From 2749ebe320ff9f77548d10fcc0a3464ac21c8e58 Mon Sep 17 00:00:00 2001 From: Cliff Wickman Date: Thu, 29 Jan 2009 15:35:26 -0600 Subject: x86: UV fix uv_flush_send_and_wait() Impact: fix possible tlb mis-flushing on UV uv_flush_send_and_wait() should return a pointer if the broadcast remote tlb shootdown requests fail. That causes the conventional IPI method of shootdown to be used. Signed-off-by: Cliff Wickman Signed-off-by: Tejun Heo --- arch/x86/kernel/tlb_uv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 89fce1b..f4b2f27 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -259,7 +259,7 @@ const struct cpumask *uv_flush_send_and_wait(int cpu, int this_blade, * the cpu's, all of which are still in the mask. */ __get_cpu_var(ptcstats).ptc_i++; - return 0; + return flush_mask; } /* -- cgit v1.1 From 552be871e67ff577ed36beb2f53d078b42304739 Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Fri, 30 Jan 2009 17:47:53 +0900 Subject: x86: pass in cpu number to switch_to_new_gdt() Impact: cleanup, prepare for xen boot fix. Xen needs to call this function very early to setup the GDT and per-cpu segments. Remove the call to smp_processor_id() and just pass in the cpu number. Signed-off-by: Brian Gerst Signed-off-by: Tejun Heo --- arch/x86/include/asm/processor.h | 2 +- arch/x86/kernel/cpu/common.c | 7 +++---- arch/x86/kernel/setup_percpu.c | 2 +- arch/x86/kernel/smpboot.c | 2 +- arch/x86/mach-voyager/voyager_smp.c | 11 ++++++----- 5 files changed, 12 insertions(+), 12 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index befa20b..1c25eb6 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -768,7 +768,7 @@ extern int sysenter_setup(void); extern struct desc_ptr early_gdt_descr; extern void cpu_set_gdt(int); -extern void switch_to_new_gdt(void); +extern void switch_to_new_gdt(int); extern void cpu_init(void); static inline unsigned long get_debugctlmsr(void) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 652fdc9..6eacd64 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -255,10 +255,9 @@ __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata; /* Current gdt points %fs at the "master" per-cpu area: after this, * it's on the real one. */ -void switch_to_new_gdt(void) +void switch_to_new_gdt(int cpu) { struct desc_ptr gdt_descr; - int cpu = smp_processor_id(); gdt_descr.address = (long)get_cpu_gdt_table(cpu); gdt_descr.size = GDT_SIZE - 1; @@ -993,7 +992,7 @@ void __cpuinit cpu_init(void) * and set up the GDT descriptor: */ - switch_to_new_gdt(); + switch_to_new_gdt(cpu); loadsegment(fs, 0); load_idt((const struct desc_ptr *)&idt_descr); @@ -1098,7 +1097,7 @@ void __cpuinit cpu_init(void) clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE); load_idt(&idt_descr); - switch_to_new_gdt(); + switch_to_new_gdt(cpu); /* * Set up and load the per-CPU TSS and LDT diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 0d1e7ac..ef91747 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -122,7 +122,7 @@ void __init setup_per_cpu_areas(void) * area. Reload any changed state for the boot CPU. */ if (cpu == boot_cpu_id) - switch_to_new_gdt(); + switch_to_new_gdt(cpu); DBG("PERCPU: cpu %4d %p\n", cpu, ptr); } diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index f9dbcff..612d3c74 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1185,7 +1185,7 @@ out: void __init native_smp_prepare_boot_cpu(void) { int me = smp_processor_id(); - switch_to_new_gdt(); + switch_to_new_gdt(me); /* already set me in cpu_online_mask in boot_cpu_init() */ cpumask_set_cpu(me, cpu_callout_mask); per_cpu(cpu_state, me) = CPU_ONLINE; diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 331cd6d..58c7cac 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -1746,12 +1746,13 @@ static void __init voyager_smp_prepare_cpus(unsigned int max_cpus) static void __cpuinit voyager_smp_prepare_boot_cpu(void) { - switch_to_new_gdt(); + int cpu = smp_processor_id(); + switch_to_new_gdt(cpu); - cpu_set(smp_processor_id(), cpu_online_map); - cpu_set(smp_processor_id(), cpu_callout_map); - cpu_set(smp_processor_id(), cpu_possible_map); - cpu_set(smp_processor_id(), cpu_present_map); + cpu_set(cpu, cpu_online_map); + cpu_set(cpu, cpu_callout_map); + cpu_set(cpu, cpu_possible_map); + cpu_set(cpu, cpu_present_map); } static int __cpuinit voyager_cpu_up(unsigned int cpu) -- cgit v1.1 From 11e3a840cd5b731cdd8f6f956dfae78a8046d09c Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 30 Jan 2009 17:47:54 +0900 Subject: x86: split loading percpu segments from loading gdt Impact: split out a function, no functional change Xen needs to be able to access percpu data from very early on. For various reasons, it cannot also load the gdt at that time. It does, however, have a pefectly functional gdt at that point, so there's no pressing need to reload the gdt. Split the function to load the segment registers off, so Xen can call it directly. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Tejun Heo --- arch/x86/include/asm/processor.h | 1 + arch/x86/kernel/cpu/common.c | 18 ++++++++++++------ 2 files changed, 13 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 1c25eb6..656d02e 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -769,6 +769,7 @@ extern struct desc_ptr early_gdt_descr; extern void cpu_set_gdt(int); extern void switch_to_new_gdt(int); +extern void load_percpu_segment(int); extern void cpu_init(void); static inline unsigned long get_debugctlmsr(void) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 6eacd64..0f73ea4 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -253,6 +253,16 @@ static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c) __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata; +void load_percpu_segment(int cpu) +{ +#ifdef CONFIG_X86_32 + loadsegment(fs, __KERNEL_PERCPU); +#else + loadsegment(gs, 0); + wrmsrl(MSR_GS_BASE, (unsigned long)per_cpu(irq_stack_union.gs_base, cpu)); +#endif +} + /* Current gdt points %fs at the "master" per-cpu area: after this, * it's on the real one. */ void switch_to_new_gdt(int cpu) @@ -263,12 +273,8 @@ void switch_to_new_gdt(int cpu) gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); /* Reload the per-cpu base */ -#ifdef CONFIG_X86_32 - loadsegment(fs, __KERNEL_PERCPU); -#else - loadsegment(gs, 0); - wrmsrl(MSR_GS_BASE, (unsigned long)per_cpu(irq_stack_union.gs_base, cpu)); -#endif + + load_percpu_segment(cpu); } static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {}; -- cgit v1.1 From 795f99b61d20c34cb04d17d8906b32f745a635ec Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 30 Jan 2009 17:47:54 +0900 Subject: xen: setup percpu data pointers Impact: fix xen booting We need to access percpu data fairly early, so set up the percpu registers as soon as possible. We only need to load the appropriate segment register. We already have a GDT, but its hard to change it early because we need to manipulate the pagetable to do so, and that hasn't been set up yet. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Tejun Heo --- arch/x86/xen/enlighten.c | 3 +++ arch/x86/xen/smp.c | 5 ++++- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index bef941f..fe19c88 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1647,6 +1647,9 @@ asmlinkage void __init xen_start_kernel(void) have_vcpu_info_placement = 0; #endif + /* setup percpu state */ + load_percpu_segment(0); + xen_smp_init(); /* Get mfn list */ diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index 7735e3d..88d5d5e 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -170,7 +170,8 @@ static void __init xen_smp_prepare_boot_cpu(void) /* We've switched to the "real" per-cpu gdt, so make sure the old memory can be recycled */ - make_lowmem_page_readwrite(&per_cpu_var(gdt_page)); + make_lowmem_page_readwrite(__per_cpu_load + + (unsigned long)&per_cpu_var(gdt_page)); xen_setup_vcpu_info_placement(); } @@ -235,6 +236,8 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle) ctxt->user_regs.ss = __KERNEL_DS; #ifdef CONFIG_X86_32 ctxt->user_regs.fs = __KERNEL_PERCPU; +#else + ctxt->gs_base_kernel = per_cpu_offset(cpu); #endif ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle; ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */ -- cgit v1.1 From ef3892bd63420380d115f755d351d2071f1f805f Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 2 Feb 2009 18:16:19 -0800 Subject: x86, percpu: fix kexec with vmlinux Impact: fix regression with kexec with vmlinux Split data.init into data.init, percpu, data.init2 sections instead of let data.init wrap percpu secion. Thus kexec loading will be happy, because sections will not overlap. Before the patch we have: Elf file type is EXEC (Executable file) Entry point 0x200000 There are 6 program headers, starting at offset 64 Program Headers: Type Offset VirtAddr PhysAddr FileSiz MemSiz Flags Align LOAD 0x0000000000200000 0xffffffff80200000 0x0000000000200000 0x0000000000ca6000 0x0000000000ca6000 R E 200000 LOAD 0x0000000000ea6000 0xffffffff80ea6000 0x0000000000ea6000 0x000000000014dfe0 0x000000000014dfe0 RWE 200000 LOAD 0x0000000001000000 0xffffffffff600000 0x0000000000ff4000 0x0000000000000888 0x0000000000000888 RWE 200000 LOAD 0x00000000011f6000 0xffffffff80ff6000 0x0000000000ff6000 0x0000000000073086 0x0000000000a2d938 RWE 200000 LOAD 0x0000000001400000 0x0000000000000000 0x000000000106a000 0x00000000001d2ce0 0x00000000001d2ce0 RWE 200000 NOTE 0x00000000009e2c1c 0xffffffff809e2c1c 0x00000000009e2c1c 0x0000000000000024 0x0000000000000024 4 Section to Segment mapping: Segment Sections... 00 .text .notes __ex_table .rodata __bug_table .pci_fixup .builtin_fw __ksymtab __ksymtab_gpl __ksymtab_strings __init_rodata __param 01 .data .init.rodata .data.cacheline_aligned .data.read_mostly 02 .vsyscall_0 .vsyscall_fn .vsyscall_gtod_data .vsyscall_1 .vsyscall_2 .vgetcpu_mode .jiffies 03 .data.init_task .smp_locks .init.text .init.data .init.setup .initcall.init .con_initcall.init .x86_cpu_dev.init .altinstructions .altinstr_replacement .exit.text .init.ramfs .bss 04 .data.percpu 05 .notes After patch we've got: Elf file type is EXEC (Executable file) Entry point 0x200000 There are 7 program headers, starting at offset 64 Program Headers: Type Offset VirtAddr PhysAddr FileSiz MemSiz Flags Align LOAD 0x0000000000200000 0xffffffff80200000 0x0000000000200000 0x0000000000ca6000 0x0000000000ca6000 R E 200000 LOAD 0x0000000000ea6000 0xffffffff80ea6000 0x0000000000ea6000 0x000000000014dfe0 0x000000000014dfe0 RWE 200000 LOAD 0x0000000001000000 0xffffffffff600000 0x0000000000ff4000 0x0000000000000888 0x0000000000000888 RWE 200000 LOAD 0x00000000011f6000 0xffffffff80ff6000 0x0000000000ff6000 0x0000000000073086 0x0000000000073086 RWE 200000 LOAD 0x0000000001400000 0x0000000000000000 0x000000000106a000 0x00000000001d2ce0 0x00000000001d2ce0 RWE 200000 LOAD 0x000000000163d000 0xffffffff8123d000 0x000000000123d000 0x0000000000000000 0x00000000007e6938 RWE 200000 NOTE 0x00000000009e2c1c 0xffffffff809e2c1c 0x00000000009e2c1c 0x0000000000000024 0x0000000000000024 4 Section to Segment mapping: Segment Sections... 00 .text .notes __ex_table .rodata __bug_table .pci_fixup .builtin_fw __ksymtab __ksymtab_gpl __ksymtab_strings __init_rodata __param 01 .data .init.rodata .data.cacheline_aligned .data.read_mostly 02 .vsyscall_0 .vsyscall_fn .vsyscall_gtod_data .vsyscall_1 .vsyscall_2 .vgetcpu_mode .jiffies 03 .data.init_task .smp_locks .init.text .init.data .init.setup .initcall.init .con_initcall.init .x86_cpu_dev.init .altinstructions .altinstr_replacement .exit.text .init.ramfs 04 .data.percpu 05 .bss 06 .notes Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index c974099..07f62d2 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -22,6 +22,7 @@ PHDRS { #ifdef CONFIG_SMP percpu PT_LOAD FLAGS(7); /* RWE */ #endif + data.init2 PT_LOAD FLAGS(7); /* RWE */ note PT_NOTE FLAGS(0); /* ___ */ } SECTIONS @@ -215,7 +216,7 @@ SECTIONS /* * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the * output PHDR, so the next output section - __data_nosave - should - * switch it back to data.init. Also, pda should be at the head of + * start another section data.init2. Also, pda should be at the head of * percpu area. Preallocate it and define the percpu offset symbol * so that it can be accessed as a percpu variable. */ @@ -232,7 +233,7 @@ SECTIONS __nosave_begin = .; .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) - } :data.init /* switch back to data.init, see PERCPU_VADDR() above */ + } :data.init2 /* use another section data.init2, see PERCPU_VADDR() above */ . = ALIGN(PAGE_SIZE); __nosave_end = .; -- cgit v1.1