From 41edafdb78feac1d1f8823846209975fde990633 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 28 Jan 2009 14:35:02 -0800 Subject: x86/pvops: add a paravirt_ident functions to allow special patching Impact: Optimization Several paravirt ops implementations simply return their arguments, the most obvious being the make_pte/pte_val class of operations on native. On 32-bit, the identity function is literally a no-op, as the calling convention uses the same registers for the first argument and return. On 64-bit, it can be implemented with a single "mov". This patch adds special identity functions for 32 and 64 bit argument, and machinery to recognize them and replace them with either nops or a mov as appropriate. At the moment, the only users for the identity functions are the pagetable entry conversion functions. The result is a measureable improvement on pagetable-heavy benchmarks (2-3%, reducing the pvops overhead from 5 to 2%). Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/kernel/paravirt.c | 75 ++++++++++++++++++++++++++++++++----- arch/x86/kernel/paravirt_patch_32.c | 12 ++++++ arch/x86/kernel/paravirt_patch_64.c | 15 ++++++++ 3 files changed, 93 insertions(+), 9 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 202514be5923..dd25e2b1593b 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -44,6 +44,17 @@ void _paravirt_nop(void) { } +/* identity function, which can be inlined */ +u32 _paravirt_ident_32(u32 x) +{ + return x; +} + +u64 _paravirt_ident_64(u64 x) +{ + return x; +} + static void __init default_banner(void) { printk(KERN_INFO "Booting paravirtualized kernel on %s\n", @@ -138,9 +149,16 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, if (opfunc == NULL) /* If there's no function, patch it with a ud2a (BUG) */ ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a)); - else if (opfunc == paravirt_nop) + else if (opfunc == _paravirt_nop) /* If the operation is a nop, then nop the callsite */ ret = paravirt_patch_nop(); + + /* identity functions just return their single argument */ + else if (opfunc == _paravirt_ident_32) + ret = paravirt_patch_ident_32(insnbuf, len); + else if (opfunc == _paravirt_ident_64) + ret = paravirt_patch_ident_64(insnbuf, len); + else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) || type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) || type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) || @@ -373,6 +391,45 @@ struct pv_apic_ops pv_apic_ops = { #endif }; +typedef pte_t make_pte_t(pteval_t); +typedef pmd_t make_pmd_t(pmdval_t); +typedef pud_t make_pud_t(pudval_t); +typedef pgd_t make_pgd_t(pgdval_t); + +typedef pteval_t pte_val_t(pte_t); +typedef pmdval_t pmd_val_t(pmd_t); +typedef pudval_t pud_val_t(pud_t); +typedef pgdval_t pgd_val_t(pgd_t); + + +#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE) +/* 32-bit pagetable entries */ +#define paravirt_native_make_pte (make_pte_t *)_paravirt_ident_32 +#define paravirt_native_pte_val (pte_val_t *)_paravirt_ident_32 + +#define paravirt_native_make_pmd (make_pmd_t *)_paravirt_ident_32 +#define paravirt_native_pmd_val (pmd_val_t *)_paravirt_ident_32 + +#define paravirt_native_make_pud (make_pud_t *)_paravirt_ident_32 +#define paravirt_native_pud_val (pud_val_t *)_paravirt_ident_32 + +#define paravirt_native_make_pgd (make_pgd_t *)_paravirt_ident_32 +#define paravirt_native_pgd_val (pgd_val_t *)_paravirt_ident_32 +#else +/* 64-bit pagetable entries */ +#define paravirt_native_make_pte (make_pte_t *)_paravirt_ident_64 +#define paravirt_native_pte_val (pte_val_t *)_paravirt_ident_64 + +#define paravirt_native_make_pmd (make_pmd_t *)_paravirt_ident_64 +#define paravirt_native_pmd_val (pmd_val_t *)_paravirt_ident_64 + +#define paravirt_native_make_pud (make_pud_t *)_paravirt_ident_64 +#define paravirt_native_pud_val (pud_val_t *)_paravirt_ident_64 + +#define paravirt_native_make_pgd (make_pgd_t *)_paravirt_ident_64 +#define paravirt_native_pgd_val (pgd_val_t *)_paravirt_ident_64 +#endif + struct pv_mmu_ops pv_mmu_ops = { #ifndef CONFIG_X86_64 .pagetable_setup_start = native_pagetable_setup_start, @@ -424,21 +481,21 @@ struct pv_mmu_ops pv_mmu_ops = { .pmd_clear = native_pmd_clear, #endif .set_pud = native_set_pud, - .pmd_val = native_pmd_val, - .make_pmd = native_make_pmd, + .pmd_val = paravirt_native_pmd_val, + .make_pmd = paravirt_native_make_pmd, #if PAGETABLE_LEVELS == 4 - .pud_val = native_pud_val, - .make_pud = native_make_pud, + .pud_val = paravirt_native_pud_val, + .make_pud = paravirt_native_make_pud, .set_pgd = native_set_pgd, #endif #endif /* PAGETABLE_LEVELS >= 3 */ - .pte_val = native_pte_val, - .pgd_val = native_pgd_val, + .pte_val = paravirt_native_pte_val, + .pgd_val = paravirt_native_pgd_val, - .make_pte = native_make_pte, - .make_pgd = native_make_pgd, + .make_pte = paravirt_native_make_pte, + .make_pgd = paravirt_native_make_pgd, .dup_mmap = paravirt_nop, .exit_mmap = paravirt_nop, diff --git a/arch/x86/kernel/paravirt_patch_32.c b/arch/x86/kernel/paravirt_patch_32.c index 9fe644f4861d..d9f32e6d6ab6 100644 --- a/arch/x86/kernel/paravirt_patch_32.c +++ b/arch/x86/kernel/paravirt_patch_32.c @@ -12,6 +12,18 @@ DEF_NATIVE(pv_mmu_ops, read_cr3, "mov %cr3, %eax"); DEF_NATIVE(pv_cpu_ops, clts, "clts"); DEF_NATIVE(pv_cpu_ops, read_tsc, "rdtsc"); +unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len) +{ + /* arg in %eax, return in %eax */ + return 0; +} + +unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len) +{ + /* arg in %edx:%eax, return in %edx:%eax */ + return 0; +} + unsigned native_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) { diff --git a/arch/x86/kernel/paravirt_patch_64.c b/arch/x86/kernel/paravirt_patch_64.c index 061d01df9ae6..3f08f34f93eb 100644 --- a/arch/x86/kernel/paravirt_patch_64.c +++ b/arch/x86/kernel/paravirt_patch_64.c @@ -19,6 +19,21 @@ DEF_NATIVE(pv_cpu_ops, usergs_sysret64, "swapgs; sysretq"); DEF_NATIVE(pv_cpu_ops, usergs_sysret32, "swapgs; sysretl"); DEF_NATIVE(pv_cpu_ops, swapgs, "swapgs"); +DEF_NATIVE(, mov32, "mov %edi, %eax"); +DEF_NATIVE(, mov64, "mov %rdi, %rax"); + +unsigned paravirt_patch_ident_32(void *insnbuf, unsigned len) +{ + return paravirt_patch_insns(insnbuf, len, + start__mov32, end__mov32); +} + +unsigned paravirt_patch_ident_64(void *insnbuf, unsigned len) +{ + return paravirt_patch_insns(insnbuf, len, + start__mov64, end__mov64); +} + unsigned native_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) { -- cgit v1.2.2 From b8aa287f77be943e37a84fa4657e27df95269bfb Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 28 Jan 2009 14:35:03 -0800 Subject: x86: fix paravirt clobber in entry_64.S Impact: Fix latent bug The clobber is trying to say that anything except RDI is available for clobbering, but actually clobbers everything. This hasn't mattered because the clobbers were basically ignored, but subsequent patches will rely on them. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/kernel/entry_64.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index a52703864a16..e4c9710cae52 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1140,7 +1140,7 @@ ENTRY(native_load_gs_index) CFI_STARTPROC pushf CFI_ADJUST_CFA_OFFSET 8 - DISABLE_INTERRUPTS(CLBR_ANY | ~(CLBR_RDI)) + DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI) SWAPGS gs_change: movl %edi,%gs -- cgit v1.2.2 From ecb93d1ccd0aac63f03be2db3cac3fa974716f4c Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 28 Jan 2009 14:35:05 -0800 Subject: x86/paravirt: add register-saving thunks to reduce caller register pressure Impact: Optimization One of the problems with inserting a pile of C calls where previously there were none is that the register pressure is greatly increased. The C calling convention says that the caller must expect a certain set of registers may be trashed by the callee, and that the callee can use those registers without restriction. This includes the function argument registers, and several others. This patch seeks to alleviate this pressure by introducing wrapper thunks that will do the register saving/restoring, so that the callsite doesn't need to worry about it, but the callee function can be conventional compiler-generated code. In many cases (particularly performance-sensitive cases) the callee will be in assembler anyway, and need not use the compiler's calling convention. Standard calling convention is: arguments return scratch x86-32 eax edx ecx eax ? x86-64 rdi rsi rdx rcx rax r8 r9 r10 r11 The thunk preserves all argument and scratch registers. The return register is not preserved, and is available as a scratch register for unwrapped callee code (and of course the return value). Wrapped function pointers are themselves wrapped in a struct paravirt_callee_save structure, in order to get some warning from the compiler when functions with mismatched calling conventions are used. The most common paravirt ops, both statically and dynamically, are interrupt enable/disable/save/restore, so handle them first. This is particularly easy since their calls are handled specially anyway. XXX Deal with VMI. What's their calling convention? Signed-off-by: H. Peter Anvin --- arch/x86/kernel/paravirt.c | 8 ++++---- arch/x86/kernel/vsmp_64.c | 12 ++++++++---- 2 files changed, 12 insertions(+), 8 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index dd25e2b1593b..8adb6b5aa421 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -310,10 +310,10 @@ struct pv_time_ops pv_time_ops = { struct pv_irq_ops pv_irq_ops = { .init_IRQ = native_init_IRQ, - .save_fl = native_save_fl, - .restore_fl = native_restore_fl, - .irq_disable = native_irq_disable, - .irq_enable = native_irq_enable, + .save_fl = __PV_IS_CALLEE_SAVE(native_save_fl), + .restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl), + .irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable), + .irq_enable = __PV_IS_CALLEE_SAVE(native_irq_enable), .safe_halt = native_safe_halt, .halt = native_halt, #ifdef CONFIG_X86_64 diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index a688f3bfaec2..c609205df594 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -37,6 +37,7 @@ static unsigned long vsmp_save_fl(void) flags &= ~X86_EFLAGS_IF; return flags; } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_save_fl); static void vsmp_restore_fl(unsigned long flags) { @@ -46,6 +47,7 @@ static void vsmp_restore_fl(unsigned long flags) flags |= X86_EFLAGS_AC; native_restore_fl(flags); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl); static void vsmp_irq_disable(void) { @@ -53,6 +55,7 @@ static void vsmp_irq_disable(void) native_restore_fl((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable); static void vsmp_irq_enable(void) { @@ -60,6 +63,7 @@ static void vsmp_irq_enable(void) native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); } +PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_enable); static unsigned __init_or_module vsmp_patch(u8 type, u16 clobbers, void *ibuf, unsigned long addr, unsigned len) @@ -90,10 +94,10 @@ static void __init set_vsmp_pv_ops(void) cap, ctl); if (cap & ctl & (1 << 4)) { /* Setup irq ops and turn on vSMP IRQ fastpath handling */ - pv_irq_ops.irq_disable = vsmp_irq_disable; - pv_irq_ops.irq_enable = vsmp_irq_enable; - pv_irq_ops.save_fl = vsmp_save_fl; - pv_irq_ops.restore_fl = vsmp_restore_fl; + pv_irq_ops.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable); + pv_irq_ops.irq_enable = PV_CALLEE_SAVE(vsmp_irq_enable); + pv_irq_ops.save_fl = PV_CALLEE_SAVE(vsmp_save_fl); + pv_irq_ops.restore_fl = PV_CALLEE_SAVE(vsmp_restore_fl); pv_init_ops.patch = vsmp_patch; ctl &= ~(1 << 4); -- cgit v1.2.2 From da5de7c22eb705be709a57e486e7475a6969b994 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 28 Jan 2009 14:35:07 -0800 Subject: x86/paravirt: use callee-saved convention for pte_val/make_pte/etc Impact: Optimization In the native case, pte_val, make_pte, etc are all just identity functions, so there's no need to clobber a lot of registers over them. (This changes the 32-bit callee-save calling convention to return both EAX and EDX so functions can return 64-bit values.) Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/kernel/paravirt.c | 53 +++++++++++----------------------------------- 1 file changed, 12 insertions(+), 41 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 8adb6b5aa421..cea11c8e3049 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -391,43 +391,12 @@ struct pv_apic_ops pv_apic_ops = { #endif }; -typedef pte_t make_pte_t(pteval_t); -typedef pmd_t make_pmd_t(pmdval_t); -typedef pud_t make_pud_t(pudval_t); -typedef pgd_t make_pgd_t(pgdval_t); - -typedef pteval_t pte_val_t(pte_t); -typedef pmdval_t pmd_val_t(pmd_t); -typedef pudval_t pud_val_t(pud_t); -typedef pgdval_t pgd_val_t(pgd_t); - - #if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE) /* 32-bit pagetable entries */ -#define paravirt_native_make_pte (make_pte_t *)_paravirt_ident_32 -#define paravirt_native_pte_val (pte_val_t *)_paravirt_ident_32 - -#define paravirt_native_make_pmd (make_pmd_t *)_paravirt_ident_32 -#define paravirt_native_pmd_val (pmd_val_t *)_paravirt_ident_32 - -#define paravirt_native_make_pud (make_pud_t *)_paravirt_ident_32 -#define paravirt_native_pud_val (pud_val_t *)_paravirt_ident_32 - -#define paravirt_native_make_pgd (make_pgd_t *)_paravirt_ident_32 -#define paravirt_native_pgd_val (pgd_val_t *)_paravirt_ident_32 +#define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_32) #else /* 64-bit pagetable entries */ -#define paravirt_native_make_pte (make_pte_t *)_paravirt_ident_64 -#define paravirt_native_pte_val (pte_val_t *)_paravirt_ident_64 - -#define paravirt_native_make_pmd (make_pmd_t *)_paravirt_ident_64 -#define paravirt_native_pmd_val (pmd_val_t *)_paravirt_ident_64 - -#define paravirt_native_make_pud (make_pud_t *)_paravirt_ident_64 -#define paravirt_native_pud_val (pud_val_t *)_paravirt_ident_64 - -#define paravirt_native_make_pgd (make_pgd_t *)_paravirt_ident_64 -#define paravirt_native_pgd_val (pgd_val_t *)_paravirt_ident_64 +#define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_64) #endif struct pv_mmu_ops pv_mmu_ops = { @@ -481,21 +450,23 @@ struct pv_mmu_ops pv_mmu_ops = { .pmd_clear = native_pmd_clear, #endif .set_pud = native_set_pud, - .pmd_val = paravirt_native_pmd_val, - .make_pmd = paravirt_native_make_pmd, + + .pmd_val = PTE_IDENT, + .make_pmd = PTE_IDENT, #if PAGETABLE_LEVELS == 4 - .pud_val = paravirt_native_pud_val, - .make_pud = paravirt_native_make_pud, + .pud_val = PTE_IDENT, + .make_pud = PTE_IDENT, + .set_pgd = native_set_pgd, #endif #endif /* PAGETABLE_LEVELS >= 3 */ - .pte_val = paravirt_native_pte_val, - .pgd_val = paravirt_native_pgd_val, + .pte_val = PTE_IDENT, + .pgd_val = PTE_IDENT, - .make_pte = paravirt_native_make_pte, - .make_pgd = paravirt_native_make_pgd, + .make_pte = PTE_IDENT, + .make_pgd = PTE_IDENT, .dup_mmap = paravirt_nop, .exit_mmap = paravirt_nop, -- cgit v1.2.2 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/x86/kernel') diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 89fce1b6d01f..f4b2f27d19b9 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 v1.2.2 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/kernel/cpu/common.c | 7 +++---- arch/x86/kernel/setup_percpu.c | 2 +- arch/x86/kernel/smpboot.c | 2 +- 3 files changed, 5 insertions(+), 6 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 652fdc9a757a..6eacd64b602e 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 0d1e7ac439f4..ef91747bbed5 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 f9dbcff43546..612d3c74f6a3 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; -- cgit v1.2.2 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/kernel/cpu/common.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 6eacd64b602e..0f73ea423089 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 v1.2.2 From 664c7954721adfc9bd61de6ec78f89f482f1a802 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 30 Jan 2009 23:18:41 -0800 Subject: x86/vmi: fix interrupt enable/disable/save/restore calling convention. Zach says: > Enable/Disable have no clobbers at all. > Save clobbers only return value, %eax > Restore also clobbers nothing. This is precisely compatible with the calling convention, so we can just call them directly without wrapping. (Compile tested only.) Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/kernel/vmi_32.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'arch/x86/kernel') diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 1d3302cc2ddf..eb9e7347928e 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -670,10 +670,11 @@ static inline int __init activate_vmi(void) para_fill(pv_mmu_ops.write_cr2, SetCR2); para_fill(pv_mmu_ops.write_cr3, SetCR3); para_fill(pv_cpu_ops.write_cr4, SetCR4); - para_fill(pv_irq_ops.save_fl, GetInterruptMask); - para_fill(pv_irq_ops.restore_fl, SetInterruptMask); - para_fill(pv_irq_ops.irq_disable, DisableInterrupts); - para_fill(pv_irq_ops.irq_enable, EnableInterrupts); + + para_fill(pv_irq_ops.save_fl.func, GetInterruptMask); + para_fill(pv_irq_ops.restore_fl.func, SetInterruptMask); + para_fill(pv_irq_ops.irq_disable.func, DisableInterrupts); + para_fill(pv_irq_ops.irq_enable.func, EnableInterrupts); para_fill(pv_cpu_ops.wbinvd, WBINVD); para_fill(pv_cpu_ops.read_tsc, RDTSC); -- cgit v1.2.2 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') diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index c9740996430a..07f62d287ff0 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 v1.2.2