From 816afe4ff98ee10b1d30fd66361be132a0a5cee6 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 6 Aug 2012 17:29:49 +0930 Subject: [PATCH 1/7] x86/smp: Don't ever patch back to UP if we unplug cpus We still patch SMP instructions to UP variants if we boot with a single CPU, but not at any other time. In particular, not if we unplug CPUs to return to a single cpu. Paul McKenney points out: mean offline overhead is 6251/48=130.2 milliseconds. If I remove the alternatives_smp_switch() from the offline path [...] the mean offline overhead is 550/42=13.1 milliseconds Basically, we're never going to get those 120ms back, and the code is pretty messy. We get rid of: 1) The "smp-alt-once" boot option. It's actually "smp-alt-boot", the documentation is wrong. It's now the default. 2) The skip_smp_alternatives flag used by suspend. 3) arch_disable_nonboot_cpus_begin() and arch_disable_nonboot_cpus_end() which were only used to set this one flag. Signed-off-by: Rusty Russell Cc: Paul McKenney Cc: Suresh Siddha Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/87vcgwwive.fsf@rustcorp.com.au Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 3 - arch/x86/include/asm/alternative.h | 4 +- arch/x86/kernel/alternative.c | 107 +++++++--------------------- arch/x86/kernel/smpboot.c | 20 +----- arch/x86/xen/smp.c | 6 +- kernel/cpu.c | 11 --- 6 files changed, 32 insertions(+), 119 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index ad7e2e5088c1..7aef3345f739 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2638,9 +2638,6 @@ bytes respectively. Such letter suffixes can also be entirely omitted. smart2= [HW] Format: [,[,...,]] - smp-alt-once [X86-32,SMP] On a hotplug CPU system, only - attempt to substitute SMP alternatives once at boot. - smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port smsc-ircc2.ircc_sir= [HW] SIR base I/O port diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 70780689599a..444704c8e186 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -60,7 +60,7 @@ extern void alternatives_smp_module_add(struct module *mod, char *name, void *locks, void *locks_end, void *text, void *text_end); extern void alternatives_smp_module_del(struct module *mod); -extern void alternatives_smp_switch(int smp); +extern void alternatives_enable_smp(void); extern int alternatives_text_reserved(void *start, void *end); extern bool skip_smp_alternatives; #else @@ -68,7 +68,7 @@ static inline void alternatives_smp_module_add(struct module *mod, char *name, void *locks, void *locks_end, void *text, void *text_end) {} static inline void alternatives_smp_module_del(struct module *mod) {} -static inline void alternatives_smp_switch(int smp) {} +static inline void alternatives_enable_smp(void) {} static inline int alternatives_text_reserved(void *start, void *end) { return 0; diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index afb7ff79a29f..af1f326a31c4 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -23,19 +23,6 @@ #define MAX_PATCH_LEN (255-1) -#ifdef CONFIG_HOTPLUG_CPU -static int smp_alt_once; - -static int __init bootonly(char *str) -{ - smp_alt_once = 1; - return 1; -} -__setup("smp-alt-boot", bootonly); -#else -#define smp_alt_once 1 -#endif - static int __initdata_or_module debug_alternative; static int __init debug_alt(char *str) @@ -326,9 +313,6 @@ static void alternatives_smp_unlock(const s32 *start, const s32 *end, { const s32 *poff; - if (noreplace_smp) - return; - mutex_lock(&text_mutex); for (poff = start; poff < end; poff++) { u8 *ptr = (u8 *)poff + *poff; @@ -359,7 +343,7 @@ struct smp_alt_module { }; static LIST_HEAD(smp_alt_modules); static DEFINE_MUTEX(smp_alt); -static int smp_mode = 1; /* protected by smp_alt */ +static bool uniproc_patched = false; /* protected by smp_alt */ void __init_or_module alternatives_smp_module_add(struct module *mod, char *name, @@ -368,19 +352,18 @@ void __init_or_module alternatives_smp_module_add(struct module *mod, { struct smp_alt_module *smp; - if (noreplace_smp) - return; + mutex_lock(&smp_alt); + if (!uniproc_patched) + goto unlock; - if (smp_alt_once) { - if (boot_cpu_has(X86_FEATURE_UP)) - alternatives_smp_unlock(locks, locks_end, - text, text_end); - return; - } + if (num_possible_cpus() == 1) + /* Don't bother remembering, we'll never have to undo it. */ + goto smp_unlock; smp = kzalloc(sizeof(*smp), GFP_KERNEL); if (NULL == smp) - return; /* we'll run the (safe but slow) SMP code then ... */ + /* we'll run the (safe but slow) SMP code then ... */ + goto unlock; smp->mod = mod; smp->name = name; @@ -392,11 +375,10 @@ void __init_or_module alternatives_smp_module_add(struct module *mod, __func__, smp->locks, smp->locks_end, smp->text, smp->text_end, smp->name); - mutex_lock(&smp_alt); list_add_tail(&smp->next, &smp_alt_modules); - if (boot_cpu_has(X86_FEATURE_UP)) - alternatives_smp_unlock(smp->locks, smp->locks_end, - smp->text, smp->text_end); +smp_unlock: + alternatives_smp_unlock(locks, locks_end, text, text_end); +unlock: mutex_unlock(&smp_alt); } @@ -404,24 +386,18 @@ void __init_or_module alternatives_smp_module_del(struct module *mod) { struct smp_alt_module *item; - if (smp_alt_once || noreplace_smp) - return; - mutex_lock(&smp_alt); list_for_each_entry(item, &smp_alt_modules, next) { if (mod != item->mod) continue; list_del(&item->next); - mutex_unlock(&smp_alt); - DPRINTK("%s: %s\n", __func__, item->name); kfree(item); - return; + break; } mutex_unlock(&smp_alt); } -bool skip_smp_alternatives; -void alternatives_smp_switch(int smp) +void alternatives_enable_smp(void) { struct smp_alt_module *mod; @@ -436,34 +412,21 @@ void alternatives_smp_switch(int smp) pr_info("lockdep: fixing up alternatives\n"); #endif - if (noreplace_smp || smp_alt_once || skip_smp_alternatives) - return; - BUG_ON(!smp && (num_online_cpus() > 1)); + /* Why bother if there are no other CPUs? */ + BUG_ON(num_possible_cpus() == 1); mutex_lock(&smp_alt); - /* - * Avoid unnecessary switches because it forces JIT based VMs to - * throw away all cached translations, which can be quite costly. - */ - if (smp == smp_mode) { - /* nothing */ - } else if (smp) { + if (uniproc_patched) { pr_info("switching to SMP code\n"); + BUG_ON(num_online_cpus() != 1); clear_cpu_cap(&boot_cpu_data, X86_FEATURE_UP); clear_cpu_cap(&cpu_data(0), X86_FEATURE_UP); list_for_each_entry(mod, &smp_alt_modules, next) alternatives_smp_lock(mod->locks, mod->locks_end, mod->text, mod->text_end); - } else { - pr_info("switching to UP code\n"); - set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP); - set_cpu_cap(&cpu_data(0), X86_FEATURE_UP); - list_for_each_entry(mod, &smp_alt_modules, next) - alternatives_smp_unlock(mod->locks, mod->locks_end, - mod->text, mod->text_end); + uniproc_patched = false; } - smp_mode = smp; mutex_unlock(&smp_alt); } @@ -540,40 +503,22 @@ void __init alternative_instructions(void) apply_alternatives(__alt_instructions, __alt_instructions_end); - /* switch to patch-once-at-boottime-only mode and free the - * tables in case we know the number of CPUs will never ever - * change */ -#ifdef CONFIG_HOTPLUG_CPU - if (num_possible_cpus() < 2) - smp_alt_once = 1; -#endif - #ifdef CONFIG_SMP - if (smp_alt_once) { - if (1 == num_possible_cpus()) { - pr_info("switching to UP code\n"); - set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP); - set_cpu_cap(&cpu_data(0), X86_FEATURE_UP); - - alternatives_smp_unlock(__smp_locks, __smp_locks_end, - _text, _etext); - } - } else { + /* Patch to UP if other cpus not imminent. */ + if (!noreplace_smp && (num_present_cpus() == 1 || setup_max_cpus <= 1)) { + uniproc_patched = true; alternatives_smp_module_add(NULL, "core kernel", __smp_locks, __smp_locks_end, _text, _etext); - - /* Only switch to UP mode if we don't immediately boot others */ - if (num_present_cpus() == 1 || setup_max_cpus <= 1) - alternatives_smp_switch(0); } -#endif - apply_paravirt(__parainstructions, __parainstructions_end); - if (smp_alt_once) + if (!uniproc_patched || num_possible_cpus() == 1) free_init_pages("SMP alternatives", (unsigned long)__smp_locks, (unsigned long)__smp_locks_end); +#endif + + apply_paravirt(__parainstructions, __parainstructions_end); restart_nmi(); } diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 7c5a8c314c02..c80a33bc528b 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -665,7 +665,8 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle) unsigned long boot_error = 0; int timeout; - alternatives_smp_switch(1); + /* Just in case we booted with a single CPU. */ + alternatives_enable_smp(); idle->thread.sp = (unsigned long) (((struct pt_regs *) (THREAD_SIZE + task_stack_page(idle))) - 1); @@ -1053,20 +1054,6 @@ out: preempt_enable(); } -void arch_disable_nonboot_cpus_begin(void) -{ - /* - * Avoid the smp alternatives switch during the disable_nonboot_cpus(). - * In the suspend path, we will be back in the SMP mode shortly anyways. - */ - skip_smp_alternatives = true; -} - -void arch_disable_nonboot_cpus_end(void) -{ - skip_smp_alternatives = false; -} - void arch_enable_nonboot_cpus_begin(void) { set_mtrr_aps_delayed_init(); @@ -1256,9 +1243,6 @@ void native_cpu_die(unsigned int cpu) if (per_cpu(cpu_state, cpu) == CPU_DEAD) { if (system_state == SYSTEM_RUNNING) pr_info("CPU %u is now offline\n", cpu); - - if (1 == num_online_cpus()) - alternatives_smp_switch(0); return; } msleep(100); diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index f58dca7a6e52..353c50f18702 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -377,7 +377,8 @@ static int __cpuinit xen_cpu_up(unsigned int cpu, struct task_struct *idle) return rc; if (num_online_cpus() == 1) - alternatives_smp_switch(1); + /* Just in case we booted with a single CPU. */ + alternatives_enable_smp(); rc = xen_smp_intr_init(cpu); if (rc) @@ -424,9 +425,6 @@ static void xen_cpu_die(unsigned int cpu) unbind_from_irqhandler(per_cpu(xen_irq_work, cpu), NULL); xen_uninit_lock_cpu(cpu); xen_teardown_timer(cpu); - - if (num_online_cpus() == 1) - alternatives_smp_switch(0); } static void __cpuinit xen_play_dead(void) /* used only with HOTPLUG_CPU */ diff --git a/kernel/cpu.c b/kernel/cpu.c index 14d32588cccd..f6bfe3e03f6b 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -439,14 +439,6 @@ EXPORT_SYMBOL_GPL(cpu_up); #ifdef CONFIG_PM_SLEEP_SMP static cpumask_var_t frozen_cpus; -void __weak arch_disable_nonboot_cpus_begin(void) -{ -} - -void __weak arch_disable_nonboot_cpus_end(void) -{ -} - int disable_nonboot_cpus(void) { int cpu, first_cpu, error = 0; @@ -458,7 +450,6 @@ int disable_nonboot_cpus(void) * with the userspace trying to use the CPU hotplug at the same time */ cpumask_clear(frozen_cpus); - arch_disable_nonboot_cpus_begin(); printk("Disabling non-boot CPUs ...\n"); for_each_online_cpu(cpu) { @@ -474,8 +465,6 @@ int disable_nonboot_cpus(void) } } - arch_disable_nonboot_cpus_end(); - if (!error) { BUG_ON(num_online_cpus() > 1); /* Make sure the CPUs won't be enabled by someone else */ From 6eebdda35e6b18d0dddb2a44e34211bd94f0cad6 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Fri, 24 Aug 2012 23:58:47 +0400 Subject: [PATCH 2/7] x86: Drop unnecessary kernel_eflags variable on 64-bit On 64 bit x86 we save the current eflags in cpu_init for use in ret_from_fork. Strictly speaking reserved bits in EFLAGS should be read as written but in practise it is unlikely that EFLAGS could ever be extended in this way and the kernel alread clears any undefined flags early on. The equivalent 32 bit code simply hard codes 0x0202 as the new EFLAGS. This change makes 64 bit use the same mechanism to setup the initial EFLAGS on fork. Note that 64 bit resets EFLAGS before calling schedule_tail() as opposed to 32 bit which calls schedule_tail() first. Therefore the correct value for EFLAGS has opposite IF bit. Signed-off-by: Ian Campbell Signed-off-by: Cyrill Gorcunov Acked-by: Andi Kleen Acked-by: "H. Peter Anvin" Cc: Brian Gerst Cc: Peter Zijlstra Cc: Pekka Enberg Cc: Andi Kleen Link: http://lkml.kernel.org/r/20120824195847.GA31628@moon Signed-off-by: Ingo Molnar --- arch/x86/include/asm/processor.h | 1 - arch/x86/kernel/cpu/common.c | 4 ---- arch/x86/kernel/entry_64.S | 2 +- 3 files changed, 1 insertion(+), 6 deletions(-) diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index d048cad9bcad..9738b39e4eb9 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -423,7 +423,6 @@ DECLARE_INIT_PER_CPU(irq_stack_union); DECLARE_PER_CPU(char *, irq_stack_ptr); DECLARE_PER_CPU(unsigned int, irq_count); -extern unsigned long kernel_eflags; extern asmlinkage void ignore_sysret(void); #else /* X86_64 */ #ifdef CONFIG_CC_STACKPROTECTOR diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index a5fbc3c5fccc..9961e2e23709 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1116,8 +1116,6 @@ void syscall_init(void) X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL); } -unsigned long kernel_eflags; - /* * Copies of the original ist values from the tss are only accessed during * debugging, no special alignment required. @@ -1299,8 +1297,6 @@ void __cpuinit cpu_init(void) fpu_init(); xsave_init(); - raw_local_save_flags(kernel_eflags); - if (is_uv_system()) uv_cpu_init(); } diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 69babd8c834f..b1dac12dc5e6 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -440,7 +440,7 @@ ENTRY(ret_from_fork) LOCK ; btr $TIF_FORK,TI_flags(%r8) - pushq_cfi kernel_eflags(%rip) + pushq_cfi $0x0002 popfq_cfi # reset kernel eflags call schedule_tail # rdi: 'prev' task parameter From 1edfbb4153bd29bcf8d2236676238d5237972be1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 10 Sep 2012 12:04:16 +0100 Subject: [PATCH 3/7] x86/64: Adjust types of temporaries used by ffs()/fls()/fls64() The 64-bit special cases of the former two (the thrird one is 64-bit only anyway) don't need to use "long" temporaries, as the result will always fit in a 32-bit variable, and the functions return plain "int". This avoids a few REX prefixes, i.e. minimally reduces code size. Signed-off-by: Jan Beulich Cc: Linus Torvalds Link: http://lkml.kernel.org/r/504DE550020000780009A258@nat28.tlf.novell.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/bitops.h | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index 72f5009deb5a..ebaee695394e 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -417,10 +417,9 @@ static inline int ffs(int x) * We cannot do this on 32 bits because at the very least some * 486 CPUs did not behave this way. */ - long tmp = -1; asm("bsfl %1,%0" : "=r" (r) - : "rm" (x), "0" (tmp)); + : "rm" (x), "0" (-1)); #elif defined(CONFIG_X86_CMOV) asm("bsfl %1,%0\n\t" "cmovzl %2,%0" @@ -459,10 +458,9 @@ static inline int fls(int x) * We cannot do this on 32 bits because at the very least some * 486 CPUs did not behave this way. */ - long tmp = -1; asm("bsrl %1,%0" : "=r" (r) - : "rm" (x), "0" (tmp)); + : "rm" (x), "0" (-1)); #elif defined(CONFIG_X86_CMOV) asm("bsrl %1,%0\n\t" "cmovzl %2,%0" @@ -490,13 +488,13 @@ static inline int fls(int x) #ifdef CONFIG_X86_64 static __always_inline int fls64(__u64 x) { - long bitpos = -1; + int bitpos = -1; /* * AMD64 says BSRQ won't clobber the dest reg if x==0; Intel64 says the * dest reg is undefined if x==0, but their CPU architect says its * value is written to set it to the same as before. */ - asm("bsrq %1,%0" + asm("bsrq %1,%q0" : "+r" (bitpos) : "rm" (x)); return bitpos + 1; From 5870661c091e827973674cc3469b50c959008c2b Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 10 Sep 2012 12:24:43 +0100 Subject: [PATCH 4/7] x86: Prefer TZCNT over BFS Following a relatively recent compiler change, make use of the fact that for non-zero input BSF and TZCNT produce the same result, and that CPUs not knowing of TZCNT will treat the instruction as BSF (i.e. ignore what looks like a REP prefix to them). The assumption here is that TZCNT would never have worse performance than BSF. For the moment, only do this when the respective generic-CPU option is selected (as there are no specific-CPU options covering the CPUs supporting TZCNT), and don't do that when size optimization was requested. Signed-off-by: Jan Beulich Cc: Linus Torvalds Link: http://lkml.kernel.org/r/504DEA1B020000780009A277@nat28.tlf.novell.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/bitops.h | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index ebaee695394e..b2af6645ea7e 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -347,6 +347,19 @@ static int test_bit(int nr, const volatile unsigned long *addr); ? constant_test_bit((nr), (addr)) \ : variable_test_bit((nr), (addr))) +#if (defined(CONFIG_X86_GENERIC) || defined(CONFIG_GENERIC_CPU)) \ + && !defined(CONFIG_CC_OPTIMIZE_FOR_SIZE) +/* + * Since BSF and TZCNT have sufficiently similar semantics for the purposes + * for which we use them here, BMI-capable hardware will decode the prefixed + * variant as 'tzcnt ...' and may execute that faster than 'bsf ...', while + * older hardware will ignore the REP prefix and decode it as 'bsf ...'. + */ +# define BSF_PREFIX "rep;" +#else +# define BSF_PREFIX +#endif + /** * __ffs - find first set bit in word * @word: The word to search @@ -355,7 +368,7 @@ static int test_bit(int nr, const volatile unsigned long *addr); */ static inline unsigned long __ffs(unsigned long word) { - asm("bsf %1,%0" + asm(BSF_PREFIX "bsf %1,%0" : "=r" (word) : "rm" (word)); return word; @@ -369,12 +382,14 @@ static inline unsigned long __ffs(unsigned long word) */ static inline unsigned long ffz(unsigned long word) { - asm("bsf %1,%0" + asm(BSF_PREFIX "bsf %1,%0" : "=r" (word) : "r" (~word)); return word; } +#undef BSF_PREFIX + /* * __fls: find last set bit in word * @word: The word to search From e26a44a2d618a491d5c6a2a8aaf66ee03a94739f Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Tue, 18 Sep 2012 12:16:14 +0100 Subject: [PATCH 5/7] x86: Use REP BSF unconditionally Make "REP BSF" unconditional, as per the suggestion of hpa and Linus, this removes the insane BSF_PREFIX conditional and simplifies the logic. Suggested-by: "H. Peter Anvin" Suggested-by: Linus Torvalds Signed-off-by: Jan Beulich Link: http://lkml.kernel.org/r/5058741E020000780009C014@nat28.tlf.novell.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/bitops.h | 19 ++----------------- 1 file changed, 2 insertions(+), 17 deletions(-) diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index b2af6645ea7e..6dfd0195bb55 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -347,19 +347,6 @@ static int test_bit(int nr, const volatile unsigned long *addr); ? constant_test_bit((nr), (addr)) \ : variable_test_bit((nr), (addr))) -#if (defined(CONFIG_X86_GENERIC) || defined(CONFIG_GENERIC_CPU)) \ - && !defined(CONFIG_CC_OPTIMIZE_FOR_SIZE) -/* - * Since BSF and TZCNT have sufficiently similar semantics for the purposes - * for which we use them here, BMI-capable hardware will decode the prefixed - * variant as 'tzcnt ...' and may execute that faster than 'bsf ...', while - * older hardware will ignore the REP prefix and decode it as 'bsf ...'. - */ -# define BSF_PREFIX "rep;" -#else -# define BSF_PREFIX -#endif - /** * __ffs - find first set bit in word * @word: The word to search @@ -368,7 +355,7 @@ static int test_bit(int nr, const volatile unsigned long *addr); */ static inline unsigned long __ffs(unsigned long word) { - asm(BSF_PREFIX "bsf %1,%0" + asm("rep; bsf %1,%0" : "=r" (word) : "rm" (word)); return word; @@ -382,14 +369,12 @@ static inline unsigned long __ffs(unsigned long word) */ static inline unsigned long ffz(unsigned long word) { - asm(BSF_PREFIX "bsf %1,%0" + asm("rep; bsf %1,%0" : "=r" (word) : "r" (~word)); return word; } -#undef BSF_PREFIX - /* * __fls: find last set bit in word * @word: The word to search From 1b2b23d8573076a587ed2081e0d2b69691079e0e Mon Sep 17 00:00:00 2001 From: Tao Guo Date: Wed, 26 Sep 2012 04:28:22 -0400 Subject: [PATCH 6/7] x86_64: Work around old GAS bug GAS in binutils(2.16.91) could not parse parentheses within macro parameters unless fully parenthesized, and this is a workaround to make old gas work without generating below errors: arch/x86/kernel/entry_64.S: Assembler messages: arch/x86/kernel/entry_64.S:387: Error: too many positional arguments arch/x86/kernel/entry_64.S:389: Error: too many positional arguments [...] Signed-off-by: Tao Guo Reluctantly-Acked-by: Jan Beulich Cc: Linus Torvalds Cc: Andrew Morton Link: http://lkml.kernel.org/r/1348648102-12653-1-git-send-email-glorioustao@gmail.com [ Jan argues that these old GAS versions are fragile - which is so, but lets give them a chance. ] Signed-off-by: Ingo Molnar --- arch/x86/include/asm/calling.h | 48 ++++++++++++++++------------------ arch/x86/kernel/entry_64.S | 20 +++++++------- 2 files changed, 33 insertions(+), 35 deletions(-) diff --git a/arch/x86/include/asm/calling.h b/arch/x86/include/asm/calling.h index a9e3a740f697..7f8422a28a46 100644 --- a/arch/x86/include/asm/calling.h +++ b/arch/x86/include/asm/calling.h @@ -49,38 +49,36 @@ For 32-bit we have the following conventions - kernel is built with #include "dwarf2.h" /* - * 64-bit system call stack frame layout defines and helpers, for - * assembly code (note that the seemingly unnecessary parentheses - * are to prevent cpp from inserting spaces in expressions that get - * passed to macros): + * 64-bit system call stack frame layout defines and helpers, + * for assembly code: */ -#define R15 (0) -#define R14 (8) -#define R13 (16) -#define R12 (24) -#define RBP (32) -#define RBX (40) +#define R15 0 +#define R14 8 +#define R13 16 +#define R12 24 +#define RBP 32 +#define RBX 40 /* arguments: interrupts/non tracing syscalls only save up to here: */ -#define R11 (48) -#define R10 (56) -#define R9 (64) -#define R8 (72) -#define RAX (80) -#define RCX (88) -#define RDX (96) -#define RSI (104) -#define RDI (112) -#define ORIG_RAX (120) /* + error_code */ +#define R11 48 +#define R10 56 +#define R9 64 +#define R8 72 +#define RAX 80 +#define RCX 88 +#define RDX 96 +#define RSI 104 +#define RDI 112 +#define ORIG_RAX 120 /* + error_code */ /* end of arguments */ /* cpu exception frame or undefined in case of fast syscall: */ -#define RIP (128) -#define CS (136) -#define EFLAGS (144) -#define RSP (152) -#define SS (160) +#define RIP 128 +#define CS 136 +#define EFLAGS 144 +#define RSP 152 +#define SS 160 #define ARGOFFSET R11 #define SWFRAME ORIG_RAX diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index b1dac12dc5e6..2c6706167c8d 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -342,15 +342,15 @@ ENDPROC(native_usergs_sysret64) .macro SAVE_ARGS_IRQ cld /* start from rbp in pt_regs and jump over */ - movq_cfi rdi, RDI-RBP - movq_cfi rsi, RSI-RBP - movq_cfi rdx, RDX-RBP - movq_cfi rcx, RCX-RBP - movq_cfi rax, RAX-RBP - movq_cfi r8, R8-RBP - movq_cfi r9, R9-RBP - movq_cfi r10, R10-RBP - movq_cfi r11, R11-RBP + movq_cfi rdi, (RDI-RBP) + movq_cfi rsi, (RSI-RBP) + movq_cfi rdx, (RDX-RBP) + movq_cfi rcx, (RCX-RBP) + movq_cfi rax, (RAX-RBP) + movq_cfi r8, (R8-RBP) + movq_cfi r9, (R9-RBP) + movq_cfi r10, (R10-RBP) + movq_cfi r11, (R11-RBP) /* Save rbp so that we can unwind from get_irq_regs() */ movq_cfi rbp, 0 @@ -384,7 +384,7 @@ ENDPROC(native_usergs_sysret64) .endm ENTRY(save_rest) - PARTIAL_FRAME 1 REST_SKIP+8 + PARTIAL_FRAME 1 (REST_SKIP+8) movq 5*8+16(%rsp), %r11 /* save return address */ movq_cfi rbx, RBX+16 movq_cfi rbp, RBP+16 From c416ddf5b909736f5b57d348f5de159693e699ad Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 25 Sep 2012 14:51:19 +0200 Subject: [PATCH 7/7] x86: Unspaghettize do_trap() Cleanup the label maze in this function. Having a seperate function to first handle the traps that don't generate a signal makes it easier to convert into more readable conditional paths. Signed-off-by: Frederic Weisbecker Cc: Oleg Nesterov Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1348577479-2564-1-git-send-email-fweisbec@gmail.com [ Fixed 32-bit build failure. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/traps.c | 66 ++++++++++++++++++++--------------------- 1 file changed, 32 insertions(+), 34 deletions(-) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index b481341c9369..6ff771559af3 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -107,30 +107,45 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) dec_preempt_count(); } +static int __kprobes +do_trap_no_signal(struct task_struct *tsk, int trapnr, char *str, + struct pt_regs *regs, long error_code) +{ +#ifdef CONFIG_X86_32 + if (regs->flags & X86_VM_MASK) { + /* + * Traps 0, 1, 3, 4, and 5 should be forwarded to vm86. + * On nmi (interrupt 2), do_trap should not be called. + */ + if (trapnr < X86_TRAP_UD) { + if (!handle_vm86_trap((struct kernel_vm86_regs *) regs, + error_code, trapnr)) + return 0; + } + return -1; + } +#endif + if (!user_mode(regs)) { + if (!fixup_exception(regs)) { + tsk->thread.error_code = error_code; + tsk->thread.trap_nr = trapnr; + die(str, regs, error_code); + } + return 0; + } + + return -1; +} + static void __kprobes do_trap(int trapnr, int signr, char *str, struct pt_regs *regs, long error_code, siginfo_t *info) { struct task_struct *tsk = current; -#ifdef CONFIG_X86_32 - if (regs->flags & X86_VM_MASK) { - /* - * traps 0, 1, 3, 4, and 5 should be forwarded to vm86. - * On nmi (interrupt 2), do_trap should not be called. - */ - if (trapnr < X86_TRAP_UD) - goto vm86_trap; - goto trap_signal; - } -#endif - if (!user_mode(regs)) - goto kernel_trap; - -#ifdef CONFIG_X86_32 -trap_signal: -#endif + if (!do_trap_no_signal(tsk, trapnr, str, regs, error_code)) + return; /* * We want error_code and trap_nr set for userspace faults and * kernelspace faults which result in die(), but not @@ -158,23 +173,6 @@ trap_signal: force_sig_info(signr, info, tsk); else force_sig(signr, tsk); - return; - -kernel_trap: - if (!fixup_exception(regs)) { - tsk->thread.error_code = error_code; - tsk->thread.trap_nr = trapnr; - die(str, regs, error_code); - } - return; - -#ifdef CONFIG_X86_32 -vm86_trap: - if (handle_vm86_trap((struct kernel_vm86_regs *) regs, - error_code, trapnr)) - goto trap_signal; - return; -#endif } #define DO_ERROR(trapnr, signr, str, name) \