From: alvise rigo <a.rigo@virtualopensystems.com> To: "Alex Bennée" <alex.bennee@linaro.org> Cc: mttcg@listserver.greensocs.com, "Peter Maydell" <peter.maydell@linaro.org>, "Andrew Jones" <drjones@redhat.com>, "Claudio Fontana" <claudio.fontana@huawei.com>, kvm@vger.kernel.org, "Alexander Spyridakis" <a.spyridakis@virtualopensystems.com>, "Mark Burton" <mark.burton@greensocs.com>, "QEMU Developers" <qemu-devel@nongnu.org>, "KONRAD Frédéric" <fred.konrad@greensocs.com> Subject: Re: [kvm-unit-tests PATCH v5 11/11] new: arm/barrier-test for memory barriers Date: Mon, 3 Aug 2015 18:46:30 +0200 [thread overview] Message-ID: <CAH47eN1i56z9J4pDH0LoCt+Mu2v=yUtr+AQvX1szJbFcZGyDmw@mail.gmail.com> (raw) In-Reply-To: <87mvy8l5kt.fsf@linaro.org> [-- Attachment #1: Type: text/plain, Size: 1615 bytes --] On Mon, Aug 3, 2015 at 6:06 PM, Alex Bennée <alex.bennee@linaro.org> wrote: > > alvise rigo <a.rigo@virtualopensystems.com> writes: > > > On Mon, Aug 3, 2015 at 12:30 PM, Alex Bennée <alex.bennee@linaro.org> > wrote: > >> > >> alvise rigo <a.rigo@virtualopensystems.com> writes: > >> > >>> Hi Alex, > >>> > >>> Nice set of tests, they are proving to be helpful. > >>> One question below. > >>> > <snip> > >>> > >>> Why are we calling these last two instructions with the 'eq' suffix? > >>> Shouldn't we just strex r1, r0, [sptr] and then cmp r1, #0? > >> > >> Possibly, my armv7 is a little rusty. I'm just looking at tweaking this > >> test now so I'll try and clean that up. > > Please find the updated test attached. I've also included some new test > modes. In theory the barrier test by itself should still fail but it > Thanks, I will check them out. > passes on real ARMv7 as well as TCG. I'm trying to run the test on a > heavier core-ed ARMv7 to check. I suspect we get away with it on > ARMv7-on-x86_64 due to the strong ordering of the x86. > The "excl" and "acqrel" tests now run without issue (although again > plain acqrel semantics shouldn't stop a race corrupting shared_value). I suppose that, in order to have some race conditions due to a lack of a proper emulation of barriers and acqrel instructions, we need a test that does not involve atomic instructions at all, to reduce the emulation overhead as much as possible. Does this sound reasonable? > > I'll tweak the v8 versions of the test tomorrow. > > -- > Alex Bennée > > [-- Attachment #2: Type: text/html, Size: 2842 bytes --]
WARNING: multiple messages have this Message-ID (diff)
From: alvise rigo <a.rigo@virtualopensystems.com> To: "Alex Bennée" <alex.bennee@linaro.org> Cc: mttcg@listserver.greensocs.com, "Peter Maydell" <peter.maydell@linaro.org>, "Andrew Jones" <drjones@redhat.com>, "Claudio Fontana" <claudio.fontana@huawei.com>, kvm@vger.kernel.org, "Alexander Spyridakis" <a.spyridakis@virtualopensystems.com>, "Mark Burton" <mark.burton@greensocs.com>, "QEMU Developers" <qemu-devel@nongnu.org>, "KONRAD Frédéric" <fred.konrad@greensocs.com> Subject: Re: [Qemu-devel] [kvm-unit-tests PATCH v5 11/11] new: arm/barrier-test for memory barriers Date: Mon, 3 Aug 2015 18:46:30 +0200 [thread overview] Message-ID: <CAH47eN1i56z9J4pDH0LoCt+Mu2v=yUtr+AQvX1szJbFcZGyDmw@mail.gmail.com> (raw) In-Reply-To: <87mvy8l5kt.fsf@linaro.org> [-- Attachment #1: Type: text/plain, Size: 1615 bytes --] On Mon, Aug 3, 2015 at 6:06 PM, Alex Bennée <alex.bennee@linaro.org> wrote: > > alvise rigo <a.rigo@virtualopensystems.com> writes: > > > On Mon, Aug 3, 2015 at 12:30 PM, Alex Bennée <alex.bennee@linaro.org> > wrote: > >> > >> alvise rigo <a.rigo@virtualopensystems.com> writes: > >> > >>> Hi Alex, > >>> > >>> Nice set of tests, they are proving to be helpful. > >>> One question below. > >>> > <snip> > >>> > >>> Why are we calling these last two instructions with the 'eq' suffix? > >>> Shouldn't we just strex r1, r0, [sptr] and then cmp r1, #0? > >> > >> Possibly, my armv7 is a little rusty. I'm just looking at tweaking this > >> test now so I'll try and clean that up. > > Please find the updated test attached. I've also included some new test > modes. In theory the barrier test by itself should still fail but it > Thanks, I will check them out. > passes on real ARMv7 as well as TCG. I'm trying to run the test on a > heavier core-ed ARMv7 to check. I suspect we get away with it on > ARMv7-on-x86_64 due to the strong ordering of the x86. > The "excl" and "acqrel" tests now run without issue (although again > plain acqrel semantics shouldn't stop a race corrupting shared_value). I suppose that, in order to have some race conditions due to a lack of a proper emulation of barriers and acqrel instructions, we need a test that does not involve atomic instructions at all, to reduce the emulation overhead as much as possible. Does this sound reasonable? > > I'll tweak the v8 versions of the test tomorrow. > > -- > Alex Bennée > > [-- Attachment #2: Type: text/html, Size: 2842 bytes --]
next prev parent reply other threads:[~2015-08-03 16:46 UTC|newest] Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-07-31 15:53 [kvm-unit-tests PATCH v5 00/11] My current MTTCG tests Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 01/11] arm/run: set indentation defaults for emacs Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 02/11] README: add some CONTRIBUTING notes Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 03/11] configure: emit HOST=$host to config.mak Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 04/11] arm/run: introduce usingkvm var and use it Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-08-02 16:36 ` Andrew Jones 2015-08-02 16:36 ` Andrew Jones 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 05/11] lib/printf: support the %u unsigned fmt field Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 18:25 ` Andrew Jones 2015-07-31 18:25 ` Andrew Jones 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 06/11] lib/arm: add flush_tlb_page mmu function Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 18:35 ` Andrew Jones 2015-07-31 18:35 ` Andrew Jones 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 07/11] new arm/tlbflush-test: TLB torture test Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 18:51 ` Andrew Jones 2015-07-31 18:51 ` Andrew Jones 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 08/11] arm/unittests.cfg: add the tlbflush tests Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 18:53 ` Andrew Jones 2015-07-31 18:53 ` Andrew Jones 2015-07-31 15:53 ` [kvm-unit-tests PATCH v5 09/11] arm: query /dev/kvm for maximum vcpus Alex Bennée 2015-07-31 15:53 ` [Qemu-devel] " Alex Bennée 2015-07-31 19:17 ` Andrew Jones 2015-07-31 19:17 ` Andrew Jones 2015-08-02 16:40 ` Andrew Jones 2015-08-02 16:40 ` Andrew Jones 2015-07-31 15:54 ` [kvm-unit-tests PATCH v5 10/11] new: add isaac prng library from CCAN Alex Bennée 2015-07-31 15:54 ` [Qemu-devel] " Alex Bennée 2015-07-31 19:22 ` Andrew Jones 2015-07-31 19:22 ` Andrew Jones 2015-07-31 15:54 ` [kvm-unit-tests PATCH v5 11/11] new: arm/barrier-test for memory barriers Alex Bennée 2015-07-31 15:54 ` [Qemu-devel] " Alex Bennée 2015-07-31 19:30 ` Andrew Jones 2015-07-31 19:30 ` Andrew Jones 2015-08-03 10:02 ` alvise rigo 2015-08-03 10:02 ` [Qemu-devel] " alvise rigo 2015-08-03 10:30 ` Alex Bennée 2015-08-03 10:30 ` [Qemu-devel] " Alex Bennée 2015-08-03 10:34 ` alvise rigo 2015-08-03 10:34 ` [Qemu-devel] " alvise rigo 2015-08-03 16:06 ` Alex Bennée 2015-08-03 16:06 ` [Qemu-devel] " Alex Bennée 2015-08-03 16:46 ` alvise rigo [this message] 2015-08-03 16:46 ` alvise rigo 2015-08-04 7:30 ` Alex Bennée 2015-08-04 7:30 ` [Qemu-devel] " Alex Bennée 2015-08-02 16:44 ` [Qemu-devel] [kvm-unit-tests PATCH v5 00/11] My current MTTCG tests Andrew Jones 2015-08-02 16:44 ` Andrew Jones
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to='CAH47eN1i56z9J4pDH0LoCt+Mu2v=yUtr+AQvX1szJbFcZGyDmw@mail.gmail.com' \ --to=a.rigo@virtualopensystems.com \ --cc=a.spyridakis@virtualopensystems.com \ --cc=alex.bennee@linaro.org \ --cc=claudio.fontana@huawei.com \ --cc=drjones@redhat.com \ --cc=fred.konrad@greensocs.com \ --cc=kvm@vger.kernel.org \ --cc=mark.burton@greensocs.com \ --cc=mttcg@listserver.greensocs.com \ --cc=peter.maydell@linaro.org \ --cc=qemu-devel@nongnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.