From: <sidneym@codeaurora.org>
To: "'Szabolcs Nagy'" <nsz@port70.net>
Cc: "'Rich Felker'" <dalias@libc.org>, <musl@lists.openwall.com>
Subject: RE: [musl] Hexagon DSP support
Date: Fri, 24 Jul 2020 12:49:42 -0500 [thread overview]
Message-ID: <21d501d661e2$d059fd70$710df850$@codeaurora.org> (raw)
In-Reply-To: <20200723215603.GA755469@port70.net>
> -----Original Message-----
> From: Szabolcs Nagy <nsz@port70.net>
> Sent: Thursday, July 23, 2020 4:56 PM
> To: sidneym@codeaurora.org
> Cc: 'Rich Felker' <dalias@libc.org>; musl@lists.openwall.com
> Subject: Re: [musl] Hexagon DSP support
>
> * sidneym@codeaurora.org <sidneym@codeaurora.org> [2020-07-20
> 16:26:58 -0500]:
> > I removed fma/fmal/fmax/fmin/fabs from compiler-rt-builtins,
> > https://reviews.llvm.org/D82263
> > The comparison with musl can be found here:
> > https://github.com/quic/musl/compare/hexagon but I've also attached
> > the patch.
> >
> > An assert in clang when building both musl and libc-test for hexagon
> > was fixed by, https://reviews.llvm.org/D80952 prior to this change
> > -frounding-math had to be used.
> >
> > The test-results are also attached. Everything is built with the
> > tip-of-tree llvm so sometimes results vary but these are the results I
> > got from this morning's clone. The only notable difference in the
> > results would be that both fma and fmal fail and this is because of
> > the compiler-rt change. I didn't add fma to musl because it require
> > more complex assembly, along the lines you saw in an earlier version
with
> sqrt.
>
>
> the fma and sqrt failures are still not fully explained, e.g. this looks
wrong:
>
> src/math/special/fma.h:42: RN fma(0x1p+0,0x1p+0,-0x1p-1074) want 0x1p+0
> got -0x1.fffffp-43 ulperr -4503599627370496.000 = -0x1p+52 + 0x0p+0
>
> the only target specific bit in fma is a_clz_64 so i would check that.
>
> e.g. a_clz_64(1ULL << 42) should give 21 (this computation happens during
> the fma test case above).
Hexagon didn't have a_clz_64 implemented however I added this morning it and
noticed no differences. I will update the patch with that routine included.
I did notice a compiler regression in how it compiled fma and have asked a
compiler person to take a look. An older version of our internally
maintained compiler does produce the expected results for the values I used
but later versions do not. Unfortunately changing optimization levels will
produce different results as well.
Thanks,
next prev parent reply other threads:[~2020-07-24 17:50 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-15 13:19 sidneym
2020-04-15 16:30 ` Rich Felker
2020-04-15 17:50 ` sidneym
2020-04-15 18:06 ` Szabolcs Nagy
2020-04-15 18:22 ` sidneym
2020-04-16 9:36 ` Szabolcs Nagy
2020-04-16 15:34 ` Rich Felker
2020-04-16 16:26 ` sidneym
2020-04-16 16:34 ` 'Rich Felker'
2020-04-15 18:26 ` Rich Felker
2020-04-15 19:12 ` sidneym
2020-04-15 19:29 ` 'Rich Felker'
2020-04-30 22:44 ` sidneym
2020-04-30 23:51 ` Rich Felker
2020-05-05 23:37 ` sidneym
2020-05-06 0:59 ` Rich Felker
2020-06-18 16:37 ` sidneym
2020-06-18 21:42 ` Szabolcs Nagy
2020-06-19 21:58 ` sidneym
2020-06-19 22:46 ` Rich Felker
2020-06-20 0:03 ` [musl] strtok Robert Skopalík
2020-06-20 0:15 ` Rich Felker
2020-06-20 0:36 ` Robert Skopalík
2020-06-20 0:46 ` Robert Skopalík
2020-06-20 1:44 ` Rich Felker
2020-06-20 7:07 ` Patrick Oppenlander
2020-06-20 13:00 ` Robert Skopalík
2020-06-22 0:57 ` Bery Saidi
2020-06-20 2:29 ` [musl] Hexagon DSP support sidneym
2020-06-20 3:20 ` Rich Felker
2020-07-20 21:26 ` sidneym
2020-07-23 21:56 ` Szabolcs Nagy
2020-07-24 17:49 ` sidneym [this message]
2020-09-16 20:49 ` sidneym
2020-09-17 1:32 ` 'Rich Felker'
2020-09-17 22:31 ` sidneym
2020-09-18 1:08 ` Rich Felker
2020-09-18 8:10 ` Szabolcs Nagy
2020-09-20 13:12 ` sidneym
2020-09-20 17:17 ` 'Rich Felker'
2020-09-21 14:09 ` sidneym
2020-04-15 18:55 ` Fangrui Song
2021-03-09 20:25 sidneym
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='21d501d661e2$d059fd70$710df850$@codeaurora.org' \
--to=sidneym@codeaurora.org \
--cc=dalias@libc.org \
--cc=musl@lists.openwall.com \
--cc=nsz@port70.net \
/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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.vuxu.org/mirror/musl/
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).