From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45311) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ey7R4-0008Kj-SR for qemu-devel@nongnu.org; Mon, 19 Mar 2018 22:55:39 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ey7R1-0001bc-0B for qemu-devel@nongnu.org; Mon, 19 Mar 2018 22:55:38 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:52670 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1ey7R0-0001au-LE for qemu-devel@nongnu.org; Mon, 19 Mar 2018 22:55:34 -0400 Date: Tue, 20 Mar 2018 10:55:27 +0800 From: Peter Xu Message-ID: <20180320025527.GA20227@xz-mi> References: <20180316140508.863778-1-eblake@redhat.com> <20180319092618.GF2477@xz-mi> <3a7260c1-bb51-63a0-5a1b-588a7b1757b3@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PULL 00/38] QAPI patches for 2018-03-12, 2.12 softfreeze List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eric Blake Cc: Peter Maydell , QEMU Developers On Mon, Mar 19, 2018 at 10:27:41AM -0500, Eric Blake wrote: > On 03/19/2018 09:57 AM, Eric Blake wrote: > > On 03/19/2018 04:26 AM, Peter Xu wrote: > >=20 > > > > > for you to fetch changes up to > > > > > 75eb57e3ed3682f011a6694863044e8b143a9821: > > > > >=20 > > > > > =C2=A0=C2=A0 qapi: Pass '-u' when doing non-silent diff (2018-0= 3-16 > > > > > 09:00:07 -0500) > > > > >=20 > >=20 > > > > Hi. I get a bunch of test assertion failures with this: > > > >=20 >=20 > >=20 > > I haven't been able to reproduce the testsuite failures on my Linux b= ox, > > but if it's a race, then that doesn't make me all the more confident = on > > what it takes to reproduce and/or fix the race. >=20 > Okay, my simple builds on just x86_64-softmmu weren't hitting it, but m= y > 'build all binaries' tree seems to be hitting the same thing: >=20 > GTESTER check-qtest-ppcemb > qemu-system-ppcemb: chardev/char-io.c:91: io_watch_poll_finalize: Asser= tion > `iwp->src =3D=3D NULL' failed. > Broken pipe > GTester: last random seed: R02S74d45e64b38428eddd131a5c1b4c878c > make: *** [/home/eblake/qemu-tmp/tests/Makefile.include:878: > check-qtest-ppcemb] Error 1 >=20 > so I'm now testing if your squash makes a difference, now that it looks= like > I'm reproducing the problem. Exactly what I encountered. My old tests on v8 are not strong enough (less binaries, less concurrency). I reproduced that easily when I didn't specify --target-list (so all binaries), then run tests with more concurrency (-j8). However still I never reproduced the QIO problem even with that. I suspect that's a more hard-to-trigger race, and even it might not related to OOB (but I'm not sure). --=20 Peter Xu