Date: Mon, 06 Feb 2023 01:43:26 +0000 From: bugzilla-noreply@freebsd.org To: bugs@FreeBSD.org Subject: [Bug 269328] nfs client: data corruption using fspacectl and mmap Message-ID: <bug-269328-227-cTQjg2QUTO@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-269328-227@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269328 Rick Macklem <rmacklem@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Assignee|bugs@FreeBSD.org |rmacklem@FreeBSD.org --- Comment #8 from Rick Macklem <rmacklem@FreeBSD.org> --- Created attachment 239937 --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=239937&action=edit do a VOP_FSYNC() after vop_stddeallocate() to push writes of 0s This simple patch fixes the case that the test program finds. I am not 100% sure why it does fix this, but the code should always have done this since having the changes committed to the server before VOP_DEALLOCATE() returns is required. Btw, for a newer NFSv4.2 server that supports a Deallocate operation, your test program works. (I hacked my server so that Deallocate was disabled and was able to reproduce the failure with your test program.) Maybe asomers@ can test the patch and see if it fixes all the NFS cases he observes? -- You are receiving this mail because: You are the assignee for the bug.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-269328-227-cTQjg2QUTO>