public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: "Andre Vieira (lists)" <andre.simoesdiasvieira@arm.com>
To: Richard Biener <rguenther@suse.de>
Cc: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>,
	"Andre Vieira (lists) via Gcc-patches" <gcc-patches@gcc.gnu.org>,
	Richard Sandiford <richard.sandiford@arm.com>
Subject: Re: ifcvt: Fix bitpos calculation in bitfield lowering [PR107229]
Date: Thu, 13 Oct 2022 15:42:55 +0100	[thread overview]
Message-ID: <bcee7445-9bc0-a0cb-985d-99ef764fc569@arm.com> (raw)
In-Reply-To: <nycvar.YFH.7.77.849.2210131415020.18337@jbgna.fhfr.qr>


On 13/10/2022 15:15, Richard Biener wrote:
> On Thu, 13 Oct 2022, Andre Vieira (lists) wrote:
>
>> Hi Rainer,
>>
>> Thanks for reporting, I was actually expecting these! I thought about
>> pre-empting them by using a positive filter on the tests for aarch64 and
>> x86_64 as I knew those would pass, but I thought it would be better to let
>> other targets report failures since then you get a testsuite that covers more
>> targets than just what I'm able to check.
>>
>> Are there any sparc architectures that would support these or should I just
>> xfail sparc*-*-* ?
>>
>> For instance: I also saw PR107240 for which one of the write tests fails on
>> Power 7 BE. I'm suggesting adding an xfail for that one
> for the failure below we seem to require vectorizing shifts for which I
> think we have a vect_* target to check?
'vect_shift' no sparc on the list of supported targets, so that should 
do it, I'll add it when I add my fix for powerpc too.
>
>> Kind regards,
>> Andre
>>
>> On 13/10/2022 12:39, Rainer Orth wrote:
>>> Hi Andre,
>>>
>>>> The bitposition calculation for the bitfield lowering in loop if conversion
>>>> was not
>>>> taking DECL_FIELD_OFFSET into account, which meant that it would result in
>>>> wrong bitpositions for bitfields that did not end up having representations
>>>> starting at the beginning of the struct.
>>>>
>>>> Bootstrappend and regression tested on aarch64-none-linux-gnu and
>>>> x86_64-pc-linux-gnu.
>>> I tried this patch together with the one for PR tree-optimization/107226
>>> on sparc-sun-solaris2.11 to check if it cures the bootstrap failure
>>> reported in PR tree-optimization/107232.  While this restores bootstrap,
>>> several of the new tests FAIL:
>>>
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-1.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-1.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-2.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-2.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-3.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 2 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-3.c scan-tree-dump-times vect
>>> "vectorized 2 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-4.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-4.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-read-6.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-1.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-1.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-2.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-2.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-3.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-3.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-5.c -flto -ffat-lto-objects
>>> scan-tree-dump-times vect "vectorized 1 loops" 1
>>> +FAIL: gcc.dg/vect/vect-bitfield-write-5.c scan-tree-dump-times vect
>>> "vectorized 1 loops" 1
>>>
>>> For vect-bitfield-read-1.c, the dump has
>>>
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:   ==> examining pattern def
>>> statement: patt_31 = patt_30 >> 1;
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:   ==> examining statement:
>>> patt_31 = patt_30 >> 1;
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:   vect_is_simple_use:
>>> operand _ifc__27 & 4294967294, type of def: internal
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:   vect_is_simple_use:
>>> vectype vector(2) unsigned int
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:   vect_is_simple_use:
>>> operand 1, type of def: constant
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: missed:   op not supported by
>>> target.
>>> gcc.dg/vect/vect-bitfield-read-1.c:23:1: missed:   not vectorized: relevant
>>> stmt not supported: patt_31 = patt_30 >> 1;
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: missed:  bad operation or
>>> unsupported loop bound.
>>> gcc.dg/vect/vect-bitfield-read-1.c:25:23: note:  ***** Analysis  failed with
>>> vector mode V2SI
>>>
>>>   Rainer
>>>

      reply	other threads:[~2022-10-13 14:43 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-12 17:29 Andre Vieira (lists)
2022-10-13  8:14 ` Richard Biener
2022-10-13 10:15   ` Andre Vieira (lists)
2022-10-13 10:18     ` Richard Biener
2022-10-13 11:39 ` Rainer Orth
2022-10-13 13:55   ` Andre Vieira (lists)
2022-10-13 14:15     ` Richard Biener
2022-10-13 14:42       ` Andre Vieira (lists) [this message]

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=bcee7445-9bc0-a0cb-985d-99ef764fc569@arm.com \
    --to=andre.simoesdiasvieira@arm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=rguenther@suse.de \
    --cc=richard.sandiford@arm.com \
    --cc=ro@CeBiTec.Uni-Bielefeld.DE \
    /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).