r374288 - Recommit "[Clang] Pragma vectorize_width() implies vectorize(enable)"

[Sjoerd Meijer via cfe-commits]

Hi All,

Apologies for the delay in responding, I was away from keyboard (llvm dev conf).

This is a real funny patch! It is a (simple) fix that uncovers quite a few things. Anyway, thanks Michael for your analysis! As I will need to make some changes this time (this wasn't the case last time), I definitely definitely open a new review soon.

Cheers,
Sjoerd.
________________________________
From: Michael Kruse <llvm-commits at meinersbur.de>
Sent: 25 October 2019 23:18
To: Jordan Rupprecht <rupprecht at google.com>
Cc: Michael Kruse <llvm-commits at meinersbur.de>; Hans Wennborg <hans at chromium.org>; cfe-commits <cfe-commits at lists.llvm.org>; Hal Finkel <hfinkel at anl.gov>; Sjoerd Meijer <Sjoerd.Meijer at arm.com>
Subject: Re: r374288 - Recommit "[Clang] Pragma vectorize_width() implies vectorize(enable)"

@sjoerdmeijer

Before recommitting, please re-open the patch review.

Michael

Am Do., 24. Okt. 2019 um 18:45 Uhr schrieb Jordan Rupprecht
<rupprecht at google.com>:
>
> Reverted in 6d424a161bf3e52730371da0b9439ed93a8ce406 due to the issue described here. Should hopefully be a trivial fix forward.
>
> On Tue, Oct 22, 2019 at 2:46 PM Michael Kruse <llvm-commits at meinersbur.de> wrote:
>>
>> Am Mo., 21. Okt. 2019 um 23:44 Uhr schrieb Jordan Rupprecht
>> <rupprecht at google.com>:
>> > At any rate, it sounds like this is not a codegen bug at all, but just an over-eager warning?
>>
>> That interpretation is different from mine. Codgen emits the following
>> from vectorize(disable)
>>
>> !4 = !{!"llvm.loop.vectorize.enable", i1 true}
>>
>> which is is not what I'd expect.
>>
>> Michael
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.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20191029/23a4ece8/attachment-0001.html>