Date: Tue, 16 Nov 2010 12:43:45 +0000 (UTC) From: Andriy Gapon <avg@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r215398 - head/sys/x86/cpufreq Message-ID: <201011161243.oAGChjpi054218@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: avg Date: Tue Nov 16 12:43:45 2010 New Revision: 215398 URL: http://svn.freebsd.org/changeset/base/215398 Log: hwpstate: use CPU_FOREACH when binding to all available processors Also, add a comment mentioning _PSD - on some systems it's enough to put one logical CPU into a particular P-state to make other CPUs in the same domain to enter that P-state. Also, call sched_unbind() after the loop - sched_bind() automatically rebinds from previous CPU to a new one, and the new arrangement of code is safer against early loop exit. Plus one minor style nit. MFC after: 10 days Modified: head/sys/x86/cpufreq/hwpstate.c Modified: head/sys/x86/cpufreq/hwpstate.c ============================================================================== --- head/sys/x86/cpufreq/hwpstate.c Tue Nov 16 12:30:47 2010 (r215397) +++ head/sys/x86/cpufreq/hwpstate.c Tue Nov 16 12:43:45 2010 (r215398) @@ -157,7 +157,6 @@ DRIVER_MODULE(hwpstate, cpu, hwpstate_dr static int hwpstate_goto_pstate(device_t dev, int pstate) { - struct pcpu *pc; int i; uint64_t msr; int j; @@ -171,18 +170,15 @@ hwpstate_goto_pstate(device_t dev, int p if(limit > id) id = limit; - error = 0; /* * We are going to the same Px-state on all cpus. + * Probably should take _PSD into account. */ - for (i = 0; i < mp_ncpus; i++) { - /* Find each cpu. */ - pc = pcpu_find(i); - if (pc == NULL) - return (ENXIO); - thread_lock(curthread); + error = 0; + CPU_FOREACH(i) { /* Bind to each cpu. */ - sched_bind(curthread, pc->pc_cpuid); + thread_lock(curthread); + sched_bind(curthread, i); thread_unlock(curthread); HWPSTATE_DEBUG(dev, "setting P%d-state on cpu%d\n", id, PCPU_GET(cpuid)); @@ -204,10 +200,10 @@ hwpstate_goto_pstate(device_t dev, int p HWPSTATE_DEBUG(dev, "error: loop is not enough.\n"); error = ENXIO; } - thread_lock(curthread); - sched_unbind(curthread); - thread_unlock(curthread); } + thread_lock(curthread); + sched_unbind(curthread); + thread_unlock(curthread); return (error); }
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201011161243.oAGChjpi054218>