This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

strange top statistics for 3.1 cc1


I'm running RedHat 7.1 and compiling with

[lucier@curie objdir]$ /usr/local/gcc-3.1/bin/gcc -v
Reading specs from /usr/local/gcc-3.1/lib/gcc-lib/i686-pc-linux-gnu/3.1/specs
Configured with: ../configure --prefix=/usr/local/gcc-3.1 --enable-checking=no
Thread model: single
gcc version 3.1 20010824 (experimental)

The BOOT_CFLAGS were -O2 -g.

The top statistics on my machine are:

  1:51pm  up 5 days, 19:44,  3 users,  load average: 1.02, 1.14, 1.08
66 processes: 60 sleeping, 4 running, 0 zombie, 2 stopped
CPU states: 49.2% user, 50.7% system,  0.0% nice,  0.0% idle
Mem:   384516K av,  350216K used,   34300K free,     584K shrd,    1272K buff
Swap:  787200K av,  232204K used,  554996K free                  206828K cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
27425 lucier    18   0  168M 167M 69572 R    95.9 44.4  14:51 cc1
27444 lucier    14   0  1056 1056   840 R     1.5  0.2   0:00 top

Note that nearly half the CPU time is in system time; this is
the same when running xosview.

Is this accurate?  Is top crazy, or is 3.1 spending so much time
in system code?

Brad Lucier


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]