Date: Mon, 24 Apr 2006 01:59:09 GMT From: Kip Macy <kmacy@FreeBSD.org> To: Perforce Change Reviews <perforce@freebsd.org> Subject: PERFORCE change 95973 for review Message-ID: <200604240159.k3O1x9Xu047917@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=95973 Change 95973 by kmacy@kmacy_storage:sun4v_rwbuf on 2006/04/24 01:58:41 update init_mondo use Affected files ... .. //depot/projects/kmacy_sun4v/src/sys/sun4v/sun4v/mp_machdep.c#5 edit Differences ... ==== //depot/projects/kmacy_sun4v/src/sys/sun4v/sun4v/mp_machdep.c#5 (text+ko) ==== @@ -316,10 +316,10 @@ pc->pc_node = child; all_cpus |= 1 << cpuid; -#if 1 +#if 0 if (mp_ncpus == MAXCPU) #else - if (mp_ncpus == 16) + if (mp_ncpus == 4) #endif break; @@ -455,7 +455,7 @@ int i, retries; - init_mondo(d0, d1, d2, (uint64_t)ackmask); + init_mondo(d0, d1, d2, (uint64_t)pmap_kextract((vm_offset_t)ackmask)); retries = 0; @@ -511,16 +511,13 @@ cpumask_t cpus; uint64_t ackmask; - /* 1) our cpu_ipi_ast doesn't do anything - * 2) a reschedule will be triggered at the next timer interrupt + /* + * * 3) forward_wakeup appears to abuse ASTs - * 4) handling 4-way threading vs 2-way threading will just serve - * to further obfuscate forward_wakeup + * 4) handling 4-way threading vs 2-way threading should happen here + * and not in forward wakeup */ - if (ipi == IPI_AST) - return; - cpulist = PCPU_GET(cpulist); cpus = ((icpus & ~PCPU_GET(cpumask)) & ipi_ready_mask); @@ -531,7 +528,7 @@ cpulist[cpu_count] = (uint16_t)i; cpu_count++; } - + cpu_ipi_selected(cpu_count, cpulist, (u_long)tl_ipi_level, ipi, 0, &ackmask); }
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200604240159.k3O1x9Xu047917>