RSS/Atom feed Twitter
Site is read-only, email is disabled

ocl regression

This discussion is connected to the gegl-developer-list.gnome.org mailing list which is provided by the GIMP developers and not related to gimpusers.com.

This is a read-only list on gimpusers.com so this discussion thread is read-only, too.

2 of 2 messages available
Toggle history

Please log in to manage your subscriptions.

ocl regression Jan Vesely 18 Jun 23:16
  ocl regression Jan Vesely 19 Jun 13:32
Jan Vesely
2014-06-18 23:16:25 UTC (almost 10 years ago)

ocl regression

Hi,

I noticed two tests (clones.xml, gegl.xml) regressed on both intel ocl sdk and mesa/clover. bisect shows the first bad commit to be: de26038e19883243fb465ddf290a05eb23d91bae sampler: make size of cache in sampler base class adaptive

both tests fail in one pixel: wrong pixels : 1/196608 (0.00%)
max ?e : 62.833
avg ?e (wrong) : 62.833(wrong) 0.000(total) FAIL clones.xml (OpenCL)

wrong pixels : 1/79000 (0.00%) max ?e : 100.000
avg ?e (wrong) : 100.000(wrong) 0.001(total) FAIL gegl.xml (OpenCL)

regards,

Jan Vesely 
Jan Vesely
2014-06-19 13:32:02 UTC (almost 10 years ago)

ocl regression

390214ca03ea0c973037a2b84fd77ccb592f093c fixes this,

thanks

On Wed, 2014-06-18 at 19:16 -0400, Jan Vesely wrote:

Hi,

I noticed two tests (clones.xml, gegl.xml) regressed on both intel ocl sdk and mesa/clover. bisect shows the first bad commit to be: de26038e19883243fb465ddf290a05eb23d91bae sampler: make size of cache in sampler base class adaptive

both tests fail in one pixel: wrong pixels : 1/196608 (0.00%)
max ?e : 62.833
avg ?e (wrong) : 62.833(wrong) 0.000(total) FAIL clones.xml (OpenCL)

wrong pixels : 1/79000 (0.00%) max ?e : 100.000
avg ?e (wrong) : 100.000(wrong) 0.001(total) FAIL gegl.xml (OpenCL)

regards,

Jan Vesely