From 3e5d8f978435bb9ba4dfe3f4514e65e7885db1a9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 Jan 2009 20:41:35 +0900 Subject: x86: make percpu symbols zerobased on SMP [ Based on original patch from Christoph Lameter and Mike Travis. ] This patch makes percpu symbols zerobased on x86_64 SMP by adding PERCPU_VADDR() to vmlinux.lds.h which helps setting explicit vaddr on the percpu output section and using it in vmlinux_64.lds.S. A new PHDR is added as existing ones cannot contain sections near address zero. PERCPU_VADDR() also adds a new symbol __per_cpu_load which always points to the vaddr of the loaded percpu data.init region. The following adjustments have been made to accomodate the address change. * code to locate percpu gdt_page in head_64.S is updated to add the load address to the gdt_page offset. * __per_cpu_load is used in places where access to the init data area is necessary. * pda->data_offset is initialized soon after C code is entered as zero value doesn't work anymore. This patch is mostly taken from Mike Travis' "x86_64: Base percpu variables at zero" patch. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 1a614c0e6be..f50280db0df 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -19,6 +19,9 @@ PHDRS { data PT_LOAD FLAGS(7); /* RWE */ user PT_LOAD FLAGS(7); /* RWE */ data.init PT_LOAD FLAGS(7); /* RWE */ +#ifdef CONFIG_SMP + percpu PT_LOAD FLAGS(7); /* RWE */ +#endif note PT_NOTE FLAGS(0); /* ___ */ } SECTIONS @@ -208,14 +211,26 @@ SECTIONS __initramfs_end = .; #endif +#ifdef CONFIG_SMP + /* + * 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. + */ + . = ALIGN(PAGE_SIZE); + PERCPU_VADDR(0, :percpu) +#else PERCPU(PAGE_SIZE) +#endif . = ALIGN(PAGE_SIZE); __init_end = .; . = ALIGN(PAGE_SIZE); __nosave_begin = .; - .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { *(.data.nosave) } + .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { + *(.data.nosave) + } :data.init /* switch back to data.init, see PERCPU_VADDR() above */ . = ALIGN(PAGE_SIZE); __nosave_end = .; -- cgit v1.2.3 From 1a51e3a0aed18767cf2762e95456ecfeb0bca5e6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 Jan 2009 20:41:35 +0900 Subject: x86: fold pda into percpu area on SMP [ Based on original patch from Christoph Lameter and Mike Travis. ] Currently pdas and percpu areas are allocated separately. %gs points to local pda and percpu area can be reached using pda->data_offset. This patch folds pda into percpu area. Due to strange gcc requirement, pda needs to be at the beginning of the percpu area so that pda->stack_canary is at %gs:40. To achieve this, a new percpu output section macro - PERCPU_VADDR_PREALLOC() - is added and used to reserve pda sized chunk at the start of the percpu area. After this change, for boot cpu, %gs first points to pda in the data.init area and later during setup_per_cpu_areas() gets updated to point to the actual pda. This means that setup_per_cpu_areas() need to reload %gs for CPU0 while clearing pda area for other cpus as cpu0 already has modified it when control reaches setup_per_cpu_areas(). This patch also removes now unnecessary get_local_pda() and its call sites. A lot of this patch is taken from Mike Travis' "x86_64: Fold pda into per cpu area" patch. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index f50280db0df..962f21f1d4d 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -5,6 +5,7 @@ #define LOAD_OFFSET __START_KERNEL_map #include +#include #include #undef i386 /* in case the preprocessor is a 32bit one */ @@ -215,10 +216,11 @@ 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. + * switch it back to data.init. Also, pda should be at the head of + * percpu area. Preallocate it. */ . = ALIGN(PAGE_SIZE); - PERCPU_VADDR(0, :percpu) + PERCPU_VADDR_PREALLOC(0, :percpu, pda_size) #else PERCPU(PAGE_SIZE) #endif -- cgit v1.2.3 From b12d8db8fbfaed1e8222a15333a3645599636854 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 Jan 2009 20:41:35 +0900 Subject: x86: make pda a percpu variable [ Based on original patch from Christoph Lameter and Mike Travis. ] As pda is now allocated in percpu area, it can easily be made a proper percpu variable. Make it so by defining per cpu symbol from linker script and declaring it in C code for SMP and simply defining it for UP. This change cleans up code and brings SMP and UP closer a bit. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 962f21f1d4d..d2a0baa87d1 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -217,10 +217,12 @@ 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 - * percpu area. Preallocate it. + * percpu area. Preallocate it and define the percpu offset symbol + * so that it can be accessed as a percpu variable. */ . = ALIGN(PAGE_SIZE); PERCPU_VADDR_PREALLOC(0, :percpu, pda_size) + per_cpu____pda = __per_cpu_start; #else PERCPU(PAGE_SIZE) #endif -- cgit v1.2.3 From 49357d19e4fb31e28796eaff83499e7584c26878 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 13 Jan 2009 20:41:35 +0900 Subject: x86: convert pda ops to wrappers around x86 percpu accessors pda is now a percpu variable and there's no reason it can't use plain x86 percpu accessors. Add x86_test_and_clear_bit_percpu() and replace pda op implementations with wrappers around x86 percpu accessors. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index d2a0baa87d1..a09abb8fb97 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -14,7 +14,6 @@ OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64") OUTPUT_ARCH(i386:x86-64) ENTRY(phys_startup_64) jiffies_64 = jiffies; -_proxy_pda = 1; PHDRS { text PT_LOAD FLAGS(5); /* R_E */ data PT_LOAD FLAGS(7); /* RWE */ -- cgit v1.2.3 From 947e76cdc34c782fc947313d4331380686eebbad Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Mon, 19 Jan 2009 12:21:28 +0900 Subject: x86: move stack_canary into irq_stack Impact: x86_64 percpu area layout change, irq_stack now at the beginning Now that the PDA is empty except for the stack canary, it can be removed. The irqstack is moved to the start of the per-cpu section. If the stack protector is enabled, the canary overlaps the bottom 48 bytes of the irqstack. tj: * updated subject * dropped asm relocation of irq_stack_ptr * updated comments a bit * rebased on top of stack canary changes Signed-off-by: Brian Gerst Signed-off-by: Tejun Heo --- arch/x86/kernel/vmlinux_64.lds.S | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index a09abb8fb97..c9740996430 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -220,8 +220,7 @@ SECTIONS * so that it can be accessed as a percpu variable. */ . = ALIGN(PAGE_SIZE); - PERCPU_VADDR_PREALLOC(0, :percpu, pda_size) - per_cpu____pda = __per_cpu_start; + PERCPU_VADDR(0, :percpu) #else PERCPU(PAGE_SIZE) #endif @@ -262,3 +261,8 @@ SECTIONS */ ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), "kernel image bigger than KERNEL_IMAGE_SIZE") + +#ifdef CONFIG_SMP +ASSERT((per_cpu__irq_stack_union == 0), + "irq_stack_union is not at start of per-cpu area"); +#endif -- cgit v1.2.3 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/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index c9740996430..07f62d287ff 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 v1.2.3 From 2add8e235cbe0dcd672c33fc322754e15500238c Mon Sep 17 00:00:00 2001 From: Brian Gerst Date: Sun, 8 Feb 2009 09:58:39 -0500 Subject: x86: use linker to offset symbols by __per_cpu_load Impact: cleanup and bug fix Use the linker to create symbols for certain per-cpu variables that are offset by __per_cpu_load. This allows the removal of the runtime fixup of the GDT pointer, which fixes a bug with resume reported by Jiri Slaby. Reported-by: Jiri Slaby Signed-off-by: Brian Gerst Acked-by: Jiri Slaby Signed-off-by: Ingo Molnar --- arch/x86/kernel/vmlinux_64.lds.S | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 07f62d287ff..087a7f2c639 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -257,6 +257,14 @@ SECTIONS DWARF_DEBUG } + /* + * Per-cpu symbols which need to be offset from __per_cpu_load + * for the boot processor. + */ +#define INIT_PER_CPU(x) init_per_cpu__##x = per_cpu__##x + __per_cpu_load +INIT_PER_CPU(gdt_page); +INIT_PER_CPU(irq_stack_union); + /* * Build-time check on the image size: */ -- cgit v1.2.3 From 0341c14da49e7b93d2998926f6ac89a3129e3fa1 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 13 Feb 2009 11:14:01 -0800 Subject: x86: use _types.h headers in asm where available In general, the only definitions that assembly files can use are in _types.S headers (where available), so convert them. Signed-off-by: Jeremy Fitzhardinge --- arch/x86/kernel/vmlinux_64.lds.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel/vmlinux_64.lds.S') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 087a7f2c639..fbfced6f680 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -6,7 +6,7 @@ #include #include -#include +#include #undef i386 /* in case the preprocessor is a 32bit one */ -- cgit v1.2.3