Can someone explain this?



In order to track some performance issues, I've been "stracing" a program that makes gtk calls. The following is a portion of the output that I'm curious about.

The snippet below is a result of a call to  gdk_pixmap_create_from_xpm

Specifically does anyone know why the exact same _llseek call is repeated several hundred times?


...

open("/XPM/common/title/title.xpm", O_RDONLY) = 15
fstat64(15, {st_mode=S_IFREG|0664, st_size=18293, ...}) = 0
old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40313000
read(15, "/* XPM */\nstatic char *title_rec"..., 4096) = 4096
_llseek(15, -4005, [91], SEEK_CUR)      = 0
read(15, "\"   290    25      221          "..., 4096) = 4096
_llseek(15, 4187, [4187], SEEK_SET)     = 0
_llseek(15, 4187, [4187], SEEK_SET)     = 0
_llseek(15, 4187, [4187], SEEK_SET)     = 0

... Repeat the above __llseek several hundred times

_llseek(15, 4187, [4187], SEEK_SET)     = 0
shmget(IPC_PRIVATE, 14500, IPC_CREAT|0x1ff|0666) = 39190530
shmat(39190530, 0, 0)                   = 0x40843000
write(10, "\201\1\4\0\350\5\340\0\2\0V\2\0\240#\1+\1\1\0", 20) = 20
read(10, "\1\1W5\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 32) = 32
shmctl(39190530, IPC_RMID, 0)           = 0
read(15, "caO.CaO.CacacaV.M.d.M.daV.daA.d."..., 4096) = 4096
read(15, "ibiaEbiaEbi#Nbibibm#tbm#tbmbibi#"..., 4096) = 4096
read(15, "q#8#N#F#N#8#F#oaqaqaqbi#F#Fbibib"..., 4096) = 4096
read(15, "E#tbibmbibibibm#tbmbmbi.Savav#Sa"..., 4096) = 1818
write(10, "5\20\4\0\351\5\340\0\336\5\340\0\"\1\31\0007\1\5\0\352"..., 104) = 104 read(10, "\1\1]5\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 32) = 32
shmdt(0x40843000)                       = 0
close(15)                               = 0




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]