Date: Thu, 15 Dec 2022 18:46:21 +0000 From: bugzilla-noreply@freebsd.org To: testing@FreeBSD.org Subject: [Bug 268390] Kyua KTLS tests (*bad_minor, *bad_major) fail randomly Message-ID: <bug-268390-32464-5dUUgPzqhd@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-268390-32464@https.bugs.freebsd.org/bugzilla/> References: <bug-268390-32464@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D268390 --- Comment #1 from John Baldwin <jhb@FreeBSD.org> --- I have a series of reviews for this already. I haven't yet patched it to permit ECONNRESET errors from close(). However, it's curious that usleep() makes a difference. I would suspect that would not help as my understandin= g of the race is that the receiving end of the socket pair notices that the error and drops the connection before the sending end calls close() hence close() returning ECONNRESET. The usleep() should only make that worse as the error detection and drop is triggered by receiving the packets, not by the call to recvmsg(). https://github.com/CTSRD-CHERI/cheribsd/issues/1566 is the first review in = the series, though I haven't yet uploaded a workaround for spurious close() err= ors. The third patch deals with spurious errors from shutdown(). --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-268390-32464-5dUUgPzqhd>