From a2d771c036eb8c040683089ca04c36dfb93a0e60 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Fri, 29 Oct 2010 16:56:19 +0100 Subject: xen: correct size of level2_kernel_pgt sizeof(pmd_t *) is 4 bytes on 32-bit PAE leading to an allocation of only 2048 bytes. The correct size is sizeof(pmd_t) giving us a full page allocation. Signed-off-by: Ian Campbell Cc: Jeremy Fitzhardinge Signed-off-by: Jeremy Fitzhardinge --- arch/x86/xen/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index c237b810b03..21ed8d7f75a 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -2126,7 +2126,7 @@ __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, { pmd_t *kernel_pmd; - level2_kernel_pgt = extend_brk(sizeof(pmd_t *) * PTRS_PER_PMD, PAGE_SIZE); + level2_kernel_pgt = extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE); max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->pt_base) + xen_start_info->nr_pt_frames * PAGE_SIZE + -- cgit v1.2.3-18-g5258 From cf38d0ba7efdc476815768b2b999b27cfae69747 Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Mon, 1 Nov 2010 12:53:50 +0600 Subject: x86, mm: Fix section mismatch in tlb.c Mark tlb_cpuhp_notify as __cpuinit. It's basically a callback function, which is called from __cpuinit init_smp_flash(). So - it's safe. We were warned by the following warning: WARNING: arch/x86/mm/built-in.o(.text+0x356d): Section mismatch in reference from the function tlb_cpuhp_notify() to the function .cpuinit.text:calculate_tlb_offset() The function tlb_cpuhp_notify() references the function __cpuinit calculate_tlb_offset(). This is often because tlb_cpuhp_notify lacks a __cpuinit annotation or the annotation of calculate_tlb_offset is wrong. Signed-off-by: Rakib Mullick Cc: Borislav Petkov Cc: Shaohua Li LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/mm/tlb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 49358481c73..12cdbb17ad1 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -251,7 +251,7 @@ static void __cpuinit calculate_tlb_offset(void) } } -static int tlb_cpuhp_notify(struct notifier_block *n, +static int __cpuinit tlb_cpuhp_notify(struct notifier_block *n, unsigned long action, void *hcpu) { switch (action & 0xf) { -- cgit v1.2.3-18-g5258 From edde99ce05290e50ce0b3495d209e54e6349ab47 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 25 Oct 2010 03:21:24 +0200 Subject: KVM: Write protect memory after slot swap I have observed the following bug trigger: 1. userspace calls GET_DIRTY_LOG 2. kvm_mmu_slot_remove_write_access is called and makes a page ro 3. page fault happens and makes the page writeable fault is logged in the bitmap appropriately 4. kvm_vm_ioctl_get_dirty_log swaps slot pointers a lot of time passes 5. guest writes into the page 6. userspace calls GET_DIRTY_LOG At point (5), bitmap is clean and page is writeable, thus, guest modification of memory is not logged and GET_DIRTY_LOG returns an empty bitmap. The rule is that all pages are either dirty in the current bitmap, or write-protected, which is violated here. It seems that just moving kvm_mmu_slot_remove_write_access down to after the slot pointer swap should fix this bug. KVM-Stable-Tag. Signed-off-by: Michael S. Tsirkin Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 2288ad829b3..b0818f67206 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3169,10 +3169,6 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_memslots *slots, *old_slots; unsigned long *dirty_bitmap; - spin_lock(&kvm->mmu_lock); - kvm_mmu_slot_remove_write_access(kvm, log->slot); - spin_unlock(&kvm->mmu_lock); - r = -ENOMEM; dirty_bitmap = vmalloc(n); if (!dirty_bitmap) @@ -3194,6 +3190,10 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, dirty_bitmap = old_slots->memslots[log->slot].dirty_bitmap; kfree(old_slots); + spin_lock(&kvm->mmu_lock); + kvm_mmu_slot_remove_write_access(kvm, log->slot); + spin_unlock(&kvm->mmu_lock); + r = -EFAULT; if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) { vfree(dirty_bitmap); -- cgit v1.2.3-18-g5258 From eb45fda45f915c7ca3e81e005e853cb770da2642 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Mon, 25 Oct 2010 11:58:22 -0200 Subject: KVM: MMU: fix rmap_remove on non present sptes drop_spte should not attempt to rmap_remove a non present shadow pte. This fixes a BUG_ON seen on kvm-autotest. Signed-off-by: Marcelo Tosatti Reported-by: Lucas Meneghel Rodrigues Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 908ea5464a5..fb8b376bf28 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -720,7 +720,7 @@ static void rmap_remove(struct kvm *kvm, u64 *spte) } } -static void set_spte_track_bits(u64 *sptep, u64 new_spte) +static int set_spte_track_bits(u64 *sptep, u64 new_spte) { pfn_t pfn; u64 old_spte = *sptep; @@ -731,19 +731,20 @@ static void set_spte_track_bits(u64 *sptep, u64 new_spte) old_spte = __xchg_spte(sptep, new_spte); if (!is_rmap_spte(old_spte)) - return; + return 0; pfn = spte_to_pfn(old_spte); if (!shadow_accessed_mask || old_spte & shadow_accessed_mask) kvm_set_pfn_accessed(pfn); if (!shadow_dirty_mask || (old_spte & shadow_dirty_mask)) kvm_set_pfn_dirty(pfn); + return 1; } static void drop_spte(struct kvm *kvm, u64 *sptep, u64 new_spte) { - set_spte_track_bits(sptep, new_spte); - rmap_remove(kvm, sptep); + if (set_spte_track_bits(sptep, new_spte)) + rmap_remove(kvm, sptep); } static u64 *rmap_next(struct kvm *kvm, unsigned long *rmapp, u64 *spte) -- cgit v1.2.3-18-g5258 From 97e69aa62f8b5d338d6cff49be09e37cc1262838 Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Sat, 30 Oct 2010 22:54:47 +0400 Subject: KVM: x86: fix information leak to userland Structures kvm_vcpu_events, kvm_debugregs, kvm_pit_state2 and kvm_clock_data are copied to userland with some padding and reserved fields unitialized. It leads to leaking of contents of kernel stack memory. We have to initialize them to zero. In patch v1 Jan Kiszka suggested to fill reserved fields with zeros instead of memset'ting the whole struct. It makes sense as these fields are explicitly marked as padding. No more fields need zeroing. KVM-Stable-Tag. Signed-off-by: Vasiliy Kulikov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b0818f67206..463c65b8f93 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2560,6 +2560,7 @@ static void kvm_vcpu_ioctl_x86_get_vcpu_events(struct kvm_vcpu *vcpu, !kvm_exception_is_soft(vcpu->arch.exception.nr); events->exception.nr = vcpu->arch.exception.nr; events->exception.has_error_code = vcpu->arch.exception.has_error_code; + events->exception.pad = 0; events->exception.error_code = vcpu->arch.exception.error_code; events->interrupt.injected = @@ -2573,12 +2574,14 @@ static void kvm_vcpu_ioctl_x86_get_vcpu_events(struct kvm_vcpu *vcpu, events->nmi.injected = vcpu->arch.nmi_injected; events->nmi.pending = vcpu->arch.nmi_pending; events->nmi.masked = kvm_x86_ops->get_nmi_mask(vcpu); + events->nmi.pad = 0; events->sipi_vector = vcpu->arch.sipi_vector; events->flags = (KVM_VCPUEVENT_VALID_NMI_PENDING | KVM_VCPUEVENT_VALID_SIPI_VECTOR | KVM_VCPUEVENT_VALID_SHADOW); + memset(&events->reserved, 0, sizeof(events->reserved)); } static int kvm_vcpu_ioctl_x86_set_vcpu_events(struct kvm_vcpu *vcpu, @@ -2623,6 +2626,7 @@ static void kvm_vcpu_ioctl_x86_get_debugregs(struct kvm_vcpu *vcpu, dbgregs->dr6 = vcpu->arch.dr6; dbgregs->dr7 = vcpu->arch.dr7; dbgregs->flags = 0; + memset(&dbgregs->reserved, 0, sizeof(dbgregs->reserved)); } static int kvm_vcpu_ioctl_x86_set_debugregs(struct kvm_vcpu *vcpu, @@ -3106,6 +3110,7 @@ static int kvm_vm_ioctl_get_pit2(struct kvm *kvm, struct kvm_pit_state2 *ps) sizeof(ps->channels)); ps->flags = kvm->arch.vpit->pit_state.flags; mutex_unlock(&kvm->arch.vpit->pit_state.lock); + memset(&ps->reserved, 0, sizeof(ps->reserved)); return r; } @@ -3486,6 +3491,7 @@ long kvm_arch_vm_ioctl(struct file *filp, user_ns.clock = kvm->arch.kvmclock_offset + now_ns; local_irq_enable(); user_ns.flags = 0; + memset(&user_ns.pad, 0, sizeof(user_ns.pad)); r = -EFAULT; if (copy_to_user(argp, &user_ns, sizeof(user_ns))) -- cgit v1.2.3-18-g5258 From 453d9c57e27b4401bc3e98906bcac31ae8be0165 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 1 Nov 2010 14:01:13 +0100 Subject: KVM: x86: Issue smp_call_function_many with preemption disabled smp_call_function_many is specified to be called only with preemption disabled. Fulfill this requirement. Signed-off-by: Jan Kiszka Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/x86') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 463c65b8f93..cdac9e592aa 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3978,8 +3978,10 @@ int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) return X86EMUL_CONTINUE; if (kvm_x86_ops->has_wbinvd_exit()) { + preempt_disable(); smp_call_function_many(vcpu->arch.wbinvd_dirty_mask, wbinvd_ipi, NULL, 1); + preempt_enable(); cpumask_clear(vcpu->arch.wbinvd_dirty_mask); } wbinvd(); -- cgit v1.2.3-18-g5258 From 07cf2a64c2ad3408a0e12aa4cd6040b30c09381d Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Sat, 6 Nov 2010 10:06:49 +0100 Subject: xen: fix memory leak in Xen PCI MSI/MSI-X allocator. Stanse found that xen_setup_msi_irqs leaks memory when xen_allocate_pirq fails. Free the memory in that fail path. Signed-off-by: Jiri Slaby Signed-off-by: Konrad Rzeszutek Wilk Cc: xen-devel@lists.xensource.com Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: x86@kernel.org --- arch/x86/pci/xen.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index 117f5b8daf7..d7b5109f7a9 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -147,8 +147,10 @@ static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) irq = xen_allocate_pirq(v[i], 0, /* not sharable */ (type == PCI_CAP_ID_MSIX) ? "pcifront-msi-x" : "pcifront-msi"); - if (irq < 0) - return -1; + if (irq < 0) { + ret = -1; + goto free; + } ret = set_irq_msi(irq, msidesc); if (ret) @@ -164,7 +166,7 @@ error: if (ret == -ENODEV) dev_err(&dev->dev, "Xen PCI frontend has not registered" \ " MSI/MSI-X support!\n"); - +free: kfree(v); return ret; } -- cgit v1.2.3-18-g5258 From 0059b2436a86fedb2747f654f8e10a67e97d8614 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 8 Nov 2010 22:20:29 +0100 Subject: x86: Address gcc4.6 "set but not used" warnings in apic.h native_apic_msr_read() and x2apic_enabled() use rdmsr(msr, low, high), but only use the low part. gcc4.6 complains about this: .../apic.h:144:11: warning: variable 'high' set but not used [-Wunused-but-set-variable] rdmsr() is just a wrapper around rdmsrl() which splits the 64bit value into low and high, so using rdmsrl() directly solves this. [tglx: Changed the variables to u64 as suggested by Cyrill. It's less confusing and has no code impact as this is 64bit only anyway. Massaged changelog as well. ] Signed-off-by: Andi Kleen Cc: x86@kernel.org Cc: Cyrill Gorcunov LKML-Reference: <1289251229-19589-1-git-send-email-andi@firstfloor.org> Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/apic.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index 286de34b0ed..f6ce0bda3b9 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -141,13 +141,13 @@ static inline void native_apic_msr_write(u32 reg, u32 v) static inline u32 native_apic_msr_read(u32 reg) { - u32 low, high; + u64 msr; if (reg == APIC_DFR) return -1; - rdmsr(APIC_BASE_MSR + (reg >> 4), low, high); - return low; + rdmsrl(APIC_BASE_MSR + (reg >> 4), msr); + return (u32)msr; } static inline void native_x2apic_wait_icr_idle(void) @@ -181,12 +181,12 @@ extern void enable_x2apic(void); extern void x2apic_icr_write(u32 low, u32 id); static inline int x2apic_enabled(void) { - int msr, msr2; + u64 msr; if (!cpu_has_x2apic) return 0; - rdmsr(MSR_IA32_APICBASE, msr, msr2); + rdmsrl(MSR_IA32_APICBASE, msr); if (msr & X2APIC_ENABLE) return 1; return 0; -- cgit v1.2.3-18-g5258 From 8e5e9521c13ff8cf6727999999c8d88cc64b5ff7 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Tue, 9 Nov 2010 00:08:11 +0100 Subject: x86: Remove unnecessary casts of void ptr returning alloc function return values The [vk][cmz]alloc(_node) family of functions return void pointers which it's completely unnecessary/pointless to cast to other pointer types since that happens implicitly. This patch removes such casts from arch/x86. Signed-off-by: Jesper Juhl Cc: trivial@kernel.org Cc: amd64-microcode@amd64.org Cc: Andreas Herrmann LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/microcode_amd.c | 2 +- arch/x86/platform/uv/tlb_uv.c | 13 ++++++------- 2 files changed, 7 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c index e1af7c055c7..ce0cb4721c9 100644 --- a/arch/x86/kernel/microcode_amd.c +++ b/arch/x86/kernel/microcode_amd.c @@ -212,7 +212,7 @@ static int install_equiv_cpu_table(const u8 *buf) return 0; } - equiv_cpu_table = (struct equiv_cpu_entry *) vmalloc(size); + equiv_cpu_table = vmalloc(size); if (!equiv_cpu_table) { pr_err("failed to allocate equivalent CPU table\n"); return 0; diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c index 20ea20a39e2..a318194002b 100644 --- a/arch/x86/platform/uv/tlb_uv.c +++ b/arch/x86/platform/uv/tlb_uv.c @@ -1343,8 +1343,8 @@ uv_activation_descriptor_init(int node, int pnode) * each bau_desc is 64 bytes; there are 8 (UV_ITEMS_PER_DESCRIPTOR) * per cpu; and up to 32 (UV_ADP_SIZE) cpu's per uvhub */ - bau_desc = (struct bau_desc *)kmalloc_node(sizeof(struct bau_desc)* - UV_ADP_SIZE*UV_ITEMS_PER_DESCRIPTOR, GFP_KERNEL, node); + bau_desc = kmalloc_node(sizeof(struct bau_desc) * UV_ADP_SIZE + * UV_ITEMS_PER_DESCRIPTOR, GFP_KERNEL, node); BUG_ON(!bau_desc); pa = uv_gpa(bau_desc); /* need the real nasid*/ @@ -1402,9 +1402,9 @@ uv_payload_queue_init(int node, int pnode) struct bau_payload_queue_entry *pqp_malloc; struct bau_control *bcp; - pqp = (struct bau_payload_queue_entry *) kmalloc_node( - (DEST_Q_SIZE + 1) * sizeof(struct bau_payload_queue_entry), - GFP_KERNEL, node); + pqp = kmalloc_node((DEST_Q_SIZE + 1) + * sizeof(struct bau_payload_queue_entry), + GFP_KERNEL, node); BUG_ON(!pqp); pqp_malloc = pqp; @@ -1520,8 +1520,7 @@ static void __init uv_init_per_cpu(int nuvhubs) timeout_us = calculate_destination_timeout(); - uvhub_descs = (struct uvhub_desc *) - kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); + uvhub_descs = kmalloc(nuvhubs * sizeof(struct uvhub_desc), GFP_KERNEL); memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc)); uvhub_mask = kzalloc((nuvhubs+7)/8, GFP_KERNEL); for_each_present_cpu(cpu) { -- cgit v1.2.3-18-g5258 From 62b0cfc240b1d4601333912ef8760e0ca9ec2cec Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Sat, 6 Nov 2010 15:41:04 -0500 Subject: x86, UV: Update node controller MMRs A new version of the SGI UV hub node controller is being developed. A few of the MMRs (control registers) that exist on the current hub no longer exist on the new hub. Fortunately, there are alternate MMRs that are are functionally equivalent and that exist on both hubs. This patch changes the UV code to use MMRs that exist in BOTH versions of the hub node controller. Signed-off-by: Jack Steiner LKML-Reference: <20101106204056.GA27584@sgi.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/uv/uv_mmrs.h | 189 +++++++++++++++++++------------------ arch/x86/kernel/apic/x2apic_uv_x.c | 12 +-- 2 files changed, 102 insertions(+), 99 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h index b2f2d2e05ce..6d90adf4428 100644 --- a/arch/x86/include/asm/uv/uv_mmrs.h +++ b/arch/x86/include/asm/uv/uv_mmrs.h @@ -805,6 +805,78 @@ union uvh_node_present_table_u { } s; }; +/* ========================================================================= */ +/* UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR */ +/* ========================================================================= */ +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR 0x16000c8UL + +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_BASE_SHFT 24 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_BASE_MASK 0x00000000ff000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_M_ALIAS_SHFT 48 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_M_ALIAS_MASK 0x001f000000000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_ENABLE_SHFT 63 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR_ENABLE_MASK 0x8000000000000000UL + +union uvh_rh_gam_alias210_overlay_config_0_mmr_u { + unsigned long v; + struct uvh_rh_gam_alias210_overlay_config_0_mmr_s { + unsigned long rsvd_0_23: 24; /* */ + unsigned long base : 8; /* RW */ + unsigned long rsvd_32_47: 16; /* */ + unsigned long m_alias : 5; /* RW */ + unsigned long rsvd_53_62: 10; /* */ + unsigned long enable : 1; /* RW */ + } s; +}; + +/* ========================================================================= */ +/* UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR */ +/* ========================================================================= */ +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR 0x16000d8UL + +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_BASE_SHFT 24 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_BASE_MASK 0x00000000ff000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_M_ALIAS_SHFT 48 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_M_ALIAS_MASK 0x001f000000000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_ENABLE_SHFT 63 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR_ENABLE_MASK 0x8000000000000000UL + +union uvh_rh_gam_alias210_overlay_config_1_mmr_u { + unsigned long v; + struct uvh_rh_gam_alias210_overlay_config_1_mmr_s { + unsigned long rsvd_0_23: 24; /* */ + unsigned long base : 8; /* RW */ + unsigned long rsvd_32_47: 16; /* */ + unsigned long m_alias : 5; /* RW */ + unsigned long rsvd_53_62: 10; /* */ + unsigned long enable : 1; /* RW */ + } s; +}; + +/* ========================================================================= */ +/* UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR */ +/* ========================================================================= */ +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR 0x16000e8UL + +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_BASE_SHFT 24 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_BASE_MASK 0x00000000ff000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_M_ALIAS_SHFT 48 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_M_ALIAS_MASK 0x001f000000000000UL +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_ENABLE_SHFT 63 +#define UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR_ENABLE_MASK 0x8000000000000000UL + +union uvh_rh_gam_alias210_overlay_config_2_mmr_u { + unsigned long v; + struct uvh_rh_gam_alias210_overlay_config_2_mmr_s { + unsigned long rsvd_0_23: 24; /* */ + unsigned long base : 8; /* RW */ + unsigned long rsvd_32_47: 16; /* */ + unsigned long m_alias : 5; /* RW */ + unsigned long rsvd_53_62: 10; /* */ + unsigned long enable : 1; /* RW */ + } s; +}; + /* ========================================================================= */ /* UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR */ /* ========================================================================= */ @@ -856,6 +928,29 @@ union uvh_rh_gam_alias210_redirect_config_2_mmr_u { } s; }; +/* ========================================================================= */ +/* UVH_RH_GAM_CONFIG_MMR */ +/* ========================================================================= */ +#define UVH_RH_GAM_CONFIG_MMR 0x1600000UL + +#define UVH_RH_GAM_CONFIG_MMR_M_SKT_SHFT 0 +#define UVH_RH_GAM_CONFIG_MMR_M_SKT_MASK 0x000000000000003fUL +#define UVH_RH_GAM_CONFIG_MMR_N_SKT_SHFT 6 +#define UVH_RH_GAM_CONFIG_MMR_N_SKT_MASK 0x00000000000003c0UL +#define UVH_RH_GAM_CONFIG_MMR_MMIOL_CFG_SHFT 12 +#define UVH_RH_GAM_CONFIG_MMR_MMIOL_CFG_MASK 0x0000000000001000UL + +union uvh_rh_gam_config_mmr_u { + unsigned long v; + struct uvh_rh_gam_config_mmr_s { + unsigned long m_skt : 6; /* RW */ + unsigned long n_skt : 4; /* RW */ + unsigned long rsvd_10_11: 2; /* */ + unsigned long mmiol_cfg : 1; /* RW */ + unsigned long rsvd_13_63: 51; /* */ + } s; +}; + /* ========================================================================= */ /* UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR */ /* ========================================================================= */ @@ -987,97 +1082,5 @@ union uvh_rtc1_int_config_u { } s; }; -/* ========================================================================= */ -/* UVH_SI_ADDR_MAP_CONFIG */ -/* ========================================================================= */ -#define UVH_SI_ADDR_MAP_CONFIG 0xc80000UL - -#define UVH_SI_ADDR_MAP_CONFIG_M_SKT_SHFT 0 -#define UVH_SI_ADDR_MAP_CONFIG_M_SKT_MASK 0x000000000000003fUL -#define UVH_SI_ADDR_MAP_CONFIG_N_SKT_SHFT 8 -#define UVH_SI_ADDR_MAP_CONFIG_N_SKT_MASK 0x0000000000000f00UL - -union uvh_si_addr_map_config_u { - unsigned long v; - struct uvh_si_addr_map_config_s { - unsigned long m_skt : 6; /* RW */ - unsigned long rsvd_6_7: 2; /* */ - unsigned long n_skt : 4; /* RW */ - unsigned long rsvd_12_63: 52; /* */ - } s; -}; - -/* ========================================================================= */ -/* UVH_SI_ALIAS0_OVERLAY_CONFIG */ -/* ========================================================================= */ -#define UVH_SI_ALIAS0_OVERLAY_CONFIG 0xc80008UL - -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_BASE_SHFT 24 -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_BASE_MASK 0x00000000ff000000UL -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_M_ALIAS_SHFT 48 -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_M_ALIAS_MASK 0x001f000000000000UL -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_ENABLE_SHFT 63 -#define UVH_SI_ALIAS0_OVERLAY_CONFIG_ENABLE_MASK 0x8000000000000000UL - -union uvh_si_alias0_overlay_config_u { - unsigned long v; - struct uvh_si_alias0_overlay_config_s { - unsigned long rsvd_0_23: 24; /* */ - unsigned long base : 8; /* RW */ - unsigned long rsvd_32_47: 16; /* */ - unsigned long m_alias : 5; /* RW */ - unsigned long rsvd_53_62: 10; /* */ - unsigned long enable : 1; /* RW */ - } s; -}; - -/* ========================================================================= */ -/* UVH_SI_ALIAS1_OVERLAY_CONFIG */ -/* ========================================================================= */ -#define UVH_SI_ALIAS1_OVERLAY_CONFIG 0xc80010UL - -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_BASE_SHFT 24 -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_BASE_MASK 0x00000000ff000000UL -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_M_ALIAS_SHFT 48 -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_M_ALIAS_MASK 0x001f000000000000UL -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_ENABLE_SHFT 63 -#define UVH_SI_ALIAS1_OVERLAY_CONFIG_ENABLE_MASK 0x8000000000000000UL - -union uvh_si_alias1_overlay_config_u { - unsigned long v; - struct uvh_si_alias1_overlay_config_s { - unsigned long rsvd_0_23: 24; /* */ - unsigned long base : 8; /* RW */ - unsigned long rsvd_32_47: 16; /* */ - unsigned long m_alias : 5; /* RW */ - unsigned long rsvd_53_62: 10; /* */ - unsigned long enable : 1; /* RW */ - } s; -}; - -/* ========================================================================= */ -/* UVH_SI_ALIAS2_OVERLAY_CONFIG */ -/* ========================================================================= */ -#define UVH_SI_ALIAS2_OVERLAY_CONFIG 0xc80018UL - -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_BASE_SHFT 24 -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_BASE_MASK 0x00000000ff000000UL -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_M_ALIAS_SHFT 48 -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_M_ALIAS_MASK 0x001f000000000000UL -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_ENABLE_SHFT 63 -#define UVH_SI_ALIAS2_OVERLAY_CONFIG_ENABLE_MASK 0x8000000000000000UL - -union uvh_si_alias2_overlay_config_u { - unsigned long v; - struct uvh_si_alias2_overlay_config_s { - unsigned long rsvd_0_23: 24; /* */ - unsigned long base : 8; /* RW */ - unsigned long rsvd_32_47: 16; /* */ - unsigned long m_alias : 5; /* RW */ - unsigned long rsvd_53_62: 10; /* */ - unsigned long enable : 1; /* RW */ - } s; -}; - -#endif /* _ASM_X86_UV_UV_MMRS_H */ +#endif /* __ASM_UV_MMRS_X86_H__ */ diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index ed4118de249..194539aea17 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -379,14 +379,14 @@ struct redir_addr { #define DEST_SHIFT UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR_DEST_BASE_SHFT static __initdata struct redir_addr redir_addrs[] = { - {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR, UVH_SI_ALIAS0_OVERLAY_CONFIG}, - {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_1_MMR, UVH_SI_ALIAS1_OVERLAY_CONFIG}, - {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_2_MMR, UVH_SI_ALIAS2_OVERLAY_CONFIG}, + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR, UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_0_MMR}, + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_1_MMR, UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_1_MMR}, + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_2_MMR, UVH_RH_GAM_ALIAS210_OVERLAY_CONFIG_2_MMR}, }; static __init void get_lowmem_redirect(unsigned long *base, unsigned long *size) { - union uvh_si_alias0_overlay_config_u alias; + union uvh_rh_gam_alias210_overlay_config_2_mmr_u alias; union uvh_rh_gam_alias210_redirect_config_2_mmr_u redirect; int i; @@ -660,7 +660,7 @@ void uv_nmi_init(void) void __init uv_system_init(void) { - union uvh_si_addr_map_config_u m_n_config; + union uvh_rh_gam_config_mmr_u m_n_config; union uvh_node_id_u node_id; unsigned long gnode_upper, lowmem_redir_base, lowmem_redir_size; int bytes, nid, cpu, lcpu, pnode, blade, i, j, m_val, n_val; @@ -670,7 +670,7 @@ void __init uv_system_init(void) map_low_mmrs(); - m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG); + m_n_config.v = uv_read_local_mmr(UVH_RH_GAM_CONFIG_MMR ); m_val = m_n_config.s.m_skt; n_val = m_n_config.s.n_skt; mmr_base = -- cgit v1.2.3-18-g5258 From 2f62bf7d238f6dfa39faf24c746d0b8dd60f85c5 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Thu, 4 Nov 2010 15:23:58 +0000 Subject: x86: Adjust section annotations in AMD Fam10 MMCONF enabling code check_enable_amd_mmconf_dmi() gets called only for the BSP, hence everything hanging off of it can be __init*. Signed-off-by: Jan Beulich Acked-by: Yinghai Lu LKML-Reference: <4CD2DE1E0200007800020990@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/kernel/mmconf-fam10h_64.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c index 71825806cd4..6da143c2a6b 100644 --- a/arch/x86/kernel/mmconf-fam10h_64.c +++ b/arch/x86/kernel/mmconf-fam10h_64.c @@ -217,13 +217,13 @@ void __cpuinit fam10h_check_enable_mmcfg(void) wrmsrl(address, val); } -static int __devinit set_check_enable_amd_mmconf(const struct dmi_system_id *d) +static int __init set_check_enable_amd_mmconf(const struct dmi_system_id *d) { pci_probe |= PCI_CHECK_ENABLE_AMD_MMCONF; return 0; } -static const struct dmi_system_id __cpuinitconst mmconf_dmi_table[] = { +static const struct dmi_system_id __initconst mmconf_dmi_table[] = { { .callback = set_check_enable_amd_mmconf, .ident = "Sun Microsystems Machine", @@ -234,7 +234,8 @@ static const struct dmi_system_id __cpuinitconst mmconf_dmi_table[] = { {} }; -void __cpuinit check_enable_amd_mmconf_dmi(void) +/* Called from a __cpuinit function, but only on the BSP. */ +void __ref check_enable_amd_mmconf_dmi(void) { dmi_check_system(mmconf_dmi_table); } -- cgit v1.2.3-18-g5258 From 2a8dcbd6cd2270f912ca141547d9296ce08abe4a Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 7 Nov 2010 22:57:18 +0100 Subject: x86, apic: Remove double #include Remove the second inclusion. Signed-off-by: Jesper Juhl LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/apic.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 850657d1b0e..3f838d53739 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -52,7 +52,6 @@ #include #include #include -#include unsigned int num_processors; -- cgit v1.2.3-18-g5258 From 1f523bf36734375dd6e986c9f47f010d00a8caca Mon Sep 17 00:00:00 2001 From: Kusanagi Kouichi Date: Fri, 5 Nov 2010 20:04:42 +0900 Subject: x86, pvclock: Remove leftover scale_delta() function Commit 92580d64e16402762e2acc3022f065397c780425 ("x86: pvclock: Move scale_delta into common header") forgot to remove scale_delta. Signed-off-by: Kusanagi Kouichi Cc: Zachary Amsden Cc: Marcelo Tosatti Cc: Glauber Costa LKML-Reference: <20101105110444.BAF6D6FC03B@msa105.auone-net.jp> Signed-off-by: Ingo Molnar --- arch/x86/kernel/pvclock.c | 38 -------------------------------------- 1 file changed, 38 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index bab3b9e6f66..008b91eefa1 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -41,44 +41,6 @@ void pvclock_set_flags(u8 flags) valid_flags = flags; } -/* - * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction, - * yielding a 64-bit result. - */ -static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift) -{ - u64 product; -#ifdef __i386__ - u32 tmp1, tmp2; -#endif - - if (shift < 0) - delta >>= -shift; - else - delta <<= shift; - -#ifdef __i386__ - __asm__ ( - "mul %5 ; " - "mov %4,%%eax ; " - "mov %%edx,%4 ; " - "mul %5 ; " - "xor %5,%5 ; " - "add %4,%%eax ; " - "adc %5,%%edx ; " - : "=A" (product), "=r" (tmp1), "=r" (tmp2) - : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) ); -#elif defined(__x86_64__) - __asm__ ( - "mul %%rdx ; shrd $32,%%rdx,%%rax" - : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) ); -#else -#error implement me! -#endif - - return product; -} - static u64 pvclock_get_nsec_offset(struct pvclock_shadow_time *shadow) { u64 delta = native_read_tsc() - shadow->tsc_timestamp; -- cgit v1.2.3-18-g5258 From 034c6efa4616e5ff6253549e973e7fef12899324 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 1 Nov 2010 18:52:05 +0100 Subject: perf, amd: Use kmalloc_node(,__GFP_ZERO) for northbridge structure allocation Jasper suggested we use the zeroing capability of the allocators instead of calling memset ourselves. Add node affinity while we're at it. Reported-by: Jesper Juhl Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_amd.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index 46d58448c3a..e421b8cd694 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -280,11 +280,11 @@ static struct amd_nb *amd_alloc_nb(int cpu, int nb_id) struct amd_nb *nb; int i; - nb = kmalloc(sizeof(struct amd_nb), GFP_KERNEL); + nb = kmalloc_node(sizeof(struct amd_nb), GFP_KERNEL | __GFP_ZERO, + cpu_to_node(cpu)); if (!nb) return NULL; - memset(nb, 0, sizeof(*nb)); nb->nb_id = nb_id; /* -- cgit v1.2.3-18-g5258 From 9ec23a7f6d2537faf14368e066e307c06812c4ca Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Thu, 28 Oct 2010 11:32:29 -0700 Subject: xen: do not release any memory under 1M in domain 0 We already deliberately setup a 1-1 P2M for the region up to 1M in order to allow code which assumes this region is already mapped to work without having to convert everything to ioremap. Domain 0 should not return any apparently unused memory regions (reserved or otherwise) in this region to Xen since the e820 may not accurately reflect what the BIOS has stashed in this region. Signed-off-by: Ian Campbell Signed-off-by: Jeremy Fitzhardinge --- arch/x86/xen/setup.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index b1dbdaa23ec..769c4b01fa3 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -118,16 +118,18 @@ static unsigned long __init xen_return_unused_memory(unsigned long max_pfn, const struct e820map *e820) { phys_addr_t max_addr = PFN_PHYS(max_pfn); - phys_addr_t last_end = 0; + phys_addr_t last_end = ISA_END_ADDRESS; unsigned long released = 0; int i; + /* Free any unused memory above the low 1Mbyte. */ for (i = 0; i < e820->nr_map && last_end < max_addr; i++) { phys_addr_t end = e820->map[i].addr; end = min(max_addr, end); - released += xen_release_chunk(last_end, end); - last_end = e820->map[i].addr + e820->map[i].size; + if (last_end < end) + released += xen_release_chunk(last_end, end); + last_end = max(last_end, e820->map[i].addr + e820->map[i].size); } if (last_end < max_addr) @@ -164,6 +166,7 @@ char * __init xen_memory_setup(void) XENMEM_memory_map; rc = HYPERVISOR_memory_op(op, &memmap); if (rc == -ENOSYS) { + BUG_ON(xen_initial_domain()); memmap.nr_entries = 1; map[0].addr = 0ULL; map[0].size = mem_end; @@ -201,12 +204,13 @@ char * __init xen_memory_setup(void) } /* - * Even though this is normal, usable memory under Xen, reserve - * ISA memory anyway because too many things think they can poke + * In domU, the ISA region is normal, usable memory, but we + * reserve ISA memory anyway because too many things poke * about in there. * - * In a dom0 kernel, this region is identity mapped with the - * hardware ISA area, so it really is out of bounds. + * In Dom0, the host E820 information can leave gaps in the + * ISA range, which would cause us to release those pages. To + * avoid this, we unconditionally reserve them here. */ e820_add_region(ISA_START_ADDRESS, ISA_END_ADDRESS - ISA_START_ADDRESS, E820_RESERVED); -- cgit v1.2.3-18-g5258 From 4723d0f2f96e6c910f951d595067eb31e0dd2d01 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Wed, 22 Sep 2010 11:09:19 -0600 Subject: x86/PCI: coalesce overlapping host bridge windows Some BIOSes provide PCI host bridge windows that overlap, e.g., pci_root PNP0A03:00: host bridge window [mem 0xb0000000-0xffffffff] pci_root PNP0A03:00: host bridge window [mem 0xafffffff-0xdfffffff] pci_root PNP0A03:00: host bridge window [mem 0xf0000000-0xffffffff] If we simply insert these as children of iomem_resource, the second window fails because it conflicts with the first, and the third is inserted as a child of the first, i.e., b0000000-ffffffff PCI Bus 0000:00 f0000000-ffffffff PCI Bus 0000:00 When we claim PCI device resources, this can cause collisions like this if we put them in the first window: pci 0000:00:01.0: address space collision: [mem 0xff300000-0xff4fffff] conflicts with PCI Bus 0000:00 [mem 0xf0000000-0xffffffff] Host bridge windows are top-level resources by definition, so it doesn't make sense to make the third window a child of the first. This patch coalesces any host bridge windows that overlap. For the example above, the result is this single window: pci_root PNP0A03:00: host bridge window [mem 0xafffffff-0xffffffff] This fixes a 2.6.34 regression. Reference: https://bugzilla.kernel.org/show_bug.cgi?id=17011 Reported-and-tested-by: Anisse Astier Reported-and-tested-by: Pramod Dematagoda Signed-off-by: Bjorn Helgaas Signed-off-by: Jesse Barnes --- arch/x86/pci/acpi.c | 103 ++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 83 insertions(+), 20 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index 15466c096ba..0972315c386 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -138,7 +138,6 @@ setup_resource(struct acpi_resource *acpi_res, void *data) struct acpi_resource_address64 addr; acpi_status status; unsigned long flags; - struct resource *root, *conflict; u64 start, end; status = resource_to_addr(acpi_res, &addr); @@ -146,12 +145,10 @@ setup_resource(struct acpi_resource *acpi_res, void *data) return AE_OK; if (addr.resource_type == ACPI_MEMORY_RANGE) { - root = &iomem_resource; flags = IORESOURCE_MEM; if (addr.info.mem.caching == ACPI_PREFETCHABLE_MEMORY) flags |= IORESOURCE_PREFETCH; } else if (addr.resource_type == ACPI_IO_RANGE) { - root = &ioport_resource; flags = IORESOURCE_IO; } else return AE_OK; @@ -172,25 +169,90 @@ setup_resource(struct acpi_resource *acpi_res, void *data) return AE_OK; } - conflict = insert_resource_conflict(root, res); - if (conflict) { - dev_err(&info->bridge->dev, - "address space collision: host bridge window %pR " - "conflicts with %s %pR\n", - res, conflict->name, conflict); - } else { - pci_bus_add_resource(info->bus, res, 0); - info->res_num++; - if (addr.translation_offset) - dev_info(&info->bridge->dev, "host bridge window %pR " - "(PCI address [%#llx-%#llx])\n", - res, res->start - addr.translation_offset, - res->end - addr.translation_offset); + info->res_num++; + if (addr.translation_offset) + dev_info(&info->bridge->dev, "host bridge window %pR " + "(PCI address [%#llx-%#llx])\n", + res, res->start - addr.translation_offset, + res->end - addr.translation_offset); + else + dev_info(&info->bridge->dev, "host bridge window %pR\n", res); + + return AE_OK; +} + +static bool resource_contains(struct resource *res, resource_size_t point) +{ + if (res->start <= point && point <= res->end) + return true; + return false; +} + +static void coalesce_windows(struct pci_root_info *info, int type) +{ + int i, j; + struct resource *res1, *res2; + + for (i = 0; i < info->res_num; i++) { + res1 = &info->res[i]; + if (!(res1->flags & type)) + continue; + + for (j = i + 1; j < info->res_num; j++) { + res2 = &info->res[j]; + if (!(res2->flags & type)) + continue; + + /* + * I don't like throwing away windows because then + * our resources no longer match the ACPI _CRS, but + * the kernel resource tree doesn't allow overlaps. + */ + if (resource_contains(res1, res2->start) || + resource_contains(res1, res2->end) || + resource_contains(res2, res1->start) || + resource_contains(res2, res1->end)) { + res1->start = min(res1->start, res2->start); + res1->end = max(res1->end, res2->end); + dev_info(&info->bridge->dev, + "host bridge window expanded to %pR; %pR ignored\n", + res1, res2); + res2->flags = 0; + } + } + } +} + +static void add_resources(struct pci_root_info *info) +{ + int i; + struct resource *res, *root, *conflict; + + if (!pci_use_crs) + return; + + coalesce_windows(info, IORESOURCE_MEM); + coalesce_windows(info, IORESOURCE_IO); + + for (i = 0; i < info->res_num; i++) { + res = &info->res[i]; + + if (res->flags & IORESOURCE_MEM) + root = &iomem_resource; + else if (res->flags & IORESOURCE_IO) + root = &ioport_resource; else - dev_info(&info->bridge->dev, - "host bridge window %pR\n", res); + continue; + + conflict = insert_resource_conflict(root, res); + if (conflict) + dev_err(&info->bridge->dev, + "address space collision: host bridge window %pR " + "conflicts with %s %pR\n", + res, conflict->name, conflict); + else + pci_bus_add_resource(info->bus, res, 0); } - return AE_OK; } static void @@ -224,6 +286,7 @@ get_current_resources(struct acpi_device *device, int busnum, acpi_walk_resources(device->handle, METHOD_NAME__CRS, setup_resource, &info); + add_resources(&info); return; name_alloc_fail: -- cgit v1.2.3-18-g5258 From 451a3c24b0135bce54542009b5fde43846c7cf67 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 17 Nov 2010 16:26:55 +0100 Subject: BKL: remove extraneous #include The big kernel lock has been removed from all these files at some point, leaving only the #include. Remove this too as a cleanup. Signed-off-by: Arnd Bergmann Signed-off-by: Linus Torvalds --- arch/x86/ia32/sys_ia32.c | 1 - arch/x86/kernel/cpuid.c | 1 - arch/x86/kernel/msr.c | 1 - 3 files changed, 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index 849813f398e..5852519b2d0 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -28,7 +28,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/x86/kernel/cpuid.c b/arch/x86/kernel/cpuid.c index 1b7b31ab7d8..212a6a42527 100644 --- a/arch/x86/kernel/cpuid.c +++ b/arch/x86/kernel/cpuid.c @@ -33,7 +33,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/x86/kernel/msr.c b/arch/x86/kernel/msr.c index 7bf2dc4c8f7..12fcbe2c143 100644 --- a/arch/x86/kernel/msr.c +++ b/arch/x86/kernel/msr.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include #include -- cgit v1.2.3-18-g5258 From 10a6e67648d4b47769953bd24759ba9609bf00df Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 15 Nov 2010 08:07:35 -0600 Subject: kgdb,x86: fix regression in detach handling The fix from ba773f7c510c0b252145933926c636c439889207 (x86,kgdb: Fix hw breakpoint regression) was not entirely complete. The kgdb_remove_all_hw_break() function also needs to call the hw_break_release_slot() or else a breakpoint can get activated again after the debugger has detached. The kgdb test suite exposes the behavior in the form of either a hang or repetitive failure. The kernel config that exposes the problem contains all of the following: CONFIG_DEBUG_RODATA=y CONFIG_KGDB_TESTS=y CONFIG_KGDB_TESTS_ON_BOOT=y CONFIG_KGDB_TESTS_BOOT_STRING="V1F100" Reported-by: Frederic Weisbecker Signed-off-by: Jason Wessel Tested-by: Frederic Weisbecker --- arch/x86/kernel/kgdb.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c index ec592caac4b..cd21b654dec 100644 --- a/arch/x86/kernel/kgdb.c +++ b/arch/x86/kernel/kgdb.c @@ -315,14 +315,18 @@ static void kgdb_remove_all_hw_break(void) if (!breakinfo[i].enabled) continue; bp = *per_cpu_ptr(breakinfo[i].pev, cpu); - if (bp->attr.disabled == 1) + if (!bp->attr.disabled) { + arch_uninstall_hw_breakpoint(bp); + bp->attr.disabled = 1; continue; + } if (dbg_is_early) early_dr7 &= ~encode_dr7(i, breakinfo[i].len, breakinfo[i].type); - else - arch_uninstall_hw_breakpoint(bp); - bp->attr.disabled = 1; + else if (hw_break_release_slot(i)) + printk(KERN_ERR "KGDB: hw bpt remove failed %lx\n", + breakinfo[i].addr); + breakinfo[i].enabled = 0; } } -- cgit v1.2.3-18-g5258 From 0a77fe4c188e25917799f2356d4aa5e6d80c39a2 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 19 Oct 2010 18:48:35 +0200 Subject: KVM: Correct ordering of ldt reload wrt fs/gs reload If fs or gs refer to the ldt, they must be reloaded after the ldt. Reorder the code to that effect. Userspace code that uses the ldt with kvm is nonexistent, so this doesn't fix a user-visible bug. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 2 +- arch/x86/kvm/vmx.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 82e144a4e51..1ca12298ffc 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -3395,6 +3395,7 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; load_host_msrs(vcpu); + kvm_load_ldt(ldt_selector); loadsegment(fs, fs_selector); #ifdef CONFIG_X86_64 load_gs_index(gs_selector); @@ -3402,7 +3403,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) #else loadsegment(gs, gs_selector); #endif - kvm_load_ldt(ldt_selector); reload_tss(vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 8da0e45ff7c..6fe7df75bfd 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -839,8 +839,6 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) ++vmx->vcpu.stat.host_state_reload; vmx->host_state.loaded = 0; - if (vmx->host_state.fs_reload_needed) - loadsegment(fs, vmx->host_state.fs_sel); if (vmx->host_state.gs_ldt_reload_needed) { kvm_load_ldt(vmx->host_state.ldt_sel); #ifdef CONFIG_X86_64 @@ -850,6 +848,8 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) loadsegment(gs, vmx->host_state.gs_sel); #endif } + if (vmx->host_state.fs_reload_needed) + loadsegment(fs, vmx->host_state.fs_sel); reload_tss(); #ifdef CONFIG_X86_64 if (is_long_mode(&vmx->vcpu)) { -- cgit v1.2.3-18-g5258 From c8770e7ba63bb5dd8fe5f9d251275a8fa717fb78 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 11 Nov 2010 12:37:26 +0200 Subject: KVM: VMX: Fix host userspace gsbase corruption We now use load_gs_index() to load gs safely; unfortunately this also changes MSR_KERNEL_GS_BASE, which we managed separately. This resulted in confusion and breakage running 32-bit host userspace on a 64-bit kernel. Fix by - saving guest MSR_KERNEL_GS_BASE before we we reload the host's gs - doing the host save/load unconditionally, instead of only when in guest long mode Things can be cleaned up further, but this is the minmal fix for now. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/vmx.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 6fe7df75bfd..ff21fdda0c5 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -821,10 +821,9 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) #endif #ifdef CONFIG_X86_64 - if (is_long_mode(&vmx->vcpu)) { - rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_host_kernel_gs_base); + rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_host_kernel_gs_base); + if (is_long_mode(&vmx->vcpu)) wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base); - } #endif for (i = 0; i < vmx->save_nmsrs; ++i) kvm_set_shared_msr(vmx->guest_msrs[i].index, @@ -839,11 +838,14 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) ++vmx->vcpu.stat.host_state_reload; vmx->host_state.loaded = 0; +#ifdef CONFIG_X86_64 + if (is_long_mode(&vmx->vcpu)) + rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base); +#endif if (vmx->host_state.gs_ldt_reload_needed) { kvm_load_ldt(vmx->host_state.ldt_sel); #ifdef CONFIG_X86_64 load_gs_index(vmx->host_state.gs_sel); - wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs); #else loadsegment(gs, vmx->host_state.gs_sel); #endif @@ -852,10 +854,7 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) loadsegment(fs, vmx->host_state.fs_sel); reload_tss(); #ifdef CONFIG_X86_64 - if (is_long_mode(&vmx->vcpu)) { - rdmsrl(MSR_KERNEL_GS_BASE, vmx->msr_guest_kernel_gs_base); - wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_host_kernel_gs_base); - } + wrmsrl(MSR_KERNEL_GS_BASE, vmx->msr_host_kernel_gs_base); #endif if (current_thread_info()->status & TS_USEDFPU) clts(); -- cgit v1.2.3-18-g5258