[PATCH] D33205: ARM] Fix Neon vector type alignment to 64-bit

[Stephen Hines via llvm-commits]

I would request that this not change for Android. Our implementation has
been using 128-bit alignment for far too long, and I would rather see that
document updated to say 128-bit alignment to match the behavior as
implemented. Changing this in LLVM now would cause all sorts of binary
compatibility issues that we don't want to force on our users. Past
vector-related ABI issues have been grandfathered in before, so I do hope
you reconsider this patch.

Thanks,
Steve

On Tue, May 16, 2017 at 1:40 AM, Javed Absar via llvm-commits <
llvm-commits at lists.llvm.org> wrote:

> Hi James:
>
>
> Good point. I am considering the following case -
>
> struct test_struct {
>
>   int a;
>
>   int32x4_t b;
>
> };
>
>
> clang will currently give it 12 bytes of padding, even though it should have been 4 bytes according to AAPCS.
>
>
> *** Dumping AST Record Layout
>
>          0 | struct test_struct
>
>          0 |   int a
>
>         16 |   int32x4_t b
>
>            | [sizeof=32, align=16]
>
> Best Regards
>
> Javed
>
>
> p.s. James, for some reason, you comment does not appear in the
> phabricator.
>
>
>
> ------------------------------
> *From:* James Molloy <james at jamesmolloy.co.uk>
> *Sent:* 15 May 2017 18:49
> *To:* reviews+D33205+public+7fad182a923fbe49 at reviews.llvm.org; Javed
> Absar via Phabricator; Javed Absar; diana.picus at linaro.org;
> renato.golin at linaro.org
> *Cc:* junbuml at codeaurora.org; llvm-commits at lists.llvm.org;
> kanheim at a-bix.com
> *Subject:* Re: [PATCH] D33205: ARM] Fix Neon vector type alignment to
> 64-bit
>
> Hi Javed,
>
> Why reduce the preferred alignment from 128? Natural vector alignment
> means we can more often use the :128 hints on loads and stores.
>
> What's the rationale here?
>
> James
> On Mon, 15 May 2017 at 18:10, Javed Absar via Phabricator via llvm-commits
> <llvm-commits at lists.llvm.org> wrote:
>
>> javed.absar created this revision.
>> Herald added subscribers: srhines, aemerson.
>>
>> The maximum alignment for ARM NEON data types should be 64-bits as
>> specified in  ARM procedure call standard document Sec. A.2 Notes  [1].
>> This patch fixes it from its current larger natural default values.
>>
>> [1] http://infocenter.arm.com/help/topic/com.arm.doc.
>> ihi0042f/IHI0042F_aapcs.pdf
>>
>>
>> https://reviews.llvm.org/D33205
>>
>> Files:
>>   lib/Basic/Targets.cpp
>>   test/CodeGen/arm-abi-vector.c
>>   test/CodeGen/arm-neon-misc.c
>>   test/CodeGen/arm-swiftcall.c
>>   test/CodeGen/armv7k-abi.c
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>>
> IMPORTANT NOTICE: The contents of this email and any attachments are
> confidential and may also be privileged. If you are not the intended
> recipient, please notify the sender immediately and do not disclose the
> contents to any other person, use it for any purpose, or store or copy the
> information in any medium. Thank you.
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170516/19ac73ad/attachment.html>