Re: Why does glibc use AVX-512?

> On Mar 26, 2021, at 2:11 PM, Florian Weimer wrote: > > * Andy Lutomirski: > >>> On Fri, Mar 26, 2021 at 1:35 PM Florian Weimer wrote: >>> >>> I mean the immense slowdown you get if you use %xmm registers after > their %ymm counterparts (doesn't have to be %zmm, that issue is > present

Re: Why does glibc use AVX-512?

* Andy Lutomirski: > On Fri, Mar 26, 2021 at 1:35 PM Florian Weimer wrote: >> >> * Andy Lutomirski: >> >> > On Fri, Mar 26, 2021 at 12:34 PM Florian Weimer wrote: >> >> x86: Sporadic failures in tst-cpu-features-cpuinfo >> >> >> >

Re: Why does glibc use AVX-512?

On Fri, Mar 26, 2021 at 1:35 PM Florian Weimer wrote: > > * Andy Lutomirski: > > > On Fri, Mar 26, 2021 at 12:34 PM Florian Weimer wrote: > >> x86: Sporadic failures in tst-cpu-features-cpuinfo > >> > > > > It's worth noting that

Re: Why does glibc use AVX-512?

On Fri, Mar 26, 2021 at 1:35 PM Florian Weimer wrote: > > All this needs to work transparently without user intervention. We > cannot require firmware upgrades to fix the incorrect RTM reporting > issue (the bug I referenced). I think we can require software updates > which tell glibc when to

Re: Why does glibc use AVX-512?

* Andy Lutomirski: > On Fri, Mar 26, 2021 at 12:34 PM Florian Weimer wrote: >> x86: Sporadic failures in tst-cpu-features-cpuinfo >> > > It's worth noting that recent microcode updates have make RTM > considerably less likely to

Re: Why does glibc use AVX-512?

On 26/03/2021 19:47, Andy Lutomirski wrote: > On Fri, Mar 26, 2021 at 12:34 PM Florian Weimer wrote: >> * Andy Lutomirski: >> > AVX-512 cleared, and programs need to explicitly request enablement. > This would allow programs to opt into not saving/restoring across > signals or to

Re: Why does glibc use AVX-512?

On Fri, Mar 26, 2021 at 12:34 PM Florian Weimer wrote: > > * Andy Lutomirski: > > >> > AVX-512 cleared, and programs need to explicitly request enablement. > >> > This would allow programs to opt into not saving/restoring across > >> > signals or to save/restore in buffers supplied when the

Re: Why does glibc use AVX-512?

* Andy Lutomirski: >> > AVX-512 cleared, and programs need to explicitly request enablement. >> > This would allow programs to opt into not saving/restoring across >> > signals or to save/restore in buffers supplied when the feature is >> > enabled. >> >> Isn't XSAVEOPT already able to handle

Re: Why does glibc use AVX-512?

On Fri, Mar 26, 2021 at 3:08 AM Borislav Petkov wrote: > > On Thu, Mar 25, 2021 at 09:38:24PM -0700, Andy Lutomirski wrote: > > I think we should seriously consider solutions in which, for new > > tasks, XCR0 has new giant features (e.g. AMX) and possibly even > > AVX-512 cleared, and programs

Re: Why does glibc use AVX-512?

On Fri, Mar 26, 2021 at 5:12 AM Florian Weimer wrote: > > * Andy Lutomirski-alpha: > > > glibc appears to use AVX512F for memcpy by default. (Unless > > Prefer_ERMS is default-on, but I genuinely can't tell if this is the > > case. I did some searching.) The commit adding it refers to a 2016 >

RE: Why does glibc use AVX-512?

From: Andy Lutomirski > Sent: 26 March 2021 04:38 > > Hi all- > > glibc appears to use AVX512F for memcpy by default. (Unless > Prefer_ERMS is default-on, but I genuinely can't tell if this is the > case. I did some searching.) The commit adding it refers to a 2016 > email saying that it's

Re: Why does glibc use AVX-512?

* Andy Lutomirski-alpha: > glibc appears to use AVX512F for memcpy by default. (Unless > Prefer_ERMS is default-on, but I genuinely can't tell if this is the > case. I did some searching.) The commit adding it refers to a 2016 > email saying that it's 30% on KNL. As far as I know, glibc only

Re: Why does glibc use AVX-512?

On Thu, Mar 25, 2021 at 09:38:24PM -0700, Andy Lutomirski wrote: > I think we should seriously consider solutions in which, for new > tasks, XCR0 has new giant features (e.g. AMX) and possibly even > AVX-512 cleared, and programs need to explicitly request enablement. I totally agree with making

Why does glibc use AVX-512?

Hi all- glibc appears to use AVX512F for memcpy by default. (Unless Prefer_ERMS is default-on, but I genuinely can't tell if this is the case. I did some searching.) The commit adding it refers to a 2016 email saying that it's 30% on KNL. Unfortunately, AVX-512 is now available in normal