From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Jones Subject: Re: [Qemu-devel] [kvm-unit-tests PATCH v5 00/11] My current MTTCG tests Date: Sun, 2 Aug 2015 18:44:05 +0200 Message-ID: <20150802164405.GD3441@hawk.localdomain> References: <1438358041-18021-1-git-send-email-alex.bennee@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: mttcg@greensocs.com, mark.burton@greensocs.com, fred.konrad@greensocs.com, peter.maydell@linaro.org, kvm@vger.kernel.org, a.spyridakis@virtualopensystems.com, claudio.fontana@huawei.com, a.rigo@virtualopensystems.com, qemu-devel@nongnu.org To: Alex =?iso-8859-1?Q?Benn=E9e?= Return-path: Received: from mx1.redhat.com ([209.132.183.28]:59412 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751603AbbHBQoK (ORCPT ); Sun, 2 Aug 2015 12:44:10 -0400 Content-Disposition: inline In-Reply-To: <1438358041-18021-1-git-send-email-alex.bennee@linaro.org> Sender: kvm-owner@vger.kernel.org List-ID: On Fri, Jul 31, 2015 at 04:53:50PM +0100, Alex Benn=E9e wrote: > Hi, >=20 > This is the current state of my MTTCG tests based on the KVM's unit > testing framework. The earlier patches in the series have already bee= n > reviewed and will (with the exception of the emacs patch) be making > their way upstream. >=20 > There are a couple of addition to library functions: > - printf %u suppport > - flush_tlb_page for arm and arm64 > - a generic prng from CCAN >=20 > The two actual tests are: > - tlbflush-test > - barrier-test >=20 > The latter barrier test hangs the current -v6 MTTCG patch set in both > "excl" and "acqrel" modes and will make a good torture test for > Alvise's atomic patch set. I suspect the load/store ordering issues > will show up better once tested on a weak-ordered backend. I'm open t= o > suggestions for other tests worth adding to show up the issues. >=20 > The github tree can be found at: >=20 > https://github.com/stsquad/kvm-unit-tests/tree/current-mttcg-tests >=20 >=20 > Alex Benn=E9e (11): > arm/run: set indentation defaults for emacs > README: add some CONTRIBUTING notes > configure: emit HOST=3D$host to config.mak > arm/run: introduce usingkvm var and use it > lib/printf: support the %u unsigned fmt field > lib/arm: add flush_tlb_page mmu function > new arm/tlbflush-test: TLB torture test > arm/unittests.cfg: add the tlbflush tests > arm: query /dev/kvm for maximum vcpus > new: add isaac prng library from CCAN > new: arm/barrier-test for memory barriers General comment; please remove 'new' from your patch summaries. The lib/arm prefix is OK, but I've been using 'arm/arm64:' for all arm/arm64 patches, whether they're lib or tests. Thanks, drew >=20 > README | 26 ++++++ > arm/barrier-test.c | 206 +++++++++++++++++++++++++++++++++= ++++++++++ > arm/run | 19 +++- > arm/tlbflush-test.c | 194 +++++++++++++++++++++++++++++++++= +++++++ > arm/unittests.cfg | 26 +++++- > arm/utils/kvm-query.c | 41 +++++++++ > config/config-arm-common.mak | 18 +++- > configure | 2 + > lib/arm/asm/mmu.h | 11 +++ > lib/arm64/asm/mmu.h | 8 ++ > lib/printf.c | 13 +++ > lib/prng.c | 162 +++++++++++++++++++++++++++++++++= + > lib/prng.h | 82 +++++++++++++++++ > 13 files changed, 801 insertions(+), 7 deletions(-) > create mode 100644 arm/barrier-test.c > create mode 100644 arm/tlbflush-test.c > create mode 100644 arm/utils/kvm-query.c > create mode 100644 lib/prng.c > create mode 100644 lib/prng.h >=20 > --=20 > 2.5.0 >=20 >=20