:30:28 +02:00
dcache.c
direct-io.c
drop_caches.c
eventfd.c
eventpoll.c
exec.c	fix race between exit_itimers() and /proc/pid/timers	2022-07-22 10:21:18 +02:00
fcntl.c
fhandle.c
file.c
file_table.c
filesystems.c
fs-writeback.c	writeback: Fix inode->i_io_list not be protected by inode->i_lock error	2022-06-14 18:45:18 +02:00
fs_context.c
fs_parser.c
fs_pin.c
fs_struct.c
fs_types.c
fsopen.c
init.c
inode.c	writeback: Fix inode->i_io_list not be protected by inode->i_lock error	2022-06-14 18:45:18 +02:00
internal.h
io-wq.c
io-wq.h
io_uring.c	io_uring: fix provided buffer import	2022-07-12 16:42:11 +02:00
ioctl.c
Kconfig
Kconfig.binfmt
kernel_read_file.c
libfs.c
locks.c
Makefile
mbcache.c
mount.h
mpage.c
namei.c	fs: add two trivial lookup helpers	2022-06-09 10:30:56 +02:00
namespace.c	fs: hold writers when changing mount's idmapping	2022-06-09 10:29:42 +02:00
no-block.c
nsfs.c
open.c
pipe.c	pipe: Fix missing lock in pipe_resize_ring()	2022-06-06 08:48:53 +02:00
pnode.c
pnode.h
posix_acl.c
proc_namespace.c
read_write.c	fs: sendfile handles O_NONBLOCK of out_fd	2022-08-03 12:05:16 +02:00
readdir.c
remap_range.c	fs/remap: constrain dedupe of EOF blocks	2022-07-22 10:21:21 +02:00
select.c
seq_file.c	rxrpc: Fix locking issue	2022-06-09 10:30:19 +02:00
signalfd.c
splice.c
stack.c
stat.c
statfs.c
super.c
sync.c
sysctls.c
timerfd.c
userfaultfd.c	userfaultfd: provide properly masked address for huge-pages	2022-08-03 12:05:16 +02:00
utimes.c
xattr.c