Re: [PATCH] 9p: Add missing read barrier in virtio zero-copy path
Gui-Dong Han <[email protected]> Wed, 17 Jun 2026 20:55:29 +0800
| Newsgroups | dev.linux.lists.v9fs,org.kernel.vger.linux-kernel |
|---|---|
| Message-ID | <CALbr=LbuWzRo+fndQFEnhVJTfSPsEfKqwUri9ovyUe+dkXXx9g@mail.gmail.com> |
On Mon, Jun 15, 2026 at 9:13 PM Dominique Martinet <[email protected]> wrote: > > Gui-Dong Han wrote on Fri, May 29, 2026 at 03:54:41PM +0800: > > Commit 2b6e72ed747f ("9P: Add memory barriers to protect request > > fields over cb/rpc threads handoff") added a read barrier after > > p9_client_rpc() waits for req->status, pairing with the write barrier in > > p9_client_cb(). The virtio zero-copy wait path was missed. > > > > Add the same read barrier after the zero-copy wait before reading the > > completed request. > > > > Fixes: abfa034e4b8e ("fs/9p: Update zero-copy implementation in 9p") > > This doesn't make sense, the wmb was added a few years later, so I guess > this should be > Fixes: 2b6e72ed747f ("9P: Add memory barriers to protect request fields over cb/rpc threads handoff") > > With that said, a rmb here makes sense, so picking up (with fix tag > fixed up) Hi Dominique, Thanks for taking it. I was also unsure about the Fixes tag. abfa034e4b8e introduced the lockless zero-copy path, while 2b6e72ed747f introduced the wmb/rmb handoff rule completed here. Your choice makes sense to me. I understand you will fix it up when applying the patch. I can also send a v2 if you prefer. Thanks, Gui-Dong Han