linux-stable.git/fs/fuse, branch linux-6.11.y

virtiofs: use pages instead of pointer for kernel direct IO

2024-12-05T12:53:48+00:00

[ Upstream commit 41748675c0bf252b3c5f600a95830f0936d366c1 ]

When trying to insert a 10MB kernel module kept in a virtio-fs with cache
disabled, the following warning was reported:

  ------------[ cut here ]------------
  WARNING: CPU: 1 PID: 404 at mm/page_alloc.c:4551 ......
  Modules linked in:
  CPU: 1 PID: 404 Comm: insmod Not tainted 6.9.0-rc5+ #123
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) ......
  RIP: 0010:__alloc_pages+0x2bf/0x380
  ......
  Call Trace:
   
   ? __warn+0x8e/0x150
   ? __alloc_pages+0x2bf/0x380
   __kmalloc_large_node+0x86/0x160
   __kmalloc+0x33c/0x480
   virtio_fs_enqueue_req+0x240/0x6d0
   virtio_fs_wake_pending_and_unlock+0x7f/0x190
   queue_request_and_unlock+0x55/0x60
   fuse_simple_request+0x152/0x2b0
   fuse_direct_io+0x5d2/0x8c0
   fuse_file_read_iter+0x121/0x160
   __kernel_read+0x151/0x2d0
   kernel_read+0x45/0x50
   kernel_read_file+0x1a9/0x2a0
   init_module_from_file+0x6a/0xe0
   idempotent_init_module+0x175/0x230
   __x64_sys_finit_module+0x5d/0xb0
   x64_sys_call+0x1c3/0x9e0
   do_syscall_64+0x3d/0xc0
   entry_SYSCALL_64_after_hwframe+0x4b/0x53
   ......
   
  ---[ end trace 0000000000000000 ]---

The warning is triggered as follows:

1) syscall finit_module() handles the module insertion and it invokes
kernel_read_file() to read the content of the module first.

2) kernel_read_file() allocates a 10MB buffer by using vmalloc() and
passes it to kernel_read(). kernel_read() constructs a kvec iter by
using iov_iter_kvec() and passes it to fuse_file_read_iter().

3) virtio-fs disables the cache, so fuse_file_read_iter() invokes
fuse_direct_io(). As for now, the maximal read size for kvec iter is
only limited by fc->max_read. For virtio-fs, max_read is UINT_MAX, so
fuse_direct_io() doesn't split the 10MB buffer. It saves the address and
the size of the 10MB-sized buffer in out_args[0] of a fuse request and
passes the fuse request to virtio_fs_wake_pending_and_unlock().

4) virtio_fs_wake_pending_and_unlock() uses virtio_fs_enqueue_req() to
queue the request. Because virtiofs need DMA-able address, so
virtio_fs_enqueue_req() uses kmalloc() to allocate a bounce buffer for
all fuse args, copies these args into the bounce buffer and passed the
physical address of the bounce buffer to virtiofsd. The total length of
these fuse args for the passed fuse request is about 10MB, so
copy_args_to_argbuf() invokes kmalloc() with a 10MB size parameter and
it triggers the warning in __alloc_pages():

	if (WARN_ON_ONCE_GFP(order > MAX_PAGE_ORDER, gfp))
		return NULL;

5) virtio_fs_enqueue_req() will retry the memory allocation in a
kworker, but it won't help, because kmalloc() will always return NULL
due to the abnormal size and finit_module() will hang forever.

A feasible solution is to limit the value of max_read for virtio-fs, so
the length passed to kmalloc() will be limited. However it will affect
the maximal read size for normal read. And for virtio-fs write initiated
from kernel, it has the similar problem but now there is no way to limit
fc->max_write in kernel.

So instead of limiting both the values of max_read and max_write in
kernel, introducing use_pages_for_kvec_io in fuse_conn and setting it as
true in virtiofs. When use_pages_for_kvec_io is enabled, fuse will use
pages instead of pointer to pass the KVEC_IO data.

After switching to pages for KVEC_IO data, these pages will be used for
DMA through virtio-fs. If these pages are backed by vmalloc(),
{flush|invalidate}_kernel_vmap_range() are necessary to flush or
invalidate the cache before the DMA operation. So add two new fields in
fuse_args_pages to record the base address of vmalloc area and the
condition indicating whether invalidation is needed. Perform the flush
in fuse_get_user_pages() for write operations and the invalidation in
fuse_release_user_pages() for read operations.

It may seem necessary to introduce another field in fuse_conn to
indicate that these KVEC_IO pages are used for DMA, However, considering
that virtio-fs is currently the only user of use_pages_for_kvec_io, just
reuse use_pages_for_kvec_io to indicate that these pages will be used
for DMA.

Fixes: a62a8ef9d97d ("virtio-fs: add virtiofs filesystem")
Signed-off-by: Hou Tao 
Tested-by: Jingbo Xu 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Sasha Levin

fuse: update inode size after extending passthrough write

2024-11-01T01:02:38+00:00

[ Upstream commit 20121d3f58f06e977ca43eb6efe1fb23b1d2f6d9 ]

yangyun reported that libfuse test test_copy_file_range() copies zero
bytes from a newly written file when fuse passthrough is enabled.

The reason is that extending passthrough write is not updating the fuse
inode size and when vfs_copy_file_range() observes a zero size inode,
it returns without calling the filesystem copy_file_range() method.

Fix this by adjusting the fuse inode size after an extending passthrough
write.

This does not provide cache coherency of fuse inode attributes and
backing inode attributes, but it should prevent situations where fuse
inode size is too small, causing read/copy to be wrongly shortened.

Reported-by: yangyun 
Closes: https://github.com/libfuse/libfuse/issues/1048
Fixes: 57e1176e6086 ("fuse: implement read/write passthrough")
Signed-off-by: Amir Goldstein 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Sasha Levin

fs: pass offset and result to backing_file end_write() callback

2024-11-01T01:02:38+00:00

[ Upstream commit f03b296e8b516dbd63f57fc9056c1b0da1b9a0ff ]

This is needed for extending fuse inode size after fuse passthrough write.

Suggested-by: Miklos Szeredi 
Link: https://lore.kernel.org/linux-fsdevel/CAJfpegs=cvZ_NYy6Q_D42XhYS=Sjj5poM1b5TzXzOVvX=R36aA@mail.gmail.com/
Signed-off-by: Amir Goldstein 
Signed-off-by: Miklos Szeredi 
Stable-dep-of: 20121d3f58f0 ("fuse: update inode size after extending passthrough write")
Signed-off-by: Sasha Levin

fuse: use exclusive lock when FUSE_I_CACHE_IO_MODE is set

2024-10-04T14:38:27+00:00

commit 2f3d8ff457982f4055fe8f7bf19d3821ba22c376 upstream.

This may be a typo. The comment has said shared locks are
not allowed when this bit is set. If using shared lock, the
wait in `fuse_file_cached_io_open` may be forever.

Fixes: 205c1d802683 ("fuse: allow parallel dio writes with FUSE_DIRECT_IO_ALLOW_MMAP")
CC: stable@vger.kernel.org # v6.9
Signed-off-by: yangyun 
Reviewed-by: Bernd Schubert 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Greg Kroah-Hartman

Merge tag 'fuse-fixes-6.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

2024-09-03T19:32:00+00:00

Pull fuse fixes from Miklos Szeredi:

 - Fix EIO if splice and page stealing are enabled on the fuse device

 - Disable problematic combination of passthrough and writeback-cache

 - Other bug fixes found by code review

* tag 'fuse-fixes-6.11-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  fuse: disable the combination of passthrough and writeback cache
  fuse: update stats for pages in dropped aux writeback list
  fuse: clear PG_uptodate when using a stolen page
  fuse: fix memory leak in fuse_create_open
  fuse: check aborted connection before adding requests to pending list for resending
  fuse: use unsigned type for getxattr/listxattr size truncation

fuse: disable the combination of passthrough and writeback cache

2024-08-29T09:43:01+00:00

Current design and handling of passthrough is without fuse
caching and with that FUSE_WRITEBACK_CACHE is conflicting.

Fixes: 7dc4e97a4f9a ("fuse: introduce FUSE_PASSTHROUGH capability")
Cc: stable@kernel.org # v6.9
Signed-off-by: Bernd Schubert 
Acked-by: Amir Goldstein 
Reviewed-by: Josef Bacik 
Signed-off-by: Miklos Szeredi

fuse: update stats for pages in dropped aux writeback list

2024-08-28T16:10:29+00:00

In the case where the aux writeback list is dropped (e.g. the pages
have been truncated or the connection is broken), the stats for
its pages and backing device info need to be updated as well.

Fixes: e2653bd53a98 ("fuse: fix leaked aux requests")
Signed-off-by: Joanne Koong 
Reviewed-by: Josef Bacik 
Cc:  # v5.1
Signed-off-by: Miklos Szeredi

fuse: clear PG_uptodate when using a stolen page

2024-08-28T16:10:29+00:00

Originally when a stolen page was inserted into fuse's page cache by
fuse_try_move_page(), it would be marked uptodate.  Then
fuse_readpages_end() would call SetPageUptodate() again on the already
uptodate page.

Commit 413e8f014c8b ("fuse: Convert fuse_readpages_end() to use
folio_end_read()") changed that by replacing the SetPageUptodate() +
unlock_page() combination with folio_end_read(), which does mostly the
same, except it sets the uptodate flag with an xor operation, which in the
above scenario resulted in the uptodate flag being cleared, which in turn
resulted in EIO being returned on the read.

Fix by clearing PG_uptodate instead of setting it in fuse_try_move_page(),
conforming to the expectation of folio_end_read().

Reported-by: Jürg Billeter 
Debugged-by: Matthew Wilcox 
Fixes: 413e8f014c8b ("fuse: Convert fuse_readpages_end() to use folio_end_read()")
Cc:  # v6.10
Signed-off-by: Miklos Szeredi

fuse: fix memory leak in fuse_create_open

2024-08-28T16:10:29+00:00

The memory of struct fuse_file is allocated but not freed
when get_create_ext return error.

Fixes: 3e2b6fdbdc9a ("fuse: send security context of inode on file")
Cc: stable@vger.kernel.org # v5.17
Signed-off-by: yangyun 
Signed-off-by: Miklos Szeredi

fuse: check aborted connection before adding requests to pending list for resending

2024-08-28T16:10:29+00:00

There is a race condition where inflight requests will not be aborted if
they are in the middle of being re-sent when the connection is aborted.

If fuse_resend has already moved all the requests in the fpq->processing
lists to its private queue ("to_queue") and then the connection starts
and finishes aborting, these requests will be added to the pending queue
and remain on it indefinitely.

Fixes: 760eac73f9f6 ("fuse: Introduce a new notification type for resend pending requests")
Signed-off-by: Joanne Koong 
Reviewed-by: Josef Bacik 
Reviewed-by: Jingbo Xu 
Cc:  # v6.9
Signed-off-by: Miklos Szeredi