From 8ee2debce32412118cf8c239e0026ace56ea1425 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Fri, 25 Sep 2009 15:20:00 -0700 Subject: x86: Export k8 physical topology To eventually interleave emulated nodes over physical nodes, we need to know the physical topology of the machine without actually registering it. This does the k8 node setup in two parts: detection and registration. NUMA emulation can then used the physical topology detected to setup the address ranges of emulated nodes accordingly. If emulation isn't used, the k8 nodes are registered as normal. Two formals are added to the x86 NUMA setup functions: `acpi' and `k8'. These represent whether ACPI or K8 NUMA has been detected; both cannot be true at the same time. This specifies to the NUMA emulation code whether an underlying physical NUMA topology exists and which interface to use. This patch deals solely with separating the k8 setup path into Northbridge detection and registration steps and leaves the ACPI changes for a subsequent patch. The `acpi' formal is added here, however, to avoid touching all the header files again in the next patch. This approach also ensures emulated nodes will not span physical nodes so the true memory latency is not misrepresented. k8_get_nodes() may now be used to export the k8 physical topology of the machine for NUMA emulation. Signed-off-by: David Rientjes Cc: Andreas Herrmann Cc: Yinghai Lu Cc: Balbir Singh Cc: Ankita Garg Cc: Len Brown LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index e09f0e2c14b..fda0032c25c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -106,6 +106,7 @@ #include #include #include +#include #ifdef CONFIG_X86_64 #include #endif @@ -691,6 +692,9 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { void __init setup_arch(char **cmdline_p) { + int acpi = 0; + int k8 = 0; + #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); @@ -937,7 +941,11 @@ void __init setup_arch(char **cmdline_p) acpi_numa_init(); #endif - initmem_init(0, max_pfn); +#ifdef CONFIG_K8_NUMA + k8 = !k8_numa_init(0, max_pfn); +#endif + + initmem_init(0, max_pfn, acpi, k8); #ifdef CONFIG_ACPI_SLEEP /* -- cgit v1.2.3-70-g09d2 From 8716273caef7f55f39fe4fc6c69c5f9f197f41f1 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Fri, 25 Sep 2009 15:20:04 -0700 Subject: x86: Export srat physical topology This is the counterpart to "x86: export k8 physical topology" for SRAT. It is not as invasive because the acpi code already seperates node setup into detection and registration steps, with the exception of registering e820 active regions in acpi_numa_memory_affinity_init(). This is now moved to acpi_scan_nodes() if NUMA emulation is disabled or deferred. acpi_numa_init() now returns a value which specifies whether an underlying SRAT was located. If so, that topology can be used by the emulation code to interleave emulated nodes over physical nodes or to register the nodes for ACPI. acpi_get_nodes() may now be used to export the srat physical topology of the machine for NUMA emulation. Signed-off-by: David Rientjes Cc: Andreas Herrmann Cc: Yinghai Lu Cc: Balbir Singh Cc: Ankita Garg Cc: Len Brown LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/include/asm/acpi.h | 1 + arch/x86/kernel/setup.c | 5 +++-- arch/x86/mm/numa_64.c | 4 ++-- arch/x86/mm/srat_64.c | 28 +++++++++++++++++++++------- drivers/acpi/numa.c | 10 ++++++---- 5 files changed, 33 insertions(+), 15 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 4518dc50090..e3d4a0daff5 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -158,6 +158,7 @@ struct bootnode; #ifdef CONFIG_ACPI_NUMA extern int acpi_numa; +extern int acpi_get_nodes(struct bootnode *physnodes); extern int acpi_scan_nodes(unsigned long start, unsigned long end); #define NR_NODE_MEMBLKS (MAX_NUMNODES*2) extern void acpi_fake_nodes(const struct bootnode *fake_nodes, diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index fda0032c25c..f8914198270 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -938,11 +938,12 @@ void __init setup_arch(char **cmdline_p) /* * Parse SRAT to discover nodes. */ - acpi_numa_init(); + acpi = acpi_numa_init(); #endif #ifdef CONFIG_K8_NUMA - k8 = !k8_numa_init(0, max_pfn); + if (!acpi) + k8 = !k8_numa_init(0, max_pfn); #endif initmem_init(0, max_pfn, acpi, k8); diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index dad5f42dd35..d1a3d94efc8 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -540,8 +540,8 @@ void __init initmem_init(unsigned long start_pfn, unsigned long last_pfn, #endif #ifdef CONFIG_ACPI_NUMA - if (!numa_off && !acpi_scan_nodes(start_pfn << PAGE_SHIFT, - last_pfn << PAGE_SHIFT)) + if (!numa_off && acpi && !acpi_scan_nodes(start_pfn << PAGE_SHIFT, + last_pfn << PAGE_SHIFT)) return; nodes_clear(node_possible_map); nodes_clear(node_online_map); diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index dbb5381f7b3..891cbe65b2d 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -290,8 +290,6 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) printk(KERN_INFO "SRAT: Node %u PXM %u %lx-%lx\n", node, pxm, start, end); - e820_register_active_regions(node, start >> PAGE_SHIFT, - end >> PAGE_SHIFT); if (ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) { update_nodes_add(node, start, end); @@ -338,6 +336,19 @@ static int __init nodes_cover_memory(const struct bootnode *nodes) void __init acpi_numa_arch_fixup(void) {} +int __init acpi_get_nodes(struct bootnode *physnodes) +{ + int i; + int ret = 0; + + for_each_node_mask(i, nodes_parsed) { + physnodes[ret].start = nodes[i].start; + physnodes[ret].end = nodes[i].end; + ret++; + } + return ret; +} + /* Use the information discovered above to actually set up the nodes. */ int __init acpi_scan_nodes(unsigned long start, unsigned long end) { @@ -350,11 +361,6 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) for (i = 0; i < MAX_NUMNODES; i++) cutoff_node(i, start, end); - if (!nodes_cover_memory(nodes)) { - bad_srat(); - return -1; - } - memnode_shift = compute_hash_shift(node_memblk_range, num_node_memblks, memblk_nodeid); if (memnode_shift < 0) { @@ -364,6 +370,14 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) return -1; } + for_each_node_mask(i, nodes_parsed) + e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT, + nodes[i].end >> PAGE_SHIFT); + if (!nodes_cover_memory(nodes)) { + bad_srat(); + return -1; + } + /* Account for nodes with cpus and no memory */ nodes_or(node_possible_map, nodes_parsed, cpu_nodes_parsed); diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c index 202dd0c976a..2be2fb66204 100644 --- a/drivers/acpi/numa.c +++ b/drivers/acpi/numa.c @@ -283,22 +283,24 @@ acpi_table_parse_srat(enum acpi_srat_type id, int __init acpi_numa_init(void) { + int ret = 0; + /* SRAT: Static Resource Affinity Table */ if (!acpi_table_parse(ACPI_SIG_SRAT, acpi_parse_srat)) { acpi_table_parse_srat(ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY, acpi_parse_x2apic_affinity, NR_CPUS); acpi_table_parse_srat(ACPI_SRAT_TYPE_CPU_AFFINITY, acpi_parse_processor_affinity, NR_CPUS); - acpi_table_parse_srat(ACPI_SRAT_TYPE_MEMORY_AFFINITY, - acpi_parse_memory_affinity, - NR_NODE_MEMBLKS); + ret = acpi_table_parse_srat(ACPI_SRAT_TYPE_MEMORY_AFFINITY, + acpi_parse_memory_affinity, + NR_NODE_MEMBLKS); } /* SLIT: System Locality Information Table */ acpi_table_parse(ACPI_SIG_SLIT, acpi_parse_slit); acpi_numa_arch_fixup(); - return 0; + return ret; } int acpi_get_pxm(acpi_handle h) -- cgit v1.2.3-70-g09d2 From a2202aa29289db64ca7988b12343158b67b27f10 Mon Sep 17 00:00:00 2001 From: Yong Wang Date: Tue, 10 Nov 2009 09:38:24 +0800 Subject: x86: Under BIOS control, restore AP's APIC_LVTTHMR to the BSP value On platforms where the BIOS handles the thermal monitor interrupt, APIC_LVTTHMR on each logical CPU is programmed to generate a SMI and OS must not touch it. Unfortunately AP bringup sequence using INIT-SIPI-SIPI clears all the LVT entries except the mask bit. Essentially this results in all LVT entries including the thermal monitoring interrupt set to masked (clearing the bios programmed value for APIC_LVTTHMR). And this leads to kernel take over the thermal monitoring interrupt on AP's but not on BSP (leaving the bios programmed value only on BSP). As a result of this, we have seen system hangs when the thermal monitoring interrupt is generated. Fix this by reading the initial value of thermal LVT entry on BSP and if bios has taken over the control, then program the same value on all AP's and leave the thermal monitoring interrupt control on all the logical cpu's to the bios. Signed-off-by: Yong Wang Reviewed-by: Suresh Siddha Cc: Borislav Petkov Cc: Arjan van de Ven LKML-Reference: <20091110013824.GA24940@ywang-moblin2.bj.intel.com> Signed-off-by: Ingo Molnar Cc: stable@kernel.org --- arch/x86/include/asm/mce.h | 9 +++++++++ arch/x86/kernel/cpu/common.c | 2 -- arch/x86/kernel/cpu/mcheck/mce.c | 5 +++-- arch/x86/kernel/cpu/mcheck/therm_throt.c | 29 ++++++++++++++++++++++++++++- arch/x86/kernel/setup.c | 3 +++ 5 files changed, 43 insertions(+), 5 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 161485da683..858baa061cf 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -120,8 +120,10 @@ extern int mce_disabled; extern int mce_p5_enabled; #ifdef CONFIG_X86_MCE +int mcheck_init(void); void mcheck_cpu_init(struct cpuinfo_x86 *c); #else +static inline int mcheck_init(void) { return 0; } static inline void mcheck_cpu_init(struct cpuinfo_x86 *c) {} #endif @@ -215,5 +217,12 @@ extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); void intel_init_thermal(struct cpuinfo_x86 *c); void mce_log_therm_throt_event(__u64 status); + +#ifdef CONFIG_X86_THERMAL_VECTOR +extern void mcheck_intel_therm_init(void); +#else +static inline void mcheck_intel_therm_init(void) { } +#endif + #endif /* __KERNEL__ */ #endif /* _ASM_X86_MCE_H */ diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 4df69a38be5..9053be5d95c 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -837,10 +837,8 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c) boot_cpu_data.x86_capability[i] &= c->x86_capability[i]; } -#ifdef CONFIG_X86_MCE /* Init Machine Check Exception if available. */ mcheck_cpu_init(c); -#endif select_idle_routine(c); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 80801705edd..0d4102031a4 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1655,13 +1655,14 @@ static int __init mcheck_enable(char *str) } __setup("mce", mcheck_enable); -static int __init mcheck_init(void) +int __init mcheck_init(void) { atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb); + mcheck_intel_therm_init(); + return 0; } -early_initcall(mcheck_init); /* * Sysfs support diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index b3a1dba7533..7f3cf36ed12 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -49,6 +49,8 @@ static DEFINE_PER_CPU(struct thermal_state, thermal_state); static atomic_t therm_throt_en = ATOMIC_INIT(0); +static u32 lvtthmr_init __read_mostly; + #ifdef CONFIG_SYSFS #define define_therm_throt_sysdev_one_ro(_name) \ static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL) @@ -254,6 +256,18 @@ asmlinkage void smp_thermal_interrupt(struct pt_regs *regs) ack_APIC_irq(); } +void mcheck_intel_therm_init(void) +{ + /* + * This function is only called on boot CPU. Save the init thermal + * LVT value on BSP and use that value to restore APs' thermal LVT + * entry BIOS programmed later + */ + if (cpu_has(&boot_cpu_data, X86_FEATURE_ACPI) && + cpu_has(&boot_cpu_data, X86_FEATURE_ACC)) + lvtthmr_init = apic_read(APIC_LVTTHMR); +} + void intel_init_thermal(struct cpuinfo_x86 *c) { unsigned int cpu = smp_processor_id(); @@ -270,7 +284,20 @@ void intel_init_thermal(struct cpuinfo_x86 *c) * since it might be delivered via SMI already: */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); + + /* + * The initial value of thermal LVT entries on all APs always reads + * 0x10000 because APs are woken up by BSP issuing INIT-SIPI-SIPI + * sequence to them and LVT registers are reset to 0s except for + * the mask bits which are set to 1s when APs receive INIT IPI. + * Always restore the value that BIOS has programmed on AP based on + * BSP's info we saved since BIOS is always setting the same value + * for all threads/cores + */ + apic_write(APIC_LVTTHMR, lvtthmr_init); + + h = lvtthmr_init; + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { printk(KERN_DEBUG "CPU%d: Thermal monitoring handled by SMI\n", cpu); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index e09f0e2c14b..179c1f2aa45 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -109,6 +109,7 @@ #ifdef CONFIG_X86_64 #include #endif +#include /* * end_pfn only includes RAM, while max_pfn_mapped includes all e820 entries. @@ -1024,6 +1025,8 @@ void __init setup_arch(char **cmdline_p) #endif #endif x86_init.oem.banner(); + + mcheck_init(); } #ifdef CONFIG_X86_32 -- cgit v1.2.3-70-g09d2 From 196cf0d67acad70ebb2572da489d5cc7066cdd05 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 10 Nov 2009 18:27:23 -0800 Subject: x86: Make sure wakeup trampoline code is below 1MB Instead of using bootmem, try find_e820_area()/reserve_early(), and call acpi_reserve_memory() early, to allocate the wakeup trampoline code area below 1M. This is more reliable, and it also removes a dependency on bootmem. -v2: change function name to acpi_reserve_wakeup_memory(), as suggested by Rafael. Signed-off-by: Yinghai Lu Acked-by: H. Peter Anvin Acked-by: Rafael J. Wysocki Cc: pm list Cc: Len Brown Cc: Linus Torvalds LKML-Reference: <4AFA210B.3020207@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/acpi.h | 2 +- arch/x86/kernel/acpi/sleep.c | 15 +++++++++------ arch/x86/kernel/setup.c | 13 +++++++------ 3 files changed, 17 insertions(+), 13 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index e3d4a0daff5..60d2b2db0bc 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -118,7 +118,7 @@ extern void acpi_restore_state_mem(void); extern unsigned long acpi_wakeup_address; /* early initialization routine */ -extern void acpi_reserve_bootmem(void); +extern void acpi_reserve_wakeup_memory(void); /* * Check if the CPU can handle C2 and deeper diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index ca93638ba43..4a411450dfa 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -119,29 +119,32 @@ void acpi_restore_state_mem(void) /** - * acpi_reserve_bootmem - do _very_ early ACPI initialisation + * acpi_reserve_wakeup_memory - do _very_ early ACPI initialisation * * We allocate a page from the first 1MB of memory for the wakeup * routine for when we come back from a sleep state. The * runtime allocator allows specification of <16MB pages, but not * <1MB pages. */ -void __init acpi_reserve_bootmem(void) +void __init acpi_reserve_wakeup_memory(void) { + unsigned long mem; + if ((&wakeup_code_end - &wakeup_code_start) > WAKEUP_SIZE) { printk(KERN_ERR "ACPI: Wakeup code way too big, S3 disabled.\n"); return; } - acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE); + mem = find_e820_area(0, 1<<20, WAKEUP_SIZE, PAGE_SIZE); - if (!acpi_realmode) { + if (mem == -1L) { printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); return; } - - acpi_wakeup_address = virt_to_phys((void *)acpi_realmode); + acpi_realmode = (unsigned long) phys_to_virt(mem); + acpi_wakeup_address = mem; + reserve_early(mem, mem + WAKEUP_SIZE, "ACPI WAKEUP"); } diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index f8914198270..0a6e94ab833 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -897,6 +897,13 @@ void __init setup_arch(char **cmdline_p) reserve_brk(); +#ifdef CONFIG_ACPI_SLEEP + /* + * Reserve low memory region for sleep support. + * even before init_memory_mapping + */ + acpi_reserve_wakeup_memory(); +#endif init_gbpages(); /* max_pfn_mapped is updated here */ @@ -948,12 +955,6 @@ void __init setup_arch(char **cmdline_p) initmem_init(0, max_pfn, acpi, k8); -#ifdef CONFIG_ACPI_SLEEP - /* - * Reserve low memory region for sleep support. - */ - acpi_reserve_bootmem(); -#endif /* * Find and reserve possible boot-time SMP configuration: */ -- cgit v1.2.3-70-g09d2 From 4763ed4d45522b876c97e1f7f4b659d211f75571 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 13 Nov 2009 15:28:16 -0800 Subject: x86, mm: Clean up and simplify NX enablement The 32- and 64-bit code used very different mechanisms for enabling NX, but even the 32-bit code was enabling NX in head_32.S if it is available. Furthermore, we had a bewildering collection of tests for the available of NX. This patch: a) merges the 32-bit set_nx() and the 64-bit check_efer() function into a single x86_configure_nx() function. EFER control is left to the head code. b) eliminates the nx_enabled variable entirely. Things that need to test for NX enablement can verify __supported_pte_mask directly, and cpu_has_nx gives the supported status of NX. Signed-off-by: H. Peter Anvin Cc: Tejun Heo Cc: Brian Gerst Cc: Yinghai Lu Cc: Pekka Enberg Cc: Vegard Nossum Cc: Jeremy Fitzhardinge Cc: Chris Wright LKML-Reference: <1258154897-6770-5-git-send-email-hpa@zytor.com> Acked-by: Kees Cook --- arch/x86/include/asm/proto.h | 2 +- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/setup.c | 8 ++------ arch/x86/mm/init.c | 4 ++-- arch/x86/mm/setup_nx.c | 43 ++++++------------------------------------- arch/x86/xen/enlighten.c | 4 +--- 6 files changed, 13 insertions(+), 50 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/proto.h b/arch/x86/include/asm/proto.h index 621f56d7312..add7f18f17a 100644 --- a/arch/x86/include/asm/proto.h +++ b/arch/x86/include/asm/proto.h @@ -16,7 +16,7 @@ extern void ia32_sysenter_target(void); extern void syscall32_cpu_init(void); -extern void check_efer(void); +extern void x86_configure_nx(void); extern int reboot_force; diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index cc25c2b4a56..18346da8c59 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1136,7 +1136,7 @@ void __cpuinit cpu_init(void) wrmsrl(MSR_KERNEL_GS_BASE, 0); barrier(); - check_efer(); + x86_configure_nx(); if (cpu != 0) enable_x2apic(); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 0a6e94ab833..23b7f46bf84 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -787,21 +787,17 @@ void __init setup_arch(char **cmdline_p) strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; -#ifdef CONFIG_X86_64 /* * Must call this twice: Once just to detect whether hardware doesn't * support NX (so that the early EHCI debug console setup can safely * call set_fixmap(), and then again after parsing early parameters to * honor the respective command line option. */ - check_efer(); -#endif + x86_configure_nx(); parse_early_param(); -#ifdef CONFIG_X86_64 - check_efer(); -#endif + x86_configure_nx(); /* Must be before kernel pagetables are setup */ vmi_activate(); diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 73ffd5536f6..27ec2c23fd4 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -146,8 +146,8 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, use_gbpages = direct_gbpages; #endif - set_nx(); - if (nx_enabled) + /* XXX: replace this with Kees' improved messages */ + if (__supported_pte_mask & _PAGE_NX) printk(KERN_INFO "NX (Execute Disable) protection: active\n"); /* Enable PSE if available */ diff --git a/arch/x86/mm/setup_nx.c b/arch/x86/mm/setup_nx.c index 513d8ed5d2e..355818b087b 100644 --- a/arch/x86/mm/setup_nx.c +++ b/arch/x86/mm/setup_nx.c @@ -3,10 +3,8 @@ #include #include +#include -int nx_enabled; - -#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) static int disable_nx __cpuinitdata; /* @@ -22,48 +20,19 @@ static int __init noexec_setup(char *str) if (!str) return -EINVAL; if (!strncmp(str, "on", 2)) { - __supported_pte_mask |= _PAGE_NX; disable_nx = 0; } else if (!strncmp(str, "off", 3)) { disable_nx = 1; - __supported_pte_mask &= ~_PAGE_NX; } + x86_configure_nx(); return 0; } early_param("noexec", noexec_setup); -#endif - -#ifdef CONFIG_X86_PAE -void __init set_nx(void) -{ - unsigned int v[4], l, h; - - if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) { - cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]); - - if ((v[3] & (1 << 20)) && !disable_nx) { - rdmsr(MSR_EFER, l, h); - l |= EFER_NX; - wrmsr(MSR_EFER, l, h); - nx_enabled = 1; - __supported_pte_mask |= _PAGE_NX; - } - } -} -#else -void set_nx(void) -{ -} -#endif -#ifdef CONFIG_X86_64 -void __cpuinit check_efer(void) +void __cpuinit x86_configure_nx(void) { - unsigned long efer; - - rdmsrl(MSR_EFER, efer); - if (!(efer & EFER_NX) || disable_nx) + if (cpu_has_nx && !disable_nx) + __supported_pte_mask |= _PAGE_NX; + else __supported_pte_mask &= ~_PAGE_NX; } -#endif - diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 3439616d69f..c5e805d4a78 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1082,10 +1082,8 @@ asmlinkage void __init xen_start_kernel(void) __supported_pte_mask |= _PAGE_IOMAP; -#ifdef CONFIG_X86_64 /* Work out if we support NX */ - check_efer(); -#endif + x86_configure_nx(); xen_setup_features(); -- cgit v1.2.3-70-g09d2 From 4b0f3b81eb33ef18283aa71440cccfede1753ae0 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 13 Nov 2009 15:28:17 -0800 Subject: x86, mm: Report state of NX protections during boot It is possible for x86_64 systems to lack the NX bit either due to the hardware lacking support or the BIOS having turned off the CPU capability, so NX status should be reported. Additionally, anyone booting NX-capable CPUs in 32bit mode without PAE will lack NX functionality, so this change provides feedback for that case as well. Signed-off-by: Kees Cook Signed-off-by: H. Peter Anvin LKML-Reference: <1258154897-6770-6-git-send-email-hpa@zytor.com> --- arch/x86/include/asm/proto.h | 1 + arch/x86/kernel/setup.c | 11 ++++++----- arch/x86/mm/init.c | 4 ---- arch/x86/mm/setup_nx.c | 22 ++++++++++++++++++++++ 4 files changed, 29 insertions(+), 9 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/proto.h b/arch/x86/include/asm/proto.h index add7f18f17a..450c56bcd4f 100644 --- a/arch/x86/include/asm/proto.h +++ b/arch/x86/include/asm/proto.h @@ -17,6 +17,7 @@ extern void ia32_sysenter_target(void); extern void syscall32_cpu_init(void); extern void x86_configure_nx(void); +extern void x86_report_nx(void); extern int reboot_force; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 23b7f46bf84..d2043a00abc 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -788,16 +788,17 @@ void __init setup_arch(char **cmdline_p) *cmdline_p = command_line; /* - * Must call this twice: Once just to detect whether hardware doesn't - * support NX (so that the early EHCI debug console setup can safely - * call set_fixmap(), and then again after parsing early parameters to - * honor the respective command line option. + * x86_configure_nx() is called before parse_early_param() to detect + * whether hardware doesn't support NX (so that the early EHCI debug + * console setup can safely call set_fixmap()). It may then be called + * again from within noexec_setup() during parsing early parameters + * to honor the respective command line option. */ x86_configure_nx(); parse_early_param(); - x86_configure_nx(); + x86_report_nx(); /* Must be before kernel pagetables are setup */ vmi_activate(); diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 27ec2c23fd4..d406c523901 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -146,10 +146,6 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, use_gbpages = direct_gbpages; #endif - /* XXX: replace this with Kees' improved messages */ - if (__supported_pte_mask & _PAGE_NX) - printk(KERN_INFO "NX (Execute Disable) protection: active\n"); - /* Enable PSE if available */ if (cpu_has_pse) set_in_cr4(X86_CR4_PSE); diff --git a/arch/x86/mm/setup_nx.c b/arch/x86/mm/setup_nx.c index 355818b087b..a3250aa3408 100644 --- a/arch/x86/mm/setup_nx.c +++ b/arch/x86/mm/setup_nx.c @@ -36,3 +36,25 @@ void __cpuinit x86_configure_nx(void) else __supported_pte_mask &= ~_PAGE_NX; } + +void __init x86_report_nx(void) +{ + if (!cpu_has_nx) { + printk(KERN_NOTICE "Notice: NX (Execute Disable) protection " + "missing in CPU or disabled in BIOS!\n"); + } else { +#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + if (disable_nx) { + printk(KERN_INFO "NX (Execute Disable) protection: " + "disabled by kernel command line option\n"); + } else { + printk(KERN_INFO "NX (Execute Disable) protection: " + "active\n"); + } +#else + /* 32bit non-PAE kernel, NX cannot be used */ + printk(KERN_NOTICE "Notice: NX (Execute Disable) protection " + "cannot be enabled: non-PAE kernel!\n"); +#endif + } +} -- cgit v1.2.3-70-g09d2 From 44280733e71ad15377735b42d8538c109c94d7e3 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 22 Nov 2009 17:18:49 -0800 Subject: x86: Change crash kernel to reserve via reserve_early() use find_e820_area()/reserve_early() instead. -v2: address Eric's request, to restore original semantics. will fail, if the provided address can not be used. Signed-off-by: Yinghai Lu Acked-by: Eric W. Biederman LKML-Reference: <4B09E2F9.7040403@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 57 +++++++++++++------------------------------------ 1 file changed, 15 insertions(+), 42 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index d2043a00abc..e3eae5965e4 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -487,42 +487,11 @@ static void __init reserve_early_setup_data(void) #ifdef CONFIG_KEXEC -/** - * Reserve @size bytes of crashkernel memory at any suitable offset. - * - * @size: Size of the crashkernel memory to reserve. - * Returns the base address on success, and -1ULL on failure. - */ -static -unsigned long long __init find_and_reserve_crashkernel(unsigned long long size) -{ - const unsigned long long alignment = 16<<20; /* 16M */ - unsigned long long start = 0LL; - - while (1) { - int ret; - - start = find_e820_area(start, ULONG_MAX, size, alignment); - if (start == -1ULL) - return start; - - /* try to reserve it */ - ret = reserve_bootmem_generic(start, size, BOOTMEM_EXCLUSIVE); - if (ret >= 0) - return start; - - start += alignment; - } -} - static inline unsigned long long get_total_mem(void) { unsigned long long total; - total = max_low_pfn - min_low_pfn; -#ifdef CONFIG_HIGHMEM - total += highend_pfn - highstart_pfn; -#endif + total = max_pfn - min_low_pfn; return total << PAGE_SHIFT; } @@ -542,21 +511,25 @@ static void __init reserve_crashkernel(void) /* 0 means: find the address automatically */ if (crash_base <= 0) { - crash_base = find_and_reserve_crashkernel(crash_size); + const unsigned long long alignment = 16<<20; /* 16M */ + + crash_base = find_e820_area(alignment, ULONG_MAX, crash_size, + alignment); if (crash_base == -1ULL) { - pr_info("crashkernel reservation failed. " - "No suitable area found.\n"); + pr_info("crashkernel reservation failed - No suitable area found.\n"); return; } } else { - ret = reserve_bootmem_generic(crash_base, crash_size, - BOOTMEM_EXCLUSIVE); - if (ret < 0) { - pr_info("crashkernel reservation failed - " - "memory is in use\n"); + unsigned long long start; + + start = find_e820_area(crash_base, ULONG_MAX, crash_size, + 1<<20); + if (start != crash_base) { + pr_info("crashkernel reservation failed - memory is in use.\n"); return; } } + reserve_early(crash_base, crash_base + crash_size, "CRASH KERNEL"); printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " "for crashkernel (System RAM: %ldMB)\n", @@ -927,6 +900,8 @@ void __init setup_arch(char **cmdline_p) reserve_initrd(); + reserve_crashkernel(); + vsmp_init(); io_delay_init(); @@ -957,8 +932,6 @@ void __init setup_arch(char **cmdline_p) */ find_smp_config(); - reserve_crashkernel(); - #ifdef CONFIG_X86_64 /* * dma32_reserve_bootmem() allocates bootmem which may conflict -- cgit v1.2.3-70-g09d2 From b24c2a925a9837cccf54d50aeac22ba0cbc15455 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Tue, 24 Nov 2009 02:48:18 -0800 Subject: x86: Move find_smp_config() earlier and avoid bootmem usage Move the find_smp_config() call to before bootmem is initialized. Use reserve_early() instead of reserve_bootmem() in it. This simplifies the code, we only need to call find_smp_config() once and can remove the now unneeded reserve parameter from x86_init_mpparse::find_smp_config. We thus also reduce x86's dependency on bootmem allocations. Signed-off-by: Yinghai Lu LKML-Reference: <4B0BB9F2.70907@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mpspec.h | 11 +++-------- arch/x86/include/asm/x86_init.h | 2 +- arch/x86/kernel/apic/numaq_32.c | 5 ----- arch/x86/kernel/mpparse.c | 44 +++++++++++------------------------------ arch/x86/kernel/setup.c | 10 +++++----- arch/x86/kernel/visws_quirks.c | 2 +- arch/x86/mm/k8topology_64.c | 12 ----------- 7 files changed, 22 insertions(+), 64 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/include/asm/mpspec.h b/arch/x86/include/asm/mpspec.h index 79c94500c0b..644cf1a50bf 100644 --- a/arch/x86/include/asm/mpspec.h +++ b/arch/x86/include/asm/mpspec.h @@ -71,12 +71,7 @@ static inline void early_get_smp_config(void) static inline void find_smp_config(void) { - x86_init.mpparse.find_smp_config(1); -} - -static inline void early_find_smp_config(void) -{ - x86_init.mpparse.find_smp_config(0); + x86_init.mpparse.find_smp_config(); } #ifdef CONFIG_X86_MPPARSE @@ -89,7 +84,7 @@ extern void default_mpc_oem_bus_info(struct mpc_bus *m, char *str); # else # define default_mpc_oem_bus_info NULL # endif -extern void default_find_smp_config(unsigned int reserve); +extern void default_find_smp_config(void); extern void default_get_smp_config(unsigned int early); #else static inline void early_reserve_e820_mpc_new(void) { } @@ -97,7 +92,7 @@ static inline void early_reserve_e820_mpc_new(void) { } #define default_mpc_apic_id NULL #define default_smp_read_mpc_oem NULL #define default_mpc_oem_bus_info NULL -#define default_find_smp_config x86_init_uint_noop +#define default_find_smp_config x86_init_noop #define default_get_smp_config x86_init_uint_noop #endif diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index 024cf3c1fd8..97e5fb4f3bd 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -26,7 +26,7 @@ struct x86_init_mpparse { void (*smp_read_mpc_oem)(struct mpc_table *mpc); void (*mpc_oem_pci_bus)(struct mpc_bus *m); void (*mpc_oem_bus_info)(struct mpc_bus *m, char *name); - void (*find_smp_config)(unsigned int reserve); + void (*find_smp_config)(void); void (*get_smp_config)(unsigned int early); }; diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index efa00e2b850..9c0629ceb52 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -263,11 +263,6 @@ static void __init smp_read_mpc_oem(struct mpc_table *mpc) static __init void early_check_numaq(void) { - /* - * Find possible boot-time SMP configuration: - */ - early_find_smp_config(); - /* * get boot-time SMP configuration: */ diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 5be95ef4ffe..35a57c963df 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -667,36 +667,18 @@ void __init default_get_smp_config(unsigned int early) */ } -static void __init smp_reserve_bootmem(struct mpf_intel *mpf) +static void __init smp_reserve_memory(struct mpf_intel *mpf) { unsigned long size = get_mpc_size(mpf->physptr); -#ifdef CONFIG_X86_32 - /* - * We cannot access to MPC table to compute table size yet, - * as only few megabytes from the bottom is mapped now. - * PC-9800's MPC table places on the very last of physical - * memory; so that simply reserving PAGE_SIZE from mpf->physptr - * yields BUG() in reserve_bootmem. - * also need to make sure physptr is below than max_low_pfn - * we don't need reserve the area above max_low_pfn - */ - unsigned long end = max_low_pfn * PAGE_SIZE; - if (mpf->physptr < end) { - if (mpf->physptr + size > end) - size = end - mpf->physptr; - reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); - } -#else - reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); -#endif + reserve_early(mpf->physptr, mpf->physptr+size, "MP-table mpc"); } -static int __init smp_scan_config(unsigned long base, unsigned long length, - unsigned reserve) +static int __init smp_scan_config(unsigned long base, unsigned long length) { unsigned int *bp = phys_to_virt(base); struct mpf_intel *mpf; + unsigned long mem; apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n", bp, length); @@ -717,12 +699,10 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, printk(KERN_INFO "found SMP MP-table at [%p] %llx\n", mpf, (u64)virt_to_phys(mpf)); - if (!reserve) - return 1; - reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf), - BOOTMEM_DEFAULT); + mem = virt_to_phys(mpf); + reserve_early(mem, mem + sizeof(*mpf), "MP-table mpf"); if (mpf->physptr) - smp_reserve_bootmem(mpf); + smp_reserve_memory(mpf); return 1; } @@ -732,7 +712,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, return 0; } -void __init default_find_smp_config(unsigned int reserve) +void __init default_find_smp_config(void) { unsigned int address; @@ -744,9 +724,9 @@ void __init default_find_smp_config(unsigned int reserve) * 2) Scan the top 1K of base RAM * 3) Scan the 64K of bios */ - if (smp_scan_config(0x0, 0x400, reserve) || - smp_scan_config(639 * 0x400, 0x400, reserve) || - smp_scan_config(0xF0000, 0x10000, reserve)) + if (smp_scan_config(0x0, 0x400) || + smp_scan_config(639 * 0x400, 0x400) || + smp_scan_config(0xF0000, 0x10000)) return; /* * If it is an SMP machine we should know now, unless the @@ -767,7 +747,7 @@ void __init default_find_smp_config(unsigned int reserve) address = get_bios_ebda(); if (address) - smp_scan_config(address, 0x400, reserve); + smp_scan_config(address, 0x400); } #ifdef CONFIG_X86_IO_APIC diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index e3eae5965e4..cdb6a8a506d 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -913,6 +913,11 @@ void __init setup_arch(char **cmdline_p) early_acpi_boot_init(); + /* + * Find and reserve possible boot-time SMP configuration: + */ + find_smp_config(); + #ifdef CONFIG_ACPI_NUMA /* * Parse SRAT to discover nodes. @@ -927,11 +932,6 @@ void __init setup_arch(char **cmdline_p) initmem_init(0, max_pfn, acpi, k8); - /* - * Find and reserve possible boot-time SMP configuration: - */ - find_smp_config(); - #ifdef CONFIG_X86_64 /* * dma32_reserve_bootmem() allocates bootmem which may conflict diff --git a/arch/x86/kernel/visws_quirks.c b/arch/x86/kernel/visws_quirks.c index f068553a1b1..1498efa964b 100644 --- a/arch/x86/kernel/visws_quirks.c +++ b/arch/x86/kernel/visws_quirks.c @@ -197,7 +197,7 @@ static void __init MP_processor_info(struct mpc_cpu *m) apic_version[m->apicid] = ver; } -static void __init visws_find_smp_config(unsigned int reserve) +static void __init visws_find_smp_config(void) { struct mpc_cpu *mp = phys_to_virt(CO_CPU_TAB_PHYS); unsigned short ncpus = readw(phys_to_virt(CO_CPU_NUM_PHYS)); diff --git a/arch/x86/mm/k8topology_64.c b/arch/x86/mm/k8topology_64.c index b9e2dbfe55c..970ed579d4e 100644 --- a/arch/x86/mm/k8topology_64.c +++ b/arch/x86/mm/k8topology_64.c @@ -57,18 +57,6 @@ static __init void early_get_boot_cpu_id(void) * need to get boot_cpu_id so can use that to create apicid_to_node * in k8_scan_nodes() */ - /* - * Find possible boot-time SMP configuration: - */ -#ifdef CONFIG_X86_MPPARSE - early_find_smp_config(); -#endif -#ifdef CONFIG_ACPI - /* - * Read APIC information from ACPI tables. - */ - early_acpi_boot_init(); -#endif #ifdef CONFIG_X86_MPPARSE /* * get boot-time SMP configuration: -- cgit v1.2.3-70-g09d2 From 9eaa192d8988d621217a9e6071cd403fd6010496 Mon Sep 17 00:00:00 2001 From: "Helight.Xu" Date: Mon, 30 Nov 2009 18:33:51 +0800 Subject: x86: Fix a section mismatch in arch/x86/kernel/setup.c copy_edd() should be __init. warning msg: WARNING: vmlinux.o(.text+0x7759): Section mismatch in reference from the function copy_edd() to the variable .init.data:boot_params The function copy_edd() references the variable __initdata boot_params. This is often because copy_edd lacks a __initdata annotation or the annotation of boot_params is wrong. Signed-off-by: ZhenwenXu LKML-Reference: <4B139F8F.4000907@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/setup.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/setup.c') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index e09f0e2c14b..e020b2d03b4 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -247,7 +247,7 @@ EXPORT_SYMBOL(edd); * from boot_params into a safe place. * */ -static inline void copy_edd(void) +static inline void __init copy_edd(void) { memcpy(edd.mbr_signature, boot_params.edd_mbr_sig_buffer, sizeof(edd.mbr_signature)); @@ -256,7 +256,7 @@ static inline void copy_edd(void) edd.edd_info_nr = boot_params.eddbuf_entries; } #else -static inline void copy_edd(void) +static inline void __init copy_edd(void) { } #endif -- cgit v1.2.3-70-g09d2