[x265] [PATCH] asm: idct[8x8] sse2 12232.53 -> 3480 over c code, 3555 -> over intrinsic
chen
chenm003 at 163.com
Tue Dec 9 03:48:35 CET 2014
If you look into ABI documents, you will be found that the Linux/BSD/Mac have a Red Zone range from [rsp-128] to [rsp-8], but in Windows, we haven't it.
在 2014-12-09 08:42:51,dave <dtyx265 at gmail.com> 写道:
On 12/08/2014 03:35 PM, chen wrote:
This is x64 only version, but you didn't check enviorment.
I will add a check.
And you use rsp without alloc space.
Sorry. For amd there is 128 bytes of space beyond the end of the stack that is free to use if there are no calls, at least in linux, of which not all 128 bytes are used. It looks like some of the other primitives also use this space. Is there another primitive that allocates space that I can use as an example?
At 2014-12-09 07:09:04,dtyx265 at gmail.com wrote:
># HG changeset patch
># User David T Yuen <dtyx265 at gmail.com>
># Date 1418080004 28800
># Node ID c16530ca89e4f55c2c1fd17bd2cc4640589b096b
># Parent 53f7efef5ebda6d5ff03e868f2b702c385d72ddd
>asm: idct[8x8] sse2 12232.53 -> 3480 over c code, 3555 -> over intrinsic
>
_______________________________________________
x265-devel mailing list
x265-devel at videolan.orghttps://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20141209/ccfda52d/attachment.html>
More information about the x265-devel
mailing list