[PATCH] Fix constant folding for one lane vectors (PR15611)

[Silviu Baranga]

Thanks! Committed in r180254.

- Silviu

From: Bob Wilson [mailto:bob.wilson at apple.com]
Sent: 25 April 2013 06:16
To: Silviu Baranga
Cc: llvm-commits at cs.uiuc.edu; LLVM
Subject: Re: [PATCH] Fix constant folding for one lane vectors (PR15611)

This looks good to me.

On Apr 22, 2013, at 3:59 AM, Silviu Baranga <silbar01 at arm.com<mailto:silbar01 at arm.com>> wrote:


Ping!

- Silviu


-----Original Message-----
From: Silviu Baranga [mailto:silbar01 at arm.com]
Sent: 12 April 2013 15:29
To: llvm-commits at cs.uiuc.edu<mailto:llvm-commits at cs.uiuc.edu>
Subject: [PATCH] Fix constant folding for one lane vectors (PR15611)

Hi,

The attached patch fixes constant folding for one lane vectors,
which is the root cause for PR15611. The faulty code path is
taken by scalarizing vector code during type legalization.

The constant folding code will currently return a scalar when
taking one lane input vectors, which can cause an assertion failure
later.

The changed code was wrong because if the size of the output vector
is 1, we can still be in the case where we need to return a one lane
vector.

Please review!

Thanks,
Silviu
<v1-constant-fold.diff>_______________________________________________
llvm-commits mailing list
llvm-commits at cs.uiuc.edu<mailto:llvm-commits at cs.uiuc.edu>
http://lists.cs.uiuc.edu/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.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130425/1c3f9069/attachment.html>