From a956bd6f8583326b18348ab1452b4686778f785d Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 12 Apr 2012 16:48:01 +0200 Subject: x86, microcode: Fix sysfs warning during module unload on unsupported CPUs Loading the microcode driver on an unsupported CPU and subsequently unloading the driver causes WARNING: at fs/sysfs/group.c:138 mc_device_remove+0x5f/0x70 [microcode]() Hardware name: 01972NG sysfs group ffffffffa00013d0 not found for kobject 'cpu0' Modules linked in: snd_hda_codec_hdmi snd_hda_codec_conexant snd_hda_intel btusb snd_hda_codec bluetooth thinkpad_acpi rfkill microcode(-) [last unloaded: cfg80211] Pid: 4560, comm: modprobe Not tainted 3.4.0-rc2-00002-g258f742 #5 Call Trace: [] ? warn_slowpath_common+0x7b/0xc0 [] ? warn_slowpath_fmt+0x45/0x50 [] ? sysfs_remove_group+0x34/0x120 [] ? mc_device_remove+0x5f/0x70 [microcode] [] ? subsys_interface_unregister+0x69/0xa0 [] ? mutex_lock+0x16/0x40 [] ? microcode_exit+0x50/0x92 [microcode] [] ? sys_delete_module+0x16d/0x260 [] ? wait_iff_congested+0x45/0x110 [] ? page_fault+0x1f/0x30 [] ? system_call_fastpath+0x16/0x1b on recent kernels. This is due to commit 8a25a2fd126c ("cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem") which renders commit 6c53cbfced04 ("x86, microcode: Correct sysdev_add error path") useless. See http://marc.info/?l=linux-kernel&m=133416246406478 Avoid above warning by restoring the old driver behaviour before 6c53cbfced04 ("x86, microcode: Correct sysdev_add error path"). Cc: stable@vger.kernel.org Cc: Tigran Aivazian Signed-off-by: Andreas Herrmann Acked-by: Greg Kroah-Hartman Link: http://lkml.kernel.org/r/20120411163849.GE4794@alberich.amd.com Signed-off-by: Borislav Petkov --- arch/x86/kernel/microcode_core.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index 87a0f86..d389e74 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -419,10 +419,8 @@ static int mc_device_add(struct device *dev, struct subsys_interface *sif) if (err) return err; - if (microcode_init_cpu(cpu) == UCODE_ERROR) { - sysfs_remove_group(&dev->kobj, &mc_attr_group); + if (microcode_init_cpu(cpu) == UCODE_ERROR) return -EINVAL; - } return err; } -- cgit v1.1 From 283c1f2558ef4a4411fe908364b15b73b6ab44cf Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 12 Apr 2012 16:51:57 +0200 Subject: x86, microcode: Ensure that module is only loaded on supported AMD CPUs Exit early when there's no support for a particular CPU family. Also, fixup the "no support for this CPU vendor" to be issued only when the driver is attempted to be loaded on an unsupported vendor. Cc: stable@vger.kernel.org Cc: Tigran Aivazian Signed-off-by: Andreas Herrmann Acked-by: Greg Kroah-Hartman Link: http://lkml.kernel.org/r/20120411163849.GE4794@alberich.amd.com [Boris: add a commit msg because Andreas is lazy] Signed-off-by: Borislav Petkov --- arch/x86/kernel/microcode_amd.c | 12 +++++++----- arch/x86/kernel/microcode_core.c | 6 +++--- 2 files changed, 10 insertions(+), 8 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index 73465aa..8a2ce8f 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -82,11 +82,6 @@ static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig) { struct cpuinfo_x86 *c = &cpu_data(cpu); - if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { - pr_warning("CPU%d: family %d not supported\n", cpu, c->x86); - return -1; - } - csig->rev = c->microcode; pr_info("CPU%d: patch_level=0x%08x\n", cpu, csig->rev); @@ -380,6 +375,13 @@ static struct microcode_ops microcode_amd_ops = { struct microcode_ops * __init init_amd_microcode(void) { + struct cpuinfo_x86 *c = &cpu_data(0); + + if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) { + pr_warning("AMD CPU family 0x%x not supported\n", c->x86); + return NULL; + } + patch = (void *)get_zeroed_page(GFP_KERNEL); if (!patch) return NULL; diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c index d389e74..c9bda6d 100644 --- a/arch/x86/kernel/microcode_core.c +++ b/arch/x86/kernel/microcode_core.c @@ -526,11 +526,11 @@ static int __init microcode_init(void) microcode_ops = init_intel_microcode(); else if (c->x86_vendor == X86_VENDOR_AMD) microcode_ops = init_amd_microcode(); - - if (!microcode_ops) { + else pr_err("no support for this CPU vendor\n"); + + if (!microcode_ops) return -ENODEV; - } microcode_pdev = platform_device_register_simple("microcode", -1, NULL, 0); -- cgit v1.1 From d7de8649f34d45041409d1af4ba4a521971a9075 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Wed, 11 Apr 2012 17:12:38 +0200 Subject: x86/amd: Remove broken links from comment and kernel message Signed-off-by: Andreas Herrmann Link: http://lkml.kernel.org/r/20120411151238.GA4794@alberich.amd.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 0a44b90..1248f9c 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -26,7 +26,8 @@ * contact AMD for precise details and a CPU swap. * * See http://www.multimania.com/poulot/k6bug.html - * http://www.amd.com/K6/k6docs/revgd.html + * and section 2.6.2 of "AMD-K6 Processor Revision Guide - Model 6" + * (Publication # 21266 Issue Date: August 1998) * * The following test is erm.. interesting. AMD neglected to up * the chip setting when fixing the bug but they also tweaked some @@ -94,7 +95,6 @@ static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c) "system stability may be impaired when more than 32 MB are used.\n"); else printk(KERN_CONT "probably OK (after B9730xxxx).\n"); - printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n"); } /* K6 with old style WHCR */ -- cgit v1.1 From 68894632afb2729a1d8785c877840953894c7283 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Mon, 2 Apr 2012 18:06:48 +0200 Subject: x86/platform: Remove incorrect error message in x86_default_fixup_cpu_id() It's only called from amd.c:srat_detect_node(). The introduced condition for calling the fixup code is true for all AMD multi-node processors, e.g. Magny-Cours and Interlagos. There we have 2 NUMA nodes on one socket. Thus there are cores having different numa-node-id but with equal phys_proc_id. There is no point to print error messages in such a situation. The confusing/misleading error message was introduced with commit 64be4c1c2428e148de6081af235e2418e6a66dda ("x86: Add x86_init platform override to fix up NUMA core numbering"). Remove the default fixup function (especially the error message) and replace it by a NULL pointer check, move the Numascale-specific condition for calling the fixup into the fixup-function itself and slightly adapt the comment. Signed-off-by: Andreas Herrmann Acked-by: Borislav Petkov Cc: Cc: Cc: Cc: Link: http://lkml.kernel.org/r/20120402160648.GR27684@alberich.amd.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/apic_numachip.c | 7 +++++-- arch/x86/kernel/cpu/amd.c | 7 ++++--- arch/x86/kernel/cpu/common.c | 9 --------- arch/x86/kernel/x86_init.c | 1 - 4 files changed, 9 insertions(+), 15 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/apic/apic_numachip.c b/arch/x86/kernel/apic/apic_numachip.c index 899803e..23e7542 100644 --- a/arch/x86/kernel/apic/apic_numachip.c +++ b/arch/x86/kernel/apic/apic_numachip.c @@ -207,8 +207,11 @@ static void __init map_csrs(void) static void fixup_cpu_id(struct cpuinfo_x86 *c, int node) { - c->phys_proc_id = node; - per_cpu(cpu_llc_id, smp_processor_id()) = node; + + if (c->phys_proc_id != node) { + c->phys_proc_id = node; + per_cpu(cpu_llc_id, smp_processor_id()) = node; + } } static int __init numachip_system_init(void) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 1248f9c..1c67ca1 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -353,10 +353,11 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) node = per_cpu(cpu_llc_id, cpu); /* - * If core numbers are inconsistent, it's likely a multi-fabric platform, - * so invoke platform-specific handler + * On multi-fabric platform (e.g. Numascale NumaChip) a + * platform-specific handler needs to be called to fixup some + * IDs of the CPU. */ - if (c->phys_proc_id != node) + if (x86_cpuinit.fixup_cpu_id) x86_cpuinit.fixup_cpu_id(c, node); if (!node_online(node)) { diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 67e2583..cf79302 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1163,15 +1163,6 @@ static void dbg_restore_debug_regs(void) #endif /* ! CONFIG_KGDB */ /* - * Prints an error where the NUMA and configured core-number mismatch and the - * platform didn't override this to fix it up - */ -void __cpuinit x86_default_fixup_cpu_id(struct cpuinfo_x86 *c, int node) -{ - pr_err("NUMA core number %d differs from configured core number %d\n", node, c->phys_proc_id); -} - -/* * cpu_init() initializes state that is per-CPU. Some data is already * initialized (naturally) in the bootstrap process, such as the GDT * and IDT. We reload them nevertheless, this function acts as a diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index e9f265f..9cf71d0 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -93,7 +93,6 @@ struct x86_init_ops x86_init __initdata = { struct x86_cpuinit_ops x86_cpuinit __cpuinitdata = { .early_percpu_clock_init = x86_init_noop, .setup_percpu_clockev = setup_secondary_APIC_clock, - .fixup_cpu_id = x86_default_fixup_cpu_id, }; static void default_nmi_init(void) { }; -- cgit v1.1 From 089f9fba56faf33cc6dd2a6442b7ac92c58b8209 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 16 Apr 2012 22:48:15 +0200 Subject: i387: ptrace breaks the lazy-fpu-restore logic Starting from 7e16838d "i387: support lazy restore of FPU state" we assume that fpu_owner_task doesn't need restore_fpu_checking() on the context switch, its FPU state should match what we already have in the FPU on this CPU. However, debugger can change the tracee's FPU state, in this case we should reset fpu.last_cpu to ensure fpu_lazy_restore() can't return true. Change init_fpu() to do this, it is called by user_regset->set() methods. Reported-by: Jan Kratochvil Suggested-by: Linus Torvalds Signed-off-by: Oleg Nesterov Link: http://lkml.kernel.org/r/20120416204815.GB24884@redhat.com Cc: v3.3 Signed-off-by: H. Peter Anvin --- arch/x86/kernel/i387.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 7734bcb..2d6e649 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -235,6 +235,7 @@ int init_fpu(struct task_struct *tsk) if (tsk_used_math(tsk)) { if (HAVE_HWFP && tsk == current) unlazy_fpu(tsk); + tsk->thread.fpu.last_cpu = ~0; return 0; } -- cgit v1.1 From cbf2829b61c136edcba302a5e1b6b40e97d32c00 Mon Sep 17 00:00:00 2001 From: Bryan O'Donoghue Date: Wed, 18 Apr 2012 17:37:39 +0100 Subject: x86, apic: APIC code touches invalid MSR on P5 class machines Current APIC code assumes MSR_IA32_APICBASE is present for all systems. Pentium Classic P5 and friends didn't have this MSR. MSR_IA32_APICBASE was introduced as an architectural MSR by Intel @ P6. Code paths that can touch this MSR invalidly are when vendor == Intel && cpu-family == 5 and APIC bit is set in CPUID - or when you simply pass lapic on the kernel command line, on a P5. The below patch stops Linux incorrectly interfering with the MSR_IA32_APICBASE for P5 class machines. Other code paths exist that touch the MSR - however those paths are not currently reachable for a conformant P5. Signed-off-by: Bryan O'Donoghue Link: http://lkml.kernel.org/r/4F8EEDD3.1080404@linux.intel.com Signed-off-by: H. Peter Anvin Cc: --- arch/x86/kernel/apic/apic.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 11544d8..edc2448 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1637,9 +1637,11 @@ static int __init apic_verify(void) mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; /* The BIOS may have set up the APIC at some other address */ - rdmsr(MSR_IA32_APICBASE, l, h); - if (l & MSR_IA32_APICBASE_ENABLE) - mp_lapic_addr = l & MSR_IA32_APICBASE_BASE; + if (boot_cpu_data.x86 >= 6) { + rdmsr(MSR_IA32_APICBASE, l, h); + if (l & MSR_IA32_APICBASE_ENABLE) + mp_lapic_addr = l & MSR_IA32_APICBASE_BASE; + } pr_info("Found and enabled local APIC!\n"); return 0; @@ -1657,13 +1659,15 @@ int __init apic_force_enable(unsigned long addr) * MSR. This can only be done in software for Intel P6 or later * and AMD K7 (Model > 1) or later. */ - rdmsr(MSR_IA32_APICBASE, l, h); - if (!(l & MSR_IA32_APICBASE_ENABLE)) { - pr_info("Local APIC disabled by BIOS -- reenabling.\n"); - l &= ~MSR_IA32_APICBASE_BASE; - l |= MSR_IA32_APICBASE_ENABLE | addr; - wrmsr(MSR_IA32_APICBASE, l, h); - enabled_via_apicbase = 1; + if (boot_cpu_data.x86 >= 6) { + rdmsr(MSR_IA32_APICBASE, l, h); + if (!(l & MSR_IA32_APICBASE_ENABLE)) { + pr_info("Local APIC disabled by BIOS -- reenabling.\n"); + l &= ~MSR_IA32_APICBASE_BASE; + l |= MSR_IA32_APICBASE_ENABLE | addr; + wrmsr(MSR_IA32_APICBASE, l, h); + enabled_via_apicbase = 1; + } } return apic_verify(); } @@ -2209,10 +2213,12 @@ static void lapic_resume(void) * FIXME! This will be wrong if we ever support suspend on * SMP! We'll need to do this as part of the CPU restore! */ - rdmsr(MSR_IA32_APICBASE, l, h); - l &= ~MSR_IA32_APICBASE_BASE; - l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; - wrmsr(MSR_IA32_APICBASE, l, h); + if (boot_cpu_data.x86 >= 6) { + rdmsr(MSR_IA32_APICBASE, l, h); + l &= ~MSR_IA32_APICBASE_BASE; + l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; + wrmsr(MSR_IA32_APICBASE, l, h); + } } maxlvt = lapic_get_maxlvt(); -- cgit v1.1 From a720b2dd2470a52345df11dca8d6c1466599f812 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 19 Apr 2012 12:35:08 +0200 Subject: x86, intel_cacheinfo: Fix error return code in amd_set_l3_disable_slot() If the L3 disable slot is already in use, return -EEXIST instead of -EINVAL. The caller, store_cache_disable(), checks this return value to print an appropriate warning. Also, we want to signal with -EEXIST that the current index we're disabling has actually been already disabled on the node: $ echo 12 > /sys/devices/system/cpu/cpu3/cache/index3/cache_disable_0 $ echo 12 > /sys/devices/system/cpu/cpu3/cache/index3/cache_disable_0 -bash: echo: write error: File exists $ echo 12 > /sys/devices/system/cpu/cpu3/cache/index3/cache_disable_1 -bash: echo: write error: File exists $ echo 12 > /sys/devices/system/cpu/cpu5/cache/index3/cache_disable_1 -bash: echo: write error: File exists The old code would say -bash: echo: write error: Invalid argument for disable slot 1 when playing the example above with no output in dmesg, which is clearly misleading. Reported-by: Dan Carpenter Signed-off-by: Srivatsa S. Bhat Link: http://lkml.kernel.org/r/20120419070053.GB16645@elgon.mountain [Boris: add testing for the other index too] Signed-off-by: Borislav Petkov --- arch/x86/kernel/cpu/intel_cacheinfo.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 73d08ed..b8f3653 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -433,14 +433,14 @@ int amd_set_l3_disable_slot(struct amd_northbridge *nb, int cpu, unsigned slot, /* check if @slot is already used or the index is already disabled */ ret = amd_get_l3_disable_slot(nb, slot); if (ret >= 0) - return -EINVAL; + return -EEXIST; if (index > nb->l3_cache.indices) return -EINVAL; /* check whether the other slot has disabled the same index already */ if (index == amd_get_l3_disable_slot(nb, !slot)) - return -EINVAL; + return -EEXIST; amd_l3_disable_index(nb, cpu, slot, index); @@ -468,8 +468,8 @@ static ssize_t store_cache_disable(struct _cpuid4_info *this_leaf, err = amd_set_l3_disable_slot(this_leaf->base.nb, cpu, slot, val); if (err) { if (err == -EEXIST) - printk(KERN_WARNING "L3 disable slot %d in use!\n", - slot); + pr_warning("L3 slot %d in use/index already disabled!\n", + slot); return err; } return count; -- cgit v1.1 From 2a14e541ed87bca0c125b82961ca3c6f808607d2 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Sun, 22 Apr 2012 23:03:17 -0400 Subject: ACPI: Convert wake_sleep_flags to a value instead of function With commit a2ef5c4fd44ce3922435139393b89f2cce47f576 "ACPI: Move module parameter gts and bfs to sleep.c" the wake_sleep_flags is required when calling acpi_enter_sleep_state, which means that if there are functions outside the sleep.c code they can't get the wake_sleep_flags values. This converts the function in to a exported value and converts the module config operands to a function. Acked-by: Rafael J. Wysocki Acked-by: Lin Ming [v2: Parameters can be turned on/off dynamically] [v3: unsigned char -> u8] [v4: val -> kp->arg] Signed-off-by: Konrad Rzeszutek Wilk Link: http://lkml.kernel.org/r/1335150198-21899-2-git-send-email-konrad.wilk@oracle.com Signed-off-by: H. Peter Anvin --- arch/x86/kernel/acpi/sleep.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/acpi/sleep.h b/arch/x86/kernel/acpi/sleep.h index 416d4be..fe5fdda 100644 --- a/arch/x86/kernel/acpi/sleep.h +++ b/arch/x86/kernel/acpi/sleep.h @@ -9,6 +9,8 @@ extern long saved_magic; extern int wakeup_pmode_return; +extern u8 wake_sleep_flags; + extern unsigned long acpi_copy_wakeup_routine(unsigned long); extern void wakeup_long64(void); -- cgit v1.1 From cd74257b974d6d26442c97891c4d05772748b177 Mon Sep 17 00:00:00 2001 From: Konrad Rzeszutek Wilk Date: Sun, 22 Apr 2012 23:03:18 -0400 Subject: x86, acpi: Call acpi_enter_sleep_state via an asmlinkage C function from assembler With commit a2ef5c4fd44ce3922435139393b89f2cce47f576 "ACPI: Move module parameter gts and bfs to sleep.c" the wake_sleep_flags is required when calling acpi_enter_sleep_state. The assembler code in wakeup_*.S did not do that. One solution is to call it from assembler and stick the wake_sleep_flags on the stack (for 32-bit) or in %esi (for 64-bit). hpa and rafael both suggested however to create a wrapper function to call acpi_enter_sleep_state and call said wrapper function ("acpi_enter_s3") from assembler. For 32-bit, the acpi_enter_s3 ends up looking as so: push %ebp mov %esp,%ebp sub $0x8,%esp movzbl 0xc1809314,%eax [wake_sleep_flags] movl $0x3,(%esp) mov %eax,0x4(%esp) call 0xc12d1fa0 leave ret And 64-bit: movzbl 0x9afde1(%rip),%esi [wake_sleep_flags] push %rbp mov $0x3,%edi mov %rsp,%rbp callq 0xffffffff812e9800 leaveq retq Reviewed-by: H. Peter Anvin Suggested-by: H. Peter Anvin [v2: Remove extra assembler operations, per hpa review] Signed-off-by: Konrad Rzeszutek Wilk Link: http://lkml.kernel.org/r/1335150198-21899-3-git-send-email-konrad.wilk@oracle.com Signed-off-by: H. Peter Anvin --- arch/x86/kernel/acpi/sleep.c | 4 ++++ arch/x86/kernel/acpi/sleep.h | 2 ++ arch/x86/kernel/acpi/wakeup_32.S | 4 +--- arch/x86/kernel/acpi/wakeup_64.S | 4 +--- 4 files changed, 8 insertions(+), 6 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 103b6ab..146a49c 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -24,6 +24,10 @@ unsigned long acpi_realmode_flags; static char temp_stack[4096]; #endif +asmlinkage void acpi_enter_s3(void) +{ + acpi_enter_sleep_state(3, wake_sleep_flags); +} /** * acpi_suspend_lowlevel - save kernel state * diff --git a/arch/x86/kernel/acpi/sleep.h b/arch/x86/kernel/acpi/sleep.h index fe5fdda..d68677a 100644 --- a/arch/x86/kernel/acpi/sleep.h +++ b/arch/x86/kernel/acpi/sleep.h @@ -3,6 +3,7 @@ */ #include +#include extern unsigned long saved_video_mode; extern long saved_magic; @@ -10,6 +11,7 @@ extern long saved_magic; extern int wakeup_pmode_return; extern u8 wake_sleep_flags; +extern asmlinkage void acpi_enter_s3(void); extern unsigned long acpi_copy_wakeup_routine(unsigned long); extern void wakeup_long64(void); diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S index 13ab7205..7261083 100644 --- a/arch/x86/kernel/acpi/wakeup_32.S +++ b/arch/x86/kernel/acpi/wakeup_32.S @@ -74,9 +74,7 @@ restore_registers: ENTRY(do_suspend_lowlevel) call save_processor_state call save_registers - pushl $3 - call acpi_enter_sleep_state - addl $4, %esp + call acpi_enter_s3 # In case of S3 failure, we'll emerge here. Jump # to ret_point to recover diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S index 8ea5164..014d1d2 100644 --- a/arch/x86/kernel/acpi/wakeup_64.S +++ b/arch/x86/kernel/acpi/wakeup_64.S @@ -71,9 +71,7 @@ ENTRY(do_suspend_lowlevel) movq %rsi, saved_rsi addq $8, %rsp - movl $3, %edi - xorl %eax, %eax - call acpi_enter_sleep_state + call acpi_enter_s3 /* in case something went wrong, restore the machine status and go on */ jmp resume_point -- cgit v1.1 From ea0dcf903e7d76aa5d483d876215fedcfdfe140f Mon Sep 17 00:00:00 2001 From: Greg Pearson Date: Tue, 24 Apr 2012 18:23:56 -0600 Subject: x86/apic: Use x2apic physical mode based on FADT setting Provide systems that do not support x2apic cluster mode a mechanism to select x2apic physical mode using the FADT FORCE_APIC_PHYSICAL_DESTINATION_MODE bit. Changes from v1: (based on Suresh's comments) - removed #ifdef CONFIG_ACPI - removed #include Signed-off-by: Greg Pearson Acked-by: Suresh Siddha Link: http://lkml.kernel.org/r/1335313436-32020-1-git-send-email-greg.pearson@hp.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/x2apic_phys.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index 8a778db..991e315 100644 --- a/arch/x86/kernel/apic/x2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -24,6 +24,12 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { if (x2apic_phys) return x2apic_enabled(); + else if ((acpi_gbl_FADT.header.revision >= FADT2_REVISION_ID) && + (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL) && + x2apic_enabled()) { + printk(KERN_DEBUG "System requires x2apic physical mode\n"); + return 1; + } else return 0; } -- cgit v1.1 From f7f286a910221ae18b21c18d9d0f4cd88965829f Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 3 Apr 2012 12:13:07 +0200 Subject: x86/amd: Re-enable CPU topology extensions in case BIOS has disabled it BIOS will switch off the corresponding feature flag on family 15h models 10h-1fh non-desktop CPUs. The topology extension CPUID leafs are required to detect which cores belong to the same compute unit. (thread siblings mask is set accordingly and also correct information about L1i and L2 cache sharing depends on this). W/o this patch we wouldn't see which cores belong to the same compute unit and also cache sharing information for L1i and L2 would be incorrect on such systems. Signed-off-by: Andreas Herrmann Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 1c67ca1..146bb62 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -580,6 +580,24 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) } } + /* re-enable TopologyExtensions if switched off by BIOS */ + if ((c->x86 == 0x15) && + (c->x86_model >= 0x10) && (c->x86_model <= 0x1f) && + !cpu_has(c, X86_FEATURE_TOPOEXT)) { + u64 val; + + if (!rdmsrl_amd_safe(0xc0011005, &val)) { + val |= 1ULL << 54; + wrmsrl_amd_safe(0xc0011005, val); + rdmsrl(0xc0011005, val); + if (val & (1ULL << 54)) { + set_cpu_cap(c, X86_FEATURE_TOPOEXT); + printk(KERN_INFO FW_INFO "CPU: Re-enabling " + "disabled Topology Extensions Support\n"); + } + } + } + cpu_detect_cache_sizes(c); /* Multi core CPU? */ -- cgit v1.1 From 62c49cc976af84cb0ffcb5ec07ee88da1a94e222 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 2 May 2012 15:04:02 +0300 Subject: KVM: Do not take reference to mm during async #PF It turned to be totally unneeded. The reason the code was introduced is so that KVM can prefault swapped in page, but prefault can fail even if mm is pinned since page table can change anyway. KVM handles this situation correctly though and does not inject spurious page faults. Fixes: "INFO: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected" warning while running LTP inside a KVM guest using the recent -next kernel. Reported-by: Sasha Levin Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/kernel/kvm.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index b8ba6e4..e554e5a 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -79,7 +79,6 @@ struct kvm_task_sleep_node { u32 token; int cpu; bool halted; - struct mm_struct *mm; }; static struct kvm_task_sleep_head { @@ -126,9 +125,7 @@ void kvm_async_pf_task_wait(u32 token) n.token = token; n.cpu = smp_processor_id(); - n.mm = current->active_mm; n.halted = idle || preempt_count() > 1; - atomic_inc(&n.mm->mm_count); init_waitqueue_head(&n.wq); hlist_add_head(&n.link, &b->list); spin_unlock(&b->lock); @@ -161,9 +158,6 @@ EXPORT_SYMBOL_GPL(kvm_async_pf_task_wait); static void apf_task_wake_one(struct kvm_task_sleep_node *n) { hlist_del_init(&n->link); - if (!n->mm) - return; - mmdrop(n->mm); if (n->halted) smp_send_reschedule(n->cpu); else if (waitqueue_active(&n->wq)) @@ -207,7 +201,7 @@ again: * async PF was not yet handled. * Add dummy entry for the token. */ - n = kmalloc(sizeof(*n), GFP_ATOMIC); + n = kzalloc(sizeof(*n), GFP_ATOMIC); if (!n) { /* * Allocation failed! Busy wait while other cpu @@ -219,7 +213,6 @@ again: } n->token = token; n->cpu = smp_processor_id(); - n->mm = NULL; init_waitqueue_head(&n->wq); hlist_add_head(&n->link, &b->list); } else -- cgit v1.1 From febb72a6e4cc6c8cffcc1ea649a3fb364f1ea432 Mon Sep 17 00:00:00 2001 From: Larry Finger Date: Sun, 6 May 2012 19:40:03 -0500 Subject: IA32 emulation: Fix build problem for modular ia32 a.out support Commit ce7e5d2d19bc ("x86: fix broken TASK_SIZE for ia32_aout") breaks kernel builds when "CONFIG_IA32_AOUT=m" with ERROR: "set_personality_ia32" [arch/x86/ia32/ia32_aout.ko] undefined! make[1]: *** [__modpost] Error 1 The entry point needs to be exported. Signed-off-by: Larry Finger Acked-by: Al Viro Signed-off-by: Linus Torvalds --- arch/x86/kernel/process_64.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 733ca39..43d8b48 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -423,6 +423,7 @@ void set_personality_ia32(bool x32) current_thread_info()->status |= TS_COMPAT; } } +EXPORT_SYMBOL_GPL(set_personality_ia32); unsigned long get_wchan(struct task_struct *p) { -- cgit v1.1 From d5e28005a1d2e67833852f4c9ea8ec206ea3ff85 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 27 Apr 2012 10:54:35 -0700 Subject: percpu, x86: don't use PMD_SIZE as embedded atom_size on 32bit With the embed percpu first chunk allocator, x86 uses either PAGE_SIZE or PMD_SIZE for atom_size. PMD_SIZE is used when CPU supports PSE so that percpu areas are aligned to PMD mappings and possibly allow using PMD mappings in vmalloc areas in the future. Using larger atom_size doesn't waste actual memory; however, it does require larger vmalloc space allocation later on for !first chunks. With reasonably sized vmalloc area, PMD_SIZE shouldn't be a problem but x86_32 at this point is anything but reasonable in terms of address space and using larger atom_size reportedly leads to frequent percpu allocation failures on certain setups. As there is no reason to not use PMD_SIZE on x86_64 as vmalloc space is aplenty and most x86_64 configurations support PSE, fix the issue by always using PMD_SIZE on x86_64 and PAGE_SIZE on x86_32. v2: drop cpu_has_pse test and make x86_64 always use PMD_SIZE and x86_32 PAGE_SIZE as suggested by hpa. Signed-off-by: Tejun Heo Reported-by: Yanmin Zhang Reported-by: ShuoX Liu Acked-by: H. Peter Anvin LKML-Reference: <4F97BA98.6010001@intel.com> Cc: stable@vger.kernel.org --- arch/x86/kernel/setup_percpu.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 71f4727..5a98aa2 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -185,10 +185,22 @@ void __init setup_per_cpu_areas(void) #endif rc = -EINVAL; if (pcpu_chosen_fc != PCPU_FC_PAGE) { - const size_t atom_size = cpu_has_pse ? PMD_SIZE : PAGE_SIZE; const size_t dyn_size = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE - PERCPU_FIRST_CHUNK_RESERVE; + size_t atom_size; + /* + * On 64bit, use PMD_SIZE for atom_size so that embedded + * percpu areas are aligned to PMD. This, in the future, + * can also allow using PMD mappings in vmalloc area. Use + * PAGE_SIZE on 32bit as vmalloc space is highly contended + * and large vmalloc area allocs can easily fail. + */ +#ifdef CONFIG_X86_64 + atom_size = PMD_SIZE; +#else + atom_size = PAGE_SIZE; +#endif rc = pcpu_embed_first_chunk(PERCPU_FIRST_CHUNK_RESERVE, dyn_size, atom_size, pcpu_cpu_distance, -- cgit v1.1