| Message ID | 1692048.1681857607@warthog.procyon.org.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | cifs: Fix unbuffered read | expand |
David Howells <dhowells@redhat.com> writes: > If read() is done in an unbuffered manner, such that, say, > cifs_strict_readv() goes through cifs_user_readv() and thence > __cifs_readv(), it doesn't recognise the EOF and keeps indicating to > userspace that it returning full buffers of data. > > This is due to ctx->iter being advanced in cifs_send_async_read() as the > buffer is split up amongst a number of rdata objects. The iterator count > is then used in collect_uncached_read_data() in the non-DIO case to set the > total length read - and thus the return value of sys_read(). But since the > iterator normally gets used up completely during splitting, ctx->total_len > gets overridden to the full amount. > > However, prior to that in collect_uncached_read_data(), we've gone through > the list of rdatas and added up the amount of data we actually received > (which we then throw away). > > Fix this by removing the bit that overrides the amount read in the non-DIO > case and just going with the total added up in the aforementioned loop. > > This was observed by mounting a cifs share with multiple channels, e.g.: > > mount //192.168.6.1/test /test/ -o user=shares,pass=...,max_channels=6 > > and then reading a 1MiB file on the share: > > strace cat /xfstest.test/1M >/dev/null > > Through strace, the same data can be seen being read again and again. > > Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") > Signed-off-by: David Howells <dhowells@redhat.com> > cc: Steve French <smfrench@gmail.com> > cc: Paulo Alcantara <pc@manguebit.com> > cc: Jérôme Glisse <jglisse@redhat.com> > cc: Long Li <longli@microsoft.com> > cc: Enzo Matsumiya <ematsumiya@suse.de> > cc: Shyam Prasad N <nspmangalore@gmail.com> > cc: Rohith Surabattula <rohiths.msft@gmail.com> > cc: Jeff Layton <jlayton@kernel.org> > cc: linux-cifs@vger.kernel.org > --- > fs/cifs/file.c | 4 ---- > 1 file changed, 4 deletions(-) Acked-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Updated to add Paulo's Acked-by and also attached the other fix. Let
me know if any additional feedback/review/testing results
cifs: Reapply lost fix from commit 30b2b2196d6e
Reapply the fix from
30b2b2196d6e ("cifs: do not include page data when checking signature")
that got lost in the iteratorisation of the cifs driver.
On Tue, Apr 18, 2023 at 5:40 PM David Howells <dhowells@redhat.com> wrote:
>
>
> If read() is done in an unbuffered manner, such that, say,
> cifs_strict_readv() goes through cifs_user_readv() and thence
> __cifs_readv(), it doesn't recognise the EOF and keeps indicating to
> userspace that it returning full buffers of data.
>
> This is due to ctx->iter being advanced in cifs_send_async_read() as the
> buffer is split up amongst a number of rdata objects. The iterator count
> is then used in collect_uncached_read_data() in the non-DIO case to set the
> total length read - and thus the return value of sys_read(). But since the
> iterator normally gets used up completely during splitting, ctx->total_len
> gets overridden to the full amount.
>
> However, prior to that in collect_uncached_read_data(), we've gone through
> the list of rdatas and added up the amount of data we actually received
> (which we then throw away).
>
> Fix this by removing the bit that overrides the amount read in the non-DIO
> case and just going with the total added up in the aforementioned loop.
>
> This was observed by mounting a cifs share with multiple channels, e.g.:
>
> mount //192.168.6.1/test /test/ -o user=shares,pass=...,max_channels=6
>
> and then reading a 1MiB file on the share:
>
> strace cat /xfstest.test/1M >/dev/null
>
> Through strace, the same data can be seen being read again and again.
>
> Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list")
> Signed-off-by: David Howells <dhowells@redhat.com>
> cc: Steve French <smfrench@gmail.com>
> cc: Paulo Alcantara <pc@manguebit.com>
> cc: Jérôme Glisse <jglisse@redhat.com>
> cc: Long Li <longli@microsoft.com>
> cc: Enzo Matsumiya <ematsumiya@suse.de>
> cc: Shyam Prasad N <nspmangalore@gmail.com>
> cc: Rohith Surabattula <rohiths.msft@gmail.com>
> cc: Jeff Layton <jlayton@kernel.org>
> cc: linux-cifs@vger.kernel.org
> ---
> fs/cifs/file.c | 4 ----
> 1 file changed, 4 deletions(-)
>
> diff --git a/fs/cifs/file.c b/fs/cifs/file.c
> index 321f9b7c84c9..f8877dc91cc5 100644
> --- a/fs/cifs/file.c
> +++ b/fs/cifs/file.c
> @@ -4010,7 +4010,6 @@ static void
> collect_uncached_read_data(struct cifs_aio_ctx *ctx)
> {
> struct cifs_readdata *rdata, *tmp;
> - struct iov_iter *to = &ctx->iter;
> struct cifs_sb_info *cifs_sb;
> int rc;
>
> @@ -4076,9 +4075,6 @@ collect_uncached_read_data(struct cifs_aio_ctx *ctx)
> kref_put(&rdata->refcount, cifs_readdata_release);
> }
>
> - if (!ctx->direct_io)
> - ctx->total_len = ctx->len - iov_iter_count(to);
> -
> /* mask nodata case */
> if (rc == -ENODATA)
> rc = 0;
>
diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 321f9b7c84c9..f8877dc91cc5 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -4010,7 +4010,6 @@ static void collect_uncached_read_data(struct cifs_aio_ctx *ctx) { struct cifs_readdata *rdata, *tmp; - struct iov_iter *to = &ctx->iter; struct cifs_sb_info *cifs_sb; int rc; @@ -4076,9 +4075,6 @@ collect_uncached_read_data(struct cifs_aio_ctx *ctx) kref_put(&rdata->refcount, cifs_readdata_release); } - if (!ctx->direct_io) - ctx->total_len = ctx->len - iov_iter_count(to); - /* mask nodata case */ if (rc == -ENODATA) rc = 0;
If read() is done in an unbuffered manner, such that, say, cifs_strict_readv() goes through cifs_user_readv() and thence __cifs_readv(), it doesn't recognise the EOF and keeps indicating to userspace that it returning full buffers of data. This is due to ctx->iter being advanced in cifs_send_async_read() as the buffer is split up amongst a number of rdata objects. The iterator count is then used in collect_uncached_read_data() in the non-DIO case to set the total length read - and thus the return value of sys_read(). But since the iterator normally gets used up completely during splitting, ctx->total_len gets overridden to the full amount. However, prior to that in collect_uncached_read_data(), we've gone through the list of rdatas and added up the amount of data we actually received (which we then throw away). Fix this by removing the bit that overrides the amount read in the non-DIO case and just going with the total added up in the aforementioned loop. This was observed by mounting a cifs share with multiple channels, e.g.: mount //192.168.6.1/test /test/ -o user=shares,pass=...,max_channels=6 and then reading a 1MiB file on the share: strace cat /xfstest.test/1M >/dev/null Through strace, the same data can be seen being read again and again. Fixes: d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") Signed-off-by: David Howells <dhowells@redhat.com> cc: Steve French <smfrench@gmail.com> cc: Paulo Alcantara <pc@manguebit.com> cc: Jérôme Glisse <jglisse@redhat.com> cc: Long Li <longli@microsoft.com> cc: Enzo Matsumiya <ematsumiya@suse.de> cc: Shyam Prasad N <nspmangalore@gmail.com> cc: Rohith Surabattula <rohiths.msft@gmail.com> cc: Jeff Layton <jlayton@kernel.org> cc: linux-cifs@vger.kernel.org --- fs/cifs/file.c | 4 ---- 1 file changed, 4 deletions(-)