I did a "valgrind" using IgorBeg's "stupid test" and trying to activate the 4 preview submenus. I hope it's useful. https://www.dropbox.com/s/8jf1mufh0ewhjzs/val.log?dl=0
Andrea: thank you very much for sending the results as running Valgrind is time-consuming and requires a lot of patience! The impressive results as shown below is out of over 6 GB used only 600,000+ bytes - that is *only 1/100 of 1%* (a little over half a MB for over 6 GB!)* lost on a quitting Cinelerra-gg.* HEAP SUMMARY: in use at exit: 694,680 bytes in 6,052 blocks total heap usage: 1,063,358 allocs, 1,057,306 frees, 6,180,470,729 bytes allocated GG has glanced at it already and will take a closer look as there are some "Conditional jumps" that he may be able to get rid of. Also, some memory leaks in ffmpeg that he has to look into: ==1322== 40 (32 direct, 8 indirect) bytes in 1 blocks are definitely lost in loss record 88 of 404 ... ==1322== by 0x7134A0: init (af_aformat.c:101) ==1322== by 0x18757DF: avfilter_init_str (avfilter.c:1018) On Fri, Mar 15, 2019 at 3:20 AM Andrea paz <[email protected]> wrote:
I did a "valgrind" using IgorBeg's "stupid test" and trying to activate the 4 preview submenus. I hope it's useful.
Andrea, Fix for the conditional jumps and the av audio fffmpeg lost bytes were checked into GIT in the last couple of days. gg/Phyllis On Fri, Mar 15, 2019 at 8:19 AM Phyllis Smith <[email protected]> wrote:
Andrea: thank you very much for sending the results as running Valgrind is time-consuming and requires a lot of patience!
The impressive results as shown below is out of over 6 GB used only 600,000+ bytes - that is *only 1/100 of 1%* (a little over half a MB for over 6 GB!)* lost on a quitting Cinelerra-gg.* HEAP SUMMARY: in use at exit: 694,680 bytes in 6,052 blocks total heap usage: 1,063,358 allocs, 1,057,306 frees, 6,180,470,729 bytes allocated
GG has glanced at it already and will take a closer look as there are some "Conditional jumps" that he may be able to get rid of. Also, some memory leaks in ffmpeg that he has to look into:
==1322== 40 (32 direct, 8 indirect) bytes in 1 blocks are definitely lost in loss record 88 of 404 ... ==1322== by 0x7134A0: init (af_aformat.c:101) ==1322== by 0x18757DF: avfilter_init_str (avfilter.c:1018)
On Fri, Mar 15, 2019 at 3:20 AM Andrea paz <[email protected]> wrote:
I did a "valgrind" using IgorBeg's "stupid test" and trying to activate the 4 preview submenus. I hope it's useful.
New release with fix leak memory, new valgrind: https://www.dropbox.com/s/8jf1mufh0ewhjzs/val.log?dl=0
participants (2)
-
Andrea paz -
Phyllis Smith