Agabaria, Mohammed via llvm-dev
2017-Jun-12 10:41 UTC
[llvm-dev] Enable vectorizer-maximize-bandwidth by default?
Guys, Just to clarify that with the current fix in SLM there is no need to wait for other issues to be fixed (minor issue). So you can move on with your patch. From: Agabaria, Mohammed Sent: Wednesday, June 07, 2017 15:24 To: Zaks, Ayal <ayal.zaks at intel.com>; Chandler Carruth <chandlerc at gmail.com>; Flamedoge <code.kchoi at gmail.com>; Dehao Chen <dehao at google.com> Cc: llvm-dev <llvm-dev at lists.llvm.org> Subject: RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? Guys I have uploaded a fix on phabricator which fix the issue in SLM: https://reviews.llvm.org/D33983 From: Zaks, Ayal Sent: Tuesday, June 06, 2017 22:32 To: Chandler Carruth <chandlerc at gmail.com<mailto:chandlerc at gmail.com>>; Agabaria, Mohammed <mohammed.agabaria at intel.com<mailto:mohammed.agabaria at intel.com>>; Flamedoge <code.kchoi at gmail.com<mailto:code.kchoi at gmail.com>>; Dehao Chen <dehao at google.com<mailto:dehao at google.com>> Cc: llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> Subject: RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? Chandler> Have you considered contributing these benchmarks to the LLVM test suite? These benchmarks are not ours to contribute; they’re from the EEMBC standard, whose availability is similar in nature to that of SPEC. From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of Chandler Carruth via llvm-dev Sent: Tuesday, May 30, 2017 23:07 To: Agabaria, Mohammed <mohammed.agabaria at intel.com<mailto:mohammed.agabaria at intel.com>>; Flamedoge <code.kchoi at gmail.com<mailto:code.kchoi at gmail.com>>; Dehao Chen <dehao at google.com<mailto:dehao at google.com>> Cc: llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> Subject: Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? On Tue, May 30, 2017 at 1:40 AM Agabaria, Mohammed via llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> wrote: We’re seeing nice improvements but also significant degradations on IA, which we would like to investigate before the patch is committed. Major degradations we see: networking ip_pktcheckb1m -6.80 % ip_pktcheckb2m -6.74 % ip_pktcheckb4m -7.57 % ip_pktcheckb512k -6.58 % Telecom autcor00data_1 -78.02 % autcor00data_2 -76.80 % autcor00data_3 -77.00 % (on Atom) We still working on creating reproducers. In general we support this patch, just want to have a chance to investigate the issues. We need a few days for that. I mean, OK... but keep in mind that Dehao's original email went out over a week ago, so this patch has already been held up a while. As these benchmarks aren't readily available, we also can't do anything to help until a test case is posted. Have you considered contributing these benchmarks to the LLVM test suite? From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of Chandler Carruth via llvm-dev Sent: Tuesday, May 30, 2017 11:27 To: Flamedoge <code.kchoi at gmail.com<mailto:code.kchoi at gmail.com>>; Dehao Chen <dehao at google.com<mailto:dehao at google.com>> Cc: llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> Subject: Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? If you care about such hardware, please run benchmarks with the flag? Dehao has made this flag available. It is important that those who care about particular hardware provid ebenchmark results. Not everyone in the community will have access to particular hardware variants. On Tue, May 30, 2017 at 1:14 AM Flamedoge <code.kchoi at gmail.com<mailto:code.kchoi at gmail.com>> wrote: Thank you for running these. May I suggest testing on AVX2 capable hardware? That would be Intel Haswell, AMD Carrizo and up. I'm not sure what "vectorizer-maximize-bandwidth" implies, but doubling the vector lanes may help light up parallel regions. Kevin On Tue, May 30, 2017 at 12:58 AM, Chandler Carruth via llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> wrote: On Fri, May 19, 2017 at 4:01 PM Adam Nemet via llvm-dev <llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org>> wrote: I will run it on Cyclone/AArch64 next week. FYI, we're still waiting on these Adam... _______________________________________________ LLVM Developers mailing list llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. _______________________________________________ LLVM Developers mailing list llvm-dev at lists.llvm.org<mailto:llvm-dev at lists.llvm.org> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170612/7f1f8e07/attachment-0001.html>
Dehao Chen via llvm-dev
2017-Jun-13 00:43 UTC
[llvm-dev] Enable vectorizer-maximize-bandwidth by default?
Thanks for the update. I think overall consensus is supportive for this change. If no one objects by then end of tomorrow (Tuesday), I will submit the patch to make vectorizer-maximize-bandwidth on by default. Thanks, Dehao On Mon, Jun 12, 2017 at 3:41 AM, Agabaria, Mohammed < mohammed.agabaria at intel.com> wrote:> Guys, Just to clarify that with the current fix in SLM there is no need to > wait for other issues to be fixed (minor issue). > > So you can move on with your patch. > > > > *From:* Agabaria, Mohammed > *Sent:* Wednesday, June 07, 2017 15:24 > *To:* Zaks, Ayal <ayal.zaks at intel.com>; Chandler Carruth < > chandlerc at gmail.com>; Flamedoge <code.kchoi at gmail.com>; Dehao Chen < > dehao at google.com> > *Cc:* llvm-dev <llvm-dev at lists.llvm.org> > *Subject:* RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? > > > > Guys I have uploaded a fix on phabricator which fix the issue in SLM: > https://reviews.llvm.org/D33983 > > > > *From:* Zaks, Ayal > *Sent:* Tuesday, June 06, 2017 22:32 > *To:* Chandler Carruth <chandlerc at gmail.com>; Agabaria, Mohammed < > mohammed.agabaria at intel.com>; Flamedoge <code.kchoi at gmail.com>; Dehao > Chen <dehao at google.com> > *Cc:* llvm-dev <llvm-dev at lists.llvm.org> > *Subject:* RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? > > > > Chandler> Have you considered contributing these benchmarks to the LLVM > test suite? > > > > These benchmarks are not ours to contribute; they’re from the EEMBC > standard, whose availability is similar in nature to that of SPEC. > > > > > > *From:* llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org > <llvm-dev-bounces at lists.llvm.org>] *On Behalf Of *Chandler Carruth via > llvm-dev > *Sent:* Tuesday, May 30, 2017 23:07 > *To:* Agabaria, Mohammed <mohammed.agabaria at intel.com>; Flamedoge < > code.kchoi at gmail.com>; Dehao Chen <dehao at google.com> > *Cc:* llvm-dev <llvm-dev at lists.llvm.org> > *Subject:* Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? > > > > On Tue, May 30, 2017 at 1:40 AM Agabaria, Mohammed via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > We’re seeing nice improvements but also significant degradations on IA, > which we would like to investigate before the patch is committed. > > > > Major degradations we see: > > > > networking > > ip_pktcheckb1m -6.80 % > > ip_pktcheckb2m -6.74 % > > ip_pktcheckb4m -7.57 % > > ip_pktcheckb512k -6.58 % > > Telecom > > autcor00data_1 -78.02 % > > autcor00data_2 -76.80 % > > autcor00data_3 -77.00 % > > (on Atom) > > > > We still working on creating reproducers. > > In general we support this patch, just want to have a chance to > investigate the issues. We need a few days for that. > > > > I mean, OK... but keep in mind that Dehao's original email went out over a > week ago, so this patch has already been held up a while. As these > benchmarks aren't readily available, we also can't do anything to help > until a test case is posted. > > > > Have you considered contributing these benchmarks to the LLVM test suite? > > > > > > > > > > *From:* llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org > <llvm-dev-bounces at lists.llvm.org>] *On Behalf Of *Chandler Carruth via > llvm-dev > *Sent:* Tuesday, May 30, 2017 11:27 > *To:* Flamedoge <code.kchoi at gmail.com>; Dehao Chen <dehao at google.com> > *Cc:* llvm-dev <llvm-dev at lists.llvm.org> > *Subject:* Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by default? > > > > If you care about such hardware, please run benchmarks with the flag? > > > > Dehao has made this flag available. It is important that those who care > about particular hardware provid ebenchmark results. Not everyone in the > community will have access to particular hardware variants. > > > > On Tue, May 30, 2017 at 1:14 AM Flamedoge <code.kchoi at gmail.com> wrote: > > Thank you for running these. > > > > May I suggest testing on AVX2 capable hardware? That would be Intel > Haswell, AMD Carrizo and up. > > I'm not sure what "vectorizer-maximize-bandwidth" implies, but doubling > the vector lanes may help light up parallel regions. > > > > Kevin > > > > On Tue, May 30, 2017 at 12:58 AM, Chandler Carruth via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > On Fri, May 19, 2017 at 4:01 PM Adam Nemet via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > I will run it on Cyclone/AArch64 next week. > > > > FYI, we're still waiting on these Adam... > > > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev > > --------------------------------------------------------------------- > Intel Israel (74) Limited > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev > > --------------------------------------------------------------------- > Intel Israel (74) Limited > > This e-mail and any attachments may contain confidential material for > the sole use of the intended recipient(s). Any review or distribution > by others is strictly prohibited. If you are not the intended > recipient, please contact the sender and delete all copies. >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170612/0e63af4b/attachment.html>
Dehao Chen via llvm-dev
2017-Jun-21 22:03 UTC
[llvm-dev] Enable vectorizer-maximize-bandwidth by default?
The patch is committed in r305960. Please let me know if there is unexpected perf impact after this change. Cheers, Dehao On Mon, Jun 12, 2017 at 5:43 PM, Dehao Chen <dehao at google.com> wrote:> Thanks for the update. I think overall consensus is supportive for this > change. If no one objects by then end of tomorrow (Tuesday), I will submit > the patch to make vectorizer-maximize-bandwidth on by default. > > Thanks, > Dehao > > > On Mon, Jun 12, 2017 at 3:41 AM, Agabaria, Mohammed < > mohammed.agabaria at intel.com> wrote: > >> Guys, Just to clarify that with the current fix in SLM there is no need >> to wait for other issues to be fixed (minor issue). >> >> So you can move on with your patch. >> >> >> >> *From:* Agabaria, Mohammed >> *Sent:* Wednesday, June 07, 2017 15:24 >> *To:* Zaks, Ayal <ayal.zaks at intel.com>; Chandler Carruth < >> chandlerc at gmail.com>; Flamedoge <code.kchoi at gmail.com>; Dehao Chen < >> dehao at google.com> >> *Cc:* llvm-dev <llvm-dev at lists.llvm.org> >> *Subject:* RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by >> default? >> >> >> >> Guys I have uploaded a fix on phabricator which fix the issue in SLM: >> https://reviews.llvm.org/D33983 >> >> >> >> *From:* Zaks, Ayal >> *Sent:* Tuesday, June 06, 2017 22:32 >> *To:* Chandler Carruth <chandlerc at gmail.com>; Agabaria, Mohammed < >> mohammed.agabaria at intel.com>; Flamedoge <code.kchoi at gmail.com>; Dehao >> Chen <dehao at google.com> >> *Cc:* llvm-dev <llvm-dev at lists.llvm.org> >> *Subject:* RE: [llvm-dev] Enable vectorizer-maximize-bandwidth by >> default? >> >> >> >> Chandler> Have you considered contributing these benchmarks to the LLVM >> test suite? >> >> >> >> These benchmarks are not ours to contribute; they’re from the EEMBC >> standard, whose availability is similar in nature to that of SPEC. >> >> >> >> >> >> *From:* llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org >> <llvm-dev-bounces at lists.llvm.org>] *On Behalf Of *Chandler Carruth via >> llvm-dev >> *Sent:* Tuesday, May 30, 2017 23:07 >> *To:* Agabaria, Mohammed <mohammed.agabaria at intel.com>; Flamedoge < >> code.kchoi at gmail.com>; Dehao Chen <dehao at google.com> >> *Cc:* llvm-dev <llvm-dev at lists.llvm.org> >> *Subject:* Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by >> default? >> >> >> >> On Tue, May 30, 2017 at 1:40 AM Agabaria, Mohammed via llvm-dev < >> llvm-dev at lists.llvm.org> wrote: >> >> We’re seeing nice improvements but also significant degradations on IA, >> which we would like to investigate before the patch is committed. >> >> >> >> Major degradations we see: >> >> >> >> networking >> >> ip_pktcheckb1m -6.80 % >> >> ip_pktcheckb2m -6.74 % >> >> ip_pktcheckb4m -7.57 % >> >> ip_pktcheckb512k -6.58 % >> >> Telecom >> >> autcor00data_1 -78.02 % >> >> autcor00data_2 -76.80 % >> >> autcor00data_3 -77.00 % >> >> (on Atom) >> >> >> >> We still working on creating reproducers. >> >> In general we support this patch, just want to have a chance to >> investigate the issues. We need a few days for that. >> >> >> >> I mean, OK... but keep in mind that Dehao's original email went out over >> a week ago, so this patch has already been held up a while. As these >> benchmarks aren't readily available, we also can't do anything to help >> until a test case is posted. >> >> >> >> Have you considered contributing these benchmarks to the LLVM test suite? >> >> >> >> >> >> >> >> >> >> *From:* llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org >> <llvm-dev-bounces at lists.llvm.org>] *On Behalf Of *Chandler Carruth via >> llvm-dev >> *Sent:* Tuesday, May 30, 2017 11:27 >> *To:* Flamedoge <code.kchoi at gmail.com>; Dehao Chen <dehao at google.com> >> *Cc:* llvm-dev <llvm-dev at lists.llvm.org> >> *Subject:* Re: [llvm-dev] Enable vectorizer-maximize-bandwidth by >> default? >> >> >> >> If you care about such hardware, please run benchmarks with the flag? >> >> >> >> Dehao has made this flag available. It is important that those who care >> about particular hardware provid ebenchmark results. Not everyone in the >> community will have access to particular hardware variants. >> >> >> >> On Tue, May 30, 2017 at 1:14 AM Flamedoge <code.kchoi at gmail.com> wrote: >> >> Thank you for running these. >> >> >> >> May I suggest testing on AVX2 capable hardware? That would be Intel >> Haswell, AMD Carrizo and up. >> >> I'm not sure what "vectorizer-maximize-bandwidth" implies, but doubling >> the vector lanes may help light up parallel regions. >> >> >> >> Kevin >> >> >> >> On Tue, May 30, 2017 at 12:58 AM, Chandler Carruth via llvm-dev < >> llvm-dev at lists.llvm.org> wrote: >> >> On Fri, May 19, 2017 at 4:01 PM Adam Nemet via llvm-dev < >> llvm-dev at lists.llvm.org> wrote: >> >> I will run it on Cyclone/AArch64 next week. >> >> >> >> FYI, we're still waiting on these Adam... >> >> >> >> _______________________________________________ >> LLVM Developers mailing list >> llvm-dev at lists.llvm.org >> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev >> >> --------------------------------------------------------------------- >> Intel Israel (74) Limited >> >> This e-mail and any attachments may contain confidential material for >> the sole use of the intended recipient(s). Any review or distribution >> by others is strictly prohibited. If you are not the intended >> recipient, please contact the sender and delete all copies. >> >> _______________________________________________ >> LLVM Developers mailing list >> llvm-dev at lists.llvm.org >> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev >> >> --------------------------------------------------------------------- >> Intel Israel (74) Limited >> >> This e-mail and any attachments may contain confidential material for >> the sole use of the intended recipient(s). Any review or distribution >> by others is strictly prohibited. If you are not the intended >> recipient, please contact the sender and delete all copies. >> > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170621/183bc514/attachment.html>