Date: Sat, 23 Nov 1996 15:31:29 -0700 From: Steve Passe <smp@csn.net> To: dg@Root.COM Cc: freebsd-smp@freefall.freebsd.org, Peter Wemm <peter@spinner.dialix.com> Subject: Re: SMP -current merge Message-ID: <199611232231.PAA19920@clem.systemsix.com> In-Reply-To: Your message of "Sat, 23 Nov 1996 14:17:52 PST." <199611232217.OAA01895@root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, > Uhh, the "idle" queue shouldn't have anything on it unless there is an >"idprio" process running. If this patch truely makes a difference, then >there is something very wrong with the handling of whichidqs somewhere. If there are 2 CPUS smp_kickoff() creates 2 idleprocs, and thus there are 2 possible processess that can be in the idleq. When there are no rtq procs and no norq procs to run, the 1st CPU is inside smp_idleloop of cpuidleX, and the 2nd CPU is in smp_idleloop of cpuidleY, and thus there are no procs in the idleq. When the 1st CPU is busy doing something usefull (running something other than an idleproc), the 2nd CPU will be in smp_idleloop of either cpuidleX or cpuidleY, and the other cpuidle[XY] will be in the idleq. So the 2nd CPU sees the idleproc that it ISN'T currently in and switches to that one, and on and on .... -- Steve Passe | powered by smp@csn.net | FreeBSD -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.6.2 mQCNAzHe7tEAAAEEAM274wAEEdP+grIrV6UtBt54FB5ufifFRA5ujzflrvlF8aoE 04it5BsUPFi3jJLfvOQeydbegexspPXL6kUejYt2OeptHuroIVW5+y2M2naTwqtX WVGeBP6s2q/fPPAS+g+sNZCpVBTbuinKa/C4Q6HJ++M9AyzIq5EuvO0a8Rr9AAUR tBlTdGV2ZSBQYXNzZSA8c21wQGNzbi5uZXQ+ =ds99 -----END PGP PUBLIC KEY BLOCK-----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611232231.PAA19920>