From b5908548537ccd3ada258ca5348df7ffc93e5a06 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 10 Nov 2010 22:29:49 -0500 Subject: tracing: Force arch_local_irq_* notrace for paravirt When running ktest.pl randconfig tests, I would sometimes trigger a lockdep annotation bug (possible reason: unannotated irqs-on). This triggering happened right after function tracer self test was executed. After doing a config bisect I found that this was caused with having function tracer, paravirt guest, prove locking, and rcu torture all enabled. The rcu torture just enhanced the likelyhood of triggering the bug. Prove locking was needed, since it was the thing that was bugging. Function tracer would trace and disable interrupts in all sorts of funny places. paravirt guest would turn arch_local_irq_* into functions that would be traced. Besides the fact that tracing arch_local_irq_* is just a bad idea, this is what is happening. The bug happened simply in the local_irq_restore() code: if (raw_irqs_disabled_flags(flags)) { \ raw_local_irq_restore(flags); \ trace_hardirqs_off(); \ } else { \ trace_hardirqs_on(); \ raw_local_irq_restore(flags); \ } \ The raw_local_irq_restore() was defined as arch_local_irq_restore(). Now imagine, we are about to enable interrupts. We go into the else case and call trace_hardirqs_on() which tells lockdep that we are enabling interrupts, so it sets the current->hardirqs_enabled = 1. Then we call raw_local_irq_restore() which calls arch_local_irq_restore() which gets traced! Now in the function tracer we disable interrupts with local_irq_save(). This is fine, but flags is stored that we have interrupts disabled. When the function tracer calls local_irq_restore() it does it, but this time with flags set as disabled, so we go into the if () path. This keeps interrupts disabled and calls trace_hardirqs_off() which sets current->hardirqs_enabled = 0. When the tracer is finished and proceeds with the original code, we enable interrupts but leave current->hardirqs_enabled as 0. Which now breaks lockdeps internal processing. Cc: Thomas Gleixner Signed-off-by: Steven Rostedt --- arch/x86/include/asm/paravirt.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index 18e3b8a8709..ef9975812c7 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -824,27 +824,27 @@ static __always_inline void arch_spin_unlock(struct arch_spinlock *lock) #define __PV_IS_CALLEE_SAVE(func) \ ((struct paravirt_callee_save) { func }) -static inline unsigned long arch_local_save_flags(void) +static inline notrace unsigned long arch_local_save_flags(void) { return PVOP_CALLEE0(unsigned long, pv_irq_ops.save_fl); } -static inline void arch_local_irq_restore(unsigned long f) +static inline notrace void arch_local_irq_restore(unsigned long f) { PVOP_VCALLEE1(pv_irq_ops.restore_fl, f); } -static inline void arch_local_irq_disable(void) +static inline notrace void arch_local_irq_disable(void) { PVOP_VCALLEE0(pv_irq_ops.irq_disable); } -static inline void arch_local_irq_enable(void) +static inline notrace void arch_local_irq_enable(void) { PVOP_VCALLEE0(pv_irq_ops.irq_enable); } -static inline unsigned long arch_local_irq_save(void) +static inline notrace unsigned long arch_local_irq_save(void) { unsigned long f; -- cgit v1.2.3-70-g09d2 From 7e77506a5918d82cafa2ffa783ab57c23f9e9817 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Thu, 30 Sep 2010 12:37:26 +0100 Subject: xen: implement XENMEM_machphys_mapping This hypercall allows Xen to specify a non-default location for the machine to physical mapping. This capability is used when running a 32 bit domain 0 on a 64 bit hypervisor to shrink the hypervisor hole to exactly the size required. [ Impact: add Xen hypercall definitions ] Signed-off-by: Ian Campbell Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Stefano Stabellini --- arch/x86/include/asm/xen/interface.h | 6 +++--- arch/x86/include/asm/xen/interface_32.h | 5 +++++ arch/x86/include/asm/xen/interface_64.h | 13 +------------ arch/x86/include/asm/xen/page.h | 7 ++++--- arch/x86/xen/enlighten.c | 7 +++++++ arch/x86/xen/mmu.c | 14 ++++++++++++++ include/xen/interface/memory.h | 13 +++++++++++++ 7 files changed, 47 insertions(+), 18 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/xen/interface.h b/arch/x86/include/asm/xen/interface.h index e8506c1f0c5..1c10c88ee4e 100644 --- a/arch/x86/include/asm/xen/interface.h +++ b/arch/x86/include/asm/xen/interface.h @@ -61,9 +61,9 @@ DEFINE_GUEST_HANDLE(void); #define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START) #endif -#ifndef machine_to_phys_mapping -#define machine_to_phys_mapping ((unsigned long *)HYPERVISOR_VIRT_START) -#endif +#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START) +#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END) +#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>__MACH2PHYS_SHIFT) /* Maximum number of virtual CPUs in multi-processor guests. */ #define MAX_VIRT_CPUS 32 diff --git a/arch/x86/include/asm/xen/interface_32.h b/arch/x86/include/asm/xen/interface_32.h index 42a7e004ae5..8413688b257 100644 --- a/arch/x86/include/asm/xen/interface_32.h +++ b/arch/x86/include/asm/xen/interface_32.h @@ -32,6 +32,11 @@ /* And the trap vector is... */ #define TRAP_INSTR "int $0x82" +#define __MACH2PHYS_VIRT_START 0xF5800000 +#define __MACH2PHYS_VIRT_END 0xF6800000 + +#define __MACH2PHYS_SHIFT 2 + /* * Virtual addresses beyond this are not modifiable by guest OSes. The * machine->physical mapping table starts at this address, read-only. diff --git a/arch/x86/include/asm/xen/interface_64.h b/arch/x86/include/asm/xen/interface_64.h index 100d2662b97..839a4811cf9 100644 --- a/arch/x86/include/asm/xen/interface_64.h +++ b/arch/x86/include/asm/xen/interface_64.h @@ -39,18 +39,7 @@ #define __HYPERVISOR_VIRT_END 0xFFFF880000000000 #define __MACH2PHYS_VIRT_START 0xFFFF800000000000 #define __MACH2PHYS_VIRT_END 0xFFFF804000000000 - -#ifndef HYPERVISOR_VIRT_START -#define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START) -#define HYPERVISOR_VIRT_END mk_unsigned_long(__HYPERVISOR_VIRT_END) -#endif - -#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START) -#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END) -#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>3) -#ifndef machine_to_phys_mapping -#define machine_to_phys_mapping ((unsigned long *)HYPERVISOR_VIRT_START) -#endif +#define __MACH2PHYS_SHIFT 3 /* * int HYPERVISOR_set_segment_base(unsigned int which, unsigned long base) diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index dd8c1414b3d..8760cc60a21 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #include @@ -35,6 +36,8 @@ typedef struct xpaddr { #define MAX_DOMAIN_PAGES \ ((unsigned long)((u64)CONFIG_XEN_MAX_DOMAIN_MEMORY * 1024 * 1024 * 1024 / PAGE_SIZE)) +extern unsigned long *machine_to_phys_mapping; +extern unsigned int machine_to_phys_order; extern unsigned long get_phys_to_machine(unsigned long pfn); extern bool set_phys_to_machine(unsigned long pfn, unsigned long mfn); @@ -69,10 +72,8 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn) if (xen_feature(XENFEAT_auto_translated_physmap)) return mfn; -#if 0 if (unlikely((mfn >> machine_to_phys_order) != 0)) - return max_mapnr; -#endif + return ~0; pfn = 0; /* diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 235c0f4d386..bd355493461 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -75,6 +75,11 @@ DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info); enum xen_domain_type xen_domain_type = XEN_NATIVE; EXPORT_SYMBOL_GPL(xen_domain_type); +unsigned long *machine_to_phys_mapping = (void *)MACH2PHYS_VIRT_START; +EXPORT_SYMBOL(machine_to_phys_mapping); +unsigned int machine_to_phys_order; +EXPORT_SYMBOL(machine_to_phys_order); + struct start_info *xen_start_info; EXPORT_SYMBOL_GPL(xen_start_info); @@ -1097,6 +1102,8 @@ asmlinkage void __init xen_start_kernel(void) xen_domain_type = XEN_PV_DOMAIN; + xen_setup_machphys_mapping(); + /* Install Xen paravirt ops */ pv_info = xen_info; pv_init_ops = xen_init_ops; diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 21ed8d7f75a..bd2713a8257 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -2034,6 +2034,20 @@ static __init void xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) set_page_prot(pmd, PAGE_KERNEL_RO); } +void __init xen_setup_machphys_mapping(void) +{ + struct xen_machphys_mapping mapping; + unsigned long machine_to_phys_nr_ents; + + if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) { + machine_to_phys_mapping = (unsigned long *)mapping.v_start; + machine_to_phys_nr_ents = mapping.max_mfn + 1; + } else { + machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES; + } + machine_to_phys_order = fls(machine_to_phys_nr_ents - 1); +} + #ifdef CONFIG_X86_64 static void convert_pfn_mfn(void *v) { diff --git a/include/xen/interface/memory.h b/include/xen/interface/memory.h index d7a6c13bde6..eac3ce15371 100644 --- a/include/xen/interface/memory.h +++ b/include/xen/interface/memory.h @@ -140,6 +140,19 @@ struct xen_machphys_mfn_list { }; DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mfn_list); +/* + * Returns the location in virtual address space of the machine_to_phys + * mapping table. Architectures which do not have a m2p table, or which do not + * map it by default into guest address space, do not implement this command. + * arg == addr of xen_machphys_mapping_t. + */ +#define XENMEM_machphys_mapping 12 +struct xen_machphys_mapping { + unsigned long v_start, v_end; /* Start and end virtual addresses. */ + unsigned long max_mfn; /* Maximum MFN that can be looked up. */ +}; +DEFINE_GUEST_HANDLE_STRUCT(xen_machphys_mapping_t); + /* * Sets the GPFN at which a particular page appears in the specified guest's * pseudophysical address space. -- cgit v1.2.3-70-g09d2 From 8191c9f69202d4dbc66063cb92059b8a58640d34 Mon Sep 17 00:00:00 2001 From: Dimitri Sivanich Date: Tue, 16 Nov 2010 16:23:52 -0600 Subject: x86: UV: Address interrupt/IO port operation conflict This patch for SGI UV systems addresses a problem whereby interrupt transactions being looped back from a local IOH, through the hub to a local CPU can (erroneously) conflict with IO port operations and other transactions. To workaound this we set a high bit in the APIC IDs used for interrupts. This bit appears to be ignored by the sockets, but it avoids the conflict in the hub. Signed-off-by: Dimitri Sivanich LKML-Reference: <20101116222352.GA8155@sgi.com> Signed-off-by: Ingo Molnar ___ arch/x86/include/asm/uv/uv_hub.h | 4 ++++ arch/x86/include/asm/uv/uv_mmrs.h | 19 ++++++++++++++++++- arch/x86/kernel/apic/x2apic_uv_x.c | 25 +++++++++++++++++++++++-- arch/x86/platform/uv/tlb_uv.c | 2 +- arch/x86/platform/uv/uv_time.c | 4 +++- 5 files changed, 49 insertions(+), 5 deletions(-) --- arch/x86/include/asm/uv/uv_hub.h | 4 ++++ arch/x86/include/asm/uv/uv_mmrs.h | 19 ++++++++++++++++++- arch/x86/kernel/apic/x2apic_uv_x.c | 25 +++++++++++++++++++++++-- arch/x86/platform/uv/tlb_uv.c | 2 +- arch/x86/platform/uv/uv_time.c | 4 +++- 5 files changed, 49 insertions(+), 5 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h index e969f691cbf..a501741c233 100644 --- a/arch/x86/include/asm/uv/uv_hub.h +++ b/arch/x86/include/asm/uv/uv_hub.h @@ -199,6 +199,8 @@ union uvh_apicid { #define UVH_APICID 0x002D0E00L #define UV_APIC_PNODE_SHIFT 6 +#define UV_APICID_HIBIT_MASK 0xffff0000 + /* Local Bus from cpu's perspective */ #define LOCAL_BUS_BASE 0x1c00000 #define LOCAL_BUS_SIZE (4 * 1024 * 1024) @@ -491,8 +493,10 @@ static inline void uv_set_cpu_scir_bits(int cpu, unsigned char value) } } +extern unsigned int uv_apicid_hibits; static unsigned long uv_hub_ipi_value(int apicid, int vector, int mode) { + apicid |= uv_apicid_hibits; return (1UL << UVH_IPI_INT_SEND_SHFT) | ((apicid) << UVH_IPI_INT_APIC_ID_SHFT) | (mode << UVH_IPI_INT_DELIVERY_MODE_SHFT) | diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h index 6d90adf4428..20cafeac745 100644 --- a/arch/x86/include/asm/uv/uv_mmrs.h +++ b/arch/x86/include/asm/uv/uv_mmrs.h @@ -5,7 +5,7 @@ * * SGI UV MMR definitions * - * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2007-2010 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_X86_UV_UV_MMRS_H @@ -753,6 +753,23 @@ union uvh_lb_bau_sb_descriptor_base_u { } s; }; +/* ========================================================================= */ +/* UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK */ +/* ========================================================================= */ +#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK 0x320130UL +#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_32 0x009f0 + +#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_SHFT 0 +#define UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK_BIT_ENABLES_MASK 0x00000000ffffffffUL + +union uvh_lb_target_physical_apic_id_mask_u { + unsigned long v; + struct uvh_lb_target_physical_apic_id_mask_s { + unsigned long bit_enables : 32; /* RW */ + unsigned long rsvd_32_63 : 32; /* */ + } s; +}; + /* ========================================================================= */ /* UVH_NODE_ID */ /* ========================================================================= */ diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 194539aea17..c1c52c341f4 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -44,6 +44,8 @@ static u64 gru_start_paddr, gru_end_paddr; static union uvh_apicid uvh_apicid; int uv_min_hub_revision_id; EXPORT_SYMBOL_GPL(uv_min_hub_revision_id); +unsigned int uv_apicid_hibits; +EXPORT_SYMBOL_GPL(uv_apicid_hibits); static DEFINE_SPINLOCK(uv_nmi_lock); static inline bool is_GRU_range(u64 start, u64 end) @@ -85,6 +87,23 @@ static void __init early_get_apic_pnode_shift(void) uvh_apicid.s.pnode_shift = UV_APIC_PNODE_SHIFT; } +/* + * Add an extra bit as dictated by bios to the destination apicid of + * interrupts potentially passing through the UV HUB. This prevents + * a deadlock between interrupts and IO port operations. + */ +static void __init uv_set_apicid_hibit(void) +{ + union uvh_lb_target_physical_apic_id_mask_u apicid_mask; + unsigned long *mmr; + + mmr = early_ioremap(UV_LOCAL_MMR_BASE | + UVH_LB_TARGET_PHYSICAL_APIC_ID_MASK, sizeof(*mmr)); + apicid_mask.v = *mmr; + early_iounmap(mmr, sizeof(*mmr)); + uv_apicid_hibits = apicid_mask.s.bit_enables & UV_APICID_HIBIT_MASK; +} + static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { int nodeid; @@ -102,6 +121,7 @@ static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) __get_cpu_var(x2apic_extra_bits) = nodeid << (uvh_apicid.s.pnode_shift - 1); uv_system_type = UV_NON_UNIQUE_APIC; + uv_set_apicid_hibit(); return 1; } } @@ -155,6 +175,7 @@ static int __cpuinit uv_wakeup_secondary(int phys_apicid, unsigned long start_ri int pnode; pnode = uv_apicid_to_pnode(phys_apicid); + phys_apicid |= uv_apicid_hibits; val = (1UL << UVH_IPI_INT_SEND_SHFT) | (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | @@ -236,7 +257,7 @@ static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask) int cpu = cpumask_first(cpumask); if ((unsigned)cpu < nr_cpu_ids) - return per_cpu(x86_cpu_to_apicid, cpu); + return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits; else return BAD_APICID; } @@ -255,7 +276,7 @@ uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, if (cpumask_test_cpu(cpu, cpu_online_mask)) break; } - return per_cpu(x86_cpu_to_apicid, cpu); + return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits; } static unsigned int x2apic_get_apic_id(unsigned long x) diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c index a318194002b..ba9caa808a9 100644 --- a/arch/x86/platform/uv/tlb_uv.c +++ b/arch/x86/platform/uv/tlb_uv.c @@ -1455,7 +1455,7 @@ static void __init uv_init_uvhub(int uvhub, int vector) * the below initialization can't be in firmware because the * messaging IRQ will be determined by the OS */ - apicid = uvhub_to_first_apicid(uvhub); + apicid = uvhub_to_first_apicid(uvhub) | uv_apicid_hibits; uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, ((apicid << 32) | vector)); } diff --git a/arch/x86/platform/uv/uv_time.c b/arch/x86/platform/uv/uv_time.c index 56e421bc379..9daf5d1af9f 100644 --- a/arch/x86/platform/uv/uv_time.c +++ b/arch/x86/platform/uv/uv_time.c @@ -89,6 +89,7 @@ static void uv_rtc_send_IPI(int cpu) apicid = cpu_physical_id(cpu); pnode = uv_apicid_to_pnode(apicid); + apicid |= uv_apicid_hibits; val = (1UL << UVH_IPI_INT_SEND_SHFT) | (apicid << UVH_IPI_INT_APIC_ID_SHFT) | (X86_PLATFORM_IPI_VECTOR << UVH_IPI_INT_VECTOR_SHFT); @@ -107,6 +108,7 @@ static int uv_intr_pending(int pnode) static int uv_setup_intr(int cpu, u64 expires) { u64 val; + unsigned long apicid = cpu_physical_id(cpu) | uv_apicid_hibits; int pnode = uv_cpu_to_pnode(cpu); uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, @@ -117,7 +119,7 @@ static int uv_setup_intr(int cpu, u64 expires) UVH_EVENT_OCCURRED0_RTC1_MASK); val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) | - ((u64)cpu_physical_id(cpu) << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT); + ((u64)apicid << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT); /* Set configuration */ uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, val); -- cgit v1.2.3-70-g09d2 From 37db6c8f1d0c4b8f01dc049f3a893b725288660f Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Tue, 16 Nov 2010 08:25:08 +0000 Subject: x86-64: Fix and clean up AMD Fam10 MMCONF enabling Candidate memory ranges were not calculated properly (start addresses got needlessly rounded down, and end addresses didn't get rounded up at all), address comparison for secondary CPUs was done on only part of the address, and disabled status wasn't tracked properly. Signed-off-by: Jan Beulich Acked-by: Yinghai Lu Acked-by: Andreas Herrmann LKML-Reference: <4CE24DF40200007800022737@vpn.id2.novell.com> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/msr-index.h | 2 +- arch/x86/kernel/mmconf-fam10h_64.c | 64 ++++++++++++++++++-------------------- 2 files changed, 31 insertions(+), 35 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 3ea3dc48704..6b89f5e8602 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -128,7 +128,7 @@ #define FAM10H_MMIO_CONF_ENABLE (1<<0) #define FAM10H_MMIO_CONF_BUSRANGE_MASK 0xf #define FAM10H_MMIO_CONF_BUSRANGE_SHIFT 2 -#define FAM10H_MMIO_CONF_BASE_MASK 0xfffffff +#define FAM10H_MMIO_CONF_BASE_MASK 0xfffffffULL #define FAM10H_MMIO_CONF_BASE_SHIFT 20 #define MSR_FAM10H_NODE_ID 0xc001100c diff --git a/arch/x86/kernel/mmconf-fam10h_64.c b/arch/x86/kernel/mmconf-fam10h_64.c index 6da143c2a6b..ac861b8348e 100644 --- a/arch/x86/kernel/mmconf-fam10h_64.c +++ b/arch/x86/kernel/mmconf-fam10h_64.c @@ -25,7 +25,6 @@ struct pci_hostbridge_probe { }; static u64 __cpuinitdata fam10h_pci_mmconf_base; -static int __cpuinitdata fam10h_pci_mmconf_base_status; static struct pci_hostbridge_probe pci_probes[] __cpuinitdata = { { 0, 0x18, PCI_VENDOR_ID_AMD, 0x1200 }, @@ -44,10 +43,12 @@ static int __cpuinit cmp_range(const void *x1, const void *x2) return start1 - start2; } -/*[47:0] */ -/* need to avoid (0xfd<<32) and (0xfe<<32), ht used space */ +#define MMCONF_UNIT (1ULL << FAM10H_MMIO_CONF_BASE_SHIFT) +#define MMCONF_MASK (~(MMCONF_UNIT - 1)) +#define MMCONF_SIZE (MMCONF_UNIT << 8) +/* need to avoid (0xfd<<32), (0xfe<<32), and (0xff<<32), ht used space */ #define FAM10H_PCI_MMCONF_BASE (0xfcULL<<32) -#define BASE_VALID(b) ((b != (0xfdULL << 32)) && (b != (0xfeULL << 32))) +#define BASE_VALID(b) ((b) + MMCONF_SIZE <= (0xfdULL<<32) || (b) >= (1ULL<<40)) static void __cpuinit get_fam10h_pci_mmconf_base(void) { int i; @@ -64,12 +65,11 @@ static void __cpuinit get_fam10h_pci_mmconf_base(void) struct range range[8]; /* only try to get setting from BSP */ - /* -1 or 1 */ - if (fam10h_pci_mmconf_base_status) + if (fam10h_pci_mmconf_base) return; if (!early_pci_allowed()) - goto fail; + return; found = 0; for (i = 0; i < ARRAY_SIZE(pci_probes); i++) { @@ -91,7 +91,7 @@ static void __cpuinit get_fam10h_pci_mmconf_base(void) } if (!found) - goto fail; + return; /* SYS_CFG */ address = MSR_K8_SYSCFG; @@ -99,16 +99,16 @@ static void __cpuinit get_fam10h_pci_mmconf_base(void) /* TOP_MEM2 is not enabled? */ if (!(val & (1<<21))) { - tom2 = 0; + tom2 = 1ULL << 32; } else { /* TOP_MEM2 */ address = MSR_K8_TOP_MEM2; rdmsrl(address, val); - tom2 = val & (0xffffULL<<32); + tom2 = max(val & 0xffffff800000ULL, 1ULL << 32); } if (base <= tom2) - base = tom2 + (1ULL<<32); + base = (tom2 + 2 * MMCONF_UNIT - 1) & MMCONF_MASK; /* * need to check if the range is in the high mmio range that is @@ -123,11 +123,11 @@ static void __cpuinit get_fam10h_pci_mmconf_base(void) if (!(reg & 3)) continue; - start = (((u64)reg) << 8) & (0xffULL << 32); /* 39:16 on 31:8*/ + start = (u64)(reg & 0xffffff00) << 8; /* 39:16 on 31:8*/ reg = read_pci_config(bus, slot, 1, 0x84 + (i << 3)); - end = (((u64)reg) << 8) & (0xffULL << 32); /* 39:16 on 31:8*/ + end = ((u64)(reg & 0xffffff00) << 8) | 0xffff; /* 39:16 on 31:8*/ - if (!end) + if (end < tom2) continue; range[hi_mmio_num].start = start; @@ -143,32 +143,27 @@ static void __cpuinit get_fam10h_pci_mmconf_base(void) if (range[hi_mmio_num - 1].end < base) goto out; - if (range[0].start > base) + if (range[0].start > base + MMCONF_SIZE) goto out; /* need to find one window */ - base = range[0].start - (1ULL << 32); + base = (range[0].start & MMCONF_MASK) - MMCONF_UNIT; if ((base > tom2) && BASE_VALID(base)) goto out; - base = range[hi_mmio_num - 1].end + (1ULL << 32); - if ((base > tom2) && BASE_VALID(base)) + base = (range[hi_mmio_num - 1].end + MMCONF_UNIT) & MMCONF_MASK; + if (BASE_VALID(base)) goto out; /* need to find window between ranges */ - if (hi_mmio_num > 1) - for (i = 0; i < hi_mmio_num - 1; i++) { - if (range[i + 1].start > (range[i].end + (1ULL << 32))) { - base = range[i].end + (1ULL << 32); - if ((base > tom2) && BASE_VALID(base)) - goto out; - } + for (i = 1; i < hi_mmio_num; i++) { + base = (range[i - 1].end + MMCONF_UNIT) & MMCONF_MASK; + val = range[i].start & MMCONF_MASK; + if (val >= base + MMCONF_SIZE && BASE_VALID(base)) + goto out; } - -fail: - fam10h_pci_mmconf_base_status = -1; return; + out: fam10h_pci_mmconf_base = base; - fam10h_pci_mmconf_base_status = 1; } void __cpuinit fam10h_check_enable_mmcfg(void) @@ -190,11 +185,10 @@ void __cpuinit fam10h_check_enable_mmcfg(void) /* only trust the one handle 256 buses, if acpi=off */ if (!acpi_pci_disabled || busnbits >= 8) { - u64 base; - base = val & (0xffffULL << 32); - if (fam10h_pci_mmconf_base_status <= 0) { + u64 base = val & MMCONF_MASK; + + if (!fam10h_pci_mmconf_base) { fam10h_pci_mmconf_base = base; - fam10h_pci_mmconf_base_status = 1; return; } else if (fam10h_pci_mmconf_base == base) return; @@ -206,8 +200,10 @@ void __cpuinit fam10h_check_enable_mmcfg(void) * with 256 buses */ get_fam10h_pci_mmconf_base(); - if (fam10h_pci_mmconf_base_status <= 0) + if (!fam10h_pci_mmconf_base) { + pci_probe &= ~PCI_CHECK_ENABLE_AMD_MMCONF; return; + } printk(KERN_INFO "Enable MMCONFIG on AMD Family 10h\n"); val &= ~((FAM10H_MMIO_CONF_BASE_MASK< Date: Wed, 24 Nov 2010 12:57:18 -0800 Subject: arch/x86/include/asm/fixmap.h: mark __set_fixmap_offset as __always_inline When compiling arch/x86/kernel/early_printk_mrst.c with i386 allmodconfig, gcc-4.1.0 generates an out-of-line copy of __set_fixmap_offset() which contains a reference to __this_fixmap_does_not_exist which the compiler cannot elide. Marking __set_fixmap_offset() as __always_inline prevents this. Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Feng Tang Acked-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/fixmap.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index 4d293dced62..9479a037419 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -216,8 +216,8 @@ static inline unsigned long virt_to_fix(const unsigned long vaddr) } /* Return an pointer with offset calculated */ -static inline unsigned long __set_fixmap_offset(enum fixed_addresses idx, - phys_addr_t phys, pgprot_t flags) +static __always_inline unsigned long +__set_fixmap_offset(enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags) { __set_fixmap(idx, phys, flags); return fix_to_virt(idx) + (phys & (PAGE_SIZE - 1)); -- cgit v1.2.3-70-g09d2 From e7a3481c0246c8e45e79c629efd63b168e91fcda Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 25 Oct 2010 16:53:46 -0700 Subject: x86/pvclock: Zero last_value on resume If the guest domain has been suspend/resumed or migrated, then the system clock backing the pvclock clocksource may revert to a smaller value (ie, can be non-monotonic across the migration/save-restore). Make sure we zero last_value in that case so that the domain continues to see clock updates. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pvclock.h | 1 + arch/x86/kernel/pvclock.c | 5 +++++ arch/x86/xen/time.c | 2 ++ 3 files changed, 8 insertions(+) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/pvclock.h b/arch/x86/include/asm/pvclock.h index 7f7e577a0e3..31d84acc151 100644 --- a/arch/x86/include/asm/pvclock.h +++ b/arch/x86/include/asm/pvclock.h @@ -11,6 +11,7 @@ unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src); void pvclock_read_wallclock(struct pvclock_wall_clock *wall, struct pvclock_vcpu_time_info *vcpu, struct timespec *ts); +void pvclock_resume(void); /* * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction, diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 008b91eefa1..42eb3300dfc 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -83,6 +83,11 @@ unsigned long pvclock_tsc_khz(struct pvclock_vcpu_time_info *src) static atomic64_t last_value = ATOMIC64_INIT(0); +void pvclock_resume(void) +{ + atomic64_set(&last_value, 0); +} + cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src) { struct pvclock_shadow_time shadow; diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index b2bb5aa3b05..5da5e53fb94 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -426,6 +426,8 @@ void xen_timer_resume(void) { int cpu; + pvclock_resume(); + if (xen_clockevent != &xen_vcpuop_clockevent) return; -- cgit v1.2.3-70-g09d2 From 73c1160ce377d8fc6d84cb630ebf9658808bec49 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Wed, 1 Dec 2010 12:17:44 +0100 Subject: KVM: enlarge number of possible CPUID leaves Currently the number of CPUID leaves KVM handles is limited to 40. My desktop machine (AthlonII) already has 35 and future CPUs will expand this well beyond the limit. Extend the limit to 80 to make room for future processors. KVM-Stable-Tag. Signed-off-by: Andre Przywara Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 9e6fe391094..f702f82aa1e 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -79,7 +79,7 @@ #define KVM_NUM_MMU_PAGES (1 << KVM_MMU_HASH_SHIFT) #define KVM_MIN_FREE_MMU_PAGES 5 #define KVM_REFILL_PAGES 25 -#define KVM_MAX_CPUID_ENTRIES 40 +#define KVM_MAX_CPUID_ENTRIES 80 #define KVM_NR_FIXED_MTRR_REGION 88 #define KVM_NR_VAR_MTRR 8 -- cgit v1.2.3-70-g09d2 From a2c606d53ab71dee6410f10ef0adf67321d60e06 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 16 Dec 2010 10:39:02 -0700 Subject: x86: avoid high BIOS area when allocating address space This prevents allocation of the last 2MB before 4GB. The experiment described here shows Windows 7 ignoring the last 1MB: https://bugzilla.kernel.org/show_bug.cgi?id=23542#c27 This patch ignores the top 2MB instead of just 1MB because H. Peter Anvin says "There will be ROM at the top of the 32-bit address space; it's a fact of the architecture, and on at least older systems it was common to have a shadow 1 MiB below." Acked-by: H. Peter Anvin Signed-off-by: Bjorn Helgaas Signed-off-by: Jesse Barnes --- arch/x86/include/asm/e820.h | 3 +++ arch/x86/kernel/resource.c | 3 ++- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/x86/include/asm') diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 5be1542fbfa..e99d55d74df 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -72,6 +72,9 @@ struct e820map { #define BIOS_BEGIN 0x000a0000 #define BIOS_END 0x00100000 +#define BIOS_ROM_BASE 0xffe00000 +#define BIOS_ROM_END 0xffffffff + #ifdef __KERNEL__ /* see comment in arch/x86/kernel/e820.c */ extern struct e820map e820; diff --git a/arch/x86/kernel/resource.c b/arch/x86/kernel/resource.c index 89638af2ff1..2a26819bb6a 100644 --- a/arch/x86/kernel/resource.c +++ b/arch/x86/kernel/resource.c @@ -37,10 +37,11 @@ static void remove_e820_regions(struct resource *avail) void arch_remove_reservations(struct resource *avail) { - /* Trim out BIOS area (low 1MB) and E820 regions */ + /* Trim out BIOS areas (low 1MB and high 2MB) and E820 regions */ if (avail->flags & IORESOURCE_MEM) { if (avail->start < BIOS_END) avail->start = BIOS_END; + resource_clip(avail, BIOS_ROM_BASE, BIOS_ROM_END); remove_e820_regions(avail); } -- cgit v1.2.3-70-g09d2