Merge branch 'core/percpu' into x86/core

1 files changed, 283 insertions, 160 deletions

diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 65efd42454be..e99206831459 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -30,12 +30,13 @@
  *	1C(%esp) - %ds
  *	20(%esp) - %es
  *	24(%esp) - %fs
- *	28(%esp) - orig_eax
- *	2C(%esp) - %eip
- *	30(%esp) - %cs
- *	34(%esp) - %eflags
- *	38(%esp) - %oldesp
- *	3C(%esp) - %oldss
+ *	28(%esp) - %gs		saved iff !CONFIG_X86_32_LAZY_GS
+ *	2C(%esp) - orig_eax
+ *	30(%esp) - %eip
+ *	34(%esp) - %cs
+ *	38(%esp) - %eflags
+ *	3C(%esp) - %oldesp
+ *	40(%esp) - %oldss
  *
  * "current" is in register %ebx during any slow entries.
  */
@@ -101,121 +102,221 @@
 #define resume_userspace_sig	resume_userspace
 #endif
 
-#define SAVE_ALL \
-	cld; \
-	pushl %fs; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	/*CFI_REL_OFFSET fs, 0;*/\
-	pushl %es; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	/*CFI_REL_OFFSET es, 0;*/\
-	pushl %ds; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	/*CFI_REL_OFFSET ds, 0;*/\
-	pushl %eax; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET eax, 0;\
-	pushl %ebp; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET ebp, 0;\
-	pushl %edi; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET edi, 0;\
-	pushl %esi; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET esi, 0;\
-	pushl %edx; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET edx, 0;\
-	pushl %ecx; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET ecx, 0;\
-	pushl %ebx; \
-	CFI_ADJUST_CFA_OFFSET 4;\
-	CFI_REL_OFFSET ebx, 0;\
-	movl $(__USER_DS), %edx; \
-	movl %edx, %ds; \
-	movl %edx, %es; \
-	movl $(__KERNEL_PERCPU), %edx; \
+/*
+ * User gs save/restore
+ *
+ * %gs is used for userland TLS and kernel only uses it for stack
+ * canary which is required to be at %gs:20 by gcc.  Read the comment
+ * at the top of stackprotector.h for more info.
+ *
+ * Local labels 98 and 99 are used.
+ */
+#ifdef CONFIG_X86_32_LAZY_GS
+
+ /* unfortunately push/pop can't be no-op */
+.macro PUSH_GS
+	pushl $0
+	CFI_ADJUST_CFA_OFFSET 4
+.endm
+.macro POP_GS pop=0
+	addl $(4 + \pop), %esp
+	CFI_ADJUST_CFA_OFFSET -(4 + \pop)
+.endm
+.macro POP_GS_EX
+.endm
+
+ /* all the rest are no-op */
+.macro PTGS_TO_GS
+.endm
+.macro PTGS_TO_GS_EX
+.endm
+.macro GS_TO_REG reg
+.endm
+.macro REG_TO_PTGS reg
+.endm
+.macro SET_KERNEL_GS reg
+.endm
+
+#else	/* CONFIG_X86_32_LAZY_GS */
+
+.macro PUSH_GS
+	pushl %gs
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET gs, 0*/
+.endm
+
+.macro POP_GS pop=0
+98:	popl %gs
+	CFI_ADJUST_CFA_OFFSET -4
+	/*CFI_RESTORE gs*/
+  .if \pop <> 0
+	add $\pop, %esp
+	CFI_ADJUST_CFA_OFFSET -\pop
+  .endif
+.endm
+.macro POP_GS_EX
+.pushsection .fixup, "ax"
+99:	movl $0, (%esp)
+	jmp 98b
+.section __ex_table, "a"
+	.align 4
+	.long 98b, 99b
+.popsection
+.endm
+
+.macro PTGS_TO_GS
+98:	mov PT_GS(%esp), %gs
+.endm
+.macro PTGS_TO_GS_EX
+.pushsection .fixup, "ax"
+99:	movl $0, PT_GS(%esp)
+	jmp 98b
+.section __ex_table, "a"
+	.align 4
+	.long 98b, 99b
+.popsection
+.endm
+
+.macro GS_TO_REG reg
+	movl %gs, \reg
+	/*CFI_REGISTER gs, \reg*/
+.endm
+.macro REG_TO_PTGS reg
+	movl \reg, PT_GS(%esp)
+	/*CFI_REL_OFFSET gs, PT_GS*/
+.endm
+.macro SET_KERNEL_GS reg
+	movl $(__KERNEL_STACK_CANARY), \reg
+	movl \reg, %gs
+.endm
+
+#endif	/* CONFIG_X86_32_LAZY_GS */
+
+.macro SAVE_ALL
+	cld
+	PUSH_GS
+	pushl %fs
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET fs, 0;*/
+	pushl %es
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET es, 0;*/
+	pushl %ds
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET ds, 0;*/
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET eax, 0
+	pushl %ebp
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebp, 0
+	pushl %edi
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edi, 0
+	pushl %esi
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET esi, 0
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx, 0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx, 0
+	pushl %ebx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebx, 0
+	movl $(__USER_DS), %edx
+	movl %edx, %ds
+	movl %edx, %es
+	movl $(__KERNEL_PERCPU), %edx
 	movl %edx, %fs
+	SET_KERNEL_GS %edx
+.endm
 
-#define RESTORE_INT_REGS \
-	popl %ebx;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE ebx;\
-	popl %ecx;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE ecx;\
-	popl %edx;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE edx;\
-	popl %esi;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE esi;\
-	popl %edi;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE edi;\
-	popl %ebp;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	CFI_RESTORE ebp;\
-	popl %eax;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
+.macro RESTORE_INT_REGS
+	popl %ebx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ebx
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	popl %esi
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE esi
+	popl %edi
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edi
+	popl %ebp
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ebp
+	popl %eax
+	CFI_ADJUST_CFA_OFFSET -4
 	CFI_RESTORE eax
+.endm
 
-#define RESTORE_REGS	\
-	RESTORE_INT_REGS; \
-1:	popl %ds;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	/*CFI_RESTORE ds;*/\
-2:	popl %es;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	/*CFI_RESTORE es;*/\
-3:	popl %fs;	\
-	CFI_ADJUST_CFA_OFFSET -4;\
-	/*CFI_RESTORE fs;*/\
-.pushsection .fixup,"ax";	\
-4:	movl $0,(%esp);	\
-	jmp 1b;		\
-5:	movl $0,(%esp);	\
-	jmp 2b;		\
-6:	movl $0,(%esp);	\
-	jmp 3b;		\
-.section __ex_table,"a";\
-	.align 4;	\
-	.long 1b,4b;	\
-	.long 2b,5b;	\
-	.long 3b,6b;	\
+.macro RESTORE_REGS pop=0
+	RESTORE_INT_REGS
+1:	popl %ds
+	CFI_ADJUST_CFA_OFFSET -4
+	/*CFI_RESTORE ds;*/
+2:	popl %es
+	CFI_ADJUST_CFA_OFFSET -4
+	/*CFI_RESTORE es;*/
+3:	popl %fs
+	CFI_ADJUST_CFA_OFFSET -4
+	/*CFI_RESTORE fs;*/
+	POP_GS \pop
+.pushsection .fixup, "ax"
+4:	movl $0, (%esp)
+	jmp 1b
+5:	movl $0, (%esp)
+	jmp 2b
+6:	movl $0, (%esp)
+	jmp 3b
+.section __ex_table, "a"
+	.align 4
+	.long 1b, 4b
+	.long 2b, 5b
+	.long 3b, 6b
 .popsection
+	POP_GS_EX
+.endm
 
-#define RING0_INT_FRAME \
-	CFI_STARTPROC simple;\
-	CFI_SIGNAL_FRAME;\
-	CFI_DEF_CFA esp, 3*4;\
-	/*CFI_OFFSET cs, -2*4;*/\
+.macro RING0_INT_FRAME
+	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
+	CFI_DEF_CFA esp, 3*4
+	/*CFI_OFFSET cs, -2*4;*/
 	CFI_OFFSET eip, -3*4
+.endm
 
-#define RING0_EC_FRAME \
-	CFI_STARTPROC simple;\
-	CFI_SIGNAL_FRAME;\
-	CFI_DEF_CFA esp, 4*4;\
-	/*CFI_OFFSET cs, -2*4;*/\
+.macro RING0_EC_FRAME
+	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
+	CFI_DEF_CFA esp, 4*4
+	/*CFI_OFFSET cs, -2*4;*/
 	CFI_OFFSET eip, -3*4
+.endm
 
-#define RING0_PTREGS_FRAME \
-	CFI_STARTPROC simple;\
-	CFI_SIGNAL_FRAME;\
-	CFI_DEF_CFA esp, PT_OLDESP-PT_EBX;\
-	/*CFI_OFFSET cs, PT_CS-PT_OLDESP;*/\
-	CFI_OFFSET eip, PT_EIP-PT_OLDESP;\
-	/*CFI_OFFSET es, PT_ES-PT_OLDESP;*/\
-	/*CFI_OFFSET ds, PT_DS-PT_OLDESP;*/\
-	CFI_OFFSET eax, PT_EAX-PT_OLDESP;\
-	CFI_OFFSET ebp, PT_EBP-PT_OLDESP;\
-	CFI_OFFSET edi, PT_EDI-PT_OLDESP;\
-	CFI_OFFSET esi, PT_ESI-PT_OLDESP;\
-	CFI_OFFSET edx, PT_EDX-PT_OLDESP;\
-	CFI_OFFSET ecx, PT_ECX-PT_OLDESP;\
+.macro RING0_PTREGS_FRAME
+	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
+	CFI_DEF_CFA esp, PT_OLDESP-PT_EBX
+	/*CFI_OFFSET cs, PT_CS-PT_OLDESP;*/
+	CFI_OFFSET eip, PT_EIP-PT_OLDESP
+	/*CFI_OFFSET es, PT_ES-PT_OLDESP;*/
+	/*CFI_OFFSET ds, PT_DS-PT_OLDESP;*/
+	CFI_OFFSET eax, PT_EAX-PT_OLDESP
+	CFI_OFFSET ebp, PT_EBP-PT_OLDESP
+	CFI_OFFSET edi, PT_EDI-PT_OLDESP
+	CFI_OFFSET esi, PT_ESI-PT_OLDESP
+	CFI_OFFSET edx, PT_EDX-PT_OLDESP
+	CFI_OFFSET ecx, PT_ECX-PT_OLDESP
 	CFI_OFFSET ebx, PT_EBX-PT_OLDESP
+.endm
 
 ENTRY(ret_from_fork)
 	CFI_STARTPROC
@@ -362,6 +463,7 @@ sysenter_exit:
 	xorl %ebp,%ebp
 	TRACE_IRQS_ON
 1:	mov  PT_FS(%esp), %fs
+	PTGS_TO_GS
 	ENABLE_INTERRUPTS_SYSEXIT
 
 #ifdef CONFIG_AUDITSYSCALL
@@ -410,6 +512,7 @@ sysexit_audit:
 	.align 4
 	.long 1b,2b
 .popsection
+	PTGS_TO_GS_EX
 ENDPROC(ia32_sysenter_target)
 
 	# system call handler stub
@@ -452,8 +555,7 @@ restore_all:
 restore_nocheck:
 	TRACE_IRQS_IRET
 restore_nocheck_notrace:
-	RESTORE_REGS
-	addl $4, %esp			# skip orig_eax/error_code
+	RESTORE_REGS 4			# skip orig_eax/error_code
 	CFI_ADJUST_CFA_OFFSET -4
 irq_return:
 	INTERRUPT_RETURN
@@ -595,28 +697,50 @@ syscall_badsys:
 END(syscall_badsys)
 	CFI_ENDPROC
 
-#define FIXUP_ESPFIX_STACK \
-	/* since we are on a wrong stack, we cant make it a C code :( */ \
-	PER_CPU(gdt_page, %ebx); \
-	GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \
-	addl %esp, %eax; \
-	pushl $__KERNEL_DS; \
-	CFI_ADJUST_CFA_OFFSET 4; \
-	pushl %eax; \
-	CFI_ADJUST_CFA_OFFSET 4; \
-	lss (%esp), %esp; \
-	CFI_ADJUST_CFA_OFFSET -8;
-#define UNWIND_ESPFIX_STACK \
-	movl %ss, %eax; \
-	/* see if on espfix stack */ \
-	cmpw $__ESPFIX_SS, %ax; \
-	jne 27f; \
-	movl $__KERNEL_DS, %eax; \
-	movl %eax, %ds; \
-	movl %eax, %es; \
-	/* switch to normal stack */ \
-	FIXUP_ESPFIX_STACK; \
-27:;
+/*
+ * System calls that need a pt_regs pointer.
+ */
+#define PTREGSCALL(name) \
+	ALIGN; \
+ptregs_##name: \
+	leal 4(%esp),%eax; \
+	jmp sys_##name;
+
+PTREGSCALL(iopl)
+PTREGSCALL(fork)
+PTREGSCALL(clone)
+PTREGSCALL(vfork)
+PTREGSCALL(execve)
+PTREGSCALL(sigaltstack)
+PTREGSCALL(sigreturn)
+PTREGSCALL(rt_sigreturn)
+PTREGSCALL(vm86)
+PTREGSCALL(vm86old)
+
+.macro FIXUP_ESPFIX_STACK
+	/* since we are on a wrong stack, we cant make it a C code :( */
+	PER_CPU(gdt_page, %ebx)
+	GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah)
+	addl %esp, %eax
+	pushl $__KERNEL_DS
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	lss (%esp), %esp
+	CFI_ADJUST_CFA_OFFSET -8
+.endm
+.macro UNWIND_ESPFIX_STACK
+	movl %ss, %eax
+	/* see if on espfix stack */
+	cmpw $__ESPFIX_SS, %ax
+	jne 27f
+	movl $__KERNEL_DS, %eax
+	movl %eax, %ds
+	movl %eax, %es
+	/* switch to normal stack */
+	FIXUP_ESPFIX_STACK
+27:
+.endm
 
 /*
  * Build the entry stubs and pointer table with some assembler magic.
@@ -1070,7 +1194,10 @@ ENTRY(page_fault)
 	CFI_ADJUST_CFA_OFFSET 4
 	ALIGN
 error_code:
-	/* the function address is in %fs's slot on the stack */
+	/* the function address is in %gs's slot on the stack */
+	pushl %fs
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET fs, 0*/
 	pushl %es
 	CFI_ADJUST_CFA_OFFSET 4
 	/*CFI_REL_OFFSET es, 0*/
@@ -1099,20 +1226,15 @@ error_code:
 	CFI_ADJUST_CFA_OFFSET 4
 	CFI_REL_OFFSET ebx, 0
 	cld
-	pushl %fs
-	CFI_ADJUST_CFA_OFFSET 4
-	/*CFI_REL_OFFSET fs, 0*/
 	movl $(__KERNEL_PERCPU), %ecx
 	movl %ecx, %fs
 	UNWIND_ESPFIX_STACK
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	/*CFI_REGISTER es, ecx*/
-	movl PT_FS(%esp), %edi		# get the function address
+	GS_TO_REG %ecx
+	movl PT_GS(%esp), %edi		# get the function address
 	movl PT_ORIG_EAX(%esp), %edx	# get the error code
 	movl $-1, PT_ORIG_EAX(%esp)	# no syscall to restart
-	mov  %ecx, PT_FS(%esp)
-	/*CFI_REL_OFFSET fs, ES*/
+	REG_TO_PTGS %ecx
+	SET_KERNEL_GS %ecx
 	movl $(__USER_DS), %ecx
 	movl %ecx, %ds
 	movl %ecx, %es
@@ -1136,26 +1258,27 @@ END(page_fault)
  * by hand onto the new stack - while updating the return eip past
  * the instruction that would have done it for sysenter.
  */
-#define FIX_STACK(offset, ok, label)		\
-	cmpw $__KERNEL_CS,4(%esp);		\
-	jne ok;					\
-label:						\
-	movl TSS_sysenter_sp0+offset(%esp),%esp;	\
-	CFI_DEF_CFA esp, 0;			\
-	CFI_UNDEFINED eip;			\
-	pushfl;					\
-	CFI_ADJUST_CFA_OFFSET 4;		\
-	pushl $__KERNEL_CS;			\
-	CFI_ADJUST_CFA_OFFSET 4;		\
-	pushl $sysenter_past_esp;		\
-	CFI_ADJUST_CFA_OFFSET 4;		\
+.macro FIX_STACK offset ok label
+	cmpw $__KERNEL_CS, 4(%esp)
+	jne \ok
+\label:
+	movl TSS_sysenter_sp0 + \offset(%esp), %esp
+	CFI_DEF_CFA esp, 0
+	CFI_UNDEFINED eip
+	pushfl
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl $__KERNEL_CS
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl $sysenter_past_esp
+	CFI_ADJUST_CFA_OFFSET 4
 	CFI_REL_OFFSET eip, 0
+.endm
 
 ENTRY(debug)
 	RING0_INT_FRAME
 	cmpl $ia32_sysenter_target,(%esp)
 	jne debug_stack_correct
-	FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
+	FIX_STACK 12, debug_stack_correct, debug_esp_fix_insn
 debug_stack_correct:
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
@@ -1213,7 +1336,7 @@ nmi_stack_correct:
 
 nmi_stack_fixup:
 	RING0_INT_FRAME
-	FIX_STACK(12,nmi_stack_correct, 1)
+	FIX_STACK 12, nmi_stack_correct, 1
 	jmp nmi_stack_correct
 
 nmi_debug_stack_check:
@@ -1224,7 +1347,7 @@ nmi_debug_stack_check:
 	jb nmi_stack_correct
 	cmpl $debug_esp_fix_insn,(%esp)
 	ja nmi_stack_correct
-	FIX_STACK(24,nmi_stack_correct, 1)
+	FIX_STACK 24, nmi_stack_correct, 1
 	jmp nmi_stack_correct
 
 nmi_espfix_stack: