Quoting Keith Packard <[EMAIL PROTECTED]>: > Here's a .deb which should be pretty close to 2.3.2 when it does get > released (your bug is blocking that). > > Please run your usual tests and see if it runs out of memory, even > better would be to run > > valgrind --leak-check=full fc-cache -f > > as root and see if there are any leaks reported. There should be none, > I even destroy the configuration object when fc-cache exits now. > > -keith >
unfortunately, i am still having the problem of fc-cache -f running out of memory. it even runs out of memory sooner when it is running under valgrind. i think valgrind is exiting before it does all the statistics it is supposed to do. every time fc-cache runs out of memory, i get something like this written to the virtual terminal: oom-killer: gfp_mask=0x1d2 DMA per-cpu: cpu 0 hot: low 2, high 6, batch 1 cpu 0 cold: low 0, high 2, batch 1 Normal per-cpu: cpu 0 hot: low 32, high 96, batch 16 cpu 0 cold: low 0, high 32, batch 16 HighMem per-cpu: empty Free Pages 2784kB (0kB HighMem) Active:121556 inactive:683 dirty:0 writeback:0 unstable:0 free:696 slab:2827 mapped:121912 pagetables:685 DMA free:1432kB min:20kB low:40kB high:60kB active:11920kB inactive:0kB present:16384kB protections[]: 10 358 358 Normal free:1352kB min:696kB low:1392kB high:2088kB active:474304kB inactive:2732kB present:507328kB protections[]: 0 348 348 DMA: 0*4kB 1*8kB 1*16kB 0*32kB 0*64kB 1*128kB 1*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1432kB Normal: 0*4kB 1*8kB 2*16kB 1*32kB 0*64kB 0*128kB 1*256kB 0*512kB 1*1024kB 0*2048kB 0*4096kB = 1352kB HighMem: empty Swap cache: add 0, delete 0, find 0/0, race 0+0 Out of Memory: Killed process 5121 (fc-cache). maybe there's something in there that means something to you. i ran valgrind --leak-check=full --verbose --log-file=/root/3 fc-cache -fv >/root/4 i have attatched the two logs as 3.txt and 4.txt i also ran fc-cache -fv >/root/5 just to see how far through the fonts it could go before running out of memory. i have attached this as 5.txt. hope this helps. if there are any more tests or if there are new versions you want me to try, let me know. chris. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
