Cool VL Viewer forum

Register • FAQ • Search • Login
View unanswered posts | View active topics It is currently 2025-02-18 00:29:12

Board index » Contributions

All times are UTC




Reply to topic  Page 2 of 2
  [ 16 posts ]  Go to page Previous  1, 2
Print view Previous topic | Next topic 
Some AVX2 / SSE2 optimization for llface.cpp 
Author Message
Henri Beauchamp

Joined: 2009-03-17 18:42:51
Posts: 5915
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
Yeah, it looks like a math overflow or something in that vein... The copy you sent to me is no-mod, so I cannot verify, but there might be a zero in the texture repeats per face or something like that...

EDIT: Found the bug in Kathrine's SSE2 code: a _mm_add_ps used for tvv instead of _mm_sub_ps...
2020-11-01 23:53:22
Profile WWW
ZaneZimer

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
Henri Beauchamp wrote:
Yeah, it looks like a math overflow or something in that vein... The copy you sent to me is no-mod, so I cannot verify, but there might be a zero in the texture repeats per face or something like that...
I'll send you a mod version. I wondered if you might need that to experiment. I have never given anyone anything that I have crafted before, so I'm a bit of a novice...even as 'SL-old' as I am.
2020-11-02 00:27:27
Profile
Henri Beauchamp

Joined: 2009-03-17 18:42:51
Posts: 5915
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
No need any more: found the bug (see my edit above). Thank you ! :D
2020-11-02 00:30:11
Profile WWW
ZaneZimer

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
Henri Beauchamp wrote:
No need any more: found the bug (see my edit above). Thank you ! :D
Oops. I guess I didn't refresh soon enough. :)
2020-11-02 00:31:47
Profile
Henri Beauchamp

Joined: 2009-03-17 18:42:51
Posts: 5915
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
Fixed patch:


Attachments:
llface-opts-fixed.patch.gz [1.34 KiB]
Downloaded 315 times
2020-11-02 00:32:18
Profile WWW
ZaneZimer

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Post Re: Some AVX2 / SSE2 optimization for llface.cpp
Henri Beauchamp wrote:
Fixed patch:
Thanks, I'll give it a try.

*Edit: I have applied the new patch, built and can verify that corrects the texture orientation issue that I had.
2020-11-02 00:33:41
Profile
Display posts from previous:  Sort by  
Reply to topic   Page 2 of 2
  [ 16 posts ]  Go to page Previous  1, 2

Board index » Contributions

All times are UTC


Who is online

Users browsing this forum: No registered users and 2 guests

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software.