From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 31 Oct 2019 11:47:32 -0400 From: Vivek Goyal Message-ID: <20191031154732.GC7308@redhat.com> References: <20191023122523.1816-1-misono.tomohiro@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20191023122523.1816-1-misono.tomohiro@jp.fujitsu.com> Subject: Re: [Virtio-fs] [PATCH] virtiofsd: Fix data corruption with O_APPEND wirte in writeback mode List-Id: Development discussions about virtio-fs List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Misono Tomohiro Cc: virtio-fs@redhat.com, qemu-devel@nongnu.org On Wed, Oct 23, 2019 at 09:25:23PM +0900, Misono Tomohiro wrote: > When writeback mode is enabled (-o writeback), O_APPEND handling is > done in kernel. Therefore virtiofsd clears O_APPEND flag when open. > Otherwise O_APPEND flag takes precedence over pwrite() and write > data may corrupt. > > Currently clearing O_APPEND flag is done in lo_open(), but we also > need the same operation in lo_create(). So, factor out the flag > update operation in lo_open() to update_open_flags() and call it > in both lo_open() and lo_create(). > > This fixes the failure of xfstest generic/069 in writeback mode > (which tests O_APPEND write data integrity). > > Signed-off-by: Misono Tomohiro Reviewed-by: Vivek Goyal Thanks Vivek > --- > contrib/virtiofsd/passthrough_ll.c | 56 +++++++++++++++--------------- > 1 file changed, 28 insertions(+), 28 deletions(-) > > diff --git a/contrib/virtiofsd/passthrough_ll.c b/contrib/virtiofsd/passthrough_ll.c > index e8892c3c32..79fb78ecce 100644 > --- a/contrib/virtiofsd/passthrough_ll.c > +++ b/contrib/virtiofsd/passthrough_ll.c > @@ -1733,6 +1733,32 @@ static void lo_releasedir(fuse_req_t req, fuse_ino_t ino, struct fuse_file_info > fuse_reply_err(req, 0); > } > > +static void update_open_flags(int writeback, struct fuse_file_info *fi) > +{ > + /* With writeback cache, kernel may send read requests even > + when userspace opened write-only */ > + if (writeback && (fi->flags & O_ACCMODE) == O_WRONLY) { > + fi->flags &= ~O_ACCMODE; > + fi->flags |= O_RDWR; > + } > + > + /* With writeback cache, O_APPEND is handled by the kernel. > + This breaks atomicity (since the file may change in the > + underlying filesystem, so that the kernel's idea of the > + end of the file isn't accurate anymore). In this example, > + we just accept that. A more rigorous filesystem may want > + to return an error here */ > + if (writeback && (fi->flags & O_APPEND)) > + fi->flags &= ~O_APPEND; > + > + /* > + * O_DIRECT in guest should not necessarily mean bypassing page > + * cache on host as well. If somebody needs that behavior, it > + * probably should be a configuration knob in daemon. > + */ > + fi->flags &= ~O_DIRECT; > +} > + > static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name, > mode_t mode, struct fuse_file_info *fi) > { > @@ -1760,12 +1786,7 @@ static void lo_create(fuse_req_t req, fuse_ino_t parent, const char *name, > if (err) > goto out; > > - /* > - * O_DIRECT in guest should not necessarily mean bypassing page > - * cache on host as well. If somebody needs that behavior, it > - * probably should be a configuration knob in daemon. > - */ > - fi->flags &= ~O_DIRECT; > + update_open_flags(lo->writeback, fi); > > fd = openat(parent_inode->fd, name, > (fi->flags | O_CREAT) & ~O_NOFOLLOW, mode); > @@ -1966,28 +1987,7 @@ static void lo_open(fuse_req_t req, fuse_ino_t ino, struct fuse_file_info *fi) > > fuse_log(FUSE_LOG_DEBUG, "lo_open(ino=%" PRIu64 ", flags=%d)\n", ino, fi->flags); > > - /* With writeback cache, kernel may send read requests even > - when userspace opened write-only */ > - if (lo->writeback && (fi->flags & O_ACCMODE) == O_WRONLY) { > - fi->flags &= ~O_ACCMODE; > - fi->flags |= O_RDWR; > - } > - > - /* With writeback cache, O_APPEND is handled by the kernel. > - This breaks atomicity (since the file may change in the > - underlying filesystem, so that the kernel's idea of the > - end of the file isn't accurate anymore). In this example, > - we just accept that. A more rigorous filesystem may want > - to return an error here */ > - if (lo->writeback && (fi->flags & O_APPEND)) > - fi->flags &= ~O_APPEND; > - > - /* > - * O_DIRECT in guest should not necessarily mean bypassing page > - * cache on host as well. If somebody needs that behavior, it > - * probably should be a configuration knob in daemon. > - */ > - fi->flags &= ~O_DIRECT; > + update_open_flags(lo->writeback, fi); > > sprintf(buf, "%i", lo_fd(req, ino)); > fd = openat(lo->proc_self_fd, buf, fi->flags & ~O_NOFOLLOW); > -- > 2.21.0 > > _______________________________________________ > Virtio-fs mailing list > Virtio-fs@redhat.com > https://www.redhat.com/mailman/listinfo/virtio-fs