Hi Michael, Thanks for the input, I had forgotten to disable/undef DEBUG_PERF, which I was using to get a rough idea of the time taken by the logic using the processor performance counter on x86. The logic doesnt use x86intrinsic for any other purpose, so disabling DEBUG_PERF will fix it.
I can release an updated patch with the DEBUG_PERF undef'd by default, to avoid this issue. On Thu, Jul 2, 2020 at 6:28 PM Michael Niedermayer <mich...@niedermayer.cc> wrote: > On Wed, Jul 01, 2020 at 10:13:48PM +0530, hanishkvc wrote: > > v04-20200701IST2132, fbdetile > > > > Optimised Generic Detile logic to detile multiple tiles in parallel, > > logically speaking. This reduces cross check overheads to some extent > > and thus improves speed a bit. While in a hardware or multicore setup, > > it can be used to parallelise the detiling operations in a true sense. > > > > Add a additional level of subtiling wrt Tile-Yf, which I had missed out > > in the dirChangesList based tiling configuration, which is used by my > > generic detiling logic. THe overhead due to this additional subtiling > > is compensated by the speed gains due to parallel detiling. > > > > NOTE: This is a consolidated patch, it contains previous versions also > > > > v03-20200629IST2208 fbdetile > > > > Added a generic detiling logic, which can be easily configured to > > detile many different tiling schemes. > > > > The same is inturn used to detile Intel Tile-Yf layout. > > > > v02-20200627IST2331 > > > > Unrolled Intel Legacy Tile-Y detiling logic. > > > > Also a consolidated patch file, instead of the previous development > > flow based multiple patch files. > > > > v01-20200627IST1308 > > > > Implemented Intel Legacy Tile-X and Tile-Y detiling logic > > > > NOTES: > > > > This video filter allows framebuffers which are tiled to be detiled > > using logic running on the cpu, into a linear layout. > > > > Currently it supports Intel Legacy Tile-X and Tile-Y layout detiling, > > as well as the newer Intel Tile-Yf layouts. > > > > THis should help one to work with frames captured (say using kmsgrab) > > on laptops having Intel GPU. This can be done live while capturing > > itself, or it can be applied later as a seperate pass. > > > > Tile-X conversion logic has been explicitly cross checked, with Tile-X > > based frames. However Tile-Y and Tile-Yf conv logics havent been tested > > with Tile-Y | Tile-Yf based frames, but it should potentially get the > > job done, based on my current understanding of these layout formats. A > > minimal test has been done by seeing how a multicolor linear framebuffer > > gets converted into a patterned layout, depending on the detile walk. > > --- > > Changelog | 1 + > > doc/filters.texi | 74 +++++ > > libavfilter/Makefile | 1 + > > libavfilter/allfilters.c | 1 + > > libavfilter/vf_fbdetile.c | 568 ++++++++++++++++++++++++++++++++++++++ > > 5 files changed, 645 insertions(+) > > create mode 100644 libavfilter/vf_fbdetile.c > > This breaks build on non x86 > > src/libavfilter/vf_fbdetile.c:81:10: fatal error: x86intrin.h: No such > file or directory > #include <x86intrin.h> > ^~~~~~~~~~~~~ > > [...] > -- > Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB > > In a rich man's house there is no place to spit but his face. > -- Diogenes of Sinope > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe". -- Keep ;-) HanishKVC _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
