From 57d1c0c03c6b48b2b96870d831b9ce6b917f53ac Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 7 Oct 2011 13:36:40 +0200 Subject: perf/x86: Fix PEBS instruction unwind Masami spotted that we always try to decode the instruction stream as 64bit instructions when running a 64bit kernel, this doesn't work for ia32-compat proglets. Use TIF_IA32 to detect if we need to use the 32bit instruction decoder. Reported-by: Masami Hiramatsu Cc: stable@kernel.org Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel_ds.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index c0d238f49db8..73da6b64f5b7 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -493,6 +493,7 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) unsigned long from = cpuc->lbr_entries[0].from; unsigned long old_to, to = cpuc->lbr_entries[0].to; unsigned long ip = regs->ip; + int is_64bit = 0; /* * We don't need to fixup if the PEBS assist is fault like @@ -544,7 +545,10 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) } else kaddr = (void *)to; - kernel_insn_init(&insn, kaddr); +#ifdef CONFIG_X86_64 + is_64bit = kernel_ip(to) || !test_thread_flag(TIF_IA32); +#endif + insn_init(&insn, kaddr, is_64bit); insn_get_length(&insn); to += insn.length; } while (to < ip); -- cgit v1.2.3 From aa2bc1ade59003a379ffc485d6da2d92ea3370a6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 9 Nov 2011 17:56:37 +0100 Subject: perf: Don't use -ENOSPC for out of PMU resources People (Linus) objected to using -ENOSPC to signal not having enough resources on the PMU to satisfy the request. Use -EINVAL. Requested-by: Linus Torvalds Cc: Stephane Eranian Cc: Will Deacon Cc: Deng-Cheng Zhu Cc: David Daney Cc: Ralf Baechle Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-xv8geaz2zpbjhlx0svmpp28n@git.kernel.org [ merged to newer kernel, fixed up MIPS impact ] Signed-off-by: Ingo Molnar --- arch/arm/kernel/perf_event.c | 6 +++--- arch/mips/kernel/perf_event_mipsxx.c | 8 ++++---- arch/x86/kernel/cpu/perf_event.c | 10 +++++----- arch/x86/kernel/cpu/perf_event_p4.c | 2 +- 4 files changed, 13 insertions(+), 13 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 24e2347be6b1..ff17b17b668e 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -347,15 +347,15 @@ validate_group(struct perf_event *event) memset(&fake_pmu, 0, sizeof(fake_pmu)); if (!validate_event(&fake_pmu, leader)) - return -ENOSPC; + return -EINVAL; list_for_each_entry(sibling, &leader->sibling_list, group_entry) { if (!validate_event(&fake_pmu, sibling)) - return -ENOSPC; + return -EINVAL; } if (!validate_event(&fake_pmu, event)) - return -ENOSPC; + return -EINVAL; return 0; } diff --git a/arch/mips/kernel/perf_event_mipsxx.c b/arch/mips/kernel/perf_event_mipsxx.c index 4f2971bcf8e5..315fc0b250f8 100644 --- a/arch/mips/kernel/perf_event_mipsxx.c +++ b/arch/mips/kernel/perf_event_mipsxx.c @@ -623,7 +623,7 @@ static int mipspmu_event_init(struct perf_event *event) if (!atomic_inc_not_zero(&active_events)) { if (atomic_read(&active_events) > MIPS_MAX_HWEVENTS) { atomic_dec(&active_events); - return -ENOSPC; + return -EINVAL; } mutex_lock(&pmu_reserve_mutex); @@ -732,15 +732,15 @@ static int validate_group(struct perf_event *event) memset(&fake_cpuc, 0, sizeof(fake_cpuc)); if (!validate_event(&fake_cpuc, leader)) - return -ENOSPC; + return -EINVAL; list_for_each_entry(sibling, &leader->sibling_list, group_entry) { if (!validate_event(&fake_cpuc, sibling)) - return -ENOSPC; + return -EINVAL; } if (!validate_event(&fake_cpuc, event)) - return -ENOSPC; + return -EINVAL; return 0; } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 640891014b2a..ff0e8d498750 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -588,7 +588,7 @@ done: x86_pmu.put_event_constraints(cpuc, cpuc->event_list[i]); } } - return num ? -ENOSPC : 0; + return num ? -EINVAL : 0; } /* @@ -607,7 +607,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, if (is_x86_event(leader)) { if (n >= max_count) - return -ENOSPC; + return -EINVAL; cpuc->event_list[n] = leader; n++; } @@ -620,7 +620,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, continue; if (n >= max_count) - return -ENOSPC; + return -EINVAL; cpuc->event_list[n] = event; n++; @@ -1316,7 +1316,7 @@ static int validate_event(struct perf_event *event) c = x86_pmu.get_event_constraints(fake_cpuc, event); if (!c || !c->weight) - ret = -ENOSPC; + ret = -EINVAL; if (x86_pmu.put_event_constraints) x86_pmu.put_event_constraints(fake_cpuc, event); @@ -1341,7 +1341,7 @@ static int validate_group(struct perf_event *event) { struct perf_event *leader = event->group_leader; struct cpu_hw_events *fake_cpuc; - int ret = -ENOSPC, n; + int ret = -EINVAL, n; fake_cpuc = allocate_fake_cpuc(); if (IS_ERR(fake_cpuc)) diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 492bf1358a7c..ef484d9d0a25 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -1268,7 +1268,7 @@ reserve: } done: - return num ? -ENOSPC : 0; + return num ? -EINVAL : 0; } static __initconst const struct x86_pmu p4_pmu = { -- cgit v1.2.3 From ed13ec58bfe0d5dc95f748e6118432cb0fa283cb Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 14 Nov 2011 10:03:25 +0100 Subject: perf/x86: Enable raw event access to Intel offcore events Now that the core offcore support is fixed up (thanks Stephane) and we have sane generic events utilizing them, re-enable the raw access to the feature as well. Note that it doesn't matter if you use event 0x1b7 or 0x1bb to specify an offcore event, either one works and neither guarantees you'll end up on a particular offcore MSR. Based on original patch from: Vince Weaver . Signed-off-by: Peter Zijlstra Cc: Vince Weaver . Cc: Stephane Eranian Link: http://lkml.kernel.org/r/alpine.DEB.2.00.1108031200390.703@cl320.eecs.utk.edu Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index ff0e8d498750..2bda212a0010 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -312,12 +312,8 @@ int x86_setup_perfctr(struct perf_event *event) return -EOPNOTSUPP; } - /* - * Do not allow config1 (extended registers) to propagate, - * there's no sane user-space generalization yet: - */ if (attr->type == PERF_TYPE_RAW) - return 0; + return x86_pmu_extra_regs(event->attr.config, event); if (attr->type == PERF_TYPE_HW_CACHE) return set_ext_hw_attr(hwc, event); -- cgit v1.2.3 From 8e8da023f5af71662867729db5547dc54786093c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 4 Dec 2011 11:57:09 -0800 Subject: x86: Fix boot failures on older AMD CPU's People with old AMD chips are getting hung boots, because commit bcb80e53877c ("x86, microcode, AMD: Add microcode revision to /proc/cpuinfo") moved the microcode detection too early into "early_init_amd()". At that point we are *so* early in the booth that the exception tables haven't even been set up yet, so the whole rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy); doesn't actually work: if the rdmsr does a GP fault (due to non-existant MSR register on older CPU's), we can't fix it up yet, and the boot fails. Fix it by simply moving the code to a slightly later point in the boot (init_amd() instead of early_init_amd()), since the kernel itself doesn't even really care about the microcode patchlevel at this point (or really ever: it's made available to user space in /proc/cpuinfo, and updated if you do a microcode load). Reported-tested-and-bisected-by: Larry Finger Tested-by: Bob Tracy Acked-by: Borislav Petkov Cc: Ingo Molnar Cc: Srivatsa S. Bhat Signed-off-by: Linus Torvalds --- arch/x86/kernel/cpu/amd.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index c7e46cb35327..0bab2b18bb20 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -442,8 +442,6 @@ static void __cpuinit bsp_init_amd(struct cpuinfo_x86 *c) static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) { - u32 dummy; - early_init_amd_mc(c); /* @@ -473,12 +471,12 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_EXTD_APICID); } #endif - - rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy); } static void __cpuinit init_amd(struct cpuinfo_x86 *c) { + u32 dummy; + #ifdef CONFIG_SMP unsigned long long value; @@ -657,6 +655,8 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) checking_wrmsrl(MSR_AMD64_MCx_MASK(4), mask); } } + + rdmsr_safe(MSR_AMD64_PATCH_LEVEL, &c->microcode, &dummy); } #ifdef CONFIG_X86_32 -- cgit v1.2.3 From 6a600a8b8749566a7d81ad75dcb8bf5342b5a39a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 15 Nov 2011 10:51:15 +0100 Subject: perf, x86: Disable PEBS on SandyBridge chips Cc: Stephane Eranian Cc: stable@kernel.org Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 2be5ebe99872..8d601b18bf9f 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -1545,6 +1545,13 @@ static void intel_clovertown_quirks(void) x86_pmu.pebs_constraints = NULL; } +static void intel_sandybridge_quirks(void) +{ + printk(KERN_WARNING "PEBS disabled due to CPU errata.\n"); + x86_pmu.pebs = 0; + x86_pmu.pebs_constraints = NULL; +} + __init int intel_pmu_init(void) { union cpuid10_edx edx; @@ -1694,6 +1701,7 @@ __init int intel_pmu_init(void) break; case 42: /* SandyBridge */ + x86_pmu.quirks = intel_sandybridge_quirks; case 45: /* SandyBridge, "Romely-EP" */ memcpy(hw_cache_event_ids, snb_hw_cache_event_ids, sizeof(hw_cache_event_ids)); -- cgit v1.2.3 From 16e5294e5f8303756a179cf218e37dfb9ed34417 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Tue, 8 Nov 2011 19:20:44 +0100 Subject: perf, x86: Force IBS LVT offset assignment for family 10h On AMD family 10h we see firmware bug messages like the following: [Firmware Bug]: cpu 6, try to use APIC500 (LVT offset 0) for vector 0x10400, but the register is already in use for vector 0xf9 on another cpu [Firmware Bug]: cpu 6, IBS interrupt offset 0 not available (MSRC001103A=0x0000000000000100) [Firmware Bug]: using offset 1 for IBS interrupts [Firmware Bug]: workaround enabled for IBS LVT offset perf: AMD IBS detected (0x00000007) We always see this, since the offsets are not assigned by the BIOS for this family. Force LVT offset assignment in this case. If the OS assignment fails, fallback to BIOS settings and try to setup this. The fallback to BIOS settings weakens the family check since force_ibs_eilvt_setup() may fail e.g. in case of virtual machines. But setup may still succeed if BIOS offsets are correct. Other families don't have a workaround implemented that assigns LVT offsets. It's ok, to drop calling force_ibs_eilvt_setup() for that families. With the patch the [Firmware Bug] messages vanish. We see now: IBS: LVT offset 1 assigned perf: AMD IBS detected (0x00000007) Signed-off-by: Robert Richter Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111109162225.GO12451@erda.amd.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_amd_ibs.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c index ab6343d21825..3b8a2d30d14e 100644 --- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c +++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c @@ -199,8 +199,7 @@ static int force_ibs_eilvt_setup(void) goto out; } - pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset); - pr_err(FW_BUG "workaround enabled for IBS LVT offset\n"); + pr_info("IBS: LVT offset %d assigned\n", offset); return 0; out: @@ -265,19 +264,23 @@ perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *h static __init int amd_ibs_init(void) { u32 caps; - int ret; + int ret = -EINVAL; caps = __get_ibs_caps(); if (!caps) return -ENODEV; /* ibs not supported by the cpu */ - if (!ibs_eilvt_valid()) { - ret = force_ibs_eilvt_setup(); - if (ret) { - pr_err("Failed to setup IBS, %d\n", ret); - return ret; - } - } + /* + * Force LVT offset assignment for family 10h: The offsets are + * not assigned by the BIOS for this family, so the OS is + * responsible for doing it. If the OS assignment fails, fall + * back to BIOS settings and try to setup this. + */ + if (boot_cpu_data.x86 == 0x10) + force_ibs_eilvt_setup(); + + if (!ibs_eilvt_valid()) + goto out; get_online_cpus(); ibs_caps = caps; @@ -287,7 +290,11 @@ static __init int amd_ibs_init(void) smp_call_function(setup_APIC_ibs, NULL, 1); put_online_cpus(); - return perf_event_ibs_init(); + ret = perf_event_ibs_init(); +out: + if (ret) + pr_err("Failed to setup IBS, %d\n", ret); + return ret; } /* Since we need the pci subsystem to init ibs we can't do this earlier: */ -- cgit v1.2.3 From 644ddf588f5dba34df483a6ea8abe639cc102289 Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Tue, 18 Oct 2011 13:24:10 -0400 Subject: Add TAINT_FIRMWARE_WORKAROUND on MTRR fixup TAINT_FIRMWARE_WORKAROUND should be set when an MTRR fixup is done. Signed-off-by: Prarit Bhargava Acked-by: David Rientjes Link: http://lkml.kernel.org/r/1318958650-12447-1-git-send-email-prarit@redhat.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mtrr/generic.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index a71efcdbb092..e1fe7f42e0fb 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -547,6 +547,7 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, if (tmp != mask_lo) { printk(KERN_WARNING "mtrr: your BIOS has configured an incorrect mask, fixing it.\n"); + add_taint(TAINT_FIRMWARE_WORKAROUND); mask_lo = tmp; } } -- cgit v1.2.3 From 8dbf4a30033ff61091015f0076e872b5c8f717cc Mon Sep 17 00:00:00 2001 From: Ajaykumar Hotchandani Date: Fri, 11 Nov 2011 18:31:57 +0530 Subject: x86/mtrr: Resolve inconsistency with Intel processor manual Following is from Notes of section 11.5.3 of Intel processor manual available at: http://www.intel.com/Assets/PDF/manual/325384.pdf For the Pentium 4 and Intel Xeon processors, after the sequence of steps given above has been executed, the cache lines containing the code between the end of the WBINVD instruction and before the MTRRS have actually been disabled may be retained in the cache hierarchy. Here, to remove code from the cache completely, a second WBINVD instruction must be executed after the MTRRs have been disabled. This patch provides resolution for that. Ideally, I will like to make changes only for Pentium 4 and Xeon processors. But, I am not finding easier way to do it. And, extra wbinvd() instruction does not hurt much for other processors. Signed-off-by: Ajaykumar Hotchandani Cc: Linus Torvalds Cc: Arjan van de Ven Cc: Lucas De Marchi Link: http://lkml.kernel.org/r/4EBD1CC5.3030008@oracle.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mtrr/generic.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index e1fe7f42e0fb..97b26356e9ee 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -694,6 +694,7 @@ static void prepare_set(void) __acquires(set_atomicity_lock) /* Disable MTRRs, and set the default type to uncached */ mtrr_wrmsr(MSR_MTRRdefType, deftype_lo & ~0xcff, deftype_hi); + wbinvd(); } static void post_set(void) __releases(set_atomicity_lock) -- cgit v1.2.3