Hi Eli,

I attached a patch that simplifies a few occurrences of alloca in the file.

Cheers,
Juergen

Attachment: tmpmem_cleanup.diff
Description: Binary data


On Aug 18, 2013, at 6:10 PM, Eli Friedman <[email protected]> wrote:

Okay, thanks; then I won't worry about the LLVM side of this.

+    llvm::Type *VecTy = llvm::VectorType::get(Int64Ty, 2);
+    llvm::Type *PtrTy = Int8PtrTy;
+    Value *_One_ = llvm::ConstantInt::get(
Int32Ty, 1);
+    Value *Tmp = Builder.CreateAlloca(VecTy, One);

You can simplify this part using CodeGenFunction::CreateMemTemp.

-Eli


On Sun, Aug 18, 2013 at 12:35 PM, Craig Topper <[email protected]> wrote:
And I totally failed to reply to all when I did so.


On Sun, Aug 18, 2013 at 11:58 AM, Juergen Ributzka <[email protected]> wrote:
Sorry, forgot to mention that Craig reviewed it.
-Juergen

On Aug 18, 2013, at 11:50 AM, Eli Friedman <[email protected]> wrote:

On Sat, Aug 17, 2013 at 9:40 AM, Juergen Ributzka <[email protected]> wrote:
Author: ributzka
Date: Sat Aug 17 11:40:09 2013
New Revision: 188609

URL: http://llvm.org/viewvc/llvm-project?rev=188609&view=rev
Log:
Fix the name and the type of the argument for intrinisc
_mm256_broadcastsi128_si256 to align with the Intel documentation.

This fixes bug PR 16581 and rdar:14747994.


Who reviewed this?

-Eli


_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits




--
~Craig


_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to