Date: Thu, 05 Oct 2000 13:07:25 -0700 (PDT) From: John Baldwin <jhb@FreeBSD.org> To: Peter Wemm <peter@netplex.com.au> Cc: Greg Lehey <grog@lemis.com>, Matt Dillon <dillon@earth.backplane.com>, Daniel Eischen <eischen@vigrid.com>, John Polstra <jdp@polstra.com>, arch@FreeBSD.org, Alfred Perlstein <bright@wintelcom.net>, Chuck Paterson <cp@bsdi.com> Subject: Re: Mutexes and semaphores Message-ID: <XFMail.001005130725.jhb@FreeBSD.org> In-Reply-To: <200010040750.e947oaH19391@netplex.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On 04-Oct-00 Peter Wemm wrote: > Peter Wemm wrote: >> Chuck Paterson wrote: >> > Your right, not freeing these things ever does make things >> > lots easier. >> > >> > Chuck >> >> In the freebsd case, this is the case. Zones are never cleaned up, and >> certainly not unmapped. zfree() will however cause the first few bytes >> to be clobbered as they are reused for the freelist. > > Actually, a final version is at: > http://people.freebsd.org/~peter/getppid.diff > > This produces nice tight assembler code. The SMP reentrancy protection > adds two instructions.. one cmp and one conditional branch which gets taken > if there is a lost race. It hardly seems worth #ifdef'ing it (and it isn't > ifdefed in the diff). This diff is missing is the MPSAFE flag for the > getppid syscall, but thats trivial to add (for native syscalls and all other > syscall vectors that call getppid directly). Darn it, review my barrier_foo() functions on the new-bus list so you can use that instead of a manual assembler mb(); :-P > Cheers, > -Peter -- John Baldwin <jhb@FreeBSD.org> -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.baldwin.cx/~john/pgpkey.asc "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.001005130725.jhb>