Messages in this thread | | | Date | Wed, 29 Mar 2000 10:06:41 -0800 | From | Christopher Smith <> | Subject | Re: Slow pthread_create() under high load |
| |
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Tue, Mar 28, 2000 at 01:20:40PM -0700, yodaiken@fsmlabs.com wrote: > On Tue, Mar 28, 2000 at 12:06:12PM -0800, David Schwartz wrote: > > Define 'work'. 'getpid' is hardly the only function that uses pids. What > > about 'getppid'? As but one example, consider the following code: > Fork certainly complicates and since we don't allow fork in our realtime > pthreads environment, it did not occur to me. But surely this is no > biggie: fork must reset the cached value in getpid. The child forked process > is single threaded and until the first new thread is created, getpid can > be getpid. I was going to say that for most pthreads implementations I can think of, if fork() has a lot of overhead or must be serialized, that's ok.
- --Chris -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.1 (GNU/Linux) Comment: Processed by Mailcrypt 3.5.5 and Gnu Privacy Guard <http://www.gnupg.org/>
iD8DBQE44kYufrrCpthD+UYRAqHKAJ9N4M4hEBXsLTNlqr4xceVwGkshqgCeKPrs VE/dkDXkbl1HsyJaJEawPrI= =P+YB -----END PGP SIGNATURE-----
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |