From: Andrew Stubbs <ams@codesourcery.com>
To: Richard Biener <richard.guenther@gmail.com>
Cc: <gcc-patches@gcc.gnu.org>
Subject: Re: [committed 0/6] amdgcn: Add V32, V16, V8, V4, and V2 vectors
Date: Tue, 11 Oct 2022 12:53:30 +0100 [thread overview]
Message-ID: <00e7c525-cc0d-5144-d95c-573f75657e30@codesourcery.com> (raw)
In-Reply-To: <CAFiYyc06LbcjJZaBqwB3wwX5ExnuOwRwwYyNtAkGd5C94YQAwg@mail.gmail.com>
On 11/10/2022 12:29, Richard Biener wrote:
> On Tue, Oct 11, 2022 at 1:03 PM Andrew Stubbs <ams@codesourcery.com> wrote:
>>
>> This patch series adds additional vector sizes for the amdgcn backend.
>>
>> The hardware supports any arbitrary vector length up to 64-lanes via
>> masking, but GCC cannot (yet) make full use of them due to middle-end
>> limitations. Adding smaller "virtual" vector sizes increases the
>> complexity of the backend a little, but opens up optimization
>> opportunities for the current middle-end implementation somewhat. In
>> particular, it enables many more cases of SLP optimization.
>>
>> The patchset gives aproximately 100 addtional test PASS and a few extra
>> FAIL. However, the failures are not new issues, but rather existing
>> problems that did not show up because the code did not previously
>> vectorize. Expanding the testcase to allow 64-lane vectors shows the
>> same problems there.
>>
>> I shall backport these patches to the OG12 branch shortly.
>
> I suppose until you change the related_vector_mode hook the PR107096 issue
> will not hit you but at least it's then latent ...
How do you mean, change it?
static opt_machine_mode
gcn_related_vector_mode (machine_mode vector_mode,
scalar_mode element_mode, poly_uint64 nunits)
{
int n = nunits.to_constant ();
if (n == 0)
n = GET_MODE_NUNITS (vector_mode);
return VnMODE (n, element_mode);
}
It returns what it's asked for, always matching the number of lanes (not
the bitsize), which is most likely the most natural for GCN.
Andrew
next prev parent reply other threads:[~2022-10-11 11:53 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-11 11:02 Andrew Stubbs
2022-10-11 11:02 ` [committed 1/6] amdgcn: add multiple vector sizes Andrew Stubbs
2022-10-17 12:26 ` GCN: Restore build with GCC 4.8 (was: [committed 1/6] amdgcn: add multiple vector sizes) Thomas Schwinge
2022-10-11 11:02 ` [committed 2/6] amdgcn: Resolve insn conditions at compile time Andrew Stubbs
2022-10-11 11:02 ` [committed 3/6] amdgcn: Add vec_extract for partial vectors Andrew Stubbs
2022-10-11 11:02 ` [committed 4/6] amdgcn: vec_init for multiple vector sizes Andrew Stubbs
2022-10-11 11:02 ` [committed 5/6] amdgcn: Add vector integer negate insn Andrew Stubbs
2022-10-11 11:02 ` [committed 6/6] amdgcn: vector testsuite tweaks Andrew Stubbs
2022-10-28 7:46 ` Thomas Schwinge
2022-10-28 8:38 ` Stubbs, Andrew
2022-10-28 9:00 ` Thomas Schwinge
2022-10-11 11:29 ` [committed 0/6] amdgcn: Add V32, V16, V8, V4, and V2 vectors Richard Biener
2022-10-11 11:53 ` Andrew Stubbs [this message]
2022-10-11 11:58 ` Richard Biener
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=00e7c525-cc0d-5144-d95c-573f75657e30@codesourcery.com \
--to=ams@codesourcery.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=richard.guenther@gmail.com \
/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.
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 read-only IMAP folder(s) and NNTP newsgroup(s).