* [PATCH 1/2] remove __init modifier from header declaration @ 2007-12-18 16:52 Glauber de Oliveira Costa 2007-12-18 16:52 ` [PATCH 2/2] finish processor.h integration Glauber de Oliveira Costa 2007-12-19 10:12 ` [PATCH 1/2] remove __init modifier from header declaration Ingo Molnar 0 siblings, 2 replies; 6+ messages in thread From: Glauber de Oliveira Costa @ 2007-12-18 16:52 UTC (permalink / raw) To: linux-kernel Cc: akpm, glommer, tglx, mingo, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland, Glauber de Oliveira Costa This patch removes the __init modifier from an extern function declaration in acpi.h. Besides not being strictly needed, it requires the inclusion of linux/init.h, which is usually not even included directly, increasing header mess by a lot. Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com> --- include/asm-x86/acpi.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/include/asm-x86/acpi.h b/include/asm-x86/acpi.h index c477e57..2feb0c4 100644 --- a/include/asm-x86/acpi.h +++ b/include/asm-x86/acpi.h @@ -158,7 +158,7 @@ extern int acpi_scan_nodes(unsigned long start, unsigned long end); #ifdef CONFIG_X86_64 # define NR_NODE_MEMBLKS (MAX_NUMNODES*2) #endif -extern void __init acpi_fake_nodes(const struct bootnode *fake_nodes, +extern void acpi_fake_nodes(const struct bootnode *fake_nodes, int num_nodes); #else static inline void acpi_fake_nodes(const struct bootnode *fake_nodes, -- 1.5.0.6 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH 2/2] finish processor.h integration 2007-12-18 16:52 [PATCH 1/2] remove __init modifier from header declaration Glauber de Oliveira Costa @ 2007-12-18 16:52 ` Glauber de Oliveira Costa 2007-12-18 20:03 ` Glauber de Oliveira Costa 2007-12-19 10:12 ` [PATCH 1/2] remove __init modifier from header declaration Ingo Molnar 1 sibling, 1 reply; 6+ messages in thread From: Glauber de Oliveira Costa @ 2007-12-18 16:52 UTC (permalink / raw) To: linux-kernel Cc: akpm, glommer, tglx, mingo, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland, Glauber de Oliveira Costa What's left in processor_32.h and processor_64.h cannot be cleanly integrated. However, it's just a couple of definitions. They are moved to processor.h around ifdefs, and the original files are deleted. Note that there's much less headers included in the final version. Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com> --- include/asm-x86/processor.h | 128 +++++++++++++++++++++++++++++++++++++++- include/asm-x86/processor_32.h | 97 ------------------------------ include/asm-x86/processor_64.h | 66 -------------------- 3 files changed, 126 insertions(+), 165 deletions(-) delete mode 100644 include/asm-x86/processor_32.h delete mode 100644 include/asm-x86/processor_64.h diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h index 80c5002..decb740 100644 --- a/include/asm-x86/processor.h +++ b/include/asm-x86/processor.h @@ -11,6 +11,8 @@ struct mm_struct; #include <asm/percpu.h> #include <asm/system.h> #include <asm/percpu.h> +#include <asm/msr.h> +#include <asm/desc_defs.h> #include <linux/cpumask.h> #include <linux/cache.h> @@ -275,7 +277,11 @@ union i387_union { struct i387_soft_struct soft; }; -# include "processor_32.h" +/* + * the following now lives in the per cpu area: + * extern int cpu_llc_id[NR_CPUS]; + */ +DECLARE_PER_CPU(u8, cpu_llc_id); #else struct i387_fxsave_struct { u16 cwd; @@ -295,7 +301,7 @@ union i387_union { struct i387_fxsave_struct fxsave; }; -# include "processor_64.h" +DECLARE_PER_CPU(struct orig_ist, orig_ist); #endif extern void print_cpu_info(struct cpuinfo_x86 *); @@ -778,6 +784,124 @@ static inline void prefetchw(const void *x) } #define spin_lock_prefetch(x) prefetchw(x) +#ifdef CONFIG_X86_32 +/* + * User space process size: 3GB (default). + */ +#define TASK_SIZE (PAGE_OFFSET) + +#define INIT_THREAD { \ + .sp0 = sizeof(init_stack) + (long)&init_stack, \ + .vm86_info = NULL, \ + .sysenter_cs = __KERNEL_CS, \ + .io_bitmap_ptr = NULL, \ + .fs = __KERNEL_PERCPU, \ +} + +/* + * Note that the .io_bitmap member must be extra-big. This is because + * the CPU will access an additional byte beyond the end of the IO + * permission bitmap. The extra byte must be all 1 bits, and must + * be within the limit. + */ +#define INIT_TSS { \ + .x86_tss = { \ + .sp0 = sizeof(init_stack) + (long)&init_stack, \ + .ss0 = __KERNEL_DS, \ + .ss1 = __KERNEL_CS, \ + .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ + }, \ + .io_bitmap = { [0 ... IO_BITMAP_LONGS] = ~0 }, \ +} + +#define start_thread(regs, new_eip, new_esp) do { \ + __asm__("movl %0,%%gs": :"r" (0)); \ + regs->fs = 0; \ + set_fs(USER_DS); \ + regs->ds = __USER_DS; \ + regs->es = __USER_DS; \ + regs->ss = __USER_DS; \ + regs->cs = __USER_CS; \ + regs->ip = new_eip; \ + regs->sp = new_esp; \ +} while (0) + + +extern unsigned long thread_saved_pc(struct task_struct *tsk); + +#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) +#define KSTK_TOP(info) \ +({ \ + unsigned long *__ptr = (unsigned long *)(info); \ + (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ +}) + +/* + * The below -8 is to reserve 8 bytes on top of the ring0 stack. + * This is necessary to guarantee that the entire "struct pt_regs" + * is accessable even if the CPU haven't stored the SS/ESP registers + * on the stack (interrupt gate does not save these registers + * when switching to the same priv ring). + * Therefore beware: accessing the ss/esp fields of the + * "struct pt_regs" is possible, but they may contain the + * completely wrong values. + */ +#define task_pt_regs(task) \ +({ \ + struct pt_regs *__regs__; \ + __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ + __regs__ - 1; \ +}) + +#define KSTK_ESP(task) (task_pt_regs(task)->sp) + +#else +/* + * User space process size. 47bits minus one guard page. + */ +#define TASK_SIZE64 (0x800000000000UL - 4096) + +/* This decides where the kernel will search for a free chunk of vm + * space during mmap's. + */ +#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \ + 0xc0000000 : 0xFFFFe000) + +#define TASK_SIZE (test_thread_flag(TIF_IA32) ? \ + IA32_PAGE_OFFSET : TASK_SIZE64) +#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? \ + IA32_PAGE_OFFSET : TASK_SIZE64) + +#define INIT_THREAD { \ + .sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ +} + +#define INIT_TSS { \ + .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ +} + +#define start_thread(regs, new_rip, new_rsp) do { \ + asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \ + load_gs_index(0); \ + (regs)->ip = (new_rip); \ + (regs)->sp = (new_rsp); \ + write_pda(oldrsp, (new_rsp)); \ + (regs)->cs = __USER_CS; \ + (regs)->ss = __USER_DS; \ + (regs)->flags = 0x200; \ + set_fs(USER_DS); \ +} while (0) + +/* + * Return saved PC of a blocked thread. + * What is this good for? it will be always the scheduler or ret_from_fork. + */ +#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8)) + +#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1) +#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */ +#endif /* CONFIG_X86_64 */ + /* This decides where the kernel will search for a free chunk of vm * space during mmap's. */ diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h deleted file mode 100644 index 0d6a430..0000000 --- a/include/asm-x86/processor_32.h +++ /dev/null @@ -1,97 +0,0 @@ -/* - * Copyright (C) 1994 Linus Torvalds - */ - -#ifndef __ASM_I386_PROCESSOR_H -#define __ASM_I386_PROCESSOR_H - -#include <asm/vm86.h> -#include <asm/math_emu.h> -#include <asm/segment.h> -#include <asm/page.h> -#include <asm/types.h> -#include <asm/sigcontext.h> -#include <asm/cpufeature.h> -#include <asm/msr.h> -#include <asm/system.h> -#include <linux/threads.h> -#include <linux/init.h> -#include <asm/desc_defs.h> - -/* - * the following now lives in the per cpu area: - * extern int cpu_llc_id[NR_CPUS]; - */ -DECLARE_PER_CPU(u8, cpu_llc_id); - -/* - * User space process size: 3GB (default). - */ -#define TASK_SIZE (PAGE_OFFSET) - -#define INIT_THREAD { \ - .sp0 = sizeof(init_stack) + (long)&init_stack, \ - .vm86_info = NULL, \ - .sysenter_cs = __KERNEL_CS, \ - .io_bitmap_ptr = NULL, \ - .fs = __KERNEL_PERCPU, \ -} - -/* - * Note that the .io_bitmap member must be extra-big. This is because - * the CPU will access an additional byte beyond the end of the IO - * permission bitmap. The extra byte must be all 1 bits, and must - * be within the limit. - */ -#define INIT_TSS { \ - .x86_tss = { \ - .sp0 = sizeof(init_stack) + (long)&init_stack, \ - .ss0 = __KERNEL_DS, \ - .ss1 = __KERNEL_CS, \ - .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ - }, \ - .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \ -} - -#define start_thread(regs, new_eip, new_esp) do { \ - __asm__("movl %0,%%gs": :"r" (0)); \ - regs->fs = 0; \ - set_fs(USER_DS); \ - regs->ds = __USER_DS; \ - regs->es = __USER_DS; \ - regs->ss = __USER_DS; \ - regs->cs = __USER_CS; \ - regs->ip = new_eip; \ - regs->sp = new_esp; \ -} while (0) - - -extern unsigned long thread_saved_pc(struct task_struct *tsk); - -#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) -#define KSTK_TOP(info) \ -({ \ - unsigned long *__ptr = (unsigned long *)(info); \ - (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ -}) - -/* - * The below -8 is to reserve 8 bytes on top of the ring0 stack. - * This is necessary to guarantee that the entire "struct pt_regs" - * is accessable even if the CPU haven't stored the SS/ESP registers - * on the stack (interrupt gate does not save these registers - * when switching to the same priv ring). - * Therefore beware: accessing the ss/esp fields of the - * "struct pt_regs" is possible, but they may contain the - * completely wrong values. - */ -#define task_pt_regs(task) \ -({ \ - struct pt_regs *__regs__; \ - __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ - __regs__ - 1; \ -}) - -#define KSTK_ESP(task) (task_pt_regs(task)->sp) - -#endif /* __ASM_I386_PROCESSOR_H */ diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h deleted file mode 100644 index 04ce823..0000000 --- a/include/asm-x86/processor_64.h +++ /dev/null @@ -1,66 +0,0 @@ -/* - * Copyright (C) 1994 Linus Torvalds - */ - -#ifndef __ASM_X86_64_PROCESSOR_H -#define __ASM_X86_64_PROCESSOR_H - -#include <asm/segment.h> -#include <asm/page.h> -#include <asm/types.h> -#include <asm/sigcontext.h> -#include <asm/cpufeature.h> -#include <linux/threads.h> -#include <asm/msr.h> -#include <asm/current.h> -#include <asm/system.h> -#include <linux/personality.h> -#include <asm/desc_defs.h> - -/* - * User space process size. 47bits minus one guard page. - */ -#define TASK_SIZE64 (0x800000000000UL - 4096) - -/* This decides where the kernel will search for a free chunk of vm - * space during mmap's. - */ -#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000) - -#define TASK_SIZE (test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64) -#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64) - - - -DECLARE_PER_CPU(struct orig_ist, orig_ist); - -#define INIT_THREAD { \ - .sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ -} - -#define INIT_TSS { \ - .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ -} - -#define start_thread(regs,new_rip,new_rsp) do { \ - asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \ - load_gs_index(0); \ - (regs)->ip = (new_rip); \ - (regs)->sp = (new_rsp); \ - write_pda(oldrsp, (new_rsp)); \ - (regs)->cs = __USER_CS; \ - (regs)->ss = __USER_DS; \ - (regs)->flags = 0x200; \ - set_fs(USER_DS); \ -} while(0) - -/* - * Return saved PC of a blocked thread. - * What is this good for? it will be always the scheduler or ret_from_fork. - */ -#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8)) - -#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1) -#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */ - -#endif /* __ASM_X86_64_PROCESSOR_H */ -- 1.5.0.6 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 2/2] finish processor.h integration 2007-12-18 16:52 ` [PATCH 2/2] finish processor.h integration Glauber de Oliveira Costa @ 2007-12-18 20:03 ` Glauber de Oliveira Costa 0 siblings, 0 replies; 6+ messages in thread From: Glauber de Oliveira Costa @ 2007-12-18 20:03 UTC (permalink / raw) To: linux-kernel Cc: akpm, glommer, tglx, mingo, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland Glauber de Oliveira Costa wrote: > What's left in processor_32.h and processor_64.h cannot be cleanly > integrated. However, it's just a couple of definitions. They are moved > to processor.h around ifdefs, and the original files are deleted. Note that > there's much less headers included in the final version. > > Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com> > --- > include/asm-x86/processor.h | 128 +++++++++++++++++++++++++++++++++++++++- > include/asm-x86/processor_32.h | 97 ------------------------------ > include/asm-x86/processor_64.h | 66 -------------------- > 3 files changed, 126 insertions(+), 165 deletions(-) > delete mode 100644 include/asm-x86/processor_32.h > delete mode 100644 include/asm-x86/processor_64.h > > diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h > index 80c5002..decb740 100644 > --- a/include/asm-x86/processor.h > +++ b/include/asm-x86/processor.h > @@ -11,6 +11,8 @@ struct mm_struct; > #include <asm/percpu.h> > #include <asm/system.h> > #include <asm/percpu.h> > +#include <asm/msr.h> > +#include <asm/desc_defs.h> > #include <linux/cpumask.h> > #include <linux/cache.h> > > @@ -275,7 +277,11 @@ union i387_union { > struct i387_soft_struct soft; > }; > > -# include "processor_32.h" > +/* > + * the following now lives in the per cpu area: > + * extern int cpu_llc_id[NR_CPUS]; > + */ > +DECLARE_PER_CPU(u8, cpu_llc_id); > #else > struct i387_fxsave_struct { > u16 cwd; > @@ -295,7 +301,7 @@ union i387_union { > struct i387_fxsave_struct fxsave; > }; > > -# include "processor_64.h" > +DECLARE_PER_CPU(struct orig_ist, orig_ist); > #endif > > extern void print_cpu_info(struct cpuinfo_x86 *); > @@ -778,6 +784,124 @@ static inline void prefetchw(const void *x) > } > > #define spin_lock_prefetch(x) prefetchw(x) > +#ifdef CONFIG_X86_32 > +/* > + * User space process size: 3GB (default). > + */ > +#define TASK_SIZE (PAGE_OFFSET) > + > +#define INIT_THREAD { \ > + .sp0 = sizeof(init_stack) + (long)&init_stack, \ > + .vm86_info = NULL, \ > + .sysenter_cs = __KERNEL_CS, \ > + .io_bitmap_ptr = NULL, \ > + .fs = __KERNEL_PERCPU, \ > +} > + > +/* > + * Note that the .io_bitmap member must be extra-big. This is because > + * the CPU will access an additional byte beyond the end of the IO > + * permission bitmap. The extra byte must be all 1 bits, and must > + * be within the limit. > + */ > +#define INIT_TSS { \ > + .x86_tss = { \ > + .sp0 = sizeof(init_stack) + (long)&init_stack, \ > + .ss0 = __KERNEL_DS, \ > + .ss1 = __KERNEL_CS, \ > + .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ > + }, \ > + .io_bitmap = { [0 ... IO_BITMAP_LONGS] = ~0 }, \ > +} > + > +#define start_thread(regs, new_eip, new_esp) do { \ > + __asm__("movl %0,%%gs": :"r" (0)); \ > + regs->fs = 0; \ > + set_fs(USER_DS); \ > + regs->ds = __USER_DS; \ > + regs->es = __USER_DS; \ > + regs->ss = __USER_DS; \ > + regs->cs = __USER_CS; \ > + regs->ip = new_eip; \ > + regs->sp = new_esp; \ > +} while (0) > + > + > +extern unsigned long thread_saved_pc(struct task_struct *tsk); > + > +#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) > +#define KSTK_TOP(info) \ > +({ \ > + unsigned long *__ptr = (unsigned long *)(info); \ > + (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ > +}) > + > +/* > + * The below -8 is to reserve 8 bytes on top of the ring0 stack. > + * This is necessary to guarantee that the entire "struct pt_regs" > + * is accessable even if the CPU haven't stored the SS/ESP registers > + * on the stack (interrupt gate does not save these registers > + * when switching to the same priv ring). > + * Therefore beware: accessing the ss/esp fields of the > + * "struct pt_regs" is possible, but they may contain the > + * completely wrong values. > + */ > +#define task_pt_regs(task) \ > +({ \ > + struct pt_regs *__regs__; \ > + __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ > + __regs__ - 1; \ > +}) > + > +#define KSTK_ESP(task) (task_pt_regs(task)->sp) > + > +#else > +/* > + * User space process size. 47bits minus one guard page. > + */ > +#define TASK_SIZE64 (0x800000000000UL - 4096) > + > +/* This decides where the kernel will search for a free chunk of vm > + * space during mmap's. > + */ > +#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \ > + 0xc0000000 : 0xFFFFe000) > + > +#define TASK_SIZE (test_thread_flag(TIF_IA32) ? \ > + IA32_PAGE_OFFSET : TASK_SIZE64) > +#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? \ > + IA32_PAGE_OFFSET : TASK_SIZE64) > + > +#define INIT_THREAD { \ > + .sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ > +} > + > +#define INIT_TSS { \ > + .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ > +} > + > +#define start_thread(regs, new_rip, new_rsp) do { \ > + asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \ > + load_gs_index(0); \ > + (regs)->ip = (new_rip); \ > + (regs)->sp = (new_rsp); \ > + write_pda(oldrsp, (new_rsp)); \ > + (regs)->cs = __USER_CS; \ > + (regs)->ss = __USER_DS; \ > + (regs)->flags = 0x200; \ > + set_fs(USER_DS); \ > +} while (0) > + > +/* > + * Return saved PC of a blocked thread. > + * What is this good for? it will be always the scheduler or ret_from_fork. > + */ > +#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8)) > + > +#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1) > +#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */ > +#endif /* CONFIG_X86_64 */ > + > /* This decides where the kernel will search for a free chunk of vm > * space during mmap's. > */ > diff --git a/include/asm-x86/processor_32.h b/include/asm-x86/processor_32.h > deleted file mode 100644 > index 0d6a430..0000000 > --- a/include/asm-x86/processor_32.h > +++ /dev/null > @@ -1,97 +0,0 @@ > -/* > - * Copyright (C) 1994 Linus Torvalds > - */ > - > -#ifndef __ASM_I386_PROCESSOR_H > -#define __ASM_I386_PROCESSOR_H > - > -#include <asm/vm86.h> > -#include <asm/math_emu.h> > -#include <asm/segment.h> > -#include <asm/page.h> > -#include <asm/types.h> > -#include <asm/sigcontext.h> > -#include <asm/cpufeature.h> > -#include <asm/msr.h> > -#include <asm/system.h> > -#include <linux/threads.h> > -#include <linux/init.h> > -#include <asm/desc_defs.h> > - > -/* > - * the following now lives in the per cpu area: > - * extern int cpu_llc_id[NR_CPUS]; > - */ > -DECLARE_PER_CPU(u8, cpu_llc_id); > - > -/* > - * User space process size: 3GB (default). > - */ > -#define TASK_SIZE (PAGE_OFFSET) > - > -#define INIT_THREAD { \ > - .sp0 = sizeof(init_stack) + (long)&init_stack, \ > - .vm86_info = NULL, \ > - .sysenter_cs = __KERNEL_CS, \ > - .io_bitmap_ptr = NULL, \ > - .fs = __KERNEL_PERCPU, \ > -} > - > -/* > - * Note that the .io_bitmap member must be extra-big. This is because > - * the CPU will access an additional byte beyond the end of the IO > - * permission bitmap. The extra byte must be all 1 bits, and must > - * be within the limit. > - */ > -#define INIT_TSS { \ > - .x86_tss = { \ > - .sp0 = sizeof(init_stack) + (long)&init_stack, \ > - .ss0 = __KERNEL_DS, \ > - .ss1 = __KERNEL_CS, \ > - .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ > - }, \ > - .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \ > -} > - > -#define start_thread(regs, new_eip, new_esp) do { \ > - __asm__("movl %0,%%gs": :"r" (0)); \ > - regs->fs = 0; \ > - set_fs(USER_DS); \ > - regs->ds = __USER_DS; \ > - regs->es = __USER_DS; \ > - regs->ss = __USER_DS; \ > - regs->cs = __USER_CS; \ > - regs->ip = new_eip; \ > - regs->sp = new_esp; \ > -} while (0) > - > - > -extern unsigned long thread_saved_pc(struct task_struct *tsk); > - > -#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) > -#define KSTK_TOP(info) \ > -({ \ > - unsigned long *__ptr = (unsigned long *)(info); \ > - (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ > -}) > - > -/* > - * The below -8 is to reserve 8 bytes on top of the ring0 stack. > - * This is necessary to guarantee that the entire "struct pt_regs" > - * is accessable even if the CPU haven't stored the SS/ESP registers > - * on the stack (interrupt gate does not save these registers > - * when switching to the same priv ring). > - * Therefore beware: accessing the ss/esp fields of the > - * "struct pt_regs" is possible, but they may contain the > - * completely wrong values. > - */ > -#define task_pt_regs(task) \ > -({ \ > - struct pt_regs *__regs__; \ > - __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ > - __regs__ - 1; \ > -}) > - > -#define KSTK_ESP(task) (task_pt_regs(task)->sp) > - > -#endif /* __ASM_I386_PROCESSOR_H */ > diff --git a/include/asm-x86/processor_64.h b/include/asm-x86/processor_64.h > deleted file mode 100644 > index 04ce823..0000000 > --- a/include/asm-x86/processor_64.h > +++ /dev/null > @@ -1,66 +0,0 @@ > -/* > - * Copyright (C) 1994 Linus Torvalds > - */ > - > -#ifndef __ASM_X86_64_PROCESSOR_H > -#define __ASM_X86_64_PROCESSOR_H > - > -#include <asm/segment.h> > -#include <asm/page.h> > -#include <asm/types.h> > -#include <asm/sigcontext.h> > -#include <asm/cpufeature.h> > -#include <linux/threads.h> > -#include <asm/msr.h> > -#include <asm/current.h> > -#include <asm/system.h> > -#include <linux/personality.h> > -#include <asm/desc_defs.h> > - > -/* > - * User space process size. 47bits minus one guard page. > - */ > -#define TASK_SIZE64 (0x800000000000UL - 4096) > - > -/* This decides where the kernel will search for a free chunk of vm > - * space during mmap's. > - */ > -#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000) > - > -#define TASK_SIZE (test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64) > -#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64) > - > - > - > -DECLARE_PER_CPU(struct orig_ist, orig_ist); > - > -#define INIT_THREAD { \ > - .sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ > -} > - > -#define INIT_TSS { \ > - .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ > -} > - > -#define start_thread(regs,new_rip,new_rsp) do { \ > - asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \ > - load_gs_index(0); \ > - (regs)->ip = (new_rip); \ > - (regs)->sp = (new_rsp); \ > - write_pda(oldrsp, (new_rsp)); \ > - (regs)->cs = __USER_CS; \ > - (regs)->ss = __USER_DS; \ > - (regs)->flags = 0x200; \ > - set_fs(USER_DS); \ > -} while(0) > - > -/* > - * Return saved PC of a blocked thread. > - * What is this good for? it will be always the scheduler or ret_from_fork. > - */ > -#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.sp - 8)) > - > -#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.sp0 - 1) > -#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */ > - > -#endif /* __ASM_X86_64_PROCESSOR_H */ Okay, Ingo, this was a distraction of mine. As the patch was already ready, I just sent out after fixing, but you are obviously right, and it is not good even for changelog purposes. I'll send now a patch that includes all headers deleted from 64 and 32 bit, and in the future, remove the unused ones. The last patch is not needed, because we include init.h here, but it's not wrong, nevertheless. Please forget this one, and apply the one I'll send following this message. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/2] remove __init modifier from header declaration 2007-12-18 16:52 [PATCH 1/2] remove __init modifier from header declaration Glauber de Oliveira Costa 2007-12-18 16:52 ` [PATCH 2/2] finish processor.h integration Glauber de Oliveira Costa @ 2007-12-19 10:12 ` Ingo Molnar 2007-12-31 16:46 ` Alexey Dobriyan 1 sibling, 1 reply; 6+ messages in thread From: Ingo Molnar @ 2007-12-19 10:12 UTC (permalink / raw) To: Glauber de Oliveira Costa Cc: linux-kernel, akpm, glommer, tglx, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland * Glauber de Oliveira Costa <gcosta@redhat.com> wrote: > This patch removes the __init modifier from an extern function > declaration in acpi.h. > > Besides not being strictly needed, it requires the inclusion of > linux/init.h, which is usually not even included directly, increasing > header mess by a lot. thanks, applied. btw., people have been talking about reducing the include file mess for nearly a decade now, but it didnt get that much better - at least not in include/asm-x86/*.h. So include file dependency flattening patches would be more than welcome as well. (and unlike unification patches they have no expectation of being 100% perfect, so a natural ping-pong of fixes, until the changes are fully correct, would be natural.) Ingo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/2] remove __init modifier from header declaration 2007-12-19 10:12 ` [PATCH 1/2] remove __init modifier from header declaration Ingo Molnar @ 2007-12-31 16:46 ` Alexey Dobriyan 2008-01-01 16:57 ` Ingo Molnar 0 siblings, 1 reply; 6+ messages in thread From: Alexey Dobriyan @ 2007-12-31 16:46 UTC (permalink / raw) To: Ingo Molnar Cc: Glauber de Oliveira Costa, linux-kernel, akpm, glommer, tglx, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland On Wed, Dec 19, 2007 at 11:12:36AM +0100, Ingo Molnar wrote: > * Glauber de Oliveira Costa <gcosta@redhat.com> wrote: > > > This patch removes the __init modifier from an extern function > > declaration in acpi.h. > > > > Besides not being strictly needed, it requires the inclusion of > > linux/init.h, which is usually not even included directly, increasing > > header mess by a lot. > > thanks, applied. > > btw., people have been talking about reducing the include file mess for > nearly a decade now, Some of us are actually doing it. ;-) > but it didnt get that much better - at least not in > include/asm-x86/*.h. That's because hunting in include/linux/ is much more fruitful -- it decreases amount of code after preprocessing for everyone. Below are allyesconfig on x86_64 results for some kernels: $ wc ../*.i 79018171 229433654 2049266884 ../2.6.18.i 86568111 250674099 2245115663 ../2.6.19.i 85187296 247470579 2221334183 ../2.6.20.i <=== 88645422 242396928 2234855428 ../2.6.21.i 93897302 257990496 2377174442 ../2.6.22.i 98381373 268683402 2486321956 ../2.6.23.i E.g. we can have ~1.07% decrease in cpp output despite ~126000 lines added between .19 and .20. > So include file dependency flattening patches would > be more than welcome as well. Yup! Provided they're compile-tested sufficiently well. > (and unlike unification patches they have > no expectation of being 100% perfect, so a natural ping-pong of fixes, > until the changes are fully correct, would be natural.) That way you'll never clean anything especially during merge window. ;-) ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/2] remove __init modifier from header declaration 2007-12-31 16:46 ` Alexey Dobriyan @ 2008-01-01 16:57 ` Ingo Molnar 0 siblings, 0 replies; 6+ messages in thread From: Ingo Molnar @ 2008-01-01 16:57 UTC (permalink / raw) To: Alexey Dobriyan Cc: Glauber de Oliveira Costa, linux-kernel, akpm, glommer, tglx, ehabkost, jeremy, avi, anthony, virtualization, rusty, ak, chrisw, rostedt, hpa, zach, roland * Alexey Dobriyan <adobriyan@gmail.com> wrote: > > thanks, applied. > > > > btw., people have been talking about reducing the include file mess > > for nearly a decade now, > > Some of us are actually doing it. ;-) sorry, this was really not meant to slight your contributions in this area in any way :-) I'm just judging by the spaghetti still hanging around in asm-x86/*.h, and the 'fun' we have with paravirt and its type dependencies - and the periodic messups we have with macros. (and many macros are not inlines due to ugly dependencies.) > > So include file dependency flattening patches would be more than > > welcome as well. > > Yup! Provided they're compile-tested sufficiently well. as long as it builds/boots on your box with a single convenient .config of yours, we can stick such patches into x86.git and work out all the build failures it might cause. > > (and unlike unification patches they have no expectation of being > > 100% perfect, so a natural ping-pong of fixes, until the changes are > > fully correct, would be natural.) > > That way you'll never clean anything especially during merge window. > ;-) hm, what do you mean? Ingo ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2008-01-01 17:04 UTC | newest] Thread overview: 6+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2007-12-18 16:52 [PATCH 1/2] remove __init modifier from header declaration Glauber de Oliveira Costa 2007-12-18 16:52 ` [PATCH 2/2] finish processor.h integration Glauber de Oliveira Costa 2007-12-18 20:03 ` Glauber de Oliveira Costa 2007-12-19 10:12 ` [PATCH 1/2] remove __init modifier from header declaration Ingo Molnar 2007-12-31 16:46 ` Alexey Dobriyan 2008-01-01 16:57 ` Ingo Molnar
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox