Mail Archives: djgpp/1998/07/10/19:49:36
I profiled my program with libc_pg.a v2.01, here there are the first
lines. Any idea?
Flat profile:
Each sample counts as 0.055556 seconds.
% cumulative self self total
time seconds seconds calls us/call us/call name
56.12 33.11 33.11 __dpmi_int
18.08 43.78 10.67 mcount
15.44 52.89 9.11 __dj_movedata
2.07 54.11 1.22 _int86
1.60 55.06 0.94 561139 1.68 1.68 strtol
1.41 55.89 0.83 strcopnb
0.75 56.33 0.44 1120977 0.40 0.40 strncpy
0.66 56.72 0.39 1120496 0.35 0.35 strlen
0.28 56.89 0.17 636712 0.26 0.26 strcmp
0.28 57.06 0.17 91811 1.82 1.82 int86
0.28 57.22 0.17 51068 3.26 4.35 _read
0.28 57.39 0.17 51064 3.26 8.70 fread
0.19 57.50 0.11 409063 0.27 0.27 strcpy
0.19 57.61 0.11 91737 1.21 1.21 gettime
0.19 57.72 0.11 clave_dbf
0.19 57.83 0.11 copia_cab_a_mbl
0.19 57.94 0.11 copia_desp_a_mbl
0.19 58.06 0.11 hash
0.19 58.17 0.11 lee_fmuebles
0.19 58.28 0.11 lee_n_dbf
0.09 58.33 0.06 333283 0.17 1.85 atoi
0.09 58.39 0.06 54763 1.01 1.01 __dosmemput
0.09 58.44 0.06 51179 1.09 1.09
__FSEXT_get_function
0.09 58.50 0.06 51131 1.09 1.09 fseek
0.09 58.56 0.06 51066 1.09 5.44 _filbuf
0.09 58.61 0.06 4779 11.62 11.62 malloc
0.09 58.67 0.06 437 127.13 241.28 calcula_coste_mb
0.09 58.72 0.06 __movedata
0.09 58.78 0.06 buf_a_cabmueble
0.09 58.83 0.06 busca
0.09 58.89 0.06 comprueba_mdlmdf
0.09 58.94 0.06 espera_tecla
0.09 59.00 0.06 lee_tecla2
I had dowmladed DJGPP v2.02 (980628) and my program run fast (like the
980101 alpha) and run under W95. It seems to be faster at screen io too,
and when it managed to open/close a lot of small text files (reading
secuentially) it's 2 times faster than v2.01. I plan to intensively test
my program results comparing with the 16 bit version.
Regards, M.Alvarez (I know, my english is horrible :-)
- Raw text -