Merge branches 'x86/mm', 'x86/build', 'x86/apic' and 'x86/platform' into x86/core, to apply dependent patch

Signed-off-by: Ingo Molnar <mingo@kernel.org>

Documentation/kernel-parameters.txt

@@ -746,6 +746,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 	cpuidle.off=1	[CPU_IDLE]
 			disable the cpuidle sub-system
 
+	cpu_init_udelay=N
+			[X86] Delay for N microsec between assert and de-assert
+			of APIC INIT to start processors.  This delay occurs
+			on every CPU online, such as boot, and resume from suspend.
+			Default: 10000
+
 	cpcihp_generic=	[HW,PCI] Generic port I/O CompactPCI driver
 			Format:
 			<first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>]

Documentation/x86/mtrr.txt

@@ -1,7 +1,19 @@
 MTRR (Memory Type Range Register) control
-3 Jun 1999
-Richard Gooch
-<rgooch@atnf.csiro.au>
+
+Richard Gooch <rgooch@atnf.csiro.au> - 3 Jun 1999
+Luis R. Rodriguez <mcgrof@do-not-panic.com> - April 9, 2015
+
+===============================================================================
+Phasing out MTRR use
+
+MTRR use is replaced on modern x86 hardware with PAT. Over time the only type
+of effective MTRR that is expected to be supported will be for write-combining.
+As MTRR use is phased out device drivers should use arch_phys_wc_add() to make
+MTRR effective on non-PAT systems while a no-op on PAT enabled systems.
+
+For details refer to Documentation/x86/pat.txt.
+
+===============================================================================
 
   On Intel P6 family processors (Pentium Pro, Pentium II and later)
   the Memory Type Range Registers (MTRRs) may be used to control

Documentation/x86/pat.txt

@@ -34,6 +34,8 @@ ioremap                |    --    |    UC-     |       UC-        |
                        |          |            |                  |
 ioremap_cache          |    --    |    WB      |       WB         |
                        |          |            |                  |
+ioremap_uc             |    --    |    UC      |       UC         |
+                       |          |            |                  |
 ioremap_nocache        |    --    |    UC-     |       UC-        |
                        |          |            |                  |
 ioremap_wc             |    --    |    --      |       WC         |
@@ -102,7 +104,38 @@ wants to export a RAM region, it has to do set_memory_uc() or set_memory_wc()
 as step 0 above and also track the usage of those pages and use set_memory_wb()
 before the page is freed to free pool.
 
-
+MTRR effects on PAT / non-PAT systems
+-------------------------------------
+
+The following table provides the effects of using write-combining MTRRs when
+using ioremap*() calls on x86 for both non-PAT and PAT systems. Ideally
+mtrr_add() usage will be phased out in favor of arch_phys_wc_add() which will
+be a no-op on PAT enabled systems. The region over which a arch_phys_wc_add()
+is made, should already have been ioremapped with WC attributes or PAT entries,
+this can be done by using ioremap_wc() / set_memory_wc().  Devices which
+combine areas of IO memory desired to remain uncacheable with areas where
+write-combining is desirable should consider use of ioremap_uc() followed by
+set_memory_wc() to white-list effective write-combined areas.  Such use is
+nevertheless discouraged as the effective memory type is considered
+implementation defined, yet this strategy can be used as last resort on devices
+with size-constrained regions where otherwise MTRR write-combining would
+otherwise not be effective.
+
+----------------------------------------------------------------------
+MTRR Non-PAT   PAT    Linux ioremap value        Effective memory type
+----------------------------------------------------------------------
+                                                  Non-PAT |  PAT
+     PAT
+     |PCD
+     ||PWT
+     |||
+WC   000      WB      _PAGE_CACHE_MODE_WB            WC   |   WC
+WC   001      WC      _PAGE_CACHE_MODE_WC            WC*  |   WC
+WC   010      UC-     _PAGE_CACHE_MODE_UC_MINUS      WC*  |   UC
+WC   011      UC      _PAGE_CACHE_MODE_UC            UC   |   UC
+----------------------------------------------------------------------
+
+(*) denotes implementation defined and is discouraged
 
 Notes:
 

arch/ia64/include/asm/irq_remapping.h

@@ -1,6 +1,4 @@
 #ifndef __IA64_INTR_REMAPPING_H
 #define __IA64_INTR_REMAPPING_H
 #define irq_remapping_enabled 0
-#define dmar_alloc_hwirq	create_irq
-#define dmar_free_hwirq		destroy_irq
 #endif

arch/ia64/kernel/msi_ia64.c

@@ -165,7 +165,7 @@ static struct irq_chip dmar_msi_type = {
 	.irq_retrigger = ia64_msi_retrigger_irq,
 };
 
-static int
+static void
 msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg)
 {
 	struct irq_cfg *cfg = irq_cfg + irq;
@@ -186,21 +186,29 @@ msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg)
 		MSI_DATA_LEVEL_ASSERT |
 		MSI_DATA_DELIVERY_FIXED |
 		MSI_DATA_VECTOR(cfg->vector);
-	return 0;
 }
 
-int arch_setup_dmar_msi(unsigned int irq)
+int dmar_alloc_hwirq(int id, int node, void *arg)
 {
-	int ret;
+	int irq;
 	struct msi_msg msg;
 
-	ret = msi_compose_msg(NULL, irq, &msg);
-	if (ret < 0)
-		return ret;
-	dmar_msi_write(irq, &msg);
-	irq_set_chip_and_handler_name(irq, &dmar_msi_type, handle_edge_irq,
-				      "edge");
-	return 0;
+	irq = create_irq();
+	if (irq > 0) {
+		irq_set_handler_data(irq, arg);
+		irq_set_chip_and_handler_name(irq, &dmar_msi_type,
+					      handle_edge_irq, "edge");
+		msi_compose_msg(NULL, irq, &msg);
+		dmar_msi_write(irq, &msg);
+	}
+
+	return irq;
+}
+
+void dmar_free_hwirq(int irq)
+{
+	irq_set_handler_data(irq, NULL);
+	destroy_irq(irq);
 }
 #endif /* CONFIG_INTEL_IOMMU */
 

arch/x86/Kconfig

@@ -100,7 +100,7 @@ config X86
 	select IRQ_FORCED_THREADING
 	select HAVE_BPF_JIT if X86_64
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE
-	select HAVE_ARCH_HUGE_VMAP if X86_64 || (X86_32 && X86_PAE)
+	select HAVE_ARCH_HUGE_VMAP if X86_64 || X86_PAE
 	select ARCH_HAS_SG_CHAIN
 	select CLKEVT_I8253
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
@@ -341,7 +341,7 @@ config X86_FEATURE_NAMES
 
 config X86_X2APIC
 	bool "Support x2apic"
-	depends on X86_LOCAL_APIC && X86_64 && IRQ_REMAP
+	depends on X86_LOCAL_APIC && X86_64 && (IRQ_REMAP || HYPERVISOR_GUEST)
 	---help---
 	  This enables x2apic support on CPUs that have this feature.
 
@@ -441,6 +441,7 @@ config X86_UV
 	depends on X86_EXTENDED_PLATFORM
 	depends on NUMA
 	depends on X86_X2APIC
+	depends on PCI
 	---help---
 	  This option is needed in order to support SGI Ultraviolet systems.
 	  If you don't have one of these, you should say N here.
@@ -466,7 +467,6 @@ config X86_INTEL_CE
 	select X86_REBOOTFIXUPS
 	select OF
 	select OF_EARLY_FLATTREE
-	select IRQ_DOMAIN
 	---help---
 	  Select for the Intel CE media processor (CE4100) SOC.
 	  This option compiles in support for the CE4100 SOC for settop
@@ -851,11 +851,12 @@ config NR_CPUS
 	default "1" if !SMP
 	default "8192" if MAXSMP
 	default "32" if SMP && X86_BIGSMP
-	default "8" if SMP
+	default "8" if SMP && X86_32
+	default "64" if SMP
 	---help---
 	  This allows you to specify the maximum number of CPUs which this
 	  kernel will support.  If CPUMASK_OFFSTACK is enabled, the maximum
-	  supported value is 4096, otherwise the maximum value is 512.  The
+	  supported value is 8192, otherwise the maximum value is 512.  The
 	  minimum value which makes sense is 2.
 
 	  This is purely to save memory - each supported CPU adds
@@ -914,12 +915,12 @@ config X86_UP_IOAPIC
 config X86_LOCAL_APIC
 	def_bool y
 	depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC || PCI_MSI
-	select GENERIC_IRQ_LEGACY_ALLOC_HWIRQ
+	select IRQ_DOMAIN_HIERARCHY
+	select PCI_MSI_IRQ_DOMAIN if PCI_MSI
 
 config X86_IO_APIC
 	def_bool y
 	depends on X86_LOCAL_APIC || X86_UP_IOAPIC
-	select IRQ_DOMAIN
 
 config X86_REROUTE_FOR_BROKEN_BOOT_IRQS
 	bool "Reroute for broken boot IRQs"

arch/x86/Kconfig.debug

@@ -332,4 +332,15 @@ config X86_DEBUG_STATIC_CPU_HAS
 
 	  If unsure, say N.
 
+config PUNIT_ATOM_DEBUG
+	tristate "ATOM Punit debug driver"
+	select DEBUG_FS
+	select IOSF_MBI
+	---help---
+	  This is a debug driver, which gets the power states
+	  of all Punit North Complex devices. The power states of
+	  each device is exposed as part of the debugfs interface.
+	  The current power state can be read from
+	  /sys/kernel/debug/punit_atom/dev_power_state
+
 endmenu

arch/x86/Makefile

@@ -77,6 +77,12 @@ else
         KBUILD_AFLAGS += -m64
         KBUILD_CFLAGS += -m64
 
+        # Align jump targets to 1 byte, not the default 16 bytes:
+        KBUILD_CFLAGS += -falign-jumps=1
+
+        # Pack loops tightly as well:
+        KBUILD_CFLAGS += -falign-loops=1
+
         # Don't autogenerate traditional x87 instructions
         KBUILD_CFLAGS += $(call cc-option,-mno-80387)
         KBUILD_CFLAGS += $(call cc-option,-mno-fp-ret-in-387)
@@ -84,6 +90,9 @@ else
 	# Use -mpreferred-stack-boundary=3 if supported.
 	KBUILD_CFLAGS += $(call cc-option,-mpreferred-stack-boundary=3)
 
+	# Use -mskip-rax-setup if supported.
+	KBUILD_CFLAGS += $(call cc-option,-mskip-rax-setup)
+
         # FIXME - should be integrated in Makefile.cpu (Makefile_32.cpu)
         cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8)
         cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona)

arch/x86/ia32/ia32entry.S

@@ -77,12 +77,6 @@ ENTRY(native_usergs_sysret32)
 	swapgs
 	sysretl
 ENDPROC(native_usergs_sysret32)
-
-ENTRY(native_irq_enable_sysexit)
-	swapgs
-	sti
-	sysexit
-ENDPROC(native_irq_enable_sysexit)
 #endif
 
 /*
@@ -119,7 +113,7 @@ ENTRY(ia32_sysenter_target)
 	 * it is too small to ever cause noticeable irq latency.
 	 */
 	SWAPGS_UNSAFE_STACK
-	movq	PER_CPU_VAR(cpu_tss + TSS_sp0), %rsp
+	movq	PER_CPU_VAR(cpu_current_top_of_stack), %rsp
 	ENABLE_INTERRUPTS(CLBR_NONE)
 
 	/* Zero-extending 32-bit regs, do not remove */
@@ -142,7 +136,7 @@ ENTRY(ia32_sysenter_target)
 	pushq_cfi_reg	rsi			/* pt_regs->si */
 	pushq_cfi_reg	rdx			/* pt_regs->dx */
 	pushq_cfi_reg	rcx			/* pt_regs->cx */
-	pushq_cfi_reg	rax			/* pt_regs->ax */
+	pushq_cfi	$-ENOSYS		/* pt_regs->ax */
 	cld
 	sub	$(10*8),%rsp /* pt_regs->r8-11,bp,bx,r12-15 not saved */
 	CFI_ADJUST_CFA_OFFSET 10*8
@@ -169,8 +163,6 @@ sysenter_flags_fixed:
 	testl   $_TIF_WORK_SYSCALL_ENTRY, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS)
 	CFI_REMEMBER_STATE
 	jnz  sysenter_tracesys
-	cmpq	$(IA32_NR_syscalls-1),%rax
-	ja	ia32_badsys
 sysenter_do_call:
 	/* 32bit syscall -> 64bit C ABI argument conversion */
 	movl	%edi,%r8d	/* arg5 */
@@ -179,8 +171,11 @@ sysenter_do_call:
 	movl	%ebx,%edi	/* arg1 */
 	movl	%edx,%edx	/* arg3 (zero extension) */
 sysenter_dispatch:
+	cmpq	$(IA32_NR_syscalls-1),%rax
+	ja	1f
 	call	*ia32_sys_call_table(,%rax,8)
 	movq	%rax,RAX(%rsp)
+1:
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	testl	$_TIF_ALLWORK_MASK, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS)
@@ -247,9 +242,7 @@ sysexit_from_sys_call:
 	movl %ebx,%esi			/* 2nd arg: 1st syscall arg */
 	movl %eax,%edi			/* 1st arg: syscall number */
 	call __audit_syscall_entry
-	movl RAX(%rsp),%eax	/* reload syscall number */
-	cmpq $(IA32_NR_syscalls-1),%rax
-	ja ia32_badsys
+	movl ORIG_RAX(%rsp),%eax	/* reload syscall number */
 	movl %ebx,%edi			/* reload 1st syscall arg */
 	movl RCX(%rsp),%esi	/* reload 2nd syscall arg */
 	movl RDX(%rsp),%edx	/* reload 3rd syscall arg */
@@ -300,13 +293,10 @@ sysenter_tracesys:
 #endif
 	SAVE_EXTRA_REGS
 	CLEAR_RREGS
-	movq	$-ENOSYS,RAX(%rsp)/* ptrace can change this for a bad syscall */
 	movq	%rsp,%rdi        /* &pt_regs -> arg1 */
 	call	syscall_trace_enter
 	LOAD_ARGS32  /* reload args from stack in case ptrace changed it */
 	RESTORE_EXTRA_REGS
-	cmpq	$(IA32_NR_syscalls-1),%rax
-	ja	int_ret_from_sys_call /* sysenter_tracesys has set RAX(%rsp) */
 	jmp	sysenter_do_call
 	CFI_ENDPROC
 ENDPROC(ia32_sysenter_target)
@@ -356,7 +346,7 @@ ENTRY(ia32_cstar_target)
 	SWAPGS_UNSAFE_STACK
 	movl	%esp,%r8d
 	CFI_REGISTER	rsp,r8
-	movq	PER_CPU_VAR(kernel_stack),%rsp
+	movq	PER_CPU_VAR(cpu_current_top_of_stack),%rsp
 	ENABLE_INTERRUPTS(CLBR_NONE)
 
 	/* Zero-extending 32-bit regs, do not remove */
@@ -376,7 +366,7 @@ ENTRY(ia32_cstar_target)
 	pushq_cfi_reg	rdx			/* pt_regs->dx */
 	pushq_cfi_reg	rbp			/* pt_regs->cx */
 	movl	%ebp,%ecx
-	pushq_cfi_reg	rax			/* pt_regs->ax */
+	pushq_cfi	$-ENOSYS		/* pt_regs->ax */
 	sub	$(10*8),%rsp /* pt_regs->r8-11,bp,bx,r12-15 not saved */
 	CFI_ADJUST_CFA_OFFSET 10*8
 
@@ -392,8 +382,6 @@ ENTRY(ia32_cstar_target)
 	testl   $_TIF_WORK_SYSCALL_ENTRY, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS)
 	CFI_REMEMBER_STATE
 	jnz   cstar_tracesys
-	cmpq $IA32_NR_syscalls-1,%rax
-	ja  ia32_badsys
 cstar_do_call:
 	/* 32bit syscall -> 64bit C ABI argument conversion */
 	movl	%edi,%r8d	/* arg5 */
@@ -402,8 +390,11 @@ cstar_do_call:
 	movl	%ebx,%edi	/* arg1 */
 	movl	%edx,%edx	/* arg3 (zero extension) */
 cstar_dispatch:
+	cmpq	$(IA32_NR_syscalls-1),%rax
+	ja	1f
 	call *ia32_sys_call_table(,%rax,8)
 	movq %rax,RAX(%rsp)
+1:
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	testl $_TIF_ALLWORK_MASK, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS)
@@ -457,14 +448,11 @@ cstar_tracesys:
 	xchgl %r9d,%ebp
 	SAVE_EXTRA_REGS
 	CLEAR_RREGS r9
-	movq $-ENOSYS,RAX(%rsp)	/* ptrace can change this for a bad syscall */
 	movq %rsp,%rdi        /* &pt_regs -> arg1 */
 	call syscall_trace_enter
 	LOAD_ARGS32 1	/* reload args from stack in case ptrace changed it */
 	RESTORE_EXTRA_REGS
 	xchgl %ebp,%r9d
-	cmpq $(IA32_NR_syscalls-1),%rax
-	ja int_ret_from_sys_call /* cstar_tracesys has set RAX(%rsp) */
 	jmp cstar_do_call
 END(ia32_cstar_target)
 				
@@ -523,7 +511,7 @@ ENTRY(ia32_syscall)
 	pushq_cfi_reg	rsi			/* pt_regs->si */
 	pushq_cfi_reg	rdx			/* pt_regs->dx */
 	pushq_cfi_reg	rcx			/* pt_regs->cx */
-	pushq_cfi_reg	rax			/* pt_regs->ax */
+	pushq_cfi	$-ENOSYS		/* pt_regs->ax */
 	cld
 	sub	$(10*8),%rsp /* pt_regs->r8-11,bp,bx,r12-15 not saved */
 	CFI_ADJUST_CFA_OFFSET 10*8
@@ -531,8 +519,6 @@ ENTRY(ia32_syscall)
 	orl $TS_COMPAT, ASM_THREAD_INFO(TI_status, %rsp, SIZEOF_PTREGS)
 	testl $_TIF_WORK_SYSCALL_ENTRY, ASM_THREAD_INFO(TI_flags, %rsp, SIZEOF_PTREGS)
 	jnz ia32_tracesys
-	cmpq $(IA32_NR_syscalls-1),%rax
-	ja ia32_badsys
 ia32_do_call:
 	/* 32bit syscall -> 64bit C ABI argument conversion */
 	movl %edi,%r8d	/* arg5 */
@@ -540,9 +526,12 @@ ia32_do_call:
 	xchg %ecx,%esi	/* rsi:arg2, rcx:arg4 */
 	movl %ebx,%edi	/* arg1 */
 	movl %edx,%edx	/* arg3 (zero extension) */
+	cmpq	$(IA32_NR_syscalls-1),%rax
+	ja	1f
 	call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
 ia32_sysret:
 	movq %rax,RAX(%rsp)
+1:
 ia32_ret_from_sys_call:
 	CLEAR_RREGS
 	jmp int_ret_from_sys_call
@@ -550,23 +539,14 @@ ia32_ret_from_sys_call:
 ia32_tracesys:
 	SAVE_EXTRA_REGS
 	CLEAR_RREGS
-	movq $-ENOSYS,RAX(%rsp)	/* ptrace can change this for a bad syscall */
 	movq %rsp,%rdi        /* &pt_regs -> arg1 */
 	call syscall_trace_enter
 	LOAD_ARGS32	/* reload args from stack in case ptrace changed it */
 	RESTORE_EXTRA_REGS
-	cmpq $(IA32_NR_syscalls-1),%rax
-	ja  int_ret_from_sys_call	/* ia32_tracesys has set RAX(%rsp) */
 	jmp ia32_do_call
+	CFI_ENDPROC
 END(ia32_syscall)
 
-ia32_badsys:
-	movq $0,ORIG_RAX(%rsp)
-	movq $-ENOSYS,%rax
-	jmp ia32_sysret
-
-	CFI_ENDPROC
-	
 	.macro PTREGSCALL label, func
 	ALIGN
 GLOBAL(\label)

arch/x86/include/asm/alternative-asm.h

@@ -18,6 +18,12 @@
 	.endm
 #endif
 
+/*
+ * Issue one struct alt_instr descriptor entry (need to put it into
+ * the section .altinstructions, see below). This entry contains
+ * enough information for the alternatives patching code to patch an
+ * instruction. See apply_alternatives().
+ */
 .macro altinstruction_entry orig alt feature orig_len alt_len pad_len
 	.long \orig - .
 	.long \alt - .
@@ -27,6 +33,12 @@
 	.byte \pad_len
 .endm
 
+/*
+ * Define an alternative between two instructions. If @feature is
+ * present, early code in apply_alternatives() replaces @oldinstr with
+ * @newinstr. ".skip" directive takes care of proper instruction padding
+ * in case @newinstr is longer than @oldinstr.
+ */
 .macro ALTERNATIVE oldinstr, newinstr, feature
 140:
 	\oldinstr
@@ -55,6 +67,12 @@
  */
 #define alt_max_short(a, b)	((a) ^ (((a) ^ (b)) & -(-((a) < (b)))))
 
+
+/*
+ * Same as ALTERNATIVE macro above but for two alternatives. If CPU
+ * has @feature1, it replaces @oldinstr with @newinstr1. If CPU has
+ * @feature2, it replaces @oldinstr with @feature2.
+ */
 .macro ALTERNATIVE_2 oldinstr, newinstr1, feature1, newinstr2, feature2
 140:
 	\oldinstr

arch/x86/include/asm/apic.h

@@ -644,6 +644,12 @@ static inline void entering_ack_irq(void)
 	entering_irq();
 }
 
+static inline void ipi_entering_ack_irq(void)
+{
+	ack_APIC_irq();
+	irq_enter();
+}
+
 static inline void exiting_irq(void)
 {
 	irq_exit();

arch/x86/include/asm/asm.h

@@ -63,6 +63,31 @@
 	_ASM_ALIGN ;						\
 	_ASM_PTR (entry);					\
 	.popsection
+
+.macro ALIGN_DESTINATION
+	/* check for bad alignment of destination */
+	movl %edi,%ecx
+	andl $7,%ecx
+	jz 102f				/* already aligned */
+	subl $8,%ecx
+	negl %ecx
+	subl %ecx,%edx
+100:	movb (%rsi),%al
+101:	movb %al,(%rdi)
+	incq %rsi
+	incq %rdi
+	decl %ecx
+	jnz 100b
+102:
+	.section .fixup,"ax"
+103:	addl %ecx,%edx			/* ecx is zerorest also */
+	jmp copy_user_handle_tail
+	.previous
+
+	_ASM_EXTABLE(100b,103b)
+	_ASM_EXTABLE(101b,103b)
+	.endm
+
 #else
 # define _ASM_EXTABLE(from,to)					\
 	" .pushsection \"__ex_table\",\"a\"\n"			\

arch/x86/include/asm/atomic.h

@@ -22,7 +22,7 @@
  *
  * Atomically reads the value of @v.
  */
-static inline int atomic_read(const atomic_t *v)
+static __always_inline int atomic_read(const atomic_t *v)
 {
 	return ACCESS_ONCE((v)->counter);
 }
@@ -34,7 +34,7 @@ static inline int atomic_read(const atomic_t *v)
  *
  * Atomically sets the value of @v to @i.
  */
-static inline void atomic_set(atomic_t *v, int i)
+static __always_inline void atomic_set(atomic_t *v, int i)
 {
 	v->counter = i;
 }
@@ -46,7 +46,7 @@ static inline void atomic_set(atomic_t *v, int i)
  *
  * Atomically adds @i to @v.
  */
-static inline void atomic_add(int i, atomic_t *v)
+static __always_inline void atomic_add(int i, atomic_t *v)
 {
 	asm volatile(LOCK_PREFIX "addl %1,%0"
 		     : "+m" (v->counter)
@@ -60,7 +60,7 @@ static inline void atomic_add(int i, atomic_t *v)
  *
  * Atomically subtracts @i from @v.
  */
-static inline void atomic_sub(int i, atomic_t *v)
+static __always_inline void atomic_sub(int i, atomic_t *v)
 {
 	asm volatile(LOCK_PREFIX "subl %1,%0"
 		     : "+m" (v->counter)
@@ -76,7 +76,7 @@ static inline void atomic_sub(int i, atomic_t *v)
  * true if the result is zero, or false for all
  * other cases.
  */
-static inline int atomic_sub_and_test(int i, atomic_t *v)
+static __always_inline int atomic_sub_and_test(int i, atomic_t *v)
 {
 	GEN_BINARY_RMWcc(LOCK_PREFIX "subl", v->counter, "er", i, "%0", "e");
 }
@@ -87,7 +87,7 @@ static inline int atomic_sub_and_test(int i, atomic_t *v)
  *
  * Atomically increments @v by 1.
  */
-static inline void atomic_inc(atomic_t *v)
+static __always_inline void atomic_inc(atomic_t *v)
 {
 	asm volatile(LOCK_PREFIX "incl %0"
 		     : "+m" (v->counter));
@@ -99,7 +99,7 @@ static inline void atomic_inc(atomic_t *v)
  *
  * Atomically decrements @v by 1.
  */
-static inline void atomic_dec(atomic_t *v)
+static __always_inline void atomic_dec(atomic_t *v)
 {
 	asm volatile(LOCK_PREFIX "decl %0"
 		     : "+m" (v->counter));
@@ -113,7 +113,7 @@ static inline void atomic_dec(atomic_t *v)
  * returns true if the result is 0, or false for all other
  * cases.
  */
-static inline int atomic_dec_and_test(atomic_t *v)
+static __always_inline int atomic_dec_and_test(atomic_t *v)
 {
 	GEN_UNARY_RMWcc(LOCK_PREFIX "decl", v->counter, "%0", "e");
 }
@@ -126,7 +126,7 @@ static inline int atomic_dec_and_test(atomic_t *v)
  * and returns true if the result is zero, or false for all
  * other cases.
  */
-static inline int atomic_inc_and_test(atomic_t *v)
+static __always_inline int atomic_inc_and_test(atomic_t *v)
 {
 	GEN_UNARY_RMWcc(LOCK_PREFIX "incl", v->counter, "%0", "e");
 }
@@ -140,7 +140,7 @@ static inline int atomic_inc_and_test(atomic_t *v)
  * if the result is negative, or false when
  * result is greater than or equal to zero.
  */
-static inline int atomic_add_negative(int i, atomic_t *v)
+static __always_inline int atomic_add_negative(int i, atomic_t *v)
 {
 	GEN_BINARY_RMWcc(LOCK_PREFIX "addl", v->counter, "er", i, "%0", "s");
 }
@@ -152,7 +152,7 @@ static inline int atomic_add_negative(int i, atomic_t *v)
  *
  * Atomically adds @i to @v and returns @i + @v
  */
-static inline int atomic_add_return(int i, atomic_t *v)
+static __always_inline int atomic_add_return(int i, atomic_t *v)
 {
 	return i + xadd(&v->counter, i);
 }
@@ -164,7 +164,7 @@ static inline int atomic_add_return(int i, atomic_t *v)
  *
  * Atomically subtracts @i from @v and returns @v - @i
  */
-static inline int atomic_sub_return(int i, atomic_t *v)
+static __always_inline int atomic_sub_return(int i, atomic_t *v)
 {
 	return atomic_add_return(-i, v);
 }
@@ -172,7 +172,7 @@ static inline int atomic_sub_return(int i, atomic_t *v)
 #define atomic_inc_return(v)  (atomic_add_return(1, v))
 #define atomic_dec_return(v)  (atomic_sub_return(1, v))
 
-static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+static __always_inline int atomic_cmpxchg(atomic_t *v, int old, int new)
 {
 	return cmpxchg(&v->counter, old, new);
 }
@@ -191,7 +191,7 @@ static inline int atomic_xchg(atomic_t *v, int new)
  * Atomically adds @a to @v, so long as @v was not already @u.
  * Returns the old value of @v.
  */
-static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+static __always_inline int __atomic_add_unless(atomic_t *v, int a, int u)
 {
 	int c, old;
 	c = atomic_read(v);
@@ -213,7 +213,7 @@ static inline int __atomic_add_unless(atomic_t *v, int a, int u)
  * Atomically adds 1 to @v
  * Returns the new value of @u
  */
-static inline short int atomic_inc_short(short int *v)
+static __always_inline short int atomic_inc_short(short int *v)
 {
 	asm(LOCK_PREFIX "addw $1, %0" : "+m" (*v));
 	return *v;

arch/x86/include/asm/atomic64_64.h

@@ -40,7 +40,7 @@ static inline void atomic64_set(atomic64_t *v, long i)
  *
  * Atomically adds @i to @v.
  */
-static inline void atomic64_add(long i, atomic64_t *v)
+static __always_inline void atomic64_add(long i, atomic64_t *v)
 {
 	asm volatile(LOCK_PREFIX "addq %1,%0"
 		     : "=m" (v->counter)
@@ -81,7 +81,7 @@ static inline int atomic64_sub_and_test(long i, atomic64_t *v)
  *
  * Atomically increments @v by 1.
  */
-static inline void atomic64_inc(atomic64_t *v)
+static __always_inline void atomic64_inc(atomic64_t *v)
 {
 	asm volatile(LOCK_PREFIX "incq %0"
 		     : "=m" (v->counter)
@@ -94,7 +94,7 @@ static inline void atomic64_inc(atomic64_t *v)
  *
  * Atomically decrements @v by 1.
  */
-static inline void atomic64_dec(atomic64_t *v)
+static __always_inline void atomic64_dec(atomic64_t *v)
 {
 	asm volatile(LOCK_PREFIX "decq %0"
 		     : "=m" (v->counter)
@@ -148,7 +148,7 @@ static inline int atomic64_add_negative(long i, atomic64_t *v)
  *
  * Atomically adds @i to @v and returns @i + @v
  */
-static inline long atomic64_add_return(long i, atomic64_t *v)
+static __always_inline long atomic64_add_return(long i, atomic64_t *v)
 {
 	return i + xadd(&v->counter, i);
 }

arch/x86/include/asm/entry_arch.h

@@ -23,6 +23,8 @@ BUILD_INTERRUPT(x86_platform_ipi, X86_PLATFORM_IPI_VECTOR)
 #ifdef CONFIG_HAVE_KVM
 BUILD_INTERRUPT3(kvm_posted_intr_ipi, POSTED_INTR_VECTOR,
 		 smp_kvm_posted_intr_ipi)
+BUILD_INTERRUPT3(kvm_posted_intr_wakeup_ipi, POSTED_INTR_WAKEUP_VECTOR,
+		 smp_kvm_posted_intr_wakeup_ipi)
 #endif
 
 /*

arch/x86/include/asm/hardirq.h

@@ -14,6 +14,7 @@ typedef struct {
 #endif
 #ifdef CONFIG_HAVE_KVM
 	unsigned int kvm_posted_intr_ipis;
+	unsigned int kvm_posted_intr_wakeup_ipis;
 #endif
 	unsigned int x86_platform_ipis;	/* arch dependent */
 	unsigned int apic_perf_irqs;

arch/x86/include/asm/hpet.h

@@ -74,20 +74,16 @@ extern unsigned int hpet_readl(unsigned int a);
 extern void force_hpet_resume(void);
 
 struct irq_data;
+struct hpet_dev;
+struct irq_domain;
+
 extern void hpet_msi_unmask(struct irq_data *data);
 extern void hpet_msi_mask(struct irq_data *data);
-struct hpet_dev;
 extern void hpet_msi_write(struct hpet_dev *hdev, struct msi_msg *msg);
 extern void hpet_msi_read(struct hpet_dev *hdev, struct msi_msg *msg);
-
-#ifdef CONFIG_PCI_MSI
-extern int default_setup_hpet_msi(unsigned int irq, unsigned int id);
-#else
-static inline int default_setup_hpet_msi(unsigned int irq, unsigned int id)
-{
-	return -EINVAL;
-}
-#endif
+extern struct irq_domain *hpet_create_irq_domain(int hpet_id);
+extern int hpet_assign_irq(struct irq_domain *domain,
+			   struct hpet_dev *dev, int dev_num);
 
 #ifdef CONFIG_HPET_EMULATE_RTC
 

arch/x86/include/asm/hw_irq.h

@@ -29,6 +29,7 @@
 extern asmlinkage void apic_timer_interrupt(void);
 extern asmlinkage void x86_platform_ipi(void);
 extern asmlinkage void kvm_posted_intr_ipi(void);
+extern asmlinkage void kvm_posted_intr_wakeup_ipi(void);
 extern asmlinkage void error_interrupt(void);
 extern asmlinkage void irq_work_interrupt(void);
 
@@ -36,40 +37,6 @@ extern asmlinkage void spurious_interrupt(void);
 extern asmlinkage void thermal_interrupt(void);
 extern asmlinkage void reschedule_interrupt(void);
 
-extern asmlinkage void invalidate_interrupt(void);
-extern asmlinkage void invalidate_interrupt0(void);
-extern asmlinkage void invalidate_interrupt1(void);
-extern asmlinkage void invalidate_interrupt2(void);
-extern asmlinkage void invalidate_interrupt3(void);
-extern asmlinkage void invalidate_interrupt4(void);
-extern asmlinkage void invalidate_interrupt5(void);
-extern asmlinkage void invalidate_interrupt6(void);
-extern asmlinkage void invalidate_interrupt7(void);
-extern asmlinkage void invalidate_interrupt8(void);
-extern asmlinkage void invalidate_interrupt9(void);
-extern asmlinkage void invalidate_interrupt10(void);
-extern asmlinkage void invalidate_interrupt11(void);
-extern asmlinkage void invalidate_interrupt12(void);
-extern asmlinkage void invalidate_interrupt13(void);
-extern asmlinkage void invalidate_interrupt14(void);
-extern asmlinkage void invalidate_interrupt15(void);
-extern asmlinkage void invalidate_interrupt16(void);
-extern asmlinkage void invalidate_interrupt17(void);
-extern asmlinkage void invalidate_interrupt18(void);
-extern asmlinkage void invalidate_interrupt19(void);
-extern asmlinkage void invalidate_interrupt20(void);
-extern asmlinkage void invalidate_interrupt21(void);
-extern asmlinkage void invalidate_interrupt22(void);
-extern asmlinkage void invalidate_interrupt23(void);
-extern asmlinkage void invalidate_interrupt24(void);
-extern asmlinkage void invalidate_interrupt25(void);
-extern asmlinkage void invalidate_interrupt26(void);
-extern asmlinkage void invalidate_interrupt27(void);
-extern asmlinkage void invalidate_interrupt28(void);
-extern asmlinkage void invalidate_interrupt29(void);
-extern asmlinkage void invalidate_interrupt30(void);
-extern asmlinkage void invalidate_interrupt31(void);
-
 extern asmlinkage void irq_move_cleanup_interrupt(void);
 extern asmlinkage void reboot_interrupt(void);
 extern asmlinkage void threshold_interrupt(void);
@@ -92,55 +59,87 @@ extern void trace_call_function_single_interrupt(void);
 #define trace_irq_move_cleanup_interrupt  irq_move_cleanup_interrupt
 #define trace_reboot_interrupt  reboot_interrupt
 #define trace_kvm_posted_intr_ipi kvm_posted_intr_ipi
+#define trace_kvm_posted_intr_wakeup_ipi kvm_posted_intr_wakeup_ipi
 #endif /* CONFIG_TRACING */
 
-#ifdef CONFIG_IRQ_REMAP
-/* Intel specific interrupt remapping information */
-struct irq_2_iommu {
-	struct intel_iommu *iommu;
-	u16 irte_index;
-	u16 sub_handle;
-	u8  irte_mask;
-};
-
-/* AMD specific interrupt remapping information */
-struct irq_2_irte {
-	u16 devid; /* Device ID for IRTE table */
-	u16 index; /* Index into IRTE table*/
-};
-#endif	/* CONFIG_IRQ_REMAP */
-
 #ifdef	CONFIG_X86_LOCAL_APIC
 struct irq_data;
+struct pci_dev;
+struct msi_desc;
+
+enum irq_alloc_type {
+	X86_IRQ_ALLOC_TYPE_IOAPIC = 1,
+	X86_IRQ_ALLOC_TYPE_HPET,
+	X86_IRQ_ALLOC_TYPE_MSI,
+	X86_IRQ_ALLOC_TYPE_MSIX,
+	X86_IRQ_ALLOC_TYPE_DMAR,
+	X86_IRQ_ALLOC_TYPE_UV,
+};
 
-struct irq_cfg {
-	cpumask_var_t		domain;
-	cpumask_var_t		old_domain;
-	u8			vector;
-	u8			move_in_progress : 1;
-#ifdef CONFIG_IRQ_REMAP
-	u8			remapped : 1;
+struct irq_alloc_info {
+	enum irq_alloc_type	type;
+	u32			flags;
+	const struct cpumask	*mask;	/* CPU mask for vector allocation */
 	union {
-		struct irq_2_iommu irq_2_iommu;
-		struct irq_2_irte  irq_2_irte;
-	};
+		int		unused;
+#ifdef	CONFIG_HPET_TIMER
+		struct {
+			int		hpet_id;
+			int		hpet_index;
+			void		*hpet_data;
+		};
 #endif
-	union {
-#ifdef CONFIG_X86_IO_APIC
+#ifdef	CONFIG_PCI_MSI
 		struct {
-			struct list_head	irq_2_pin;
+			struct pci_dev	*msi_dev;
+			irq_hw_number_t	msi_hwirq;
+		};
+#endif
+#ifdef	CONFIG_X86_IO_APIC
+		struct {
+			int		ioapic_id;
+			int		ioapic_pin;
+			int		ioapic_node;
+			u32		ioapic_trigger : 1;
+			u32		ioapic_polarity : 1;
+			u32		ioapic_valid : 1;
+			struct IO_APIC_route_entry *ioapic_entry;
+		};
+#endif
+#ifdef	CONFIG_DMAR_TABLE
+		struct {
+			int		dmar_id;
+			void		*dmar_data;
+		};
+#endif
+#ifdef	CONFIG_HT_IRQ
+		struct {
+			int		ht_pos;
+			int		ht_idx;
+			struct pci_dev	*ht_dev;
+			void		*ht_update;
+		};
+#endif
+#ifdef	CONFIG_X86_UV
+		struct {
+			int		uv_limit;
+			int		uv_blade;
+			unsigned long	uv_offset;
+			char		*uv_name;
 		};
 #endif
 	};
 };
 
+struct irq_cfg {
+	unsigned int		dest_apicid;
+	u8			vector;
+};
+
 extern struct irq_cfg *irq_cfg(unsigned int irq);
 extern struct irq_cfg *irqd_cfg(struct irq_data *irq_data);
-extern struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node);
 extern void lock_vector_lock(void);
 extern void unlock_vector_lock(void);
-extern int assign_irq_vector(int, struct irq_cfg *, const struct cpumask *);
-extern void clear_irq_vector(int irq, struct irq_cfg *cfg);
 extern void setup_vector_irq(int cpu);
 #ifdef CONFIG_SMP
 extern void send_cleanup_vector(struct irq_cfg *);
@@ -150,10 +149,7 @@ static inline void send_cleanup_vector(struct irq_cfg *c) { }
 static inline void irq_complete_move(struct irq_cfg *c) { }
 #endif
 
-extern int apic_retrigger_irq(struct irq_data *data);
 extern void apic_ack_edge(struct irq_data *data);
-extern int apic_set_affinity(struct irq_data *data, const struct cpumask *mask,
-			     unsigned int *dest_id);
 #else	/*  CONFIG_X86_LOCAL_APIC */
 static inline void lock_vector_lock(void) {}
 static inline void unlock_vector_lock(void) {}
@@ -163,8 +159,7 @@ static inline void unlock_vector_lock(void) {}
 extern atomic_t irq_err_count;
 extern atomic_t irq_mis_count;
 
-/* EISA */
-extern void eisa_set_level_irq(unsigned int irq);
+extern void elcr_set_level_irq(unsigned int irq);
 
 /* SMP */
 extern __visible void smp_apic_timer_interrupt(struct pt_regs *);
@@ -178,7 +173,6 @@ extern asmlinkage void smp_irq_move_cleanup_interrupt(void);
 extern __visible void smp_reschedule_interrupt(struct pt_regs *);
 extern __visible void smp_call_function_interrupt(struct pt_regs *);
 extern __visible void smp_call_function_single_interrupt(struct pt_regs *);
-extern __visible void smp_invalidate_interrupt(struct pt_regs *);
 #endif
 
 extern char irq_entries_start[];

arch/x86/include/asm/io.h

@@ -177,6 +177,7 @@ static inline unsigned int isa_virt_to_bus(volatile void *address)
  * look at pci_iomap().
  */
 extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_uc(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size,
 				unsigned long prot_val);
@@ -338,6 +339,9 @@ extern bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
 #define IO_SPACE_LIMIT 0xffff
 
 #ifdef CONFIG_MTRR
+extern int __must_check arch_phys_wc_index(int handle);
+#define arch_phys_wc_index arch_phys_wc_index
+
 extern int __must_check arch_phys_wc_add(unsigned long base,
 					 unsigned long size);
 extern void arch_phys_wc_del(int handle);

arch/x86/include/asm/io_apic.h

@@ -95,9 +95,22 @@ struct IR_IO_APIC_route_entry {
 		index		: 15;
 } __attribute__ ((packed));
 
-#define IOAPIC_AUTO     -1
-#define IOAPIC_EDGE     0
-#define IOAPIC_LEVEL    1
+struct irq_alloc_info;
+struct ioapic_domain_cfg;
+
+#define IOAPIC_AUTO			-1
+#define IOAPIC_EDGE			0
+#define IOAPIC_LEVEL			1
+
+#define IOAPIC_MASKED			1
+#define IOAPIC_UNMASKED			0
+
+#define IOAPIC_POL_HIGH			0
+#define IOAPIC_POL_LOW			1
+
+#define IOAPIC_DEST_MODE_PHYSICAL	0
+#define IOAPIC_DEST_MODE_LOGICAL	1
+
 #define	IOAPIC_MAP_ALLOC		0x1
 #define	IOAPIC_MAP_CHECK		0x2
 
@@ -110,9 +123,6 @@ extern int nr_ioapics;
 
 extern int mpc_ioapic_id(int ioapic);
 extern unsigned int mpc_ioapic_addr(int ioapic);
-extern struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic);
-
-#define MP_MAX_IOAPIC_PIN 127
 
 /* # of MP IRQ source entries */
 extern int mp_irq_entries;
@@ -120,9 +130,6 @@ extern int mp_irq_entries;
 /* MP IRQ source entries */
 extern struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES];
 
-/* Older SiS APIC requires we rewrite the index register */
-extern int sis_apic_bug;
-
 /* 1 if "noapic" boot option passed */
 extern int skip_ioapic_setup;
 
@@ -132,6 +139,8 @@ extern int noioapicquirk;
 /* -1 if "noapic" boot option passed */
 extern int noioapicreroute;
 
+extern u32 gsi_top;
+
 extern unsigned long io_apic_irqs;
 
 #define IO_APIC_IRQ(x) (((x) >= NR_IRQS_LEGACY) || ((1 << (x)) & io_apic_irqs))
@@ -147,13 +156,6 @@ struct irq_cfg;
 extern void ioapic_insert_resources(void);
 extern int arch_early_ioapic_init(void);
 
-extern int native_setup_ioapic_entry(int, struct IO_APIC_route_entry *,
-				     unsigned int, int,
-				     struct io_apic_irq_attr *);
-extern void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg);
-
-extern void native_eoi_ioapic_pin(int apic, int pin, int vector);
-
 extern int save_ioapic_entries(void);
 extern void mask_ioapic_entries(void);
 extern int restore_ioapic_entries(void);
@@ -161,82 +163,32 @@ extern int restore_ioapic_entries(void);
 extern void setup_ioapic_ids_from_mpc(void);
 extern void setup_ioapic_ids_from_mpc_nocheck(void);
 
-struct io_apic_irq_attr {
-	int ioapic;
-	int ioapic_pin;
-	int trigger;
-	int polarity;
-};
-
-enum ioapic_domain_type {
-	IOAPIC_DOMAIN_INVALID,
-	IOAPIC_DOMAIN_LEGACY,
-	IOAPIC_DOMAIN_STRICT,
-	IOAPIC_DOMAIN_DYNAMIC,
-};
-
-struct device_node;
-struct irq_domain;
-struct irq_domain_ops;
-
-struct ioapic_domain_cfg {
-	enum ioapic_domain_type		type;
-	const struct irq_domain_ops	*ops;
-	struct device_node		*dev;
-};
-
-struct mp_ioapic_gsi{
-	u32 gsi_base;
-	u32 gsi_end;
-};
-extern u32 gsi_top;
-
 extern int mp_find_ioapic(u32 gsi);
 extern int mp_find_ioapic_pin(int ioapic, u32 gsi);
-extern u32 mp_pin_to_gsi(int ioapic, int pin);
-extern int mp_map_gsi_to_irq(u32 gsi, unsigned int flags);
+extern int mp_map_gsi_to_irq(u32 gsi, unsigned int flags,
+			     struct irq_alloc_info *info);
 extern void mp_unmap_irq(int irq);
 extern int mp_register_ioapic(int id, u32 address, u32 gsi_base,
 			      struct ioapic_domain_cfg *cfg);
 extern int mp_unregister_ioapic(u32 gsi_base);
 extern int mp_ioapic_registered(u32 gsi_base);
-extern int mp_irqdomain_map(struct irq_domain *domain, unsigned int virq,
-			    irq_hw_number_t hwirq);
-extern void mp_irqdomain_unmap(struct irq_domain *domain, unsigned int virq);
-extern int mp_set_gsi_attr(u32 gsi, int trigger, int polarity, int node);
-extern void __init pre_init_apic_IRQ0(void);
+
+extern void ioapic_set_alloc_attr(struct irq_alloc_info *info,
+				  int node, int trigger, int polarity);
 
 extern void mp_save_irq(struct mpc_intsrc *m);
 
 extern void disable_ioapic_support(void);
 
-extern void __init native_io_apic_init_mappings(void);
+extern void __init io_apic_init_mappings(void);
 extern unsigned int native_io_apic_read(unsigned int apic, unsigned int reg);
-extern void native_io_apic_write(unsigned int apic, unsigned int reg, unsigned int val);
-extern void native_io_apic_modify(unsigned int apic, unsigned int reg, unsigned int val);
 extern void native_disable_io_apic(void);
-extern void native_io_apic_print_entries(unsigned int apic, unsigned int nr_entries);
-extern void intel_ir_io_apic_print_entries(unsigned int apic, unsigned int nr_entries);
-extern int native_ioapic_set_affinity(struct irq_data *,
-				      const struct cpumask *,
-				      bool);
 
 static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg)
 {
 	return x86_io_apic_ops.read(apic, reg);
 }
 
-static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
-{
-	x86_io_apic_ops.write(apic, reg, value);
-}
-static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value)
-{
-	x86_io_apic_ops.modify(apic, reg, value);
-}
-
-extern void io_apic_eoi(unsigned int apic, unsigned int vector);
-
 extern void setup_IO_APIC(void);
 extern void enable_IO_APIC(void);
 extern void disable_IO_APIC(void);
@@ -253,8 +205,12 @@ static inline int arch_early_ioapic_init(void) { return 0; }
 static inline void print_IO_APICs(void) {}
 #define gsi_top (NR_IRQS_LEGACY)
 static inline int mp_find_ioapic(u32 gsi) { return 0; }
-static inline u32 mp_pin_to_gsi(int ioapic, int pin) { return UINT_MAX; }
-static inline int mp_map_gsi_to_irq(u32 gsi, unsigned int flags) { return gsi; }
+static inline int mp_map_gsi_to_irq(u32 gsi, unsigned int flags,
+				    struct irq_alloc_info *info)
+{
+	return gsi;
+}
+
 static inline void mp_unmap_irq(int irq) { }
 
 static inline int save_ioapic_entries(void)
@@ -268,17 +224,11 @@ static inline int restore_ioapic_entries(void)
 	return -ENOMEM;
 }
 
-static inline void mp_save_irq(struct mpc_intsrc *m) { };
+static inline void mp_save_irq(struct mpc_intsrc *m) { }
 static inline void disable_ioapic_support(void) { }
-#define native_io_apic_init_mappings	NULL
+static inline void io_apic_init_mappings(void) { }
 #define native_io_apic_read		NULL
-#define native_io_apic_write		NULL
-#define native_io_apic_modify		NULL
 #define native_disable_io_apic		NULL
-#define native_io_apic_print_entries	NULL
-#define native_ioapic_set_affinity	NULL
-#define native_setup_ioapic_entry	NULL
-#define native_eoi_ioapic_pin		NULL
 
 static inline void setup_IO_APIC(void) { }
 static inline void enable_IO_APIC(void) { }

arch/x86/include/asm/irq.h

@@ -30,6 +30,10 @@ extern void fixup_irqs(void);
 extern void irq_force_complete_move(int);
 #endif
 
+#ifdef CONFIG_HAVE_KVM
+extern void kvm_set_posted_intr_wakeup_handler(void (*handler)(void));
+#endif
+
 extern void (*x86_platform_ipi_callback)(void);
 extern void native_init_IRQ(void);
 extern bool handle_irq(unsigned irq, struct pt_regs *regs);

arch/x86/include/asm/irq_remapping.h

@@ -22,14 +22,12 @@
 #ifndef __X86_IRQ_REMAPPING_H
 #define __X86_IRQ_REMAPPING_H
 
+#include <asm/irqdomain.h>
+#include <asm/hw_irq.h>
 #include <asm/io_apic.h>
 
-struct IO_APIC_route_entry;
-struct io_apic_irq_attr;
-struct irq_chip;
 struct msi_msg;
-struct pci_dev;
-struct irq_cfg;
+struct irq_alloc_info;
 
 #ifdef CONFIG_IRQ_REMAP
 
@@ -39,22 +37,21 @@ extern int irq_remapping_enable(void);
 extern void irq_remapping_disable(void);
 extern int irq_remapping_reenable(int);
 extern int irq_remap_enable_fault_handling(void);
-extern int setup_ioapic_remapped_entry(int irq,
-				       struct IO_APIC_route_entry *entry,
-				       unsigned int destination,
-				       int vector,
-				       struct io_apic_irq_attr *attr);
-extern void free_remapped_irq(int irq);
-extern void compose_remapped_msi_msg(struct pci_dev *pdev,
-				     unsigned int irq, unsigned int dest,
-				     struct msi_msg *msg, u8 hpet_id);
-extern int setup_hpet_msi_remapped(unsigned int irq, unsigned int id);
 extern void panic_if_irq_remap(const char *msg);
-extern bool setup_remapped_irq(int irq,
-			       struct irq_cfg *cfg,
-			       struct irq_chip *chip);
 
-void irq_remap_modify_chip_defaults(struct irq_chip *chip);
+extern struct irq_domain *
+irq_remapping_get_ir_irq_domain(struct irq_alloc_info *info);
+extern struct irq_domain *
+irq_remapping_get_irq_domain(struct irq_alloc_info *info);
+
+/* Create PCI MSI/MSIx irqdomain, use @parent as the parent irqdomain. */
+extern struct irq_domain *arch_create_msi_irq_domain(struct irq_domain *parent);
+
+/* Get parent irqdomain for interrupt remapping irqdomain */
+static inline struct irq_domain *arch_get_ir_parent_domain(void)
+{
+	return x86_vector_domain;
+}
 
 #else  /* CONFIG_IRQ_REMAP */
 
@@ -64,42 +61,22 @@ static inline int irq_remapping_enable(void) { return -ENODEV; }
 static inline void irq_remapping_disable(void) { }
 static inline int irq_remapping_reenable(int eim) { return -ENODEV; }
 static inline int irq_remap_enable_fault_handling(void) { return -ENODEV; }
-static inline int setup_ioapic_remapped_entry(int irq,
-					      struct IO_APIC_route_entry *entry,
-					      unsigned int destination,
-					      int vector,
-					      struct io_apic_irq_attr *attr)
-{
-	return -ENODEV;
-}
-static inline void free_remapped_irq(int irq) { }
-static inline void compose_remapped_msi_msg(struct pci_dev *pdev,
-					    unsigned int irq, unsigned int dest,
-					    struct msi_msg *msg, u8 hpet_id)
-{
-}
-static inline int setup_hpet_msi_remapped(unsigned int irq, unsigned int id)
-{
-	return -ENODEV;
-}
 
 static inline void panic_if_irq_remap(const char *msg)
 {
 }
 
-static inline void irq_remap_modify_chip_defaults(struct irq_chip *chip)
+static inline struct irq_domain *
+irq_remapping_get_ir_irq_domain(struct irq_alloc_info *info)
 {
+	return NULL;
 }
 
-static inline bool setup_remapped_irq(int irq,
-				      struct irq_cfg *cfg,
-				      struct irq_chip *chip)
+static inline struct irq_domain *
+irq_remapping_get_irq_domain(struct irq_alloc_info *info)
 {
-	return false;
+	return NULL;
 }
-#endif /* CONFIG_IRQ_REMAP */
-
-#define dmar_alloc_hwirq()	irq_alloc_hwirq(-1)
-#define dmar_free_hwirq		irq_free_hwirq
 
+#endif /* CONFIG_IRQ_REMAP */
 #endif /* __X86_IRQ_REMAPPING_H */

arch/x86/include/asm/irq_vectors.h

@@ -47,31 +47,12 @@
 #define IRQ_MOVE_CLEANUP_VECTOR		FIRST_EXTERNAL_VECTOR
 
 #define IA32_SYSCALL_VECTOR		0x80
-#ifdef CONFIG_X86_32
-# define SYSCALL_VECTOR			0x80
-#endif
 
 /*
  * Vectors 0x30-0x3f are used for ISA interrupts.
  *   round up to the next 16-vector boundary
  */
-#define IRQ0_VECTOR			((FIRST_EXTERNAL_VECTOR + 16) & ~15)
-
-#define IRQ1_VECTOR			(IRQ0_VECTOR +  1)
-#define IRQ2_VECTOR			(IRQ0_VECTOR +  2)
-#define IRQ3_VECTOR			(IRQ0_VECTOR +  3)
-#define IRQ4_VECTOR			(IRQ0_VECTOR +  4)
-#define IRQ5_VECTOR			(IRQ0_VECTOR +  5)
-#define IRQ6_VECTOR			(IRQ0_VECTOR +  6)
-#define IRQ7_VECTOR			(IRQ0_VECTOR +  7)
-#define IRQ8_VECTOR			(IRQ0_VECTOR +  8)
-#define IRQ9_VECTOR			(IRQ0_VECTOR +  9)
-#define IRQ10_VECTOR			(IRQ0_VECTOR + 10)
-#define IRQ11_VECTOR			(IRQ0_VECTOR + 11)
-#define IRQ12_VECTOR			(IRQ0_VECTOR + 12)
-#define IRQ13_VECTOR			(IRQ0_VECTOR + 13)
-#define IRQ14_VECTOR			(IRQ0_VECTOR + 14)
-#define IRQ15_VECTOR			(IRQ0_VECTOR + 15)
+#define ISA_IRQ_VECTOR(irq)		(((FIRST_EXTERNAL_VECTOR + 16) & ~15) + irq)
 
 /*
  * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
@@ -105,6 +86,7 @@
 /* Vector for KVM to deliver posted interrupt IPI */
 #ifdef CONFIG_HAVE_KVM
 #define POSTED_INTR_VECTOR		0xf2
+#define POSTED_INTR_WAKEUP_VECTOR	0xf1
 #endif
 
 /*
@@ -155,18 +137,22 @@ static inline int invalid_vm86_irq(int irq)
  * static arrays.
  */
 
-#define NR_IRQS_LEGACY			  16
+#define NR_IRQS_LEGACY			16
 
-#define IO_APIC_VECTOR_LIMIT		( 32 * MAX_IO_APICS )
+#define CPU_VECTOR_LIMIT		(64 * NR_CPUS)
+#define IO_APIC_VECTOR_LIMIT		(32 * MAX_IO_APICS)
 
-#ifdef CONFIG_X86_IO_APIC
-# define CPU_VECTOR_LIMIT		(64 * NR_CPUS)
-# define NR_IRQS					\
+#if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_PCI_MSI)
+#define NR_IRQS						\
 	(CPU_VECTOR_LIMIT > IO_APIC_VECTOR_LIMIT ?	\
 		(NR_VECTORS + CPU_VECTOR_LIMIT)  :	\
 		(NR_VECTORS + IO_APIC_VECTOR_LIMIT))
-#else /* !CONFIG_X86_IO_APIC: */
-# define NR_IRQS			NR_IRQS_LEGACY
+#elif defined(CONFIG_X86_IO_APIC)
+#define	NR_IRQS				(NR_VECTORS + IO_APIC_VECTOR_LIMIT)
+#elif defined(CONFIG_PCI_MSI)
+#define NR_IRQS				(NR_VECTORS + CPU_VECTOR_LIMIT)
+#else
+#define NR_IRQS				NR_IRQS_LEGACY
 #endif
 
 #endif /* _ASM_X86_IRQ_VECTORS_H */

arch/x86/include/asm/irqdomain.h

@@ -0,0 +1,63 @@
+#ifndef _ASM_IRQDOMAIN_H
+#define _ASM_IRQDOMAIN_H
+
+#include <linux/irqdomain.h>
+#include <asm/hw_irq.h>
+
+#ifdef CONFIG_X86_LOCAL_APIC
+enum {
+	/* Allocate contiguous CPU vectors */
+	X86_IRQ_ALLOC_CONTIGUOUS_VECTORS		= 0x1,
+};
+
+extern struct irq_domain *x86_vector_domain;
+
+extern void init_irq_alloc_info(struct irq_alloc_info *info,
+				const struct cpumask *mask);
+extern void copy_irq_alloc_info(struct irq_alloc_info *dst,
+				struct irq_alloc_info *src);
+#endif /* CONFIG_X86_LOCAL_APIC */
+
+#ifdef CONFIG_X86_IO_APIC
+struct device_node;
+struct irq_data;
+
+enum ioapic_domain_type {
+	IOAPIC_DOMAIN_INVALID,
+	IOAPIC_DOMAIN_LEGACY,
+	IOAPIC_DOMAIN_STRICT,
+	IOAPIC_DOMAIN_DYNAMIC,
+};
+
+struct ioapic_domain_cfg {
+	enum ioapic_domain_type		type;
+	const struct irq_domain_ops	*ops;
+	struct device_node		*dev;
+};
+
+extern const struct irq_domain_ops mp_ioapic_irqdomain_ops;
+
+extern int mp_irqdomain_alloc(struct irq_domain *domain, unsigned int virq,
+			      unsigned int nr_irqs, void *arg);
+extern void mp_irqdomain_free(struct irq_domain *domain, unsigned int virq,
+			      unsigned int nr_irqs);
+extern void mp_irqdomain_activate(struct irq_domain *domain,
+				  struct irq_data *irq_data);
+extern void mp_irqdomain_deactivate(struct irq_domain *domain,
+				    struct irq_data *irq_data);
+extern int mp_irqdomain_ioapic_idx(struct irq_domain *domain);
+#endif /* CONFIG_X86_IO_APIC */
+
+#ifdef CONFIG_PCI_MSI
+extern void arch_init_msi_domain(struct irq_domain *domain);
+#else
+static inline void arch_init_msi_domain(struct irq_domain *domain) { }
+#endif
+
+#ifdef CONFIG_HT_IRQ
+extern void arch_init_htirq_domain(struct irq_domain *domain);
+#else
+static inline void arch_init_htirq_domain(struct irq_domain *domain) { }
+#endif
+
+#endif

arch/x86/include/asm/msi.h

@@ -0,0 +1,7 @@
+#ifndef _ASM_X86_MSI_H
+#define _ASM_X86_MSI_H
+#include <asm/hw_irq.h>
+
+typedef struct irq_alloc_info msi_alloc_info_t;
+
+#endif /* _ASM_X86_MSI_H */

arch/x86/include/asm/mtrr.h

@@ -31,7 +31,7 @@
  * arch_phys_wc_add and arch_phys_wc_del.
  */
 # ifdef CONFIG_MTRR
-extern u8 mtrr_type_lookup(u64 addr, u64 end);
+extern u8 mtrr_type_lookup(u64 addr, u64 end, u8 *uniform);
 extern void mtrr_save_fixed_ranges(void *);
 extern void mtrr_save_state(void);
 extern int mtrr_add(unsigned long base, unsigned long size,
@@ -48,14 +48,13 @@ extern void mtrr_aps_init(void);
 extern void mtrr_bp_restore(void);
 extern int mtrr_trim_uncached_memory(unsigned long end_pfn);
 extern int amd_special_default_mtrr(void);
-extern int phys_wc_to_mtrr_index(int handle);
 #  else
-static inline u8 mtrr_type_lookup(u64 addr, u64 end)
+static inline u8 mtrr_type_lookup(u64 addr, u64 end, u8 *uniform)
 {
 	/*
 	 * Return no-MTRRs:
 	 */
-	return 0xff;
+	return MTRR_TYPE_INVALID;
 }
 #define mtrr_save_fixed_ranges(arg) do {} while (0)
 #define mtrr_save_state() do {} while (0)
@@ -84,10 +83,6 @@ static inline int mtrr_trim_uncached_memory(unsigned long end_pfn)
 static inline void mtrr_centaur_report_mcr(int mcr, u32 lo, u32 hi)
 {
 }
-static inline int phys_wc_to_mtrr_index(int handle)
-{
-	return -1;
-}
 
 #define mtrr_ap_init() do {} while (0)
 #define mtrr_bp_init() do {} while (0)
@@ -127,4 +122,8 @@ struct mtrr_gentry32 {
 				 _IOW(MTRR_IOCTL_BASE,  9, struct mtrr_sentry32)
 #endif /* CONFIG_COMPAT */
 
+/* Bit fields for enabled in struct mtrr_state_type */
+#define MTRR_STATE_MTRR_FIXED_ENABLED	0x01
+#define MTRR_STATE_MTRR_ENABLED		0x02
+
 #endif /* _ASM_X86_MTRR_H */

arch/x86/include/asm/paravirt_types.h

@@ -160,13 +160,14 @@ struct pv_cpu_ops {
 	u64 (*read_pmc)(int counter);
 	unsigned long long (*read_tscp)(unsigned int *aux);
 
+#ifdef CONFIG_X86_32
 	/*
 	 * Atomically enable interrupts and return to userspace.  This
-	 * is only ever used to return to 32-bit processes; in a
-	 * 64-bit kernel, it's used for 32-on-64 compat processes, but
-	 * never native 64-bit processes.  (Jump, not call.)
+	 * is only used in 32-bit kernels.  64-bit kernels use
+	 * usergs_sysret32 instead.
 	 */
 	void (*irq_enable_sysexit)(void);
+#endif
 
 	/*
 	 * Switch to usermode gs and return to 64-bit usermode using

arch/x86/include/asm/pat.h

@@ -4,12 +4,7 @@
 #include <linux/types.h>
 #include <asm/pgtable_types.h>
 
-#ifdef CONFIG_X86_PAT
-extern int pat_enabled;
-#else
-static const int pat_enabled;
-#endif
-
+bool pat_enabled(void);
 extern void pat_init(void);
 void pat_init_cache_modes(void);
 

arch/x86/include/asm/pci.h

@@ -96,15 +96,10 @@ extern void pci_iommu_alloc(void);
 #ifdef CONFIG_PCI_MSI
 /* implemented in arch/x86/kernel/apic/io_apic. */
 struct msi_desc;
-void native_compose_msi_msg(struct pci_dev *pdev, unsigned int irq,
-			    unsigned int dest, struct msi_msg *msg, u8 hpet_id);
 int native_setup_msi_irqs(struct pci_dev *dev, int nvec, int type);
 void native_teardown_msi_irq(unsigned int irq);
 void native_restore_msi_irqs(struct pci_dev *dev);
-int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc,
-		  unsigned int irq_base, unsigned int irq_offset);
 #else
-#define native_compose_msi_msg		NULL
 #define native_setup_msi_irqs		NULL
 #define native_teardown_msi_irq		NULL
 #endif

arch/x86/include/asm/special_insns.h

@@ -215,6 +215,44 @@ static inline void clwb(volatile void *__p)
 		: [pax] "a" (p));
 }
 
+/**
+ * pcommit_sfence() - persistent commit and fence
+ *
+ * The PCOMMIT instruction ensures that data that has been flushed from the
+ * processor's cache hierarchy with CLWB, CLFLUSHOPT or CLFLUSH is accepted to
+ * memory and is durable on the DIMM.  The primary use case for this is
+ * persistent memory.
+ *
+ * This function shows how to properly use CLWB/CLFLUSHOPT/CLFLUSH and PCOMMIT
+ * with appropriate fencing.
+ *
+ * Example:
+ * void flush_and_commit_buffer(void *vaddr, unsigned int size)
+ * {
+ *         unsigned long clflush_mask = boot_cpu_data.x86_clflush_size - 1;
+ *         void *vend = vaddr + size;
+ *         void *p;
+ *
+ *         for (p = (void *)((unsigned long)vaddr & ~clflush_mask);
+ *              p < vend; p += boot_cpu_data.x86_clflush_size)
+ *                 clwb(p);
+ *
+ *         // SFENCE to order CLWB/CLFLUSHOPT/CLFLUSH cache flushes
+ *         // MFENCE via mb() also works
+ *         wmb();
+ *
+ *         // PCOMMIT and the required SFENCE for ordering
+ *         pcommit_sfence();
+ * }
+ *
+ * After this function completes the data pointed to by 'vaddr' has been
+ * accepted to memory and will be durable if the 'vaddr' points to persistent
+ * memory.
+ *
+ * PCOMMIT must always be ordered by an MFENCE or SFENCE, so to help simplify
+ * things we include both the PCOMMIT and the required SFENCE in the
+ * alternatives generated by pcommit_sfence().
+ */
 static inline void pcommit_sfence(void)
 {
 	alternative(ASM_NOP7,

arch/x86/include/asm/thread_info.h

@@ -177,8 +177,6 @@ struct thread_info {
  */
 #ifndef __ASSEMBLY__
 
-DECLARE_PER_CPU(unsigned long, kernel_stack);
-
 static inline struct thread_info *current_thread_info(void)
 {
 	return (struct thread_info *)(current_top_of_stack() - THREAD_SIZE);
@@ -197,9 +195,13 @@ static inline unsigned long current_stack_pointer(void)
 
 #else /* !__ASSEMBLY__ */
 
+#ifdef CONFIG_X86_64
+# define cpu_current_top_of_stack (cpu_tss + TSS_sp0)
+#endif
+
 /* Load thread_info address into "reg" */
 #define GET_THREAD_INFO(reg) \
-	_ASM_MOV PER_CPU_VAR(kernel_stack),reg ; \
+	_ASM_MOV PER_CPU_VAR(cpu_current_top_of_stack),reg ; \
 	_ASM_SUB $(THREAD_SIZE),reg ;
 
 /*

arch/x86/include/asm/uaccess_32.h

@@ -59,6 +59,10 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
 			__put_user_size(*(u32 *)from, (u32 __user *)to,
 					4, ret, 4);
 			return ret;
+		case 8:
+			__put_user_size(*(u64 *)from, (u64 __user *)to,
+					8, ret, 8);
+			return ret;
 		}
 	}
 	return __copy_to_user_ll(to, from, n);

arch/x86/include/asm/x86_init.h

@@ -171,38 +171,17 @@ struct x86_platform_ops {
 };
 
 struct pci_dev;
-struct msi_msg;
 
 struct x86_msi_ops {
 	int (*setup_msi_irqs)(struct pci_dev *dev, int nvec, int type);
-	void (*compose_msi_msg)(struct pci_dev *dev, unsigned int irq,
-				unsigned int dest, struct msi_msg *msg,
-			       u8 hpet_id);
 	void (*teardown_msi_irq)(unsigned int irq);
 	void (*teardown_msi_irqs)(struct pci_dev *dev);
 	void (*restore_msi_irqs)(struct pci_dev *dev);
-	int  (*setup_hpet_msi)(unsigned int irq, unsigned int id);
 };
 
-struct IO_APIC_route_entry;
-struct io_apic_irq_attr;
-struct irq_data;
-struct cpumask;
-
 struct x86_io_apic_ops {
-	void		(*init)   (void);
 	unsigned int	(*read)   (unsigned int apic, unsigned int reg);
-	void		(*write)  (unsigned int apic, unsigned int reg, unsigned int value);
-	void		(*modify) (unsigned int apic, unsigned int reg, unsigned int value);
 	void		(*disable)(void);
-	void		(*print_entries)(unsigned int apic, unsigned int nr_entries);
-	int		(*set_affinity)(struct irq_data *data,
-					const struct cpumask *mask,
-					bool force);
-	int		(*setup_entry)(int irq, struct IO_APIC_route_entry *entry,
-				       unsigned int destination, int vector,
-				       struct io_apic_irq_attr *attr);
-	void		(*eoi_ioapic_pin)(int apic, int pin, int vector);
 };
 
 extern struct x86_init_ops x86_init;

arch/x86/include/uapi/asm/mtrr.h

@@ -103,7 +103,7 @@ struct mtrr_state_type {
 #define MTRRIOC_GET_PAGE_ENTRY   _IOWR(MTRR_IOCTL_BASE, 8, struct mtrr_gentry)
 #define MTRRIOC_KILL_PAGE_ENTRY  _IOW(MTRR_IOCTL_BASE,  9, struct mtrr_sentry)
 
-/*  These are the region types  */
+/* MTRR memory types, which are defined in SDM */
 #define MTRR_TYPE_UNCACHABLE 0
 #define MTRR_TYPE_WRCOMB     1
 /*#define MTRR_TYPE_         2*/
@@ -113,5 +113,11 @@ struct mtrr_state_type {
 #define MTRR_TYPE_WRBACK     6
 #define MTRR_NUM_TYPES       7
 
+/*
+ * Invalid MTRR memory type.  mtrr_type_lookup() returns this value when
+ * MTRRs are disabled.  Note, this value is allocated from the reserved
+ * values (0x7-0xff) of the MTRR memory types.
+ */
+#define MTRR_TYPE_INVALID    0xff
 
 #endif /* _UAPI_ASM_X86_MTRR_H */

arch/x86/kernel/acpi/boot.c

@@ -31,12 +31,12 @@
 #include <linux/module.h>
 #include <linux/dmi.h>
 #include <linux/irq.h>
-#include <linux/irqdomain.h>
 #include <linux/slab.h>
 #include <linux/bootmem.h>
 #include <linux/ioport.h>
 #include <linux/pci.h>
 
+#include <asm/irqdomain.h>
 #include <asm/pci_x86.h>
 #include <asm/pgtable.h>
 #include <asm/io_apic.h>
@@ -400,57 +400,13 @@ static int mp_config_acpi_gsi(struct device *dev, u32 gsi, int trigger,
 	return 0;
 }
 
-static int mp_register_gsi(struct device *dev, u32 gsi, int trigger,
-			   int polarity)
-{
-	int irq, node;
-
-	if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
-		return gsi;
-
-	trigger = trigger == ACPI_EDGE_SENSITIVE ? 0 : 1;
-	polarity = polarity == ACPI_ACTIVE_HIGH ? 0 : 1;
-	node = dev ? dev_to_node(dev) : NUMA_NO_NODE;
-	if (mp_set_gsi_attr(gsi, trigger, polarity, node)) {
-		pr_warn("Failed to set pin attr for GSI%d\n", gsi);
-		return -1;
-	}
-
-	irq = mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC);
-	if (irq < 0)
-		return irq;
-
-	/* Don't set up the ACPI SCI because it's already set up */
-	if (enable_update_mptable && acpi_gbl_FADT.sci_interrupt != gsi)
-		mp_config_acpi_gsi(dev, gsi, trigger, polarity);
-
-	return irq;
-}
-
-static void mp_unregister_gsi(u32 gsi)
-{
-	int irq;
-
-	if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
-		return;
-
-	irq = mp_map_gsi_to_irq(gsi, 0);
-	if (irq > 0)
-		mp_unmap_irq(irq);
-}
-
-static struct irq_domain_ops acpi_irqdomain_ops = {
-	.map = mp_irqdomain_map,
-	.unmap = mp_irqdomain_unmap,
-};
-
 static int __init
 acpi_parse_ioapic(struct acpi_subtable_header * header, const unsigned long end)
 {
 	struct acpi_madt_io_apic *ioapic = NULL;
 	struct ioapic_domain_cfg cfg = {
 		.type = IOAPIC_DOMAIN_DYNAMIC,
-		.ops = &acpi_irqdomain_ops,
+		.ops = &mp_ioapic_irqdomain_ops,
 	};
 
 	ioapic = (struct acpi_madt_io_apic *)header;
@@ -652,7 +608,7 @@ static int acpi_register_gsi_pic(struct device *dev, u32 gsi,
 	 * Make sure all (legacy) PCI IRQs are set as level-triggered.
 	 */
 	if (trigger == ACPI_LEVEL_SENSITIVE)
-		eisa_set_level_irq(gsi);
+		elcr_set_level_irq(gsi);
 #endif
 
 	return gsi;
@@ -663,10 +619,21 @@ static int acpi_register_gsi_ioapic(struct device *dev, u32 gsi,
 				    int trigger, int polarity)
 {
 	int irq = gsi;
-
 #ifdef CONFIG_X86_IO_APIC
+	int node;
+	struct irq_alloc_info info;
+
+	node = dev ? dev_to_node(dev) : NUMA_NO_NODE;
+	trigger = trigger == ACPI_EDGE_SENSITIVE ? 0 : 1;
+	polarity = polarity == ACPI_ACTIVE_HIGH ? 0 : 1;
+	ioapic_set_alloc_attr(&info, node, trigger, polarity);
+
 	mutex_lock(&acpi_ioapic_lock);
-	irq = mp_register_gsi(dev, gsi, trigger, polarity);
+	irq = mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC, &info);
+	/* Don't set up the ACPI SCI because it's already set up */
+	if (irq >= 0 && enable_update_mptable &&
+	    acpi_gbl_FADT.sci_interrupt != gsi)
+		mp_config_acpi_gsi(dev, gsi, trigger, polarity);
 	mutex_unlock(&acpi_ioapic_lock);
 #endif
 
@@ -676,8 +643,12 @@ static int acpi_register_gsi_ioapic(struct device *dev, u32 gsi,
 static void acpi_unregister_gsi_ioapic(u32 gsi)
 {
 #ifdef CONFIG_X86_IO_APIC
+	int irq;
+
 	mutex_lock(&acpi_ioapic_lock);
-	mp_unregister_gsi(gsi);
+	irq = mp_map_gsi_to_irq(gsi, 0, NULL);
+	if (irq > 0)
+		mp_unmap_irq(irq);
 	mutex_unlock(&acpi_ioapic_lock);
 #endif
 }
@@ -786,7 +757,7 @@ int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base)
 	u64 addr;
 	struct ioapic_domain_cfg cfg = {
 		.type = IOAPIC_DOMAIN_DYNAMIC,
-		.ops = &acpi_irqdomain_ops,
+		.ops = &mp_ioapic_irqdomain_ops,
 	};
 
 	ioapic_id = acpi_get_ioapic_id(handle, gsi_base, &addr);

arch/x86/kernel/acpi/wakeup_64.S

@@ -62,7 +62,7 @@ ENTRY(do_suspend_lowlevel)
 	pushfq
 	popq	pt_regs_flags(%rax)
 
-	movq	$resume_point, saved_rip(%rip)
+	movq	$.Lresume_point, saved_rip(%rip)
 
 	movq	%rsp, saved_rsp
 	movq	%rbp, saved_rbp
@@ -75,10 +75,10 @@ ENTRY(do_suspend_lowlevel)
 	xorl	%eax, %eax
 	call	x86_acpi_enter_sleep_state
 	/* in case something went wrong, restore the machine status and go on */
-	jmp	resume_point
+	jmp	.Lresume_point
 
 	.align 4
-resume_point:
+.Lresume_point:
 	/* We don't restore %rax, it must be 0 anyway */
 	movq	$saved_context, %rax
 	movq	saved_context_cr4(%rax), %rbx

arch/x86/kernel/alternative.c

@@ -227,6 +227,15 @@ void __init arch_init_ideal_nops(void)
 #endif
 		}
 		break;
+
+	case X86_VENDOR_AMD:
+		if (boot_cpu_data.x86 > 0xf) {
+			ideal_nops = p6_nops;
+			return;
+		}
+
+		/* fall through */
+
 	default:
 #ifdef CONFIG_X86_64
 		ideal_nops = k8_nops;

arch/x86/kernel/apb_timer.c

@@ -171,10 +171,6 @@ static int __init apbt_clockevent_register(void)
 
 static void apbt_setup_irq(struct apbt_dev *adev)
 {
-	/* timer0 irq has been setup early */
-	if (adev->irq == 0)
-		return;
-
 	irq_modify_status(adev->irq, 0, IRQ_MOVE_PCNTXT);
 	irq_set_affinity(adev->irq, cpumask_of(adev->cpu));
 }

arch/x86/kernel/apic/htirq.c

@@ -3,6 +3,8 @@
  *
  * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo
  *	Moved from arch/x86/kernel/apic/io_apic.c.
+ * Jiang Liu <jiang.liu@linux.intel.com>
+ *	Add support of hierarchical irqdomain
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -14,78 +16,112 @@
 #include <linux/device.h>
 #include <linux/pci.h>
 #include <linux/htirq.h>
+#include <asm/irqdomain.h>
 #include <asm/hw_irq.h>
 #include <asm/apic.h>
 #include <asm/hypertransport.h>
 
+static struct irq_domain *htirq_domain;
+
 /*
  * Hypertransport interrupt support
  */
-static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
-{
-	struct ht_irq_msg msg;
-
-	fetch_ht_irq_msg(irq, &msg);
-
-	msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK);
-	msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK);
-
-	msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest);
-	msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest);
-
-	write_ht_irq_msg(irq, &msg);
-}
-
 static int
 ht_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	unsigned int dest;
+	struct irq_data *parent = data->parent_data;
 	int ret;
 
-	ret = apic_set_affinity(data, mask, &dest);
-	if (ret)
-		return ret;
-
-	target_ht_irq(data->irq, dest, cfg->vector);
-	return IRQ_SET_MASK_OK_NOCOPY;
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret >= 0) {
+		struct ht_irq_msg msg;
+		struct irq_cfg *cfg = irqd_cfg(data);
+
+		fetch_ht_irq_msg(data->irq, &msg);
+		msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK |
+				    HT_IRQ_LOW_DEST_ID_MASK);
+		msg.address_lo |= HT_IRQ_LOW_VECTOR(cfg->vector) |
+				  HT_IRQ_LOW_DEST_ID(cfg->dest_apicid);
+		msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK);
+		msg.address_hi |= HT_IRQ_HIGH_DEST_ID(cfg->dest_apicid);
+		write_ht_irq_msg(data->irq, &msg);
+	}
+
+	return ret;
 }
 
 static struct irq_chip ht_irq_chip = {
 	.name			= "PCI-HT",
 	.irq_mask		= mask_ht_irq,
 	.irq_unmask		= unmask_ht_irq,
-	.irq_ack		= apic_ack_edge,
+	.irq_ack		= irq_chip_ack_parent,
 	.irq_set_affinity	= ht_set_affinity,
-	.irq_retrigger		= apic_retrigger_irq,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
 	.flags			= IRQCHIP_SKIP_SET_WAKE,
 };
 
-int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
+static int htirq_domain_alloc(struct irq_domain *domain, unsigned int virq,
+			      unsigned int nr_irqs, void *arg)
 {
-	struct irq_cfg *cfg;
-	struct ht_irq_msg msg;
-	unsigned dest;
-	int err;
+	struct ht_irq_cfg *ht_cfg;
+	struct irq_alloc_info *info = arg;
+	struct pci_dev *dev;
+	irq_hw_number_t hwirq;
+	int ret;
 
-	if (disable_apic)
-		return -ENXIO;
+	if (nr_irqs > 1 || !info)
+		return -EINVAL;
 
-	cfg = irq_cfg(irq);
-	err = assign_irq_vector(irq, cfg, apic->target_cpus());
-	if (err)
-		return err;
+	dev = info->ht_dev;
+	hwirq = (info->ht_idx & 0xFF) |
+		PCI_DEVID(dev->bus->number, dev->devfn) << 8 |
+		(pci_domain_nr(dev->bus) & 0xFFFFFFFF) << 24;
+	if (irq_find_mapping(domain, hwirq) > 0)
+		return -EEXIST;
 
-	err = apic->cpu_mask_to_apicid_and(cfg->domain,
-					   apic->target_cpus(), &dest);
-	if (err)
-		return err;
+	ht_cfg = kmalloc(sizeof(*ht_cfg), GFP_KERNEL);
+	if (!ht_cfg)
+		return -ENOMEM;
 
-	msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, info);
+	if (ret < 0) {
+		kfree(ht_cfg);
+		return ret;
+	}
+
+	/* Initialize msg to a value that will never match the first write. */
+	ht_cfg->msg.address_lo = 0xffffffff;
+	ht_cfg->msg.address_hi = 0xffffffff;
+	ht_cfg->dev = info->ht_dev;
+	ht_cfg->update = info->ht_update;
+	ht_cfg->pos = info->ht_pos;
+	ht_cfg->idx = 0x10 + (info->ht_idx * 2);
+	irq_domain_set_info(domain, virq, hwirq, &ht_irq_chip, ht_cfg,
+			    handle_edge_irq, ht_cfg, "edge");
+
+	return 0;
+}
+
+static void htirq_domain_free(struct irq_domain *domain, unsigned int virq,
+			      unsigned int nr_irqs)
+{
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+
+	BUG_ON(nr_irqs != 1);
+	kfree(irq_data->chip_data);
+	irq_domain_free_irqs_top(domain, virq, nr_irqs);
+}
 
+static void htirq_domain_activate(struct irq_domain *domain,
+				  struct irq_data *irq_data)
+{
+	struct ht_irq_msg msg;
+	struct irq_cfg *cfg = irqd_cfg(irq_data);
+
+	msg.address_hi = HT_IRQ_HIGH_DEST_ID(cfg->dest_apicid);
 	msg.address_lo =
 		HT_IRQ_LOW_BASE |
-		HT_IRQ_LOW_DEST_ID(dest) |
+		HT_IRQ_LOW_DEST_ID(cfg->dest_apicid) |
 		HT_IRQ_LOW_VECTOR(cfg->vector) |
 		((apic->irq_dest_mode == 0) ?
 			HT_IRQ_LOW_DM_PHYSICAL :
@@ -95,13 +131,56 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 			HT_IRQ_LOW_MT_FIXED :
 			HT_IRQ_LOW_MT_ARBITRATED) |
 		HT_IRQ_LOW_IRQ_MASKED;
+	write_ht_irq_msg(irq_data->irq, &msg);
+}
 
-	write_ht_irq_msg(irq, &msg);
+static void htirq_domain_deactivate(struct irq_domain *domain,
+				    struct irq_data *irq_data)
+{
+	struct ht_irq_msg msg;
 
-	irq_set_chip_and_handler_name(irq, &ht_irq_chip,
-				      handle_edge_irq, "edge");
+	memset(&msg, 0, sizeof(msg));
+	write_ht_irq_msg(irq_data->irq, &msg);
+}
 
-	dev_dbg(&dev->dev, "irq %d for HT\n", irq);
+static const struct irq_domain_ops htirq_domain_ops = {
+	.alloc		= htirq_domain_alloc,
+	.free		= htirq_domain_free,
+	.activate	= htirq_domain_activate,
+	.deactivate	= htirq_domain_deactivate,
+};
 
-	return 0;
+void arch_init_htirq_domain(struct irq_domain *parent)
+{
+	if (disable_apic)
+		return;
+
+	htirq_domain = irq_domain_add_tree(NULL, &htirq_domain_ops, NULL);
+	if (!htirq_domain)
+		pr_warn("failed to initialize irqdomain for HTIRQ.\n");
+	else
+		htirq_domain->parent = parent;
+}
+
+int arch_setup_ht_irq(int idx, int pos, struct pci_dev *dev,
+		      ht_irq_update_t *update)
+{
+	struct irq_alloc_info info;
+
+	if (!htirq_domain)
+		return -ENOSYS;
+
+	init_irq_alloc_info(&info, NULL);
+	info.ht_idx = idx;
+	info.ht_pos = pos;
+	info.ht_dev = dev;
+	info.ht_update = update;
+
+	return irq_domain_alloc_irqs(htirq_domain, 1, dev_to_node(&dev->dev),
+				     &info);
+}
+
+void arch_teardown_ht_irq(unsigned int irq)
+{
+	irq_domain_free_irqs(irq, 1);
 }

arch/x86/kernel/apic/io_apic.c

@@ -18,6 +18,16 @@
  *					and Rolf G. Tews
  *					for testing these extensively
  *	Paul Diefenbaugh	:	Added full ACPI support
+ *
+ * Historical information which is worth to be preserved:
+ *
+ * - SiS APIC rmw bug:
+ *
+ *	We used to have a workaround for a bug in SiS chips which
+ *	required to rewrite the index register for a read-modify-write
+ *	operation as the chip lost the index information which was
+ *	setup for the read already. We cache the data now, so that
+ *	workaround has been removed.
  */
 
 #include <linux/mm.h>
@@ -31,13 +41,13 @@
 #include <linux/acpi.h>
 #include <linux/module.h>
 #include <linux/syscore_ops.h>
-#include <linux/irqdomain.h>
 #include <linux/freezer.h>
 #include <linux/kthread.h>
 #include <linux/jiffies.h>	/* time_after() */
 #include <linux/slab.h>
 #include <linux/bootmem.h>
 
+#include <asm/irqdomain.h>
 #include <asm/idle.h>
 #include <asm/io.h>
 #include <asm/smp.h>
@@ -63,27 +73,31 @@
 #define	for_each_ioapic_pin(idx, pin)	\
 	for_each_ioapic((idx))		\
 		for_each_pin((idx), (pin))
-
 #define for_each_irq_pin(entry, head) \
 	list_for_each_entry(entry, &head, list)
 
-/*
- *      Is the SiS APIC rmw bug present ?
- *      -1 = don't know, 0 = no, 1 = yes
- */
-int sis_apic_bug = -1;
-
 static DEFINE_RAW_SPINLOCK(ioapic_lock);
 static DEFINE_MUTEX(ioapic_mutex);
 static unsigned int ioapic_dynirq_base;
 static int ioapic_initialized;
 
-struct mp_pin_info {
+struct irq_pin_list {
+	struct list_head list;
+	int apic, pin;
+};
+
+struct mp_chip_data {
+	struct list_head irq_2_pin;
+	struct IO_APIC_route_entry entry;
 	int trigger;
 	int polarity;
-	int node;
-	int set;
 	u32 count;
+	bool isa_irq;
+};
+
+struct mp_ioapic_gsi {
+	u32 gsi_base;
+	u32 gsi_end;
 };
 
 static struct ioapic {
@@ -101,7 +115,6 @@ static struct ioapic {
 	struct mp_ioapic_gsi  gsi_config;
 	struct ioapic_domain_cfg irqdomain_cfg;
 	struct irq_domain *irqdomain;
-	struct mp_pin_info *pin_info;
 	struct resource *iomem_res;
 } ioapics[MAX_IO_APICS];
 
@@ -117,7 +130,7 @@ unsigned int mpc_ioapic_addr(int ioapic_idx)
 	return ioapics[ioapic_idx].mp_config.apicaddr;
 }
 
-struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic_idx)
+static inline struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic_idx)
 {
 	return &ioapics[ioapic_idx].gsi_config;
 }
@@ -129,11 +142,16 @@ static inline int mp_ioapic_pin_count(int ioapic)
 	return gsi_cfg->gsi_end - gsi_cfg->gsi_base + 1;
 }
 
-u32 mp_pin_to_gsi(int ioapic, int pin)
+static inline u32 mp_pin_to_gsi(int ioapic, int pin)
 {
 	return mp_ioapic_gsi_routing(ioapic)->gsi_base + pin;
 }
 
+static inline bool mp_is_legacy_irq(int irq)
+{
+	return irq >= 0 && irq < nr_legacy_irqs();
+}
+
 /*
  * Initialize all legacy IRQs and all pins on the first IOAPIC
  * if we have legacy interrupt controller. Kernel boot option "pirq="
@@ -144,12 +162,7 @@ static inline int mp_init_irq_at_boot(int ioapic, int irq)
 	if (!nr_legacy_irqs())
 		return 0;
 
-	return ioapic == 0 || (irq >= 0 && irq < nr_legacy_irqs());
-}
-
-static inline struct mp_pin_info *mp_pin_info(int ioapic_idx, int pin)
-{
-	return ioapics[ioapic_idx].pin_info + pin;
+	return ioapic == 0 || mp_is_legacy_irq(irq);
 }
 
 static inline struct irq_domain *mp_ioapic_irqdomain(int ioapic)
@@ -216,16 +229,6 @@ void mp_save_irq(struct mpc_intsrc *m)
 		panic("Max # of irq sources exceeded!!\n");
 }
 
-struct irq_pin_list {
-	struct list_head list;
-	int apic, pin;
-};
-
-static struct irq_pin_list *alloc_irq_pin_list(int node)
-{
-	return kzalloc_node(sizeof(struct irq_pin_list), GFP_KERNEL, node);
-}
-
 static void alloc_ioapic_saved_registers(int idx)
 {
 	size_t size;
@@ -247,8 +250,7 @@ static void free_ioapic_saved_registers(int idx)
 
 int __init arch_early_ioapic_init(void)
 {
-	struct irq_cfg *cfg;
-	int i, node = cpu_to_node(0);
+	int i;
 
 	if (!nr_legacy_irqs())
 		io_apic_irqs = ~0UL;
@@ -256,16 +258,6 @@ int __init arch_early_ioapic_init(void)
 	for_each_ioapic(i)
 		alloc_ioapic_saved_registers(i);
 
-	/*
-	 * For legacy IRQ's, start with assigning irq0 to irq15 to
-	 * IRQ0_VECTOR to IRQ15_VECTOR for all cpu's.
-	 */
-	for (i = 0; i < nr_legacy_irqs(); i++) {
-		cfg = alloc_irq_and_cfg_at(i, node);
-		cfg->vector = IRQ0_VECTOR + i;
-		cpumask_setall(cfg->domain);
-	}
-
 	return 0;
 }
 
@@ -283,7 +275,7 @@ static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
 		+ (mpc_ioapic_addr(idx) & ~PAGE_MASK);
 }
 
-void io_apic_eoi(unsigned int apic, unsigned int vector)
+static inline void io_apic_eoi(unsigned int apic, unsigned int vector)
 {
 	struct io_apic __iomem *io_apic = io_apic_base(apic);
 	writel(vector, &io_apic->eoi);
@@ -296,7 +288,8 @@ unsigned int native_io_apic_read(unsigned int apic, unsigned int reg)
 	return readl(&io_apic->data);
 }
 
-void native_io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
+static void io_apic_write(unsigned int apic, unsigned int reg,
+			  unsigned int value)
 {
 	struct io_apic __iomem *io_apic = io_apic_base(apic);
 
@@ -304,21 +297,6 @@ void native_io_apic_write(unsigned int apic, unsigned int reg, unsigned int valu
 	writel(value, &io_apic->data);
 }
 
-/*
- * Re-write a value: to be used for read-modify-write
- * cycles where the read already set up the index register.
- *
- * Older SiS APIC requires we rewrite the index register
- */
-void native_io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value)
-{
-	struct io_apic __iomem *io_apic = io_apic_base(apic);
-
-	if (sis_apic_bug)
-		writel(reg, &io_apic->index);
-	writel(value, &io_apic->data);
-}
-
 union entry_union {
 	struct { u32 w1, w2; };
 	struct IO_APIC_route_entry entry;
@@ -378,7 +356,7 @@ static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
 static void ioapic_mask_entry(int apic, int pin)
 {
 	unsigned long flags;
-	union entry_union eu = { .entry.mask = 1 };
+	union entry_union eu = { .entry.mask = IOAPIC_MASKED };
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
 	io_apic_write(apic, 0x10 + 2*pin, eu.w1);
@@ -391,16 +369,17 @@ static void ioapic_mask_entry(int apic, int pin)
  * shared ISA-space IRQs, so we have to support them. We are super
  * fast in the common case, and fast for shared ISA-space IRQs.
  */
-static int __add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
+static int __add_pin_to_irq_node(struct mp_chip_data *data,
+				 int node, int apic, int pin)
 {
 	struct irq_pin_list *entry;
 
 	/* don't allow duplicates */
-	for_each_irq_pin(entry, cfg->irq_2_pin)
+	for_each_irq_pin(entry, data->irq_2_pin)
 		if (entry->apic == apic && entry->pin == pin)
 			return 0;
 
-	entry = alloc_irq_pin_list(node);
+	entry = kzalloc_node(sizeof(struct irq_pin_list), GFP_ATOMIC, node);
 	if (!entry) {
 		pr_err("can not alloc irq_pin_list (%d,%d,%d)\n",
 		       node, apic, pin);
@@ -408,16 +387,16 @@ static int __add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pi
 	}
 	entry->apic = apic;
 	entry->pin = pin;
+	list_add_tail(&entry->list, &data->irq_2_pin);
 
-	list_add_tail(&entry->list, &cfg->irq_2_pin);
 	return 0;
 }
 
-static void __remove_pin_from_irq(struct irq_cfg *cfg, int apic, int pin)
+static void __remove_pin_from_irq(struct mp_chip_data *data, int apic, int pin)
 {
 	struct irq_pin_list *tmp, *entry;
 
-	list_for_each_entry_safe(entry, tmp, &cfg->irq_2_pin, list)
+	list_for_each_entry_safe(entry, tmp, &data->irq_2_pin, list)
 		if (entry->apic == apic && entry->pin == pin) {
 			list_del(&entry->list);
 			kfree(entry);
@@ -425,22 +404,23 @@ static void __remove_pin_from_irq(struct irq_cfg *cfg, int apic, int pin)
 		}
 }
 
-static void add_pin_to_irq_node(struct irq_cfg *cfg, int node, int apic, int pin)
+static void add_pin_to_irq_node(struct mp_chip_data *data,
+				int node, int apic, int pin)
 {
-	if (__add_pin_to_irq_node(cfg, node, apic, pin))
+	if (__add_pin_to_irq_node(data, node, apic, pin))
 		panic("IO-APIC: failed to add irq-pin. Can not proceed\n");
 }
 
 /*
  * Reroute an IRQ to a different pin.
  */
-static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
+static void __init replace_pin_at_irq_node(struct mp_chip_data *data, int node,
 					   int oldapic, int oldpin,
 					   int newapic, int newpin)
 {
 	struct irq_pin_list *entry;
 
-	for_each_irq_pin(entry, cfg->irq_2_pin) {
+	for_each_irq_pin(entry, data->irq_2_pin) {
 		if (entry->apic == oldapic && entry->pin == oldpin) {
 			entry->apic = newapic;
 			entry->pin = newpin;
@@ -450,32 +430,26 @@ static void __init replace_pin_at_irq_node(struct irq_cfg *cfg, int node,
 	}
 
 	/* old apic/pin didn't exist, so just add new ones */
-	add_pin_to_irq_node(cfg, node, newapic, newpin);
-}
-
-static void __io_apic_modify_irq(struct irq_pin_list *entry,
-				 int mask_and, int mask_or,
-				 void (*final)(struct irq_pin_list *entry))
-{
-	unsigned int reg, pin;
-
-	pin = entry->pin;
-	reg = io_apic_read(entry->apic, 0x10 + pin * 2);
-	reg &= mask_and;
-	reg |= mask_or;
-	io_apic_modify(entry->apic, 0x10 + pin * 2, reg);
-	if (final)
-		final(entry);
+	add_pin_to_irq_node(data, node, newapic, newpin);
 }
 
-static void io_apic_modify_irq(struct irq_cfg *cfg,
+static void io_apic_modify_irq(struct mp_chip_data *data,
 			       int mask_and, int mask_or,
 			       void (*final)(struct irq_pin_list *entry))
 {
+	union entry_union eu;
 	struct irq_pin_list *entry;
 
-	for_each_irq_pin(entry, cfg->irq_2_pin)
-		__io_apic_modify_irq(entry, mask_and, mask_or, final);
+	eu.entry = data->entry;
+	eu.w1 &= mask_and;
+	eu.w1 |= mask_or;
+	data->entry = eu.entry;
+
+	for_each_irq_pin(entry, data->irq_2_pin) {
+		io_apic_write(entry->apic, 0x10 + 2 * entry->pin, eu.w1);
+		if (final)
+			final(entry);
+	}
 }
 
 static void io_apic_sync(struct irq_pin_list *entry)
@@ -490,39 +464,31 @@ static void io_apic_sync(struct irq_pin_list *entry)
 	readl(&io_apic->data);
 }
 
-static void mask_ioapic(struct irq_cfg *cfg)
+static void mask_ioapic_irq(struct irq_data *irq_data)
 {
+	struct mp_chip_data *data = irq_data->chip_data;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
+	io_apic_modify_irq(data, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void mask_ioapic_irq(struct irq_data *data)
+static void __unmask_ioapic(struct mp_chip_data *data)
 {
-	mask_ioapic(irqd_cfg(data));
+	io_apic_modify_irq(data, ~IO_APIC_REDIR_MASKED, 0, NULL);
 }
 
-static void __unmask_ioapic(struct irq_cfg *cfg)
-{
-	io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
-}
-
-static void unmask_ioapic(struct irq_cfg *cfg)
+static void unmask_ioapic_irq(struct irq_data *irq_data)
 {
+	struct mp_chip_data *data = irq_data->chip_data;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	__unmask_ioapic(cfg);
+	__unmask_ioapic(data);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void unmask_ioapic_irq(struct irq_data *data)
-{
-	unmask_ioapic(irqd_cfg(data));
-}
-
 /*
  * IO-APIC versions below 0x20 don't support EOI register.
  * For the record, here is the information about various versions:
@@ -539,7 +505,7 @@ static void unmask_ioapic_irq(struct irq_data *data)
  * Otherwise, we simulate the EOI message manually by changing the trigger
  * mode to edge and then back to level, with RTE being masked during this.
  */
-void native_eoi_ioapic_pin(int apic, int pin, int vector)
+static void __eoi_ioapic_pin(int apic, int pin, int vector)
 {
 	if (mpc_ioapic_ver(apic) >= 0x20) {
 		io_apic_eoi(apic, vector);
@@ -551,7 +517,7 @@ void native_eoi_ioapic_pin(int apic, int pin, int vector)
 		/*
 		 * Mask the entry and change the trigger mode to edge.
 		 */
-		entry1.mask = 1;
+		entry1.mask = IOAPIC_MASKED;
 		entry1.trigger = IOAPIC_EDGE;
 
 		__ioapic_write_entry(apic, pin, entry1);
@@ -563,15 +529,14 @@ void native_eoi_ioapic_pin(int apic, int pin, int vector)
 	}
 }
 
-void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
+void eoi_ioapic_pin(int vector, struct mp_chip_data *data)
 {
-	struct irq_pin_list *entry;
 	unsigned long flags;
+	struct irq_pin_list *entry;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	for_each_irq_pin(entry, cfg->irq_2_pin)
-		x86_io_apic_ops.eoi_ioapic_pin(entry->apic, entry->pin,
-					       cfg->vector);
+	for_each_irq_pin(entry, data->irq_2_pin)
+		__eoi_ioapic_pin(entry->apic, entry->pin, vector);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
@@ -588,8 +553,8 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 	 * Make sure the entry is masked and re-read the contents to check
 	 * if it is a level triggered pin and if the remote-IRR is set.
 	 */
-	if (!entry.mask) {
-		entry.mask = 1;
+	if (entry.mask == IOAPIC_UNMASKED) {
+		entry.mask = IOAPIC_MASKED;
 		ioapic_write_entry(apic, pin, entry);
 		entry = ioapic_read_entry(apic, pin);
 	}
@@ -602,13 +567,12 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 		 * doesn't clear the remote-IRR if the trigger mode is not
 		 * set to level.
 		 */
-		if (!entry.trigger) {
+		if (entry.trigger == IOAPIC_EDGE) {
 			entry.trigger = IOAPIC_LEVEL;
 			ioapic_write_entry(apic, pin, entry);
 		}
-
 		raw_spin_lock_irqsave(&ioapic_lock, flags);
-		x86_io_apic_ops.eoi_ioapic_pin(apic, pin, entry.vector);
+		__eoi_ioapic_pin(apic, pin, entry.vector);
 		raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 	}
 
@@ -706,8 +670,8 @@ void mask_ioapic_entries(void)
 			struct IO_APIC_route_entry entry;
 
 			entry = ioapics[apic].saved_registers[pin];
-			if (!entry.mask) {
-				entry.mask = 1;
+			if (entry.mask == IOAPIC_UNMASKED) {
+				entry.mask = IOAPIC_MASKED;
 				ioapic_write_entry(apic, pin, entry);
 			}
 		}
@@ -809,11 +773,11 @@ static int EISA_ELCR(unsigned int irq)
 
 #endif
 
-/* ISA interrupts are always polarity zero edge triggered,
+/* ISA interrupts are always active high edge triggered,
  * when listed as conforming in the MP table. */
 
-#define default_ISA_trigger(idx)	(0)
-#define default_ISA_polarity(idx)	(0)
+#define default_ISA_trigger(idx)	(IOAPIC_EDGE)
+#define default_ISA_polarity(idx)	(IOAPIC_POL_HIGH)
 
 /* EISA interrupts are always polarity zero and can be edge or level
  * trigger depending on the ELCR value.  If an interrupt is listed as
@@ -823,53 +787,55 @@ static int EISA_ELCR(unsigned int irq)
 #define default_EISA_trigger(idx)	(EISA_ELCR(mp_irqs[idx].srcbusirq))
 #define default_EISA_polarity(idx)	default_ISA_polarity(idx)
 
-/* PCI interrupts are always polarity one level triggered,
+/* PCI interrupts are always active low level triggered,
  * when listed as conforming in the MP table. */
 
-#define default_PCI_trigger(idx)	(1)
-#define default_PCI_polarity(idx)	(1)
+#define default_PCI_trigger(idx)	(IOAPIC_LEVEL)
+#define default_PCI_polarity(idx)	(IOAPIC_POL_LOW)
 
 static int irq_polarity(int idx)
 {
 	int bus = mp_irqs[idx].srcbus;
-	int polarity;
 
 	/*
 	 * Determine IRQ line polarity (high active or low active):
 	 */
-	switch (mp_irqs[idx].irqflag & 3)
-	{
-		case 0: /* conforms, ie. bus-type dependent polarity */
-			if (test_bit(bus, mp_bus_not_pci))
-				polarity = default_ISA_polarity(idx);
-			else
-				polarity = default_PCI_polarity(idx);
-			break;
-		case 1: /* high active */
-		{
-			polarity = 0;
-			break;
-		}
-		case 2: /* reserved */
-		{
-			pr_warn("broken BIOS!!\n");
-			polarity = 1;
-			break;
-		}
-		case 3: /* low active */
-		{
-			polarity = 1;
-			break;
-		}
-		default: /* invalid */
-		{
-			pr_warn("broken BIOS!!\n");
-			polarity = 1;
-			break;
-		}
+	switch (mp_irqs[idx].irqflag & 0x03) {
+	case 0:
+		/* conforms to spec, ie. bus-type dependent polarity */
+		if (test_bit(bus, mp_bus_not_pci))
+			return default_ISA_polarity(idx);
+		else
+			return default_PCI_polarity(idx);
+	case 1:
+		return IOAPIC_POL_HIGH;
+	case 2:
+		pr_warn("IOAPIC: Invalid polarity: 2, defaulting to low\n");
+	case 3:
+	default: /* Pointless default required due to do gcc stupidity */
+		return IOAPIC_POL_LOW;
+	}
+}
+
+#ifdef CONFIG_EISA
+static int eisa_irq_trigger(int idx, int bus, int trigger)
+{
+	switch (mp_bus_id_to_type[bus]) {
+	case MP_BUS_PCI:
+	case MP_BUS_ISA:
+		return trigger;
+	case MP_BUS_EISA:
+		return default_EISA_trigger(idx);
 	}
-	return polarity;
+	pr_warn("IOAPIC: Invalid srcbus: %d defaulting to level\n", bus);
+	return IOAPIC_LEVEL;
 }
+#else
+static inline int eisa_irq_trigger(int idx, int bus, int trigger)
+{
+	return trigger;
+}
+#endif
 
 static int irq_trigger(int idx)
 {
@@ -879,153 +845,227 @@ static int irq_trigger(int idx)
 	/*
 	 * Determine IRQ trigger mode (edge or level sensitive):
 	 */
-	switch ((mp_irqs[idx].irqflag>>2) & 3)
-	{
-		case 0: /* conforms, ie. bus-type dependent */
-			if (test_bit(bus, mp_bus_not_pci))
-				trigger = default_ISA_trigger(idx);
-			else
-				trigger = default_PCI_trigger(idx);
-#ifdef CONFIG_EISA
-			switch (mp_bus_id_to_type[bus]) {
-				case MP_BUS_ISA: /* ISA pin */
-				{
-					/* set before the switch */
-					break;
-				}
-				case MP_BUS_EISA: /* EISA pin */
-				{
-					trigger = default_EISA_trigger(idx);
-					break;
-				}
-				case MP_BUS_PCI: /* PCI pin */
-				{
-					/* set before the switch */
-					break;
-				}
-				default:
-				{
-					pr_warn("broken BIOS!!\n");
-					trigger = 1;
-					break;
-				}
-			}
+	switch ((mp_irqs[idx].irqflag >> 2) & 0x03) {
+	case 0:
+		/* conforms to spec, ie. bus-type dependent trigger mode */
+		if (test_bit(bus, mp_bus_not_pci))
+			trigger = default_ISA_trigger(idx);
+		else
+			trigger = default_PCI_trigger(idx);
+		/* Take EISA into account */
+		return eisa_irq_trigger(idx, bus, trigger);
+	case 1:
+		return IOAPIC_EDGE;
+	case 2:
+		pr_warn("IOAPIC: Invalid trigger mode 2 defaulting to level\n");
+	case 3:
+	default: /* Pointless default required due to do gcc stupidity */
+		return IOAPIC_LEVEL;
+	}
+}
+
+void ioapic_set_alloc_attr(struct irq_alloc_info *info, int node,
+			   int trigger, int polarity)
+{
+	init_irq_alloc_info(info, NULL);
+	info->type = X86_IRQ_ALLOC_TYPE_IOAPIC;
+	info->ioapic_node = node;
+	info->ioapic_trigger = trigger;
+	info->ioapic_polarity = polarity;
+	info->ioapic_valid = 1;
+}
+
+#ifndef CONFIG_ACPI
+int acpi_get_override_irq(u32 gsi, int *trigger, int *polarity);
 #endif
-			break;
-		case 1: /* edge */
-		{
-			trigger = 0;
-			break;
-		}
-		case 2: /* reserved */
-		{
-			pr_warn("broken BIOS!!\n");
-			trigger = 1;
-			break;
-		}
-		case 3: /* level */
-		{
-			trigger = 1;
-			break;
-		}
-		default: /* invalid */
-		{
-			pr_warn("broken BIOS!!\n");
-			trigger = 0;
-			break;
+
+static void ioapic_copy_alloc_attr(struct irq_alloc_info *dst,
+				   struct irq_alloc_info *src,
+				   u32 gsi, int ioapic_idx, int pin)
+{
+	int trigger, polarity;
+
+	copy_irq_alloc_info(dst, src);
+	dst->type = X86_IRQ_ALLOC_TYPE_IOAPIC;
+	dst->ioapic_id = mpc_ioapic_id(ioapic_idx);
+	dst->ioapic_pin = pin;
+	dst->ioapic_valid = 1;
+	if (src && src->ioapic_valid) {
+		dst->ioapic_node = src->ioapic_node;
+		dst->ioapic_trigger = src->ioapic_trigger;
+		dst->ioapic_polarity = src->ioapic_polarity;
+	} else {
+		dst->ioapic_node = NUMA_NO_NODE;
+		if (acpi_get_override_irq(gsi, &trigger, &polarity) >= 0) {
+			dst->ioapic_trigger = trigger;
+			dst->ioapic_polarity = polarity;
+		} else {
+			/*
+			 * PCI interrupts are always active low level
+			 * triggered.
+			 */
+			dst->ioapic_trigger = IOAPIC_LEVEL;
+			dst->ioapic_polarity = IOAPIC_POL_LOW;
 		}
 	}
-	return trigger;
 }
 
-static int alloc_irq_from_domain(struct irq_domain *domain, u32 gsi, int pin)
+static int ioapic_alloc_attr_node(struct irq_alloc_info *info)
+{
+	return (info && info->ioapic_valid) ? info->ioapic_node : NUMA_NO_NODE;
+}
+
+static void mp_register_handler(unsigned int irq, unsigned long trigger)
+{
+	irq_flow_handler_t hdl;
+	bool fasteoi;
+
+	if (trigger) {
+		irq_set_status_flags(irq, IRQ_LEVEL);
+		fasteoi = true;
+	} else {
+		irq_clear_status_flags(irq, IRQ_LEVEL);
+		fasteoi = false;
+	}
+
+	hdl = fasteoi ? handle_fasteoi_irq : handle_edge_irq;
+	__irq_set_handler(irq, hdl, 0, fasteoi ? "fasteoi" : "edge");
+}
+
+static bool mp_check_pin_attr(int irq, struct irq_alloc_info *info)
 {
+	struct mp_chip_data *data = irq_get_chip_data(irq);
+
+	/*
+	 * setup_IO_APIC_irqs() programs all legacy IRQs with default trigger
+	 * and polarity attirbutes. So allow the first user to reprogram the
+	 * pin with real trigger and polarity attributes.
+	 */
+	if (irq < nr_legacy_irqs() && data->count == 1) {
+		if (info->ioapic_trigger != data->trigger)
+			mp_register_handler(irq, data->trigger);
+		data->entry.trigger = data->trigger = info->ioapic_trigger;
+		data->entry.polarity = data->polarity = info->ioapic_polarity;
+	}
+
+	return data->trigger == info->ioapic_trigger &&
+	       data->polarity == info->ioapic_polarity;
+}
+
+static int alloc_irq_from_domain(struct irq_domain *domain, int ioapic, u32 gsi,
+				 struct irq_alloc_info *info)
+{
+	bool legacy = false;
 	int irq = -1;
-	int ioapic = (int)(long)domain->host_data;
 	int type = ioapics[ioapic].irqdomain_cfg.type;
 
 	switch (type) {
 	case IOAPIC_DOMAIN_LEGACY:
 		/*
-		 * Dynamically allocate IRQ number for non-ISA IRQs in the first 16
-		 * GSIs on some weird platforms.
+		 * Dynamically allocate IRQ number for non-ISA IRQs in the first
+		 * 16 GSIs on some weird platforms.
 		 */
-		if (gsi < nr_legacy_irqs())
-			irq = irq_create_mapping(domain, pin);
-		else if (irq_create_strict_mappings(domain, gsi, pin, 1) == 0)
+		if (!ioapic_initialized || gsi >= nr_legacy_irqs())
 			irq = gsi;
+		legacy = mp_is_legacy_irq(irq);
 		break;
 	case IOAPIC_DOMAIN_STRICT:
-		if (irq_create_strict_mappings(domain, gsi, pin, 1) == 0)
-			irq = gsi;
+		irq = gsi;
 		break;
 	case IOAPIC_DOMAIN_DYNAMIC:
-		irq = irq_create_mapping(domain, pin);
 		break;
 	default:
 		WARN(1, "ioapic: unknown irqdomain type %d\n", type);
-		break;
+		return -1;
+	}
+
+	return __irq_domain_alloc_irqs(domain, irq, 1,
+				       ioapic_alloc_attr_node(info),
+				       info, legacy);
+}
+
+/*
+ * Need special handling for ISA IRQs because there may be multiple IOAPIC pins
+ * sharing the same ISA IRQ number and irqdomain only supports 1:1 mapping
+ * between IOAPIC pin and IRQ number. A typical IOAPIC has 24 pins, pin 0-15 are
+ * used for legacy IRQs and pin 16-23 are used for PCI IRQs (PIRQ A-H).
+ * When ACPI is disabled, only legacy IRQ numbers (IRQ0-15) are available, and
+ * some BIOSes may use MP Interrupt Source records to override IRQ numbers for
+ * PIRQs instead of reprogramming the interrupt routing logic. Thus there may be
+ * multiple pins sharing the same legacy IRQ number when ACPI is disabled.
+ */
+static int alloc_isa_irq_from_domain(struct irq_domain *domain,
+				     int irq, int ioapic, int pin,
+				     struct irq_alloc_info *info)
+{
+	struct mp_chip_data *data;
+	struct irq_data *irq_data = irq_get_irq_data(irq);
+	int node = ioapic_alloc_attr_node(info);
+
+	/*
+	 * Legacy ISA IRQ has already been allocated, just add pin to
+	 * the pin list assoicated with this IRQ and program the IOAPIC
+	 * entry. The IOAPIC entry
+	 */
+	if (irq_data && irq_data->parent_data) {
+		if (!mp_check_pin_attr(irq, info))
+			return -EBUSY;
+		if (__add_pin_to_irq_node(irq_data->chip_data, node, ioapic,
+					  info->ioapic_pin))
+			return -ENOMEM;
+	} else {
+		irq = __irq_domain_alloc_irqs(domain, irq, 1, node, info, true);
+		if (irq >= 0) {
+			irq_data = irq_domain_get_irq_data(domain, irq);
+			data = irq_data->chip_data;
+			data->isa_irq = true;
+		}
 	}
 
-	return irq > 0 ? irq : -1;
+	return irq;
 }
 
 static int mp_map_pin_to_irq(u32 gsi, int idx, int ioapic, int pin,
-			     unsigned int flags)
+			     unsigned int flags, struct irq_alloc_info *info)
 {
 	int irq;
+	bool legacy = false;
+	struct irq_alloc_info tmp;
+	struct mp_chip_data *data;
 	struct irq_domain *domain = mp_ioapic_irqdomain(ioapic);
-	struct mp_pin_info *info = mp_pin_info(ioapic, pin);
 
 	if (!domain)
-		return -1;
+		return -ENOSYS;
 
-	mutex_lock(&ioapic_mutex);
-
-	/*
-	 * Don't use irqdomain to manage ISA IRQs because there may be
-	 * multiple IOAPIC pins sharing the same ISA IRQ number and
-	 * irqdomain only supports 1:1 mapping between IOAPIC pin and
-	 * IRQ number. A typical IOAPIC has 24 pins, pin 0-15 are used
-	 * for legacy IRQs and pin 16-23 are used for PCI IRQs (PIRQ A-H).
-	 * When ACPI is disabled, only legacy IRQ numbers (IRQ0-15) are
-	 * available, and some BIOSes may use MP Interrupt Source records
-	 * to override IRQ numbers for PIRQs instead of reprogramming
-	 * the interrupt routing logic. Thus there may be multiple pins
-	 * sharing the same legacy IRQ number when ACPI is disabled.
-	 */
 	if (idx >= 0 && test_bit(mp_irqs[idx].srcbus, mp_bus_not_pci)) {
 		irq = mp_irqs[idx].srcbusirq;
-		if (flags & IOAPIC_MAP_ALLOC) {
-			if (info->count == 0 &&
-			    mp_irqdomain_map(domain, irq, pin) != 0)
-				irq = -1;
+		legacy = mp_is_legacy_irq(irq);
+	}
 
-			/* special handling for timer IRQ0 */
+	mutex_lock(&ioapic_mutex);
+	if (!(flags & IOAPIC_MAP_ALLOC)) {
+		if (!legacy) {
+			irq = irq_find_mapping(domain, pin);
 			if (irq == 0)
-				info->count++;
+				irq = -ENOENT;
 		}
 	} else {
-		irq = irq_find_mapping(domain, pin);
-		if (irq <= 0 && (flags & IOAPIC_MAP_ALLOC))
-			irq = alloc_irq_from_domain(domain, gsi, pin);
-	}
-
-	if (flags & IOAPIC_MAP_ALLOC) {
-		/* special handling for legacy IRQs */
-		if (irq < nr_legacy_irqs() && info->count == 1 &&
-		    mp_irqdomain_map(domain, irq, pin) != 0)
-			irq = -1;
-
-		if (irq > 0)
-			info->count++;
-		else if (info->count == 0)
-			info->set = 0;
+		ioapic_copy_alloc_attr(&tmp, info, gsi, ioapic, pin);
+		if (legacy)
+			irq = alloc_isa_irq_from_domain(domain, irq,
+							ioapic, pin, &tmp);
+		else if ((irq = irq_find_mapping(domain, pin)) == 0)
+			irq = alloc_irq_from_domain(domain, ioapic, gsi, &tmp);
+		else if (!mp_check_pin_attr(irq, &tmp))
+			irq = -EBUSY;
+		if (irq >= 0) {
+			data = irq_get_chip_data(irq);
+			data->count++;
+		}
 	}
-
 	mutex_unlock(&ioapic_mutex);
 
-	return irq > 0 ? irq : -1;
+	return irq;
 }
 
 static int pin_2_irq(int idx, int ioapic, int pin, unsigned int flags)
@@ -1058,10 +1098,10 @@ static int pin_2_irq(int idx, int ioapic, int pin, unsigned int flags)
 	}
 #endif
 
-	return  mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags);
+	return  mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags, NULL);
 }
 
-int mp_map_gsi_to_irq(u32 gsi, unsigned int flags)
+int mp_map_gsi_to_irq(u32 gsi, unsigned int flags, struct irq_alloc_info *info)
 {
 	int ioapic, pin, idx;
 
@@ -1074,31 +1114,24 @@ int mp_map_gsi_to_irq(u32 gsi, unsigned int flags)
 	if ((flags & IOAPIC_MAP_CHECK) && idx < 0)
 		return -1;
 
-	return mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags);
+	return mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags, info);
 }
 
 void mp_unmap_irq(int irq)
 {
-	struct irq_data *data = irq_get_irq_data(irq);
-	struct mp_pin_info *info;
-	int ioapic, pin;
+	struct irq_data *irq_data = irq_get_irq_data(irq);
+	struct mp_chip_data *data;
 
-	if (!data || !data->domain)
+	if (!irq_data || !irq_data->domain)
 		return;
 
-	ioapic = (int)(long)data->domain->host_data;
-	pin = (int)data->hwirq;
-	info = mp_pin_info(ioapic, pin);
+	data = irq_data->chip_data;
+	if (!data || data->isa_irq)
+		return;
 
 	mutex_lock(&ioapic_mutex);
-	if (--info->count == 0) {
-		info->set = 0;
-		if (irq < nr_legacy_irqs() &&
-		    ioapics[ioapic].irqdomain_cfg.type == IOAPIC_DOMAIN_LEGACY)
-			mp_irqdomain_unmap(data->domain, irq);
-		else
-			irq_dispose_mapping(irq);
-	}
+	if (--data->count == 0)
+		irq_domain_free_irqs(irq, 1);
 	mutex_unlock(&ioapic_mutex);
 }
 
@@ -1165,7 +1198,7 @@ out:
 }
 EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
 
-static struct irq_chip ioapic_chip;
+static struct irq_chip ioapic_chip, ioapic_ir_chip;
 
 #ifdef CONFIG_X86_32
 static inline int IO_APIC_irq_trigger(int irq)
@@ -1189,96 +1222,6 @@ static inline int IO_APIC_irq_trigger(int irq)
 }
 #endif
 
-static void ioapic_register_intr(unsigned int irq, struct irq_cfg *cfg,
-				 unsigned long trigger)
-{
-	struct irq_chip *chip = &ioapic_chip;
-	irq_flow_handler_t hdl;
-	bool fasteoi;
-
-	if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
-	    trigger == IOAPIC_LEVEL) {
-		irq_set_status_flags(irq, IRQ_LEVEL);
-		fasteoi = true;
-	} else {
-		irq_clear_status_flags(irq, IRQ_LEVEL);
-		fasteoi = false;
-	}
-
-	if (setup_remapped_irq(irq, cfg, chip))
-		fasteoi = trigger != 0;
-
-	hdl = fasteoi ? handle_fasteoi_irq : handle_edge_irq;
-	irq_set_chip_and_handler_name(irq, chip, hdl,
-				      fasteoi ? "fasteoi" : "edge");
-}
-
-int native_setup_ioapic_entry(int irq, struct IO_APIC_route_entry *entry,
-			      unsigned int destination, int vector,
-			      struct io_apic_irq_attr *attr)
-{
-	memset(entry, 0, sizeof(*entry));
-
-	entry->delivery_mode = apic->irq_delivery_mode;
-	entry->dest_mode     = apic->irq_dest_mode;
-	entry->dest	     = destination;
-	entry->vector	     = vector;
-	entry->mask	     = 0;			/* enable IRQ */
-	entry->trigger	     = attr->trigger;
-	entry->polarity	     = attr->polarity;
-
-	/*
-	 * Mask level triggered irqs.
-	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
-	 */
-	if (attr->trigger)
-		entry->mask = 1;
-
-	return 0;
-}
-
-static void setup_ioapic_irq(unsigned int irq, struct irq_cfg *cfg,
-				struct io_apic_irq_attr *attr)
-{
-	struct IO_APIC_route_entry entry;
-	unsigned int dest;
-
-	if (!IO_APIC_IRQ(irq))
-		return;
-
-	if (assign_irq_vector(irq, cfg, apic->target_cpus()))
-		return;
-
-	if (apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus(),
-					 &dest)) {
-		pr_warn("Failed to obtain apicid for ioapic %d, pin %d\n",
-			mpc_ioapic_id(attr->ioapic), attr->ioapic_pin);
-		clear_irq_vector(irq, cfg);
-
-		return;
-	}
-
-	apic_printk(APIC_VERBOSE,KERN_DEBUG
-		    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> "
-		    "IRQ %d Mode:%i Active:%i Dest:%d)\n",
-		    attr->ioapic, mpc_ioapic_id(attr->ioapic), attr->ioapic_pin,
-		    cfg->vector, irq, attr->trigger, attr->polarity, dest);
-
-	if (x86_io_apic_ops.setup_entry(irq, &entry, dest, cfg->vector, attr)) {
-		pr_warn("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
-			mpc_ioapic_id(attr->ioapic), attr->ioapic_pin);
-		clear_irq_vector(irq, cfg);
-
-		return;
-	}
-
-	ioapic_register_intr(irq, cfg, attr->trigger);
-	if (irq < nr_legacy_irqs())
-		legacy_pic->mask(irq);
-
-	ioapic_write_entry(attr->ioapic, attr->ioapic_pin, entry);
-}
-
 static void __init setup_IO_APIC_irqs(void)
 {
 	unsigned int ioapic, pin;
@@ -1298,106 +1241,41 @@ static void __init setup_IO_APIC_irqs(void)
 	}
 }
 
-/*
- * Set up the timer pin, possibly with the 8259A-master behind.
- */
-static void __init setup_timer_IRQ0_pin(unsigned int ioapic_idx,
-					unsigned int pin, int vector)
-{
-	struct IO_APIC_route_entry entry;
-	unsigned int dest;
-
-	memset(&entry, 0, sizeof(entry));
-
-	/*
-	 * We use logical delivery to get the timer IRQ
-	 * to the first CPU.
-	 */
-	if (unlikely(apic->cpu_mask_to_apicid_and(apic->target_cpus(),
-						  apic->target_cpus(), &dest)))
-		dest = BAD_APICID;
-
-	entry.dest_mode = apic->irq_dest_mode;
-	entry.mask = 0;			/* don't mask IRQ for edge */
-	entry.dest = dest;
-	entry.delivery_mode = apic->irq_delivery_mode;
-	entry.polarity = 0;
-	entry.trigger = 0;
-	entry.vector = vector;
-
-	/*
-	 * The timer IRQ doesn't have to know that behind the
-	 * scene we may have a 8259A-master in AEOI mode ...
-	 */
-	irq_set_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq,
-				      "edge");
-
-	/*
-	 * Add it to the IO-APIC irq-routing table:
-	 */
-	ioapic_write_entry(ioapic_idx, pin, entry);
-}
-
-void native_io_apic_print_entries(unsigned int apic, unsigned int nr_entries)
+void ioapic_zap_locks(void)
 {
-	int i;
-
-	pr_debug(" NR Dst Mask Trig IRR Pol Stat Dmod Deli Vect:\n");
-
-	for (i = 0; i <= nr_entries; i++) {
-		struct IO_APIC_route_entry entry;
-
-		entry = ioapic_read_entry(apic, i);
-
-		pr_debug(" %02x %02X  ", i, entry.dest);
-		pr_cont("%1d    %1d    %1d   %1d   %1d    "
-			"%1d    %1d    %02X\n",
-			entry.mask,
-			entry.trigger,
-			entry.irr,
-			entry.polarity,
-			entry.delivery_status,
-			entry.dest_mode,
-			entry.delivery_mode,
-			entry.vector);
-	}
+	raw_spin_lock_init(&ioapic_lock);
 }
 
-void intel_ir_io_apic_print_entries(unsigned int apic,
-				    unsigned int nr_entries)
+static void io_apic_print_entries(unsigned int apic, unsigned int nr_entries)
 {
 	int i;
+	char buf[256];
+	struct IO_APIC_route_entry entry;
+	struct IR_IO_APIC_route_entry *ir_entry = (void *)&entry;
 
-	pr_debug(" NR Indx Fmt Mask Trig IRR Pol Stat Indx2 Zero Vect:\n");
-
+	printk(KERN_DEBUG "IOAPIC %d:\n", apic);
 	for (i = 0; i <= nr_entries; i++) {
-		struct IR_IO_APIC_route_entry *ir_entry;
-		struct IO_APIC_route_entry entry;
-
 		entry = ioapic_read_entry(apic, i);
-
-		ir_entry = (struct IR_IO_APIC_route_entry *)&entry;
-
-		pr_debug(" %02x %04X ", i, ir_entry->index);
-		pr_cont("%1d   %1d    %1d    %1d   %1d   "
-			"%1d    %1d     %X    %02X\n",
-			ir_entry->format,
-			ir_entry->mask,
-			ir_entry->trigger,
-			ir_entry->irr,
-			ir_entry->polarity,
-			ir_entry->delivery_status,
-			ir_entry->index2,
-			ir_entry->zero,
-			ir_entry->vector);
+		snprintf(buf, sizeof(buf),
+			 " pin%02x, %s, %s, %s, V(%02X), IRR(%1d), S(%1d)",
+			 i,
+			 entry.mask == IOAPIC_MASKED ? "disabled" : "enabled ",
+			 entry.trigger == IOAPIC_LEVEL ? "level" : "edge ",
+			 entry.polarity == IOAPIC_POL_LOW ? "low " : "high",
+			 entry.vector, entry.irr, entry.delivery_status);
+		if (ir_entry->format)
+			printk(KERN_DEBUG "%s, remapped, I(%04X),  Z(%X)\n",
+			       buf, (ir_entry->index << 15) | ir_entry->index,
+			       ir_entry->zero);
+		else
+			printk(KERN_DEBUG "%s, %s, D(%02X), M(%1d)\n",
+			       buf,
+			       entry.dest_mode == IOAPIC_DEST_MODE_LOGICAL ?
+			       "logical " : "physical",
+			       entry.dest, entry.delivery_mode);
 	}
 }
 
-void ioapic_zap_locks(void)
-{
-	raw_spin_lock_init(&ioapic_lock);
-}
-
 static void __init print_IO_APIC(int ioapic_idx)
 {
 	union IO_APIC_reg_00 reg_00;
@@ -1451,16 +1329,13 @@ static void __init print_IO_APIC(int ioapic_idx)
 	}
 
 	printk(KERN_DEBUG ".... IRQ redirection table:\n");
-
-	x86_io_apic_ops.print_entries(ioapic_idx, reg_01.bits.entries);
+	io_apic_print_entries(ioapic_idx, reg_01.bits.entries);
 }
 
 void __init print_IO_APICs(void)
 {
 	int ioapic_idx;
-	struct irq_cfg *cfg;
 	unsigned int irq;
-	struct irq_chip *chip;
 
 	printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
 	for_each_ioapic(ioapic_idx)
@@ -1480,18 +1355,20 @@ void __init print_IO_APICs(void)
 	printk(KERN_DEBUG "IRQ to pin mappings:\n");
 	for_each_active_irq(irq) {
 		struct irq_pin_list *entry;
+		struct irq_chip *chip;
+		struct mp_chip_data *data;
 
 		chip = irq_get_chip(irq);
-		if (chip != &ioapic_chip)
+		if (chip != &ioapic_chip && chip != &ioapic_ir_chip)
 			continue;
-
-		cfg = irq_cfg(irq);
-		if (!cfg)
+		data = irq_get_chip_data(irq);
+		if (!data)
 			continue;
-		if (list_empty(&cfg->irq_2_pin))
+		if (list_empty(&data->irq_2_pin))
 			continue;
+
 		printk(KERN_DEBUG "IRQ%d ", irq);
-		for_each_irq_pin(entry, cfg->irq_2_pin)
+		for_each_irq_pin(entry, data->irq_2_pin)
 			pr_cont("-> %d:%d", entry->apic, entry->pin);
 		pr_cont("\n");
 	}
@@ -1564,15 +1441,12 @@ void native_disable_io_apic(void)
 		struct IO_APIC_route_entry entry;
 
 		memset(&entry, 0, sizeof(entry));
-		entry.mask            = 0; /* Enabled */
-		entry.trigger         = 0; /* Edge */
-		entry.irr             = 0;
-		entry.polarity        = 0; /* High */
-		entry.delivery_status = 0;
-		entry.dest_mode       = 0; /* Physical */
-		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
-		entry.vector          = 0;
-		entry.dest            = read_apic_id();
+		entry.mask		= IOAPIC_UNMASKED;
+		entry.trigger		= IOAPIC_EDGE;
+		entry.polarity		= IOAPIC_POL_HIGH;
+		entry.dest_mode		= IOAPIC_DEST_MODE_PHYSICAL;
+		entry.delivery_mode	= dest_ExtINT;
+		entry.dest		= read_apic_id();
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
@@ -1582,7 +1456,6 @@ void native_disable_io_apic(void)
 
 	if (cpu_has_apic || apic_from_smp_config())
 		disconnect_bsp_APIC(ioapic_i8259.pin != -1);
-
 }
 
 /*
@@ -1792,7 +1665,6 @@ static int __init timer_irq_works(void)
  * This is not complete - we should be able to fake
  * an edge even if it isn't on the 8259A...
  */
-
 static unsigned int startup_ioapic_irq(struct irq_data *data)
 {
 	int was_pending = 0, irq = data->irq;
@@ -1804,74 +1676,22 @@ static unsigned int startup_ioapic_irq(struct irq_data *data)
 		if (legacy_pic->irq_pending(irq))
 			was_pending = 1;
 	}
-	__unmask_ioapic(irqd_cfg(data));
+	__unmask_ioapic(data->chip_data);
 	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 
 	return was_pending;
 }
 
-/*
- * Level and edge triggered IO-APIC interrupts need different handling,
- * so we use two separate IRQ descriptors. Edge triggered IRQs can be
- * handled with the level-triggered descriptor, but that one has slightly
- * more overhead. Level-triggered interrupts cannot be handled with the
- * edge-triggered handler, without risking IRQ storms and other ugly
- * races.
- */
-
-static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
-{
-	int apic, pin;
-	struct irq_pin_list *entry;
-	u8 vector = cfg->vector;
-
-	for_each_irq_pin(entry, cfg->irq_2_pin) {
-		unsigned int reg;
-
-		apic = entry->apic;
-		pin = entry->pin;
-
-		io_apic_write(apic, 0x11 + pin*2, dest);
-		reg = io_apic_read(apic, 0x10 + pin*2);
-		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
-		reg |= vector;
-		io_apic_modify(apic, 0x10 + pin*2, reg);
-	}
-}
-
-int native_ioapic_set_affinity(struct irq_data *data,
-			       const struct cpumask *mask,
-			       bool force)
-{
-	unsigned int dest, irq = data->irq;
-	unsigned long flags;
-	int ret;
-
-	if (!config_enabled(CONFIG_SMP))
-		return -EPERM;
-
-	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	ret = apic_set_affinity(data, mask, &dest);
-	if (!ret) {
-		/* Only the high 8 bits are valid. */
-		dest = SET_APIC_LOGICAL_ID(dest);
-		__target_IO_APIC_irq(irq, dest, irqd_cfg(data));
-		ret = IRQ_SET_MASK_OK_NOCOPY;
-	}
-	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
-	return ret;
-}
-
 atomic_t irq_mis_count;
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
-static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
+static bool io_apic_level_ack_pending(struct mp_chip_data *data)
 {
 	struct irq_pin_list *entry;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	for_each_irq_pin(entry, cfg->irq_2_pin) {
+	for_each_irq_pin(entry, data->irq_2_pin) {
 		unsigned int reg;
 		int pin;
 
@@ -1888,18 +1708,17 @@ static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 	return false;
 }
 
-static inline bool ioapic_irqd_mask(struct irq_data *data, struct irq_cfg *cfg)
+static inline bool ioapic_irqd_mask(struct irq_data *data)
 {
 	/* If we are moving the irq we need to mask it */
 	if (unlikely(irqd_is_setaffinity_pending(data))) {
-		mask_ioapic(cfg);
+		mask_ioapic_irq(data);
 		return true;
 	}
 	return false;
 }
 
-static inline void ioapic_irqd_unmask(struct irq_data *data,
-				      struct irq_cfg *cfg, bool masked)
+static inline void ioapic_irqd_unmask(struct irq_data *data, bool masked)
 {
 	if (unlikely(masked)) {
 		/* Only migrate the irq if the ack has been received.
@@ -1928,31 +1747,30 @@ static inline void ioapic_irqd_unmask(struct irq_data *data,
 		 * accurate and is causing problems then it is a hardware bug
 		 * and you can go talk to the chipset vendor about it.
 		 */
-		if (!io_apic_level_ack_pending(cfg))
+		if (!io_apic_level_ack_pending(data->chip_data))
 			irq_move_masked_irq(data);
-		unmask_ioapic(cfg);
+		unmask_ioapic_irq(data);
 	}
 }
 #else
-static inline bool ioapic_irqd_mask(struct irq_data *data, struct irq_cfg *cfg)
+static inline bool ioapic_irqd_mask(struct irq_data *data)
 {
 	return false;
 }
-static inline void ioapic_irqd_unmask(struct irq_data *data,
-				      struct irq_cfg *cfg, bool masked)
+static inline void ioapic_irqd_unmask(struct irq_data *data, bool masked)
 {
 }
 #endif
 
-static void ack_ioapic_level(struct irq_data *data)
+static void ioapic_ack_level(struct irq_data *irq_data)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	int i, irq = data->irq;
+	struct irq_cfg *cfg = irqd_cfg(irq_data);
 	unsigned long v;
 	bool masked;
+	int i;
 
 	irq_complete_move(cfg);
-	masked = ioapic_irqd_mask(data, cfg);
+	masked = ioapic_irqd_mask(irq_data);
 
 	/*
 	 * It appears there is an erratum which affects at least version 0x11
@@ -2004,11 +1822,49 @@ static void ack_ioapic_level(struct irq_data *data)
 	 */
 	if (!(v & (1 << (i & 0x1f)))) {
 		atomic_inc(&irq_mis_count);
+		eoi_ioapic_pin(cfg->vector, irq_data->chip_data);
+	}
+
+	ioapic_irqd_unmask(irq_data, masked);
+}
+
+static void ioapic_ir_ack_level(struct irq_data *irq_data)
+{
+	struct mp_chip_data *data = irq_data->chip_data;
+
+	/*
+	 * Intr-remapping uses pin number as the virtual vector
+	 * in the RTE. Actual vector is programmed in
+	 * intr-remapping table entry. Hence for the io-apic
+	 * EOI we use the pin number.
+	 */
+	ack_APIC_irq();
+	eoi_ioapic_pin(data->entry.vector, data);
+}
 
-		eoi_ioapic_irq(irq, cfg);
+static int ioapic_set_affinity(struct irq_data *irq_data,
+			       const struct cpumask *mask, bool force)
+{
+	struct irq_data *parent = irq_data->parent_data;
+	struct mp_chip_data *data = irq_data->chip_data;
+	struct irq_pin_list *entry;
+	struct irq_cfg *cfg;
+	unsigned long flags;
+	int ret;
+
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	raw_spin_lock_irqsave(&ioapic_lock, flags);
+	if (ret >= 0 && ret != IRQ_SET_MASK_OK_DONE) {
+		cfg = irqd_cfg(irq_data);
+		data->entry.dest = cfg->dest_apicid;
+		data->entry.vector = cfg->vector;
+		for_each_irq_pin(entry, data->irq_2_pin)
+			__ioapic_write_entry(entry->apic, entry->pin,
+					     data->entry);
 	}
+	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
 
-	ioapic_irqd_unmask(data, cfg, masked);
+	return ret;
 }
 
 static struct irq_chip ioapic_chip __read_mostly = {
@@ -2016,10 +1872,20 @@ static struct irq_chip ioapic_chip __read_mostly = {
 	.irq_startup		= startup_ioapic_irq,
 	.irq_mask		= mask_ioapic_irq,
 	.irq_unmask		= unmask_ioapic_irq,
-	.irq_ack		= apic_ack_edge,
-	.irq_eoi		= ack_ioapic_level,
-	.irq_set_affinity	= native_ioapic_set_affinity,
-	.irq_retrigger		= apic_retrigger_irq,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_eoi		= ioapic_ack_level,
+	.irq_set_affinity	= ioapic_set_affinity,
+	.flags			= IRQCHIP_SKIP_SET_WAKE,
+};
+
+static struct irq_chip ioapic_ir_chip __read_mostly = {
+	.name			= "IR-IO-APIC",
+	.irq_startup		= startup_ioapic_irq,
+	.irq_mask		= mask_ioapic_irq,
+	.irq_unmask		= unmask_ioapic_irq,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_eoi		= ioapic_ir_ack_level,
+	.irq_set_affinity	= ioapic_set_affinity,
 	.flags			= IRQCHIP_SKIP_SET_WAKE,
 };
 
@@ -2113,12 +1979,12 @@ static inline void __init unlock_ExtINT_logic(void)
 
 	memset(&entry1, 0, sizeof(entry1));
 
-	entry1.dest_mode = 0;			/* physical delivery */
-	entry1.mask = 0;			/* unmask IRQ now */
+	entry1.dest_mode = IOAPIC_DEST_MODE_PHYSICAL;
+	entry1.mask = IOAPIC_UNMASKED;
 	entry1.dest = hard_smp_processor_id();
 	entry1.delivery_mode = dest_ExtINT;
 	entry1.polarity = entry0.polarity;
-	entry1.trigger = 0;
+	entry1.trigger = IOAPIC_EDGE;
 	entry1.vector = 0;
 
 	ioapic_write_entry(apic, pin, entry1);
@@ -2152,6 +2018,25 @@ static int __init disable_timer_pin_setup(char *arg)
 }
 early_param("disable_timer_pin_1", disable_timer_pin_setup);
 
+static int mp_alloc_timer_irq(int ioapic, int pin)
+{
+	int irq = -1;
+	struct irq_domain *domain = mp_ioapic_irqdomain(ioapic);
+
+	if (domain) {
+		struct irq_alloc_info info;
+
+		ioapic_set_alloc_attr(&info, NUMA_NO_NODE, 0, 0);
+		info.ioapic_id = mpc_ioapic_id(ioapic);
+		info.ioapic_pin = pin;
+		mutex_lock(&ioapic_mutex);
+		irq = alloc_isa_irq_from_domain(domain, 0, ioapic, pin, &info);
+		mutex_unlock(&ioapic_mutex);
+	}
+
+	return irq;
+}
+
 /*
  * This code may look a bit paranoid, but it's supposed to cooperate with
  * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
@@ -2162,7 +2047,9 @@ early_param("disable_timer_pin_1", disable_timer_pin_setup);
  */
 static inline void __init check_timer(void)
 {
-	struct irq_cfg *cfg = irq_cfg(0);
+	struct irq_data *irq_data = irq_get_irq_data(0);
+	struct mp_chip_data *data = irq_data->chip_data;
+	struct irq_cfg *cfg = irqd_cfg(irq_data);
 	int node = cpu_to_node(0);
 	int apic1, pin1, apic2, pin2;
 	unsigned long flags;
@@ -2174,7 +2061,6 @@ static inline void __init check_timer(void)
 	 * get/set the timer IRQ vector:
 	 */
 	legacy_pic->mask(0);
-	assign_irq_vector(0, cfg, apic->target_cpus());
 
 	/*
 	 * As IRQ0 is to be enabled in the 8259A, the virtual
@@ -2215,23 +2101,21 @@ static inline void __init check_timer(void)
 	}
 
 	if (pin1 != -1) {
-		/*
-		 * Ok, does IRQ0 through the IOAPIC work?
-		 */
+		/* Ok, does IRQ0 through the IOAPIC work? */
 		if (no_pin1) {
-			add_pin_to_irq_node(cfg, node, apic1, pin1);
-			setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
+			mp_alloc_timer_irq(apic1, pin1);
 		} else {
-			/* for edge trigger, setup_ioapic_irq already
-			 * leave it unmasked.
+			/*
+			 * for edge trigger, it's already unmasked,
 			 * so only need to unmask if it is level-trigger
 			 * do we really have level trigger timer?
 			 */
 			int idx;
 			idx = find_irq_entry(apic1, pin1, mp_INT);
 			if (idx != -1 && irq_trigger(idx))
-				unmask_ioapic(cfg);
+				unmask_ioapic_irq(irq_get_chip_data(0));
 		}
+		irq_domain_activate_irq(irq_data);
 		if (timer_irq_works()) {
 			if (disable_timer_pin_1 > 0)
 				clear_IO_APIC_pin(0, pin1);
@@ -2251,8 +2135,8 @@ static inline void __init check_timer(void)
 		/*
 		 * legacy devices should be connected to IO APIC #0
 		 */
-		replace_pin_at_irq_node(cfg, node, apic1, pin1, apic2, pin2);
-		setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
+		replace_pin_at_irq_node(data, node, apic1, pin1, apic2, pin2);
+		irq_domain_activate_irq(irq_data);
 		legacy_pic->unmask(0);
 		if (timer_irq_works()) {
 			apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
@@ -2329,36 +2213,35 @@ out:
 
 static int mp_irqdomain_create(int ioapic)
 {
-	size_t size;
+	struct irq_alloc_info info;
+	struct irq_domain *parent;
 	int hwirqs = mp_ioapic_pin_count(ioapic);
 	struct ioapic *ip = &ioapics[ioapic];
 	struct ioapic_domain_cfg *cfg = &ip->irqdomain_cfg;
 	struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(ioapic);
 
-	size = sizeof(struct mp_pin_info) * mp_ioapic_pin_count(ioapic);
-	ip->pin_info = kzalloc(size, GFP_KERNEL);
-	if (!ip->pin_info)
-		return -ENOMEM;
-
 	if (cfg->type == IOAPIC_DOMAIN_INVALID)
 		return 0;
 
+	init_irq_alloc_info(&info, NULL);
+	info.type = X86_IRQ_ALLOC_TYPE_IOAPIC;
+	info.ioapic_id = mpc_ioapic_id(ioapic);
+	parent = irq_remapping_get_ir_irq_domain(&info);
+	if (!parent)
+		parent = x86_vector_domain;
+
 	ip->irqdomain = irq_domain_add_linear(cfg->dev, hwirqs, cfg->ops,
 					      (void *)(long)ioapic);
-	if(!ip->irqdomain) {
-		kfree(ip->pin_info);
-		ip->pin_info = NULL;
+	if (!ip->irqdomain)
 		return -ENOMEM;
-	}
+
+	ip->irqdomain->parent = parent;
 
 	if (cfg->type == IOAPIC_DOMAIN_LEGACY ||
 	    cfg->type == IOAPIC_DOMAIN_STRICT)
 		ioapic_dynirq_base = max(ioapic_dynirq_base,
 					 gsi_cfg->gsi_end + 1);
 
-	if (gsi_cfg->gsi_base == 0)
-		irq_set_default_host(ip->irqdomain);
-
 	return 0;
 }
 
@@ -2368,8 +2251,6 @@ static void ioapic_destroy_irqdomain(int idx)
 		irq_domain_remove(ioapics[idx].irqdomain);
 		ioapics[idx].irqdomain = NULL;
 	}
-	kfree(ioapics[idx].pin_info);
-	ioapics[idx].pin_info = NULL;
 }
 
 void __init setup_IO_APIC(void)
@@ -2399,20 +2280,6 @@ void __init setup_IO_APIC(void)
 	ioapic_initialized = 1;
 }
 
-/*
- *      Called after all the initialization is done. If we didn't find any
- *      APIC bugs then we can allow the modify fast path
- */
-
-static int __init io_apic_bug_finalize(void)
-{
-	if (sis_apic_bug == -1)
-		sis_apic_bug = 0;
-	return 0;
-}
-
-late_initcall(io_apic_bug_finalize);
-
 static void resume_ioapic_id(int ioapic_idx)
 {
 	unsigned long flags;
@@ -2451,20 +2318,6 @@ static int __init ioapic_init_ops(void)
 
 device_initcall(ioapic_init_ops);
 
-static int
-io_apic_setup_irq_pin(unsigned int irq, int node, struct io_apic_irq_attr *attr)
-{
-	struct irq_cfg *cfg = alloc_irq_and_cfg_at(irq, node);
-	int ret;
-
-	if (!cfg)
-		return -EINVAL;
-	ret = __add_pin_to_irq_node(cfg, node, attr->ioapic, attr->ioapic_pin);
-	if (!ret)
-		setup_ioapic_irq(irq, cfg, attr);
-	return ret;
-}
-
 static int io_apic_get_redir_entries(int ioapic)
 {
 	union IO_APIC_reg_01	reg_01;
@@ -2692,7 +2545,7 @@ void __init setup_ioapic_dest(void)
 		else
 			mask = apic->target_cpus();
 
-		x86_io_apic_ops.set_affinity(idata, mask, false);
+		irq_set_affinity(irq, mask);
 	}
 
 }
@@ -2737,7 +2590,7 @@ static struct resource * __init ioapic_setup_resources(void)
 	return res;
 }
 
-void __init native_io_apic_init_mappings(void)
+void __init io_apic_init_mappings(void)
 {
 	unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
 	struct resource *ioapic_res;
@@ -2962,7 +2815,6 @@ int mp_unregister_ioapic(u32 gsi_base)
 {
 	int ioapic, pin;
 	int found = 0;
-	struct mp_pin_info *pin_info;
 
 	for_each_ioapic(ioapic)
 		if (ioapics[ioapic].gsi_config.gsi_base == gsi_base) {
@@ -2975,11 +2827,17 @@ int mp_unregister_ioapic(u32 gsi_base)
 	}
 
 	for_each_pin(ioapic, pin) {
-		pin_info = mp_pin_info(ioapic, pin);
-		if (pin_info->count) {
-			pr_warn("pin%d on IOAPIC%d is still in use.\n",
-				pin, ioapic);
-			return -EBUSY;
+		u32 gsi = mp_pin_to_gsi(ioapic, pin);
+		int irq = mp_map_gsi_to_irq(gsi, 0, NULL);
+		struct mp_chip_data *data;
+
+		if (irq >= 0) {
+			data = irq_get_chip_data(irq);
+			if (data && data->count) {
+				pr_warn("pin%d on IOAPIC%d is still in use.\n",
+					pin, ioapic);
+				return -EBUSY;
+			}
 		}
 	}
 
@@ -3006,108 +2864,141 @@ int mp_ioapic_registered(u32 gsi_base)
 	return 0;
 }
 
-static inline void set_io_apic_irq_attr(struct io_apic_irq_attr *irq_attr,
-					int ioapic, int ioapic_pin,
-					int trigger, int polarity)
+static void mp_irqdomain_get_attr(u32 gsi, struct mp_chip_data *data,
+				  struct irq_alloc_info *info)
 {
-	irq_attr->ioapic	= ioapic;
-	irq_attr->ioapic_pin	= ioapic_pin;
-	irq_attr->trigger	= trigger;
-	irq_attr->polarity	= polarity;
+	if (info && info->ioapic_valid) {
+		data->trigger = info->ioapic_trigger;
+		data->polarity = info->ioapic_polarity;
+	} else if (acpi_get_override_irq(gsi, &data->trigger,
+					 &data->polarity) < 0) {
+		/* PCI interrupts are always active low level triggered. */
+		data->trigger = IOAPIC_LEVEL;
+		data->polarity = IOAPIC_POL_LOW;
+	}
 }
 
-int mp_irqdomain_map(struct irq_domain *domain, unsigned int virq,
-		     irq_hw_number_t hwirq)
+static void mp_setup_entry(struct irq_cfg *cfg, struct mp_chip_data *data,
+			   struct IO_APIC_route_entry *entry)
 {
-	int ioapic = (int)(long)domain->host_data;
-	struct mp_pin_info *info = mp_pin_info(ioapic, hwirq);
-	struct io_apic_irq_attr attr;
+	memset(entry, 0, sizeof(*entry));
+	entry->delivery_mode = apic->irq_delivery_mode;
+	entry->dest_mode     = apic->irq_dest_mode;
+	entry->dest	     = cfg->dest_apicid;
+	entry->vector	     = cfg->vector;
+	entry->trigger	     = data->trigger;
+	entry->polarity	     = data->polarity;
+	/*
+	 * Mask level triggered irqs. Edge triggered irqs are masked
+	 * by the irq core code in case they fire.
+	 */
+	if (data->trigger == IOAPIC_LEVEL)
+		entry->mask = IOAPIC_MASKED;
+	else
+		entry->mask = IOAPIC_UNMASKED;
+}
 
-	/* Get default attribute if not set by caller yet */
-	if (!info->set) {
-		u32 gsi = mp_pin_to_gsi(ioapic, hwirq);
+int mp_irqdomain_alloc(struct irq_domain *domain, unsigned int virq,
+		       unsigned int nr_irqs, void *arg)
+{
+	int ret, ioapic, pin;
+	struct irq_cfg *cfg;
+	struct irq_data *irq_data;
+	struct mp_chip_data *data;
+	struct irq_alloc_info *info = arg;
 
-		if (acpi_get_override_irq(gsi, &info->trigger,
-					  &info->polarity) < 0) {
-			/*
-			 * PCI interrupts are always polarity one level
-			 * triggered.
-			 */
-			info->trigger = 1;
-			info->polarity = 1;
-		}
-		info->node = NUMA_NO_NODE;
+	if (!info || nr_irqs > 1)
+		return -EINVAL;
+	irq_data = irq_domain_get_irq_data(domain, virq);
+	if (!irq_data)
+		return -EINVAL;
 
-		/*
-		 * setup_IO_APIC_irqs() programs all legacy IRQs with default
-		 * trigger and polarity attributes. Don't set the flag for that
-		 * case so the first legacy IRQ user could reprogram the pin
-		 * with real trigger and polarity attributes.
-		 */
-		if (virq >= nr_legacy_irqs() || info->count)
-			info->set = 1;
-	}
-	set_io_apic_irq_attr(&attr, ioapic, hwirq, info->trigger,
-			     info->polarity);
+	ioapic = mp_irqdomain_ioapic_idx(domain);
+	pin = info->ioapic_pin;
+	if (irq_find_mapping(domain, (irq_hw_number_t)pin) > 0)
+		return -EEXIST;
 
-	return io_apic_setup_irq_pin(virq, info->node, &attr);
-}
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
 
-void mp_irqdomain_unmap(struct irq_domain *domain, unsigned int virq)
-{
-	struct irq_data *data = irq_get_irq_data(virq);
-	struct irq_cfg *cfg = irq_cfg(virq);
-	int ioapic = (int)(long)domain->host_data;
-	int pin = (int)data->hwirq;
+	info->ioapic_entry = &data->entry;
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, info);
+	if (ret < 0) {
+		kfree(data);
+		return ret;
+	}
+
+	INIT_LIST_HEAD(&data->irq_2_pin);
+	irq_data->hwirq = info->ioapic_pin;
+	irq_data->chip = (domain->parent == x86_vector_domain) ?
+			  &ioapic_chip : &ioapic_ir_chip;
+	irq_data->chip_data = data;
+	mp_irqdomain_get_attr(mp_pin_to_gsi(ioapic, pin), data, info);
+
+	cfg = irqd_cfg(irq_data);
+	add_pin_to_irq_node(data, ioapic_alloc_attr_node(info), ioapic, pin);
+	if (info->ioapic_entry)
+		mp_setup_entry(cfg, data, info->ioapic_entry);
+	mp_register_handler(virq, data->trigger);
+	if (virq < nr_legacy_irqs())
+		legacy_pic->mask(virq);
+
+	apic_printk(APIC_VERBOSE, KERN_DEBUG
+		    "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i Dest:%d)\n",
+		    ioapic, mpc_ioapic_id(ioapic), pin, cfg->vector,
+		    virq, data->trigger, data->polarity, cfg->dest_apicid);
 
-	ioapic_mask_entry(ioapic, pin);
-	__remove_pin_from_irq(cfg, ioapic, pin);
-	WARN_ON(!list_empty(&cfg->irq_2_pin));
-	arch_teardown_hwirq(virq);
+	return 0;
 }
 
-int mp_set_gsi_attr(u32 gsi, int trigger, int polarity, int node)
+void mp_irqdomain_free(struct irq_domain *domain, unsigned int virq,
+		       unsigned int nr_irqs)
 {
-	int ret = 0;
-	int ioapic, pin;
-	struct mp_pin_info *info;
+	struct irq_data *irq_data;
+	struct mp_chip_data *data;
 
-	ioapic = mp_find_ioapic(gsi);
-	if (ioapic < 0)
-		return -ENODEV;
-
-	pin = mp_find_ioapic_pin(ioapic, gsi);
-	info = mp_pin_info(ioapic, pin);
-	trigger = trigger ? 1 : 0;
-	polarity = polarity ? 1 : 0;
-
-	mutex_lock(&ioapic_mutex);
-	if (!info->set) {
-		info->trigger = trigger;
-		info->polarity = polarity;
-		info->node = node;
-		info->set = 1;
-	} else if (info->trigger != trigger || info->polarity != polarity) {
-		ret = -EBUSY;
+	BUG_ON(nr_irqs != 1);
+	irq_data = irq_domain_get_irq_data(domain, virq);
+	if (irq_data && irq_data->chip_data) {
+		data = irq_data->chip_data;
+		__remove_pin_from_irq(data, mp_irqdomain_ioapic_idx(domain),
+				      (int)irq_data->hwirq);
+		WARN_ON(!list_empty(&data->irq_2_pin));
+		kfree(irq_data->chip_data);
 	}
-	mutex_unlock(&ioapic_mutex);
-
-	return ret;
+	irq_domain_free_irqs_top(domain, virq, nr_irqs);
 }
 
-/* Enable IOAPIC early just for system timer */
-void __init pre_init_apic_IRQ0(void)
+void mp_irqdomain_activate(struct irq_domain *domain,
+			   struct irq_data *irq_data)
 {
-	struct io_apic_irq_attr attr = { 0, 0, 0, 0 };
+	unsigned long flags;
+	struct irq_pin_list *entry;
+	struct mp_chip_data *data = irq_data->chip_data;
 
-	printk(KERN_INFO "Early APIC setup for system timer0\n");
-#ifndef CONFIG_SMP
-	physid_set_mask_of_physid(boot_cpu_physical_apicid,
-					 &phys_cpu_present_map);
-#endif
-	setup_local_APIC();
+	raw_spin_lock_irqsave(&ioapic_lock, flags);
+	for_each_irq_pin(entry, data->irq_2_pin)
+		__ioapic_write_entry(entry->apic, entry->pin, data->entry);
+	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
 
-	io_apic_setup_irq_pin(0, 0, &attr);
-	irq_set_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq,
-				      "edge");
+void mp_irqdomain_deactivate(struct irq_domain *domain,
+			     struct irq_data *irq_data)
+{
+	/* It won't be called for IRQ with multiple IOAPIC pins associated */
+	ioapic_mask_entry(mp_irqdomain_ioapic_idx(domain),
+			  (int)irq_data->hwirq);
+}
+
+int mp_irqdomain_ioapic_idx(struct irq_domain *domain)
+{
+	return (int)(long)domain->host_data;
 }
+
+const struct irq_domain_ops mp_ioapic_irqdomain_ops = {
+	.alloc		= mp_irqdomain_alloc,
+	.free		= mp_irqdomain_free,
+	.activate	= mp_irqdomain_activate,
+	.deactivate	= mp_irqdomain_deactivate,
+};

arch/x86/kernel/apic/msi.c

@@ -3,6 +3,8 @@
  *
  * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo
  *	Moved from arch/x86/kernel/apic/io_apic.c.
+ * Jiang Liu <jiang.liu@linux.intel.com>
+ *	Convert to hierarchical irqdomain
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -14,22 +16,23 @@
 #include <linux/dmar.h>
 #include <linux/hpet.h>
 #include <linux/msi.h>
+#include <asm/irqdomain.h>
 #include <asm/msidef.h>
 #include <asm/hpet.h>
 #include <asm/hw_irq.h>
 #include <asm/apic.h>
 #include <asm/irq_remapping.h>
 
-void native_compose_msi_msg(struct pci_dev *pdev,
-			    unsigned int irq, unsigned int dest,
-			    struct msi_msg *msg, u8 hpet_id)
+static struct irq_domain *msi_default_domain;
+
+static void irq_msi_compose_msg(struct irq_data *data, struct msi_msg *msg)
 {
-	struct irq_cfg *cfg = irq_cfg(irq);
+	struct irq_cfg *cfg = irqd_cfg(data);
 
 	msg->address_hi = MSI_ADDR_BASE_HI;
 
 	if (x2apic_enabled())
-		msg->address_hi |= MSI_ADDR_EXT_DEST_ID(dest);
+		msg->address_hi |= MSI_ADDR_EXT_DEST_ID(cfg->dest_apicid);
 
 	msg->address_lo =
 		MSI_ADDR_BASE_LO |
@@ -39,7 +42,7 @@ void native_compose_msi_msg(struct pci_dev *pdev,
 		((apic->irq_delivery_mode != dest_LowestPrio) ?
 			MSI_ADDR_REDIRECTION_CPU :
 			MSI_ADDR_REDIRECTION_LOWPRI) |
-		MSI_ADDR_DEST_ID(dest);
+		MSI_ADDR_DEST_ID(cfg->dest_apicid);
 
 	msg->data =
 		MSI_DATA_TRIGGER_EDGE |
@@ -50,180 +53,201 @@ void native_compose_msi_msg(struct pci_dev *pdev,
 		MSI_DATA_VECTOR(cfg->vector);
 }
 
-static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq,
-			   struct msi_msg *msg, u8 hpet_id)
+/*
+ * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices,
+ * which implement the MSI or MSI-X Capability Structure.
+ */
+static struct irq_chip pci_msi_controller = {
+	.name			= "PCI-MSI",
+	.irq_unmask		= pci_msi_unmask_irq,
+	.irq_mask		= pci_msi_mask_irq,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_compose_msi_msg	= irq_msi_compose_msg,
+	.flags			= IRQCHIP_SKIP_SET_WAKE,
+};
+
+int native_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 {
-	struct irq_cfg *cfg;
-	int err;
-	unsigned dest;
+	struct irq_domain *domain;
+	struct irq_alloc_info info;
 
-	if (disable_apic)
-		return -ENXIO;
+	init_irq_alloc_info(&info, NULL);
+	info.type = X86_IRQ_ALLOC_TYPE_MSI;
+	info.msi_dev = dev;
 
-	cfg = irq_cfg(irq);
-	err = assign_irq_vector(irq, cfg, apic->target_cpus());
-	if (err)
-		return err;
+	domain = irq_remapping_get_irq_domain(&info);
+	if (domain == NULL)
+		domain = msi_default_domain;
+	if (domain == NULL)
+		return -ENOSYS;
 
-	err = apic->cpu_mask_to_apicid_and(cfg->domain,
-					   apic->target_cpus(), &dest);
-	if (err)
-		return err;
+	return pci_msi_domain_alloc_irqs(domain, dev, nvec, type);
+}
 
-	x86_msi.compose_msi_msg(pdev, irq, dest, msg, hpet_id);
+void native_teardown_msi_irq(unsigned int irq)
+{
+	irq_domain_free_irqs(irq, 1);
+}
 
-	return 0;
+static irq_hw_number_t pci_msi_get_hwirq(struct msi_domain_info *info,
+					 msi_alloc_info_t *arg)
+{
+	return arg->msi_hwirq;
 }
 
-static int
-msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
+static int pci_msi_prepare(struct irq_domain *domain, struct device *dev,
+			   int nvec, msi_alloc_info_t *arg)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	struct msi_msg msg;
-	unsigned int dest;
-	int ret;
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct msi_desc *desc = first_pci_msi_entry(pdev);
+
+	init_irq_alloc_info(arg, NULL);
+	arg->msi_dev = pdev;
+	if (desc->msi_attrib.is_msix) {
+		arg->type = X86_IRQ_ALLOC_TYPE_MSIX;
+	} else {
+		arg->type = X86_IRQ_ALLOC_TYPE_MSI;
+		arg->flags |= X86_IRQ_ALLOC_CONTIGUOUS_VECTORS;
+	}
 
-	ret = apic_set_affinity(data, mask, &dest);
-	if (ret)
-		return ret;
+	return 0;
+}
 
-	__get_cached_msi_msg(data->msi_desc, &msg);
+static void pci_msi_set_desc(msi_alloc_info_t *arg, struct msi_desc *desc)
+{
+	arg->msi_hwirq = pci_msi_domain_calc_hwirq(arg->msi_dev, desc);
+}
+
+static struct msi_domain_ops pci_msi_domain_ops = {
+	.get_hwirq	= pci_msi_get_hwirq,
+	.msi_prepare	= pci_msi_prepare,
+	.set_desc	= pci_msi_set_desc,
+};
 
-	msg.data &= ~MSI_DATA_VECTOR_MASK;
-	msg.data |= MSI_DATA_VECTOR(cfg->vector);
-	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
-	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
+static struct msi_domain_info pci_msi_domain_info = {
+	.flags		= MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
+			  MSI_FLAG_PCI_MSIX,
+	.ops		= &pci_msi_domain_ops,
+	.chip		= &pci_msi_controller,
+	.handler	= handle_edge_irq,
+	.handler_name	= "edge",
+};
 
-	__pci_write_msi_msg(data->msi_desc, &msg);
+void arch_init_msi_domain(struct irq_domain *parent)
+{
+	if (disable_apic)
+		return;
 
-	return IRQ_SET_MASK_OK_NOCOPY;
+	msi_default_domain = pci_msi_create_irq_domain(NULL,
+					&pci_msi_domain_info, parent);
+	if (!msi_default_domain)
+		pr_warn("failed to initialize irqdomain for MSI/MSI-x.\n");
 }
 
-/*
- * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices,
- * which implement the MSI or MSI-X Capability Structure.
- */
-static struct irq_chip msi_chip = {
-	.name			= "PCI-MSI",
+#ifdef CONFIG_IRQ_REMAP
+static struct irq_chip pci_msi_ir_controller = {
+	.name			= "IR-PCI-MSI",
 	.irq_unmask		= pci_msi_unmask_irq,
 	.irq_mask		= pci_msi_mask_irq,
-	.irq_ack		= apic_ack_edge,
-	.irq_set_affinity	= msi_set_affinity,
-	.irq_retrigger		= apic_retrigger_irq,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_vcpu_affinity	= irq_chip_set_vcpu_affinity_parent,
 	.flags			= IRQCHIP_SKIP_SET_WAKE,
 };
 
-int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc,
-		  unsigned int irq_base, unsigned int irq_offset)
-{
-	struct irq_chip *chip = &msi_chip;
-	struct msi_msg msg;
-	unsigned int irq = irq_base + irq_offset;
-	int ret;
-
-	ret = msi_compose_msg(dev, irq, &msg, -1);
-	if (ret < 0)
-		return ret;
-
-	irq_set_msi_desc_off(irq_base, irq_offset, msidesc);
-
-	/*
-	 * MSI-X message is written per-IRQ, the offset is always 0.
-	 * MSI message denotes a contiguous group of IRQs, written for 0th IRQ.
-	 */
-	if (!irq_offset)
-		pci_write_msi_msg(irq, &msg);
+static struct msi_domain_info pci_msi_ir_domain_info = {
+	.flags		= MSI_FLAG_USE_DEF_DOM_OPS | MSI_FLAG_USE_DEF_CHIP_OPS |
+			  MSI_FLAG_MULTI_PCI_MSI | MSI_FLAG_PCI_MSIX,
+	.ops		= &pci_msi_domain_ops,
+	.chip		= &pci_msi_ir_controller,
+	.handler	= handle_edge_irq,
+	.handler_name	= "edge",
+};
 
-	setup_remapped_irq(irq, irq_cfg(irq), chip);
+struct irq_domain *arch_create_msi_irq_domain(struct irq_domain *parent)
+{
+	return pci_msi_create_irq_domain(NULL, &pci_msi_ir_domain_info, parent);
+}
+#endif
 
-	irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
+#ifdef CONFIG_DMAR_TABLE
+static void dmar_msi_write_msg(struct irq_data *data, struct msi_msg *msg)
+{
+	dmar_msi_write(data->irq, msg);
+}
 
-	dev_dbg(&dev->dev, "irq %d for MSI/MSI-X\n", irq);
+static struct irq_chip dmar_msi_controller = {
+	.name			= "DMAR-MSI",
+	.irq_unmask		= dmar_msi_unmask,
+	.irq_mask		= dmar_msi_mask,
+	.irq_ack		= irq_chip_ack_parent,
+	.irq_set_affinity	= msi_domain_set_affinity,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_compose_msi_msg	= irq_msi_compose_msg,
+	.irq_write_msi_msg	= dmar_msi_write_msg,
+	.flags			= IRQCHIP_SKIP_SET_WAKE,
+};
 
-	return 0;
+static irq_hw_number_t dmar_msi_get_hwirq(struct msi_domain_info *info,
+					  msi_alloc_info_t *arg)
+{
+	return arg->dmar_id;
 }
 
-int native_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
+static int dmar_msi_init(struct irq_domain *domain,
+			 struct msi_domain_info *info, unsigned int virq,
+			 irq_hw_number_t hwirq, msi_alloc_info_t *arg)
 {
-	struct msi_desc *msidesc;
-	unsigned int irq;
-	int node, ret;
+	irq_domain_set_info(domain, virq, arg->dmar_id, info->chip, NULL,
+			    handle_edge_irq, arg->dmar_data, "edge");
 
-	/* Multiple MSI vectors only supported with interrupt remapping */
-	if (type == PCI_CAP_ID_MSI && nvec > 1)
-		return 1;
+	return 0;
+}
 
-	node = dev_to_node(&dev->dev);
+static struct msi_domain_ops dmar_msi_domain_ops = {
+	.get_hwirq	= dmar_msi_get_hwirq,
+	.msi_init	= dmar_msi_init,
+};
 
-	list_for_each_entry(msidesc, &dev->msi_list, list) {
-		irq = irq_alloc_hwirq(node);
-		if (!irq)
-			return -ENOSPC;
+static struct msi_domain_info dmar_msi_domain_info = {
+	.ops		= &dmar_msi_domain_ops,
+	.chip		= &dmar_msi_controller,
+};
 
-		ret = setup_msi_irq(dev, msidesc, irq, 0);
-		if (ret < 0) {
-			irq_free_hwirq(irq);
-			return ret;
-		}
+static struct irq_domain *dmar_get_irq_domain(void)
+{
+	static struct irq_domain *dmar_domain;
+	static DEFINE_MUTEX(dmar_lock);
 
-	}
-	return 0;
-}
+	mutex_lock(&dmar_lock);
+	if (dmar_domain == NULL)
+		dmar_domain = msi_create_irq_domain(NULL, &dmar_msi_domain_info,
+						    x86_vector_domain);
+	mutex_unlock(&dmar_lock);
 
-void native_teardown_msi_irq(unsigned int irq)
-{
-	irq_free_hwirq(irq);
+	return dmar_domain;
 }
 
-#ifdef CONFIG_DMAR_TABLE
-static int
-dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
-		      bool force)
+int dmar_alloc_hwirq(int id, int node, void *arg)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	unsigned int dest, irq = data->irq;
-	struct msi_msg msg;
-	int ret;
-
-	ret = apic_set_affinity(data, mask, &dest);
-	if (ret)
-		return ret;
+	struct irq_domain *domain = dmar_get_irq_domain();
+	struct irq_alloc_info info;
 
-	dmar_msi_read(irq, &msg);
+	if (!domain)
+		return -1;
 
-	msg.data &= ~MSI_DATA_VECTOR_MASK;
-	msg.data |= MSI_DATA_VECTOR(cfg->vector);
-	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
-	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
-	msg.address_hi = MSI_ADDR_BASE_HI | MSI_ADDR_EXT_DEST_ID(dest);
+	init_irq_alloc_info(&info, NULL);
+	info.type = X86_IRQ_ALLOC_TYPE_DMAR;
+	info.dmar_id = id;
+	info.dmar_data = arg;
 
-	dmar_msi_write(irq, &msg);
-
-	return IRQ_SET_MASK_OK_NOCOPY;
+	return irq_domain_alloc_irqs(domain, 1, node, &info);
 }
 
-static struct irq_chip dmar_msi_type = {
-	.name			= "DMAR_MSI",
-	.irq_unmask		= dmar_msi_unmask,
-	.irq_mask		= dmar_msi_mask,
-	.irq_ack		= apic_ack_edge,
-	.irq_set_affinity	= dmar_msi_set_affinity,
-	.irq_retrigger		= apic_retrigger_irq,
-	.flags			= IRQCHIP_SKIP_SET_WAKE,
-};
-
-int arch_setup_dmar_msi(unsigned int irq)
+void dmar_free_hwirq(int irq)
 {
-	int ret;
-	struct msi_msg msg;
-
-	ret = msi_compose_msg(NULL, irq, &msg, -1);
-	if (ret < 0)
-		return ret;
-	dmar_msi_write(irq, &msg);
-	irq_set_chip_and_handler_name(irq, &dmar_msi_type, handle_edge_irq,
-				      "edge");
-	return 0;
+	irq_domain_free_irqs(irq, 1);
 }
 #endif
 
@@ -231,56 +255,103 @@ int arch_setup_dmar_msi(unsigned int irq)
  * MSI message composition
  */
 #ifdef CONFIG_HPET_TIMER
+static inline int hpet_dev_id(struct irq_domain *domain)
+{
+	struct msi_domain_info *info = msi_get_domain_info(domain);
+
+	return (int)(long)info->data;
+}
 
-static int hpet_msi_set_affinity(struct irq_data *data,
-				 const struct cpumask *mask, bool force)
+static void hpet_msi_write_msg(struct irq_data *data, struct msi_msg *msg)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	struct msi_msg msg;
-	unsigned int dest;
-	int ret;
+	hpet_msi_write(data->handler_data, msg);
+}
 
-	ret = apic_set_affinity(data, mask, &dest);
-	if (ret)
-		return ret;
+static struct irq_chip hpet_msi_controller = {
+	.name = "HPET-MSI",
+	.irq_unmask = hpet_msi_unmask,
+	.irq_mask = hpet_msi_mask,
+	.irq_ack = irq_chip_ack_parent,
+	.irq_set_affinity = msi_domain_set_affinity,
+	.irq_retrigger = irq_chip_retrigger_hierarchy,
+	.irq_compose_msi_msg = irq_msi_compose_msg,
+	.irq_write_msi_msg = hpet_msi_write_msg,
+	.flags = IRQCHIP_SKIP_SET_WAKE,
+};
 
-	hpet_msi_read(data->handler_data, &msg);
+static irq_hw_number_t hpet_msi_get_hwirq(struct msi_domain_info *info,
+					  msi_alloc_info_t *arg)
+{
+	return arg->hpet_index;
+}
 
-	msg.data &= ~MSI_DATA_VECTOR_MASK;
-	msg.data |= MSI_DATA_VECTOR(cfg->vector);
-	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
-	msg.address_lo |= MSI_ADDR_DEST_ID(dest);
+static int hpet_msi_init(struct irq_domain *domain,
+			 struct msi_domain_info *info, unsigned int virq,
+			 irq_hw_number_t hwirq, msi_alloc_info_t *arg)
+{
+	irq_set_status_flags(virq, IRQ_MOVE_PCNTXT);
+	irq_domain_set_info(domain, virq, arg->hpet_index, info->chip, NULL,
+			    handle_edge_irq, arg->hpet_data, "edge");
 
-	hpet_msi_write(data->handler_data, &msg);
+	return 0;
+}
 
-	return IRQ_SET_MASK_OK_NOCOPY;
+static void hpet_msi_free(struct irq_domain *domain,
+			  struct msi_domain_info *info, unsigned int virq)
+{
+	irq_clear_status_flags(virq, IRQ_MOVE_PCNTXT);
 }
 
-static struct irq_chip hpet_msi_type = {
-	.name = "HPET_MSI",
-	.irq_unmask = hpet_msi_unmask,
-	.irq_mask = hpet_msi_mask,
-	.irq_ack = apic_ack_edge,
-	.irq_set_affinity = hpet_msi_set_affinity,
-	.irq_retrigger = apic_retrigger_irq,
-	.flags = IRQCHIP_SKIP_SET_WAKE,
+static struct msi_domain_ops hpet_msi_domain_ops = {
+	.get_hwirq	= hpet_msi_get_hwirq,
+	.msi_init	= hpet_msi_init,
+	.msi_free	= hpet_msi_free,
+};
+
+static struct msi_domain_info hpet_msi_domain_info = {
+	.ops		= &hpet_msi_domain_ops,
+	.chip		= &hpet_msi_controller,
 };
 
-int default_setup_hpet_msi(unsigned int irq, unsigned int id)
+struct irq_domain *hpet_create_irq_domain(int hpet_id)
 {
-	struct irq_chip *chip = &hpet_msi_type;
-	struct msi_msg msg;
-	int ret;
+	struct irq_domain *parent;
+	struct irq_alloc_info info;
+	struct msi_domain_info *domain_info;
+
+	if (x86_vector_domain == NULL)
+		return NULL;
+
+	domain_info = kzalloc(sizeof(*domain_info), GFP_KERNEL);
+	if (!domain_info)
+		return NULL;
+
+	*domain_info = hpet_msi_domain_info;
+	domain_info->data = (void *)(long)hpet_id;
+
+	init_irq_alloc_info(&info, NULL);
+	info.type = X86_IRQ_ALLOC_TYPE_HPET;
+	info.hpet_id = hpet_id;
+	parent = irq_remapping_get_ir_irq_domain(&info);
+	if (parent == NULL)
+		parent = x86_vector_domain;
+	else
+		hpet_msi_controller.name = "IR-HPET-MSI";
+
+	return msi_create_irq_domain(NULL, domain_info, parent);
+}
 
-	ret = msi_compose_msg(NULL, irq, &msg, id);
-	if (ret < 0)
-		return ret;
+int hpet_assign_irq(struct irq_domain *domain, struct hpet_dev *dev,
+		    int dev_num)
+{
+	struct irq_alloc_info info;
 
-	hpet_msi_write(irq_get_handler_data(irq), &msg);
-	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-	setup_remapped_irq(irq, irq_cfg(irq), chip);
+	init_irq_alloc_info(&info, NULL);
+	info.type = X86_IRQ_ALLOC_TYPE_HPET;
+	info.hpet_data = dev;
+	info.hpet_id = hpet_dev_id(domain);
+	info.hpet_index = dev_num;
 
-	irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
-	return 0;
+	return irq_domain_alloc_irqs(domain, 1, NUMA_NO_NODE, &info);
 }
 #endif

arch/x86/kernel/apic/vector.c

@@ -3,6 +3,8 @@
  *
  * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo
  *	Moved from arch/x86/kernel/apic/io_apic.c.
+ * Jiang Liu <jiang.liu@linux.intel.com>
+ *	Enable support of hierarchical irqdomains
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -11,15 +13,28 @@
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/compiler.h>
-#include <linux/irqdomain.h>
 #include <linux/slab.h>
+#include <asm/irqdomain.h>
 #include <asm/hw_irq.h>
 #include <asm/apic.h>
 #include <asm/i8259.h>
 #include <asm/desc.h>
 #include <asm/irq_remapping.h>
 
+struct apic_chip_data {
+	struct irq_cfg		cfg;
+	cpumask_var_t		domain;
+	cpumask_var_t		old_domain;
+	u8			move_in_progress : 1;
+};
+
+struct irq_domain *x86_vector_domain;
 static DEFINE_RAW_SPINLOCK(vector_lock);
+static cpumask_var_t vector_cpumask;
+static struct irq_chip lapic_controller;
+#ifdef	CONFIG_X86_IO_APIC
+static struct apic_chip_data *legacy_irq_data[NR_IRQS_LEGACY];
+#endif
 
 void lock_vector_lock(void)
 {
@@ -34,71 +49,59 @@ void unlock_vector_lock(void)
 	raw_spin_unlock(&vector_lock);
 }
 
-struct irq_cfg *irq_cfg(unsigned int irq)
+static struct apic_chip_data *apic_chip_data(struct irq_data *irq_data)
 {
-	return irq_get_chip_data(irq);
+	if (!irq_data)
+		return NULL;
+
+	while (irq_data->parent_data)
+		irq_data = irq_data->parent_data;
+
+	return irq_data->chip_data;
 }
 
 struct irq_cfg *irqd_cfg(struct irq_data *irq_data)
 {
-	return irq_data->chip_data;
+	struct apic_chip_data *data = apic_chip_data(irq_data);
+
+	return data ? &data->cfg : NULL;
 }
 
-static struct irq_cfg *alloc_irq_cfg(unsigned int irq, int node)
+struct irq_cfg *irq_cfg(unsigned int irq)
 {
-	struct irq_cfg *cfg;
+	return irqd_cfg(irq_get_irq_data(irq));
+}
 
-	cfg = kzalloc_node(sizeof(*cfg), GFP_KERNEL, node);
-	if (!cfg)
+static struct apic_chip_data *alloc_apic_chip_data(int node)
+{
+	struct apic_chip_data *data;
+
+	data = kzalloc_node(sizeof(*data), GFP_KERNEL, node);
+	if (!data)
 		return NULL;
-	if (!zalloc_cpumask_var_node(&cfg->domain, GFP_KERNEL, node))
-		goto out_cfg;
-	if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_KERNEL, node))
+	if (!zalloc_cpumask_var_node(&data->domain, GFP_KERNEL, node))
+		goto out_data;
+	if (!zalloc_cpumask_var_node(&data->old_domain, GFP_KERNEL, node))
 		goto out_domain;
-#ifdef	CONFIG_X86_IO_APIC
-	INIT_LIST_HEAD(&cfg->irq_2_pin);
-#endif
-	return cfg;
+	return data;
 out_domain:
-	free_cpumask_var(cfg->domain);
-out_cfg:
-	kfree(cfg);
+	free_cpumask_var(data->domain);
+out_data:
+	kfree(data);
 	return NULL;
 }
 
-struct irq_cfg *alloc_irq_and_cfg_at(unsigned int at, int node)
+static void free_apic_chip_data(struct apic_chip_data *data)
 {
-	int res = irq_alloc_desc_at(at, node);
-	struct irq_cfg *cfg;
-
-	if (res < 0) {
-		if (res != -EEXIST)
-			return NULL;
-		cfg = irq_cfg(at);
-		if (cfg)
-			return cfg;
+	if (data) {
+		free_cpumask_var(data->domain);
+		free_cpumask_var(data->old_domain);
+		kfree(data);
 	}
-
-	cfg = alloc_irq_cfg(at, node);
-	if (cfg)
-		irq_set_chip_data(at, cfg);
-	else
-		irq_free_desc(at);
-	return cfg;
-}
-
-static void free_irq_cfg(unsigned int at, struct irq_cfg *cfg)
-{
-	if (!cfg)
-		return;
-	irq_set_chip_data(at, NULL);
-	free_cpumask_var(cfg->domain);
-	free_cpumask_var(cfg->old_domain);
-	kfree(cfg);
 }
 
-static int
-__assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
+static int __assign_irq_vector(int irq, struct apic_chip_data *d,
+			       const struct cpumask *mask)
 {
 	/*
 	 * NOTE! The local APIC isn't very good at handling
@@ -114,36 +117,33 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 	static int current_vector = FIRST_EXTERNAL_VECTOR + VECTOR_OFFSET_START;
 	static int current_offset = VECTOR_OFFSET_START % 16;
 	int cpu, err;
-	cpumask_var_t tmp_mask;
 
-	if (cfg->move_in_progress)
+	if (d->move_in_progress)
 		return -EBUSY;
 
-	if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC))
-		return -ENOMEM;
-
 	/* Only try and allocate irqs on cpus that are present */
 	err = -ENOSPC;
-	cpumask_clear(cfg->old_domain);
+	cpumask_clear(d->old_domain);
 	cpu = cpumask_first_and(mask, cpu_online_mask);
 	while (cpu < nr_cpu_ids) {
 		int new_cpu, vector, offset;
 
-		apic->vector_allocation_domain(cpu, tmp_mask, mask);
+		apic->vector_allocation_domain(cpu, vector_cpumask, mask);
 
-		if (cpumask_subset(tmp_mask, cfg->domain)) {
+		if (cpumask_subset(vector_cpumask, d->domain)) {
 			err = 0;
-			if (cpumask_equal(tmp_mask, cfg->domain))
+			if (cpumask_equal(vector_cpumask, d->domain))
 				break;
 			/*
 			 * New cpumask using the vector is a proper subset of
 			 * the current in use mask. So cleanup the vector
 			 * allocation for the members that are not used anymore.
 			 */
-			cpumask_andnot(cfg->old_domain, cfg->domain, tmp_mask);
-			cfg->move_in_progress =
-			   cpumask_intersects(cfg->old_domain, cpu_online_mask);
-			cpumask_and(cfg->domain, cfg->domain, tmp_mask);
+			cpumask_andnot(d->old_domain, d->domain,
+				       vector_cpumask);
+			d->move_in_progress =
+			   cpumask_intersects(d->old_domain, cpu_online_mask);
+			cpumask_and(d->domain, d->domain, vector_cpumask);
 			break;
 		}
 
@@ -157,16 +157,18 @@ next:
 		}
 
 		if (unlikely(current_vector == vector)) {
-			cpumask_or(cfg->old_domain, cfg->old_domain, tmp_mask);
-			cpumask_andnot(tmp_mask, mask, cfg->old_domain);
-			cpu = cpumask_first_and(tmp_mask, cpu_online_mask);
+			cpumask_or(d->old_domain, d->old_domain,
+				   vector_cpumask);
+			cpumask_andnot(vector_cpumask, mask, d->old_domain);
+			cpu = cpumask_first_and(vector_cpumask,
+						cpu_online_mask);
 			continue;
 		}
 
 		if (test_bit(vector, used_vectors))
 			goto next;
 
-		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask) {
+		for_each_cpu_and(new_cpu, vector_cpumask, cpu_online_mask) {
 			if (per_cpu(vector_irq, new_cpu)[vector] >
 			    VECTOR_UNDEFINED)
 				goto next;
@@ -174,55 +176,73 @@ next:
 		/* Found one! */
 		current_vector = vector;
 		current_offset = offset;
-		if (cfg->vector) {
-			cpumask_copy(cfg->old_domain, cfg->domain);
-			cfg->move_in_progress =
-			   cpumask_intersects(cfg->old_domain, cpu_online_mask);
+		if (d->cfg.vector) {
+			cpumask_copy(d->old_domain, d->domain);
+			d->move_in_progress =
+			   cpumask_intersects(d->old_domain, cpu_online_mask);
 		}
-		for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask)
+		for_each_cpu_and(new_cpu, vector_cpumask, cpu_online_mask)
 			per_cpu(vector_irq, new_cpu)[vector] = irq;
-		cfg->vector = vector;
-		cpumask_copy(cfg->domain, tmp_mask);
+		d->cfg.vector = vector;
+		cpumask_copy(d->domain, vector_cpumask);
 		err = 0;
 		break;
 	}
-	free_cpumask_var(tmp_mask);
+
+	if (!err) {
+		/* cache destination APIC IDs into cfg->dest_apicid */
+		err = apic->cpu_mask_to_apicid_and(mask, d->domain,
+						   &d->cfg.dest_apicid);
+	}
 
 	return err;
 }
 
-int assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
+static int assign_irq_vector(int irq, struct apic_chip_data *data,
+			     const struct cpumask *mask)
 {
 	int err;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&vector_lock, flags);
-	err = __assign_irq_vector(irq, cfg, mask);
+	err = __assign_irq_vector(irq, data, mask);
 	raw_spin_unlock_irqrestore(&vector_lock, flags);
 	return err;
 }
 
-void clear_irq_vector(int irq, struct irq_cfg *cfg)
+static int assign_irq_vector_policy(int irq, int node,
+				    struct apic_chip_data *data,
+				    struct irq_alloc_info *info)
+{
+	if (info && info->mask)
+		return assign_irq_vector(irq, data, info->mask);
+	if (node != NUMA_NO_NODE &&
+	    assign_irq_vector(irq, data, cpumask_of_node(node)) == 0)
+		return 0;
+	return assign_irq_vector(irq, data, apic->target_cpus());
+}
+
+static void clear_irq_vector(int irq, struct apic_chip_data *data)
 {
 	int cpu, vector;
 	unsigned long flags;
 
 	raw_spin_lock_irqsave(&vector_lock, flags);
-	BUG_ON(!cfg->vector);
+	BUG_ON(!data->cfg.vector);
 
-	vector = cfg->vector;
-	for_each_cpu_and(cpu, cfg->domain, cpu_online_mask)
+	vector = data->cfg.vector;
+	for_each_cpu_and(cpu, data->domain, cpu_online_mask)
 		per_cpu(vector_irq, cpu)[vector] = VECTOR_UNDEFINED;
 
-	cfg->vector = 0;
-	cpumask_clear(cfg->domain);
+	data->cfg.vector = 0;
+	cpumask_clear(data->domain);
 
-	if (likely(!cfg->move_in_progress)) {
+	if (likely(!data->move_in_progress)) {
 		raw_spin_unlock_irqrestore(&vector_lock, flags);
 		return;
 	}
 
-	for_each_cpu_and(cpu, cfg->old_domain, cpu_online_mask) {
+	for_each_cpu_and(cpu, data->old_domain, cpu_online_mask) {
 		for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS;
 		     vector++) {
 			if (per_cpu(vector_irq, cpu)[vector] != irq)
@@ -231,10 +251,95 @@ void clear_irq_vector(int irq, struct irq_cfg *cfg)
 			break;
 		}
 	}
-	cfg->move_in_progress = 0;
+	data->move_in_progress = 0;
 	raw_spin_unlock_irqrestore(&vector_lock, flags);
 }
 
+void init_irq_alloc_info(struct irq_alloc_info *info,
+			 const struct cpumask *mask)
+{
+	memset(info, 0, sizeof(*info));
+	info->mask = mask;
+}
+
+void copy_irq_alloc_info(struct irq_alloc_info *dst, struct irq_alloc_info *src)
+{
+	if (src)
+		*dst = *src;
+	else
+		memset(dst, 0, sizeof(*dst));
+}
+
+static void x86_vector_free_irqs(struct irq_domain *domain,
+				 unsigned int virq, unsigned int nr_irqs)
+{
+	struct irq_data *irq_data;
+	int i;
+
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(x86_vector_domain, virq + i);
+		if (irq_data && irq_data->chip_data) {
+			clear_irq_vector(virq + i, irq_data->chip_data);
+			free_apic_chip_data(irq_data->chip_data);
+#ifdef	CONFIG_X86_IO_APIC
+			if (virq + i < nr_legacy_irqs())
+				legacy_irq_data[virq + i] = NULL;
+#endif
+			irq_domain_reset_irq_data(irq_data);
+		}
+	}
+}
+
+static int x86_vector_alloc_irqs(struct irq_domain *domain, unsigned int virq,
+				 unsigned int nr_irqs, void *arg)
+{
+	struct irq_alloc_info *info = arg;
+	struct apic_chip_data *data;
+	struct irq_data *irq_data;
+	int i, err;
+
+	if (disable_apic)
+		return -ENXIO;
+
+	/* Currently vector allocator can't guarantee contiguous allocations */
+	if ((info->flags & X86_IRQ_ALLOC_CONTIGUOUS_VECTORS) && nr_irqs > 1)
+		return -ENOSYS;
+
+	for (i = 0; i < nr_irqs; i++) {
+		irq_data = irq_domain_get_irq_data(domain, virq + i);
+		BUG_ON(!irq_data);
+#ifdef	CONFIG_X86_IO_APIC
+		if (virq + i < nr_legacy_irqs() && legacy_irq_data[virq + i])
+			data = legacy_irq_data[virq + i];
+		else
+#endif
+			data = alloc_apic_chip_data(irq_data->node);
+		if (!data) {
+			err = -ENOMEM;
+			goto error;
+		}
+
+		irq_data->chip = &lapic_controller;
+		irq_data->chip_data = data;
+		irq_data->hwirq = virq + i;
+		err = assign_irq_vector_policy(virq, irq_data->node, data,
+					       info);
+		if (err)
+			goto error;
+	}
+
+	return 0;
+
+error:
+	x86_vector_free_irqs(domain, virq, i + 1);
+	return err;
+}
+
+static const struct irq_domain_ops x86_vector_domain_ops = {
+	.alloc	= x86_vector_alloc_irqs,
+	.free	= x86_vector_free_irqs,
+};
+
 int __init arch_probe_nr_irqs(void)
 {
 	int nr;
@@ -258,8 +363,43 @@ int __init arch_probe_nr_irqs(void)
 	return nr_legacy_irqs();
 }
 
+#ifdef	CONFIG_X86_IO_APIC
+static void init_legacy_irqs(void)
+{
+	int i, node = cpu_to_node(0);
+	struct apic_chip_data *data;
+
+	/*
+	 * For legacy IRQ's, start with assigning irq0 to irq15 to
+	 * ISA_IRQ_VECTOR(i) for all cpu's.
+	 */
+	for (i = 0; i < nr_legacy_irqs(); i++) {
+		data = legacy_irq_data[i] = alloc_apic_chip_data(node);
+		BUG_ON(!data);
+
+		data->cfg.vector = ISA_IRQ_VECTOR(i);
+		cpumask_setall(data->domain);
+		irq_set_chip_data(i, data);
+	}
+}
+#else
+static void init_legacy_irqs(void) { }
+#endif
+
 int __init arch_early_irq_init(void)
 {
+	init_legacy_irqs();
+
+	x86_vector_domain = irq_domain_add_tree(NULL, &x86_vector_domain_ops,
+						NULL);
+	BUG_ON(x86_vector_domain == NULL);
+	irq_set_default_host(x86_vector_domain);
+
+	arch_init_msi_domain(x86_vector_domain);
+	arch_init_htirq_domain(x86_vector_domain);
+
+	BUG_ON(!alloc_cpumask_var(&vector_cpumask, GFP_KERNEL));
+
 	return arch_early_ioapic_init();
 }
 
@@ -267,7 +407,7 @@ static void __setup_vector_irq(int cpu)
 {
 	/* Initialize vector_irq on a new cpu */
 	int irq, vector;
-	struct irq_cfg *cfg;
+	struct apic_chip_data *data;
 
 	/*
 	 * vector_lock will make sure that we don't run into irq vector
@@ -277,13 +417,13 @@ static void __setup_vector_irq(int cpu)
 	raw_spin_lock(&vector_lock);
 	/* Mark the inuse vectors */
 	for_each_active_irq(irq) {
-		cfg = irq_cfg(irq);
-		if (!cfg)
+		data = apic_chip_data(irq_get_irq_data(irq));
+		if (!data)
 			continue;
 
-		if (!cpumask_test_cpu(cpu, cfg->domain))
+		if (!cpumask_test_cpu(cpu, data->domain))
 			continue;
-		vector = cfg->vector;
+		vector = data->cfg.vector;
 		per_cpu(vector_irq, cpu)[vector] = irq;
 	}
 	/* Mark the free vectors */
@@ -292,8 +432,8 @@ static void __setup_vector_irq(int cpu)
 		if (irq <= VECTOR_UNDEFINED)
 			continue;
 
-		cfg = irq_cfg(irq);
-		if (!cpumask_test_cpu(cpu, cfg->domain))
+		data = apic_chip_data(irq_get_irq_data(irq));
+		if (!cpumask_test_cpu(cpu, data->domain))
 			per_cpu(vector_irq, cpu)[vector] = VECTOR_UNDEFINED;
 	}
 	raw_spin_unlock(&vector_lock);
@@ -314,20 +454,20 @@ void setup_vector_irq(int cpu)
 	 * legacy vector to irq mapping:
 	 */
 	for (irq = 0; irq < nr_legacy_irqs(); irq++)
-		per_cpu(vector_irq, cpu)[IRQ0_VECTOR + irq] = irq;
+		per_cpu(vector_irq, cpu)[ISA_IRQ_VECTOR(irq)] = irq;
 
 	__setup_vector_irq(cpu);
 }
 
-int apic_retrigger_irq(struct irq_data *data)
+static int apic_retrigger_irq(struct irq_data *irq_data)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
+	struct apic_chip_data *data = apic_chip_data(irq_data);
 	unsigned long flags;
 	int cpu;
 
 	raw_spin_lock_irqsave(&vector_lock, flags);
-	cpu = cpumask_first_and(cfg->domain, cpu_online_mask);
-	apic->send_IPI_mask(cpumask_of(cpu), cfg->vector);
+	cpu = cpumask_first_and(data->domain, cpu_online_mask);
+	apic->send_IPI_mask(cpumask_of(cpu), data->cfg.vector);
 	raw_spin_unlock_irqrestore(&vector_lock, flags);
 
 	return 1;
@@ -340,73 +480,76 @@ void apic_ack_edge(struct irq_data *data)
 	ack_APIC_irq();
 }
 
-/*
- * Either sets data->affinity to a valid value, and returns
- * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and
- * leaves data->affinity untouched.
- */
-int apic_set_affinity(struct irq_data *data, const struct cpumask *mask,
-		      unsigned int *dest_id)
+static int apic_set_affinity(struct irq_data *irq_data,
+			     const struct cpumask *dest, bool force)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-	unsigned int irq = data->irq;
-	int err;
+	struct apic_chip_data *data = irq_data->chip_data;
+	int err, irq = irq_data->irq;
 
 	if (!config_enabled(CONFIG_SMP))
 		return -EPERM;
 
-	if (!cpumask_intersects(mask, cpu_online_mask))
+	if (!cpumask_intersects(dest, cpu_online_mask))
 		return -EINVAL;
 
-	err = assign_irq_vector(irq, cfg, mask);
-	if (err)
-		return err;
-
-	err = apic->cpu_mask_to_apicid_and(mask, cfg->domain, dest_id);
+	err = assign_irq_vector(irq, data, dest);
 	if (err) {
-		if (assign_irq_vector(irq, cfg, data->affinity))
+		struct irq_data *top = irq_get_irq_data(irq);
+
+		if (assign_irq_vector(irq, data, top->affinity))
 			pr_err("Failed to recover vector for irq %d\n", irq);
 		return err;
 	}
 
-	cpumask_copy(data->affinity, mask);
-
-	return 0;
+	return IRQ_SET_MASK_OK;
 }
 
+static struct irq_chip lapic_controller = {
+	.irq_ack		= apic_ack_edge,
+	.irq_set_affinity	= apic_set_affinity,
+	.irq_retrigger		= apic_retrigger_irq,
+};
+
 #ifdef CONFIG_SMP
-void send_cleanup_vector(struct irq_cfg *cfg)
+static void __send_cleanup_vector(struct apic_chip_data *data)
 {
 	cpumask_var_t cleanup_mask;
 
 	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
 		unsigned int i;
 
-		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
+		for_each_cpu_and(i, data->old_domain, cpu_online_mask)
 			apic->send_IPI_mask(cpumask_of(i),
 					    IRQ_MOVE_CLEANUP_VECTOR);
 	} else {
-		cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask);
+		cpumask_and(cleanup_mask, data->old_domain, cpu_online_mask);
 		apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
 		free_cpumask_var(cleanup_mask);
 	}
-	cfg->move_in_progress = 0;
+	data->move_in_progress = 0;
+}
+
+void send_cleanup_vector(struct irq_cfg *cfg)
+{
+	struct apic_chip_data *data;
+
+	data = container_of(cfg, struct apic_chip_data, cfg);
+	if (data->move_in_progress)
+		__send_cleanup_vector(data);
 }
 
 asmlinkage __visible void smp_irq_move_cleanup_interrupt(void)
 {
 	unsigned vector, me;
 
-	ack_APIC_irq();
-	irq_enter();
-	exit_idle();
+	entering_ack_irq();
 
 	me = smp_processor_id();
 	for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
 		int irq;
 		unsigned int irr;
 		struct irq_desc *desc;
-		struct irq_cfg *cfg;
+		struct apic_chip_data *data;
 
 		irq = __this_cpu_read(vector_irq[vector]);
 
@@ -417,8 +560,8 @@ asmlinkage __visible void smp_irq_move_cleanup_interrupt(void)
 		if (!desc)
 			continue;
 
-		cfg = irq_cfg(irq);
-		if (!cfg)
+		data = apic_chip_data(&desc->irq_data);
+		if (!data)
 			continue;
 
 		raw_spin_lock(&desc->lock);
@@ -427,10 +570,11 @@ asmlinkage __visible void smp_irq_move_cleanup_interrupt(void)
 		 * Check if the irq migration is in progress. If so, we
 		 * haven't received the cleanup request yet for this irq.
 		 */
-		if (cfg->move_in_progress)
+		if (data->move_in_progress)
 			goto unlock;
 
-		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
+		if (vector == data->cfg.vector &&
+		    cpumask_test_cpu(me, data->domain))
 			goto unlock;
 
 		irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
@@ -450,20 +594,21 @@ unlock:
 		raw_spin_unlock(&desc->lock);
 	}
 
-	irq_exit();
+	exiting_irq();
 }
 
 static void __irq_complete_move(struct irq_cfg *cfg, unsigned vector)
 {
 	unsigned me;
+	struct apic_chip_data *data;
 
-	if (likely(!cfg->move_in_progress))
+	data = container_of(cfg, struct apic_chip_data, cfg);
+	if (likely(!data->move_in_progress))
 		return;
 
 	me = smp_processor_id();
-
-	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
-		send_cleanup_vector(cfg);
+	if (vector == data->cfg.vector && cpumask_test_cpu(me, data->domain))
+		__send_cleanup_vector(data);
 }
 
 void irq_complete_move(struct irq_cfg *cfg)
@@ -475,46 +620,11 @@ void irq_force_complete_move(int irq)
 {
 	struct irq_cfg *cfg = irq_cfg(irq);
 
-	if (!cfg)
-		return;
-
-	__irq_complete_move(cfg, cfg->vector);
+	if (cfg)
+		__irq_complete_move(cfg, cfg->vector);
 }
 #endif
 
-/*
- * Dynamic irq allocate and deallocation. Should be replaced by irq domains!
- */
-int arch_setup_hwirq(unsigned int irq, int node)
-{
-	struct irq_cfg *cfg;
-	unsigned long flags;
-	int ret;
-
-	cfg = alloc_irq_cfg(irq, node);
-	if (!cfg)
-		return -ENOMEM;
-
-	raw_spin_lock_irqsave(&vector_lock, flags);
-	ret = __assign_irq_vector(irq, cfg, apic->target_cpus());
-	raw_spin_unlock_irqrestore(&vector_lock, flags);
-
-	if (!ret)
-		irq_set_chip_data(irq, cfg);
-	else
-		free_irq_cfg(irq, cfg);
-	return ret;
-}
-
-void arch_teardown_hwirq(unsigned int irq)
-{
-	struct irq_cfg *cfg = irq_cfg(irq);
-
-	free_remapped_irq(irq);
-	clear_irq_vector(irq, cfg);
-	free_irq_cfg(irq, cfg);
-}
-
 static void __init print_APIC_field(int base)
 {
 	int i;

arch/x86/kernel/apic/x2apic_phys.c

@@ -21,11 +21,13 @@ early_param("x2apic_phys", set_x2apic_phys_mode);
 
 static bool x2apic_fadt_phys(void)
 {
+#ifdef CONFIG_ACPI
 	if ((acpi_gbl_FADT.header.revision >= FADT2_REVISION_ID) &&
 		(acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) {
 		printk(KERN_DEBUG "System requires x2apic physical mode\n");
 		return true;
 	}
+#endif
 	return false;
 }
 

arch/x86/kernel/asm-offsets.c

@@ -41,6 +41,25 @@ void common(void) {
 	OFFSET(pbe_orig_address, pbe, orig_address);
 	OFFSET(pbe_next, pbe, next);
 
+#if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION)
+	BLANK();
+	OFFSET(IA32_SIGCONTEXT_ax, sigcontext_ia32, ax);
+	OFFSET(IA32_SIGCONTEXT_bx, sigcontext_ia32, bx);
+	OFFSET(IA32_SIGCONTEXT_cx, sigcontext_ia32, cx);
+	OFFSET(IA32_SIGCONTEXT_dx, sigcontext_ia32, dx);
+	OFFSET(IA32_SIGCONTEXT_si, sigcontext_ia32, si);
+	OFFSET(IA32_SIGCONTEXT_di, sigcontext_ia32, di);
+	OFFSET(IA32_SIGCONTEXT_bp, sigcontext_ia32, bp);
+	OFFSET(IA32_SIGCONTEXT_sp, sigcontext_ia32, sp);
+	OFFSET(IA32_SIGCONTEXT_ip, sigcontext_ia32, ip);
+
+	BLANK();
+	OFFSET(TI_sysenter_return, thread_info, sysenter_return);
+
+	BLANK();
+	OFFSET(IA32_RT_SIGFRAME_sigcontext, rt_sigframe_ia32, uc.uc_mcontext);
+#endif
+
 #ifdef CONFIG_PARAVIRT
 	BLANK();
 	OFFSET(PARAVIRT_enabled, pv_info, paravirt_enabled);
@@ -49,7 +68,9 @@ void common(void) {
 	OFFSET(PV_IRQ_irq_disable, pv_irq_ops, irq_disable);
 	OFFSET(PV_IRQ_irq_enable, pv_irq_ops, irq_enable);
 	OFFSET(PV_CPU_iret, pv_cpu_ops, iret);
+#ifdef CONFIG_X86_32
 	OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit);
+#endif
 	OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0);
 	OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2);
 #endif

arch/x86/kernel/asm-offsets_32.c

@@ -17,17 +17,6 @@ void foo(void);
 
 void foo(void)
 {
-	OFFSET(IA32_SIGCONTEXT_ax, sigcontext, ax);
-	OFFSET(IA32_SIGCONTEXT_bx, sigcontext, bx);
-	OFFSET(IA32_SIGCONTEXT_cx, sigcontext, cx);
-	OFFSET(IA32_SIGCONTEXT_dx, sigcontext, dx);
-	OFFSET(IA32_SIGCONTEXT_si, sigcontext, si);
-	OFFSET(IA32_SIGCONTEXT_di, sigcontext, di);
-	OFFSET(IA32_SIGCONTEXT_bp, sigcontext, bp);
-	OFFSET(IA32_SIGCONTEXT_sp, sigcontext, sp);
-	OFFSET(IA32_SIGCONTEXT_ip, sigcontext, ip);
-	BLANK();
-
 	OFFSET(CPUINFO_x86, cpuinfo_x86, x86);
 	OFFSET(CPUINFO_x86_vendor, cpuinfo_x86, x86_vendor);
 	OFFSET(CPUINFO_x86_model, cpuinfo_x86, x86_model);
@@ -37,10 +26,6 @@ void foo(void)
 	OFFSET(CPUINFO_x86_vendor_id, cpuinfo_x86, x86_vendor_id);
 	BLANK();
 
-	OFFSET(TI_sysenter_return, thread_info, sysenter_return);
-	OFFSET(TI_cpu, thread_info, cpu);
-	BLANK();
-
 	OFFSET(PT_EBX, pt_regs, bx);
 	OFFSET(PT_ECX, pt_regs, cx);
 	OFFSET(PT_EDX, pt_regs, dx);
@@ -60,9 +45,6 @@ void foo(void)
 	OFFSET(PT_OLDSS,  pt_regs, ss);
 	BLANK();
 
-	OFFSET(IA32_RT_SIGFRAME_sigcontext, rt_sigframe, uc.uc_mcontext);
-	BLANK();
-
 	OFFSET(saved_context_gdt_desc, saved_context, gdt_desc);
 	BLANK();
 

arch/x86/kernel/asm-offsets_64.c

@@ -29,27 +29,6 @@ int main(void)
 	BLANK();
 #endif
 
-#ifdef CONFIG_IA32_EMULATION
-	OFFSET(TI_sysenter_return, thread_info, sysenter_return);
-	BLANK();
-
-#define ENTRY(entry) OFFSET(IA32_SIGCONTEXT_ ## entry, sigcontext_ia32, entry)
-	ENTRY(ax);
-	ENTRY(bx);
-	ENTRY(cx);
-	ENTRY(dx);
-	ENTRY(si);
-	ENTRY(di);
-	ENTRY(bp);
-	ENTRY(sp);
-	ENTRY(ip);
-	BLANK();
-#undef ENTRY
-
-	OFFSET(IA32_RT_SIGFRAME_sigcontext, rt_sigframe_ia32, uc.uc_mcontext);
-	BLANK();
-#endif
-
 #define ENTRY(entry) OFFSET(pt_regs_ ## entry, pt_regs, entry)
 	ENTRY(bx);
 	ENTRY(cx);

arch/x86/kernel/cpu/common.c

@@ -1155,10 +1155,6 @@ static __init int setup_disablecpuid(char *arg)
 }
 __setup("clearcpuid=", setup_disablecpuid);
 
-DEFINE_PER_CPU(unsigned long, kernel_stack) =
-	(unsigned long)&init_thread_union + THREAD_SIZE;
-EXPORT_PER_CPU_SYMBOL(kernel_stack);
-
 #ifdef CONFIG_X86_64
 struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table };
 struct desc_ptr debug_idt_descr = { NR_VECTORS * 16 - 1,

arch/x86/kernel/cpu/mshyperv.c

@@ -39,14 +39,12 @@ void hyperv_vector_handler(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
-	irq_enter();
-	exit_idle();
-
+	entering_irq();
 	inc_irq_stat(irq_hv_callback_count);
 	if (vmbus_handler)
 		vmbus_handler();
 
-	irq_exit();
+	exiting_irq();
 	set_irq_regs(old_regs);
 }
 

arch/x86/kernel/cpu/mtrr/cleanup.c

@@ -98,7 +98,8 @@ x86_get_mtrr_mem_range(struct range *range, int nr_range,
 			continue;
 		base = range_state[i].base_pfn;
 		if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed &&
-		    (mtrr_state.enabled & 1)) {
+		    (mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED) &&
+		    (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) {
 			/* Var MTRR contains UC entry below 1M? Skip it: */
 			printk(BIOS_BUG_MSG, i);
 			if (base + size <= (1<<(20-PAGE_SHIFT)))

arch/x86/kernel/cpu/mtrr/generic.c

@@ -102,59 +102,76 @@ static int check_type_overlap(u8 *prev, u8 *curr)
 	return 0;
 }
 
-/*
- * Error/Semi-error returns:
- * 0xFF - when MTRR is not enabled
- * *repeat == 1 implies [start:end] spanned across MTRR range and type returned
- *		corresponds only to [start:*partial_end].
- *		Caller has to lookup again for [*partial_end:end].
+/**
+ * mtrr_type_lookup_fixed - look up memory type in MTRR fixed entries
+ *
+ * Return the MTRR fixed memory type of 'start'.
+ *
+ * MTRR fixed entries are divided into the following ways:
+ *  0x00000 - 0x7FFFF : This range is divided into eight 64KB sub-ranges
+ *  0x80000 - 0xBFFFF : This range is divided into sixteen 16KB sub-ranges
+ *  0xC0000 - 0xFFFFF : This range is divided into sixty-four 4KB sub-ranges
+ *
+ * Return Values:
+ * MTRR_TYPE_(type)  - Matched memory type
+ * MTRR_TYPE_INVALID - Unmatched
+ */
+static u8 mtrr_type_lookup_fixed(u64 start, u64 end)
+{
+	int idx;
+
+	if (start >= 0x100000)
+		return MTRR_TYPE_INVALID;
+
+	/* 0x0 - 0x7FFFF */
+	if (start < 0x80000) {
+		idx = 0;
+		idx += (start >> 16);
+		return mtrr_state.fixed_ranges[idx];
+	/* 0x80000 - 0xBFFFF */
+	} else if (start < 0xC0000) {
+		idx = 1 * 8;
+		idx += ((start - 0x80000) >> 14);
+		return mtrr_state.fixed_ranges[idx];
+	}
+
+	/* 0xC0000 - 0xFFFFF */
+	idx = 3 * 8;
+	idx += ((start - 0xC0000) >> 12);
+	return mtrr_state.fixed_ranges[idx];
+}
+
+/**
+ * mtrr_type_lookup_variable - look up memory type in MTRR variable entries
+ *
+ * Return Value:
+ * MTRR_TYPE_(type) - Matched memory type or default memory type (unmatched)
+ *
+ * Output Arguments:
+ * repeat - Set to 1 when [start:end] spanned across MTRR range and type
+ *	    returned corresponds only to [start:*partial_end].  Caller has
+ *	    to lookup again for [*partial_end:end].
+ *
+ * uniform - Set to 1 when an MTRR covers the region uniformly, i.e. the
+ *	     region is fully covered by a single MTRR entry or the default
+ *	     type.
  */
-static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat)
+static u8 mtrr_type_lookup_variable(u64 start, u64 end, u64 *partial_end,
+				    int *repeat, u8 *uniform)
 {
 	int i;
 	u64 base, mask;
 	u8 prev_match, curr_match;
 
 	*repeat = 0;
-	if (!mtrr_state_set)
-		return 0xFF;
-
-	if (!mtrr_state.enabled)
-		return 0xFF;
+	*uniform = 1;
 
-	/* Make end inclusive end, instead of exclusive */
+	/* Make end inclusive instead of exclusive */
 	end--;
 
-	/* Look in fixed ranges. Just return the type as per start */
-	if (mtrr_state.have_fixed && (start < 0x100000)) {
-		int idx;
-
-		if (start < 0x80000) {
-			idx = 0;
-			idx += (start >> 16);
-			return mtrr_state.fixed_ranges[idx];
-		} else if (start < 0xC0000) {
-			idx = 1 * 8;
-			idx += ((start - 0x80000) >> 14);
-			return mtrr_state.fixed_ranges[idx];
-		} else if (start < 0x1000000) {
-			idx = 3 * 8;
-			idx += ((start - 0xC0000) >> 12);
-			return mtrr_state.fixed_ranges[idx];
-		}
-	}
-
-	/*
-	 * Look in variable ranges
-	 * Look of multiple ranges matching this address and pick type
-	 * as per MTRR precedence
-	 */
-	if (!(mtrr_state.enabled & 2))
-		return mtrr_state.def_type;
-
-	prev_match = 0xFF;
+	prev_match = MTRR_TYPE_INVALID;
 	for (i = 0; i < num_var_ranges; ++i) {
-		unsigned short start_state, end_state;
+		unsigned short start_state, end_state, inclusive;
 
 		if (!(mtrr_state.var_ranges[i].mask_lo & (1 << 11)))
 			continue;
@@ -166,20 +183,29 @@ static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat)
 
 		start_state = ((start & mask) == (base & mask));
 		end_state = ((end & mask) == (base & mask));
+		inclusive = ((start < base) && (end > base));
 
-		if (start_state != end_state) {
+		if ((start_state != end_state) || inclusive) {
 			/*
 			 * We have start:end spanning across an MTRR.
-			 * We split the region into
-			 * either
-			 * (start:mtrr_end) (mtrr_end:end)
-			 * or
-			 * (start:mtrr_start) (mtrr_start:end)
+			 * We split the region into either
+			 *
+			 * - start_state:1
+			 * (start:mtrr_end)(mtrr_end:end)
+			 * - end_state:1
+			 * (start:mtrr_start)(mtrr_start:end)
+			 * - inclusive:1
+			 * (start:mtrr_start)(mtrr_start:mtrr_end)(mtrr_end:end)
+			 *
 			 * depending on kind of overlap.
-			 * Return the type for first region and a pointer to
-			 * the start of second region so that caller will
-			 * lookup again on the second region.
-			 * Note: This way we handle multiple overlaps as well.
+			 *
+			 * Return the type of the first region and a pointer
+			 * to the start of next region so that caller will be
+			 * advised to lookup again after having adjusted start
+			 * and end.
+			 *
+			 * Note: This way we handle overlaps with multiple
+			 * entries and the default type properly.
 			 */
 			if (start_state)
 				*partial_end = base + get_mtrr_size(mask);
@@ -193,59 +219,94 @@ static u8 __mtrr_type_lookup(u64 start, u64 end, u64 *partial_end, int *repeat)
 
 			end = *partial_end - 1; /* end is inclusive */
 			*repeat = 1;
+			*uniform = 0;
 		}
 
 		if ((start & mask) != (base & mask))
 			continue;
 
 		curr_match = mtrr_state.var_ranges[i].base_lo & 0xff;
-		if (prev_match == 0xFF) {
+		if (prev_match == MTRR_TYPE_INVALID) {
 			prev_match = curr_match;
 			continue;
 		}
 
+		*uniform = 0;
 		if (check_type_overlap(&prev_match, &curr_match))
 			return curr_match;
 	}
 
-	if (mtrr_tom2) {
-		if (start >= (1ULL<<32) && (end < mtrr_tom2))
-			return MTRR_TYPE_WRBACK;
-	}
-
-	if (prev_match != 0xFF)
+	if (prev_match != MTRR_TYPE_INVALID)
 		return prev_match;
 
 	return mtrr_state.def_type;
 }
 
-/*
- * Returns the effective MTRR type for the region
- * Error return:
- * 0xFF - when MTRR is not enabled
+/**
+ * mtrr_type_lookup - look up memory type in MTRR
+ *
+ * Return Values:
+ * MTRR_TYPE_(type)  - The effective MTRR type for the region
+ * MTRR_TYPE_INVALID - MTRR is disabled
+ *
+ * Output Argument:
+ * uniform - Set to 1 when an MTRR covers the region uniformly, i.e. the
+ *	     region is fully covered by a single MTRR entry or the default
+ *	     type.
  */
-u8 mtrr_type_lookup(u64 start, u64 end)
+u8 mtrr_type_lookup(u64 start, u64 end, u8 *uniform)
 {
-	u8 type, prev_type;
+	u8 type, prev_type, is_uniform = 1, dummy;
 	int repeat;
 	u64 partial_end;
 
-	type = __mtrr_type_lookup(start, end, &partial_end, &repeat);
+	if (!mtrr_state_set)
+		return MTRR_TYPE_INVALID;
+
+	if (!(mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED))
+		return MTRR_TYPE_INVALID;
+
+	/*
+	 * Look up the fixed ranges first, which take priority over
+	 * the variable ranges.
+	 */
+	if ((start < 0x100000) &&
+	    (mtrr_state.have_fixed) &&
+	    (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) {
+		is_uniform = 0;
+		type = mtrr_type_lookup_fixed(start, end);
+		goto out;
+	}
+
+	/*
+	 * Look up the variable ranges.  Look of multiple ranges matching
+	 * this address and pick type as per MTRR precedence.
+	 */
+	type = mtrr_type_lookup_variable(start, end, &partial_end,
+					 &repeat, &is_uniform);
 
 	/*
 	 * Common path is with repeat = 0.
 	 * However, we can have cases where [start:end] spans across some
-	 * MTRR range. Do repeated lookups for that case here.
+	 * MTRR ranges and/or the default type.  Do repeated lookups for
+	 * that case here.
 	 */
 	while (repeat) {
 		prev_type = type;
 		start = partial_end;
-		type = __mtrr_type_lookup(start, end, &partial_end, &repeat);
+		is_uniform = 0;
+		type = mtrr_type_lookup_variable(start, end, &partial_end,
+						 &repeat, &dummy);
 
 		if (check_type_overlap(&prev_type, &type))
-			return type;
+			goto out;
 	}
 
+	if (mtrr_tom2 && (start >= (1ULL<<32)) && (end < mtrr_tom2))
+		type = MTRR_TYPE_WRBACK;
+
+out:
+	*uniform = is_uniform;
 	return type;
 }
 
@@ -347,7 +408,9 @@ static void __init print_mtrr_state(void)
 		 mtrr_attrib_to_str(mtrr_state.def_type));
 	if (mtrr_state.have_fixed) {
 		pr_debug("MTRR fixed ranges %sabled:\n",
-			 mtrr_state.enabled & 1 ? "en" : "dis");
+			((mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED) &&
+			 (mtrr_state.enabled & MTRR_STATE_MTRR_FIXED_ENABLED)) ?
+			 "en" : "dis");
 		print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0);
 		for (i = 0; i < 2; ++i)
 			print_fixed(0x80000 + i * 0x20000, 0x04000,
@@ -360,7 +423,7 @@ static void __init print_mtrr_state(void)
 		print_fixed_last();
 	}
 	pr_debug("MTRR variable ranges %sabled:\n",
-		 mtrr_state.enabled & 2 ? "en" : "dis");
+		 mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED ? "en" : "dis");
 	high_width = (__ffs64(size_or_mask) - (32 - PAGE_SHIFT) + 3) / 4;
 
 	for (i = 0; i < num_var_ranges; ++i) {
@@ -382,7 +445,7 @@ static void __init print_mtrr_state(void)
 }
 
 /* Grab all of the MTRR state for this CPU into *state */
-void __init get_mtrr_state(void)
+bool __init get_mtrr_state(void)
 {
 	struct mtrr_var_range *vrs;
 	unsigned long flags;
@@ -426,6 +489,8 @@ void __init get_mtrr_state(void)
 
 	post_set();
 	local_irq_restore(flags);
+
+	return !!(mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED);
 }
 
 /* Some BIOS's are messed up and don't set all MTRRs the same! */

arch/x86/kernel/cpu/mtrr/main.c

@@ -59,6 +59,12 @@
 #define MTRR_TO_PHYS_WC_OFFSET 1000
 
 u32 num_var_ranges;
+static bool __mtrr_enabled;
+
+static bool mtrr_enabled(void)
+{
+	return __mtrr_enabled;
+}
 
 unsigned int mtrr_usage_table[MTRR_MAX_VAR_RANGES];
 static DEFINE_MUTEX(mtrr_mutex);
@@ -286,7 +292,7 @@ int mtrr_add_page(unsigned long base, unsigned long size,
 	int i, replace, error;
 	mtrr_type ltype;
 
-	if (!mtrr_if)
+	if (!mtrr_enabled())
 		return -ENXIO;
 
 	error = mtrr_if->validate_add_page(base, size, type);
@@ -435,6 +441,8 @@ static int mtrr_check(unsigned long base, unsigned long size)
 int mtrr_add(unsigned long base, unsigned long size, unsigned int type,
 	     bool increment)
 {
+	if (!mtrr_enabled())
+		return -ENODEV;
 	if (mtrr_check(base, size))
 		return -EINVAL;
 	return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
@@ -463,8 +471,8 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
 	unsigned long lbase, lsize;
 	int error = -EINVAL;
 
-	if (!mtrr_if)
-		return -ENXIO;
+	if (!mtrr_enabled())
+		return -ENODEV;
 
 	max = num_var_ranges;
 	/* No CPU hotplug when we change MTRR entries */
@@ -523,6 +531,8 @@ int mtrr_del_page(int reg, unsigned long base, unsigned long size)
  */
 int mtrr_del(int reg, unsigned long base, unsigned long size)
 {
+	if (!mtrr_enabled())
+		return -ENODEV;
 	if (mtrr_check(base, size))
 		return -EINVAL;
 	return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
@@ -538,6 +548,9 @@ EXPORT_SYMBOL(mtrr_del);
  * attempts to add a WC MTRR covering size bytes starting at base and
  * logs an error if this fails.
  *
+ * The called should provide a power of two size on an equivalent
+ * power of two boundary.
+ *
  * Drivers must store the return value to pass to mtrr_del_wc_if_needed,
  * but drivers should not try to interpret that return value.
  */
@@ -545,7 +558,7 @@ int arch_phys_wc_add(unsigned long base, unsigned long size)
 {
 	int ret;
 
-	if (pat_enabled)
+	if (pat_enabled() || !mtrr_enabled())
 		return 0;  /* Success!  (We don't need to do anything.) */
 
 	ret = mtrr_add(base, size, MTRR_TYPE_WRCOMB, true);
@@ -577,7 +590,7 @@ void arch_phys_wc_del(int handle)
 EXPORT_SYMBOL(arch_phys_wc_del);
 
 /*
- * phys_wc_to_mtrr_index - translates arch_phys_wc_add's return value
+ * arch_phys_wc_index - translates arch_phys_wc_add's return value
  * @handle: Return value from arch_phys_wc_add
  *
  * This will turn the return value from arch_phys_wc_add into an mtrr
@@ -587,14 +600,14 @@ EXPORT_SYMBOL(arch_phys_wc_del);
  * in printk line.  Alas there is an illegitimate use in some ancient
  * drm ioctls.
  */
-int phys_wc_to_mtrr_index(int handle)
+int arch_phys_wc_index(int handle)
 {
 	if (handle < MTRR_TO_PHYS_WC_OFFSET)
 		return -1;
 	else
 		return handle - MTRR_TO_PHYS_WC_OFFSET;
 }
-EXPORT_SYMBOL_GPL(phys_wc_to_mtrr_index);
+EXPORT_SYMBOL_GPL(arch_phys_wc_index);
 
 /*
  * HACK ALERT!
@@ -734,10 +747,12 @@ void __init mtrr_bp_init(void)
 	}
 
 	if (mtrr_if) {
+		__mtrr_enabled = true;
 		set_num_var_ranges();
 		init_table();
 		if (use_intel()) {
-			get_mtrr_state();
+			/* BIOS may override */
+			__mtrr_enabled = get_mtrr_state();
 
 			if (mtrr_cleanup(phys_addr)) {
 				changed_by_mtrr_cleanup = 1;
@@ -745,10 +760,16 @@ void __init mtrr_bp_init(void)
 			}
 		}
 	}
+
+	if (!mtrr_enabled())
+		pr_info("MTRR: Disabled\n");
 }
 
 void mtrr_ap_init(void)
 {
+	if (!mtrr_enabled())
+		return;
+
 	if (!use_intel() || mtrr_aps_delayed_init)
 		return;
 	/*
@@ -774,6 +795,9 @@ void mtrr_save_state(void)
 {
 	int first_cpu;
 
+	if (!mtrr_enabled())
+		return;
+
 	get_online_cpus();
 	first_cpu = cpumask_first(cpu_online_mask);
 	smp_call_function_single(first_cpu, mtrr_save_fixed_ranges, NULL, 1);
@@ -782,6 +806,8 @@ void mtrr_save_state(void)
 
 void set_mtrr_aps_delayed_init(void)
 {
+	if (!mtrr_enabled())
+		return;
 	if (!use_intel())
 		return;
 
@@ -793,7 +819,7 @@ void set_mtrr_aps_delayed_init(void)
  */
 void mtrr_aps_init(void)
 {
-	if (!use_intel())
+	if (!use_intel() || !mtrr_enabled())
 		return;
 
 	/*
@@ -810,7 +836,7 @@ void mtrr_aps_init(void)
 
 void mtrr_bp_restore(void)
 {
-	if (!use_intel())
+	if (!use_intel() || !mtrr_enabled())
 		return;
 
 	mtrr_if->set_all();
@@ -818,7 +844,7 @@ void mtrr_bp_restore(void)
 
 static int __init mtrr_init_finialize(void)
 {
-	if (!mtrr_if)
+	if (!mtrr_enabled())
 		return 0;
 
 	if (use_intel()) {

arch/x86/kernel/cpu/mtrr/mtrr.h

@@ -51,7 +51,7 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt);
 
 void fill_mtrr_var_range(unsigned int index,
 		u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi);
-void get_mtrr_state(void);
+bool get_mtrr_state(void);
 
 extern void set_mtrr_ops(const struct mtrr_ops *ops);
 

arch/x86/kernel/devicetree.c

@@ -4,7 +4,6 @@
 #include <linux/bootmem.h>
 #include <linux/export.h>
 #include <linux/io.h>
-#include <linux/irqdomain.h>
 #include <linux/interrupt.h>
 #include <linux/list.h>
 #include <linux/of.h>
@@ -17,6 +16,7 @@
 #include <linux/of_pci.h>
 #include <linux/initrd.h>
 
+#include <asm/irqdomain.h>
 #include <asm/hpet.h>
 #include <asm/apic.h>
 #include <asm/pci_x86.h>
@@ -196,38 +196,31 @@ static struct of_ioapic_type of_ioapic_type[] =
 	},
 };
 
-static int ioapic_xlate(struct irq_domain *domain,
-			struct device_node *controller,
-			const u32 *intspec, u32 intsize,
-			irq_hw_number_t *out_hwirq, u32 *out_type)
+static int dt_irqdomain_alloc(struct irq_domain *domain, unsigned int virq,
+			      unsigned int nr_irqs, void *arg)
 {
+	struct of_phandle_args *irq_data = (void *)arg;
 	struct of_ioapic_type *it;
-	u32 line, idx, gsi;
+	struct irq_alloc_info tmp;
 
-	if (WARN_ON(intsize < 2))
+	if (WARN_ON(irq_data->args_count < 2))
 		return -EINVAL;
-
-	line = intspec[0];
-
-	if (intspec[1] >= ARRAY_SIZE(of_ioapic_type))
+	if (irq_data->args[1] >= ARRAY_SIZE(of_ioapic_type))
 		return -EINVAL;
 
-	it = &of_ioapic_type[intspec[1]];
+	it = &of_ioapic_type[irq_data->args[1]];
+	ioapic_set_alloc_attr(&tmp, NUMA_NO_NODE, it->trigger, it->polarity);
+	tmp.ioapic_id = mpc_ioapic_id(mp_irqdomain_ioapic_idx(domain));
+	tmp.ioapic_pin = irq_data->args[0];
 
-	idx = (u32)(long)domain->host_data;
-	gsi = mp_pin_to_gsi(idx, line);
-	if (mp_set_gsi_attr(gsi, it->trigger, it->polarity, cpu_to_node(0)))
-		return -EBUSY;
-
-	*out_hwirq = line;
-	*out_type = it->out_type;
-	return 0;
+	return mp_irqdomain_alloc(domain, virq, nr_irqs, &tmp);
 }
 
-const struct irq_domain_ops ioapic_irq_domain_ops = {
-	.map = mp_irqdomain_map,
-	.unmap = mp_irqdomain_unmap,
-	.xlate = ioapic_xlate,
+static const struct irq_domain_ops ioapic_irq_domain_ops = {
+	.alloc		= dt_irqdomain_alloc,
+	.free		= mp_irqdomain_free,
+	.activate	= mp_irqdomain_activate,
+	.deactivate	= mp_irqdomain_deactivate,
 };
 
 static void __init dtb_add_ioapic(struct device_node *dn)

arch/x86/kernel/entry_64.S

@@ -216,7 +216,7 @@ ENTRY(system_call)
 GLOBAL(system_call_after_swapgs)
 
 	movq	%rsp,PER_CPU_VAR(rsp_scratch)
-	movq	PER_CPU_VAR(kernel_stack),%rsp
+	movq	PER_CPU_VAR(cpu_current_top_of_stack),%rsp
 
 	/* Construct struct pt_regs on stack */
 	pushq_cfi $__USER_DS			/* pt_regs->ss */
@@ -419,26 +419,27 @@ syscall_return:
 	 * a completely clean 64-bit userspace context.
 	 */
 	movq RCX(%rsp),%rcx
-	cmpq %rcx,RIP(%rsp)		/* RCX == RIP */
+	movq RIP(%rsp),%r11
+	cmpq %rcx,%r11			/* RCX == RIP */
 	jne opportunistic_sysret_failed
 
 	/*
 	 * On Intel CPUs, SYSRET with non-canonical RCX/RIP will #GP
 	 * in kernel space.  This essentially lets the user take over
-	 * the kernel, since userspace controls RSP.  It's not worth
-	 * testing for canonicalness exactly -- this check detects any
-	 * of the 17 high bits set, which is true for non-canonical
-	 * or kernel addresses.  (This will pessimize vsyscall=native.
-	 * Big deal.)
+	 * the kernel, since userspace controls RSP.
 	 *
-	 * If virtual addresses ever become wider, this will need
+	 * If width of "canonical tail" ever becomes variable, this will need
 	 * to be updated to remain correct on both old and new CPUs.
 	 */
 	.ifne __VIRTUAL_MASK_SHIFT - 47
 	.error "virtual address width changed -- SYSRET checks need update"
 	.endif
-	shr $__VIRTUAL_MASK_SHIFT, %rcx
-	jnz opportunistic_sysret_failed
+	/* Change top 16 bits to be the sign-extension of 47th bit */
+	shl	$(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx
+	sar	$(64 - (__VIRTUAL_MASK_SHIFT+1)), %rcx
+	/* If this changed %rcx, it was not canonical */
+	cmpq	%rcx, %r11
+	jne	opportunistic_sysret_failed
 
 	cmpq $__USER_CS,CS(%rsp)	/* CS must match SYSRET */
 	jne opportunistic_sysret_failed
@@ -475,8 +476,8 @@ syscall_return:
 	 */
 syscall_return_via_sysret:
 	CFI_REMEMBER_STATE
-	/* r11 is already restored (see code above) */
-	RESTORE_C_REGS_EXCEPT_R11
+	/* rcx and r11 are already restored (see code above) */
+	RESTORE_C_REGS_EXCEPT_RCX_R11
 	movq RSP(%rsp),%rsp
 	USERGS_SYSRET64
 	CFI_RESTORE_STATE
@@ -533,40 +534,27 @@ GLOBAL(stub_execveat)
 	CFI_ENDPROC
 END(stub_execveat)
 
-#ifdef CONFIG_X86_X32_ABI
+#if defined(CONFIG_X86_X32_ABI) || defined(CONFIG_IA32_EMULATION)
 	.align	8
 GLOBAL(stub_x32_execve)
+GLOBAL(stub32_execve)
 	CFI_STARTPROC
 	DEFAULT_FRAME 0, 8
 	call	compat_sys_execve
 	jmp	return_from_execve
 	CFI_ENDPROC
+END(stub32_execve)
 END(stub_x32_execve)
 	.align	8
 GLOBAL(stub_x32_execveat)
-	CFI_STARTPROC
-	DEFAULT_FRAME 0, 8
-	call	compat_sys_execveat
-	jmp	return_from_execve
-	CFI_ENDPROC
-END(stub_x32_execveat)
-#endif
-
-#ifdef CONFIG_IA32_EMULATION
-	.align	8
-GLOBAL(stub32_execve)
-	CFI_STARTPROC
-	call	compat_sys_execve
-	jmp	return_from_execve
-	CFI_ENDPROC
-END(stub32_execve)
-	.align	8
 GLOBAL(stub32_execveat)
 	CFI_STARTPROC
+	DEFAULT_FRAME 0, 8
 	call	compat_sys_execveat
 	jmp	return_from_execve
 	CFI_ENDPROC
 END(stub32_execveat)
+END(stub_x32_execveat)
 #endif
 
 /*
@@ -622,7 +610,7 @@ ENTRY(ret_from_fork)
 
 	RESTORE_EXTRA_REGS
 
-	testl $3,CS(%rsp)			# from kernel_thread?
+	testb	$3, CS(%rsp)			# from kernel_thread?
 
 	/*
 	 * By the time we get here, we have no idea whether our pt_regs,
@@ -686,8 +674,8 @@ END(irq_entries_start)
 
 	leaq -RBP(%rsp),%rdi	/* arg1 for \func (pointer to pt_regs) */
 
-	testl $3, CS-RBP(%rsp)
-	je 1f
+	testb	$3, CS-RBP(%rsp)
+	jz	1f
 	SWAPGS
 1:
 	/*
@@ -741,8 +729,8 @@ ret_from_intr:
 	CFI_DEF_CFA_REGISTER	rsp
 	CFI_ADJUST_CFA_OFFSET	RBP
 
-	testl $3,CS(%rsp)
-	je retint_kernel
+	testb	$3, CS(%rsp)
+	jz	retint_kernel
 	/* Interrupt came from user space */
 
 	GET_THREAD_INFO(%rcx)
@@ -928,6 +916,8 @@ apicinterrupt X86_PLATFORM_IPI_VECTOR \
 #ifdef CONFIG_HAVE_KVM
 apicinterrupt3 POSTED_INTR_VECTOR \
 	kvm_posted_intr_ipi smp_kvm_posted_intr_ipi
+apicinterrupt3 POSTED_INTR_WAKEUP_VECTOR \
+	kvm_posted_intr_wakeup_ipi smp_kvm_posted_intr_wakeup_ipi
 #endif
 
 #ifdef CONFIG_X86_MCE_THRESHOLD
@@ -989,7 +979,7 @@ ENTRY(\sym)
 	.if \paranoid
 	.if \paranoid == 1
 	CFI_REMEMBER_STATE
-	testl $3, CS(%rsp)		/* If coming from userspace, switch */
+	testb	$3, CS(%rsp)		/* If coming from userspace, switch */
 	jnz 1f				/* stacks. */
 	.endif
 	call paranoid_entry
@@ -1202,17 +1192,17 @@ ENTRY(xen_failsafe_callback)
 	/*CFI_REL_OFFSET ds,DS*/
 	CFI_REL_OFFSET r11,8
 	CFI_REL_OFFSET rcx,0
-	movw %ds,%cx
+	movl %ds,%ecx
 	cmpw %cx,0x10(%rsp)
 	CFI_REMEMBER_STATE
 	jne 1f
-	movw %es,%cx
+	movl %es,%ecx
 	cmpw %cx,0x18(%rsp)
 	jne 1f
-	movw %fs,%cx
+	movl %fs,%ecx
 	cmpw %cx,0x20(%rsp)
 	jne 1f
-	movw %gs,%cx
+	movl %gs,%ecx
 	cmpw %cx,0x28(%rsp)
 	jne 1f
 	/* All segments match their saved values => Category 2 (Bad IRET). */
@@ -1330,8 +1320,8 @@ ENTRY(error_entry)
 	SAVE_C_REGS 8
 	SAVE_EXTRA_REGS 8
 	xorl %ebx,%ebx
-	testl $3,CS+8(%rsp)
-	je error_kernelspace
+	testb	$3, CS+8(%rsp)
+	jz	error_kernelspace
 error_swapgs:
 	SWAPGS
 error_sti:
@@ -1382,7 +1372,7 @@ ENTRY(error_exit)
 	TRACE_IRQS_OFF
 	GET_THREAD_INFO(%rcx)
 	testl %eax,%eax
-	jne retint_kernel
+	jnz retint_kernel
 	LOCKDEP_SYS_EXIT_IRQ
 	movl TI_flags(%rcx),%edx
 	movl $_TIF_WORK_MASK,%edi
@@ -1627,7 +1617,6 @@ end_repeat_nmi:
 	je 1f
 	movq %r12, %cr2
 1:
-	
 	testl %ebx,%ebx				/* swapgs needed? */
 	jnz nmi_restore
 nmi_swapgs:

arch/x86/kernel/head_32.S

@@ -547,7 +547,7 @@ ENTRY(early_idt_handler)
 	cld
 
 	cmpl $2,(%esp)		# X86_TRAP_NMI
-	je is_nmi		# Ignore NMI
+	je .Lis_nmi		# Ignore NMI
 
 	cmpl $2,%ss:early_recursion_flag
 	je hlt_loop
@@ -600,7 +600,7 @@ ex_entry:
 	pop %ecx
 	pop %eax
 	decl %ss:early_recursion_flag
-is_nmi:
+.Lis_nmi:
 	addl $8,%esp		/* drop vector number and error code */
 	iret
 ENDPROC(early_idt_handler)

arch/x86/kernel/head_64.S

@@ -344,7 +344,7 @@ ENTRY(early_idt_handler)
 	cld
 
 	cmpl $2,(%rsp)		# X86_TRAP_NMI
-	je is_nmi		# Ignore NMI
+	je .Lis_nmi		# Ignore NMI
 
 	cmpl $2,early_recursion_flag(%rip)
 	jz  1f
@@ -409,7 +409,7 @@ ENTRY(early_idt_handler)
 	popq %rcx
 	popq %rax
 	decl early_recursion_flag(%rip)
-is_nmi:
+.Lis_nmi:
 	addq $16,%rsp		# drop vector number and error code
 	INTERRUPT_RETURN
 ENDPROC(early_idt_handler)

arch/x86/kernel/hpet.c

@@ -12,6 +12,7 @@
 #include <linux/pm.h>
 #include <linux/io.h>
 
+#include <asm/irqdomain.h>
 #include <asm/fixmap.h>
 #include <asm/hpet.h>
 #include <asm/time.h>
@@ -305,8 +306,6 @@ static void hpet_legacy_clockevent_register(void)
 	printk(KERN_DEBUG "hpet clockevent registered\n");
 }
 
-static int hpet_setup_msi_irq(unsigned int irq);
-
 static void hpet_set_mode(enum clock_event_mode mode,
 			  struct clock_event_device *evt, int timer)
 {
@@ -357,7 +356,7 @@ static void hpet_set_mode(enum clock_event_mode mode,
 			hpet_enable_legacy_int();
 		} else {
 			struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt);
-			hpet_setup_msi_irq(hdev->irq);
+			irq_domain_activate_irq(irq_get_irq_data(hdev->irq));
 			disable_irq(hdev->irq);
 			irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu));
 			enable_irq(hdev->irq);
@@ -423,6 +422,7 @@ static int hpet_legacy_next_event(unsigned long delta,
 
 static DEFINE_PER_CPU(struct hpet_dev *, cpu_hpet_dev);
 static struct hpet_dev	*hpet_devs;
+static struct irq_domain *hpet_domain;
 
 void hpet_msi_unmask(struct irq_data *data)
 {
@@ -473,31 +473,6 @@ static int hpet_msi_next_event(unsigned long delta,
 	return hpet_next_event(delta, evt, hdev->num);
 }
 
-static int hpet_setup_msi_irq(unsigned int irq)
-{
-	if (x86_msi.setup_hpet_msi(irq, hpet_blockid)) {
-		irq_free_hwirq(irq);
-		return -EINVAL;
-	}
-	return 0;
-}
-
-static int hpet_assign_irq(struct hpet_dev *dev)
-{
-	unsigned int irq = irq_alloc_hwirq(-1);
-
-	if (!irq)
-		return -EINVAL;
-
-	irq_set_handler_data(irq, dev);
-
-	if (hpet_setup_msi_irq(irq))
-		return -EINVAL;
-
-	dev->irq = irq;
-	return 0;
-}
-
 static irqreturn_t hpet_interrupt_handler(int irq, void *data)
 {
 	struct hpet_dev *dev = (struct hpet_dev *)data;
@@ -540,9 +515,6 @@ static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu)
 	if (!(hdev->flags & HPET_DEV_VALID))
 		return;
 
-	if (hpet_setup_msi_irq(hdev->irq))
-		return;
-
 	hdev->cpu = cpu;
 	per_cpu(cpu_hpet_dev, cpu) = hdev;
 	evt->name = hdev->name;
@@ -574,7 +546,7 @@ static void hpet_msi_capability_lookup(unsigned int start_timer)
 	unsigned int id;
 	unsigned int num_timers;
 	unsigned int num_timers_used = 0;
-	int i;
+	int i, irq;
 
 	if (hpet_msi_disable)
 		return;
@@ -587,6 +559,10 @@ static void hpet_msi_capability_lookup(unsigned int start_timer)
 	num_timers++; /* Value read out starts from 0 */
 	hpet_print_config();
 
+	hpet_domain = hpet_create_irq_domain(hpet_blockid);
+	if (!hpet_domain)
+		return;
+
 	hpet_devs = kzalloc(sizeof(struct hpet_dev) * num_timers, GFP_KERNEL);
 	if (!hpet_devs)
 		return;
@@ -601,15 +577,16 @@ static void hpet_msi_capability_lookup(unsigned int start_timer)
 		if (!(cfg & HPET_TN_FSB_CAP))
 			continue;
 
+		irq = hpet_assign_irq(hpet_domain, hdev, hdev->num);
+		if (irq < 0)
+			continue;
+
+		sprintf(hdev->name, "hpet%d", i);
+		hdev->num = i;
+		hdev->irq = irq;
 		hdev->flags = 0;
 		if (cfg & HPET_TN_PERIODIC_CAP)
 			hdev->flags |= HPET_DEV_PERI_CAP;
-		hdev->num = i;
-
-		sprintf(hdev->name, "hpet%d", i);
-		if (hpet_assign_irq(hdev))
-			continue;
-
 		hdev->flags |= HPET_DEV_FSB_CAP;
 		hdev->flags |= HPET_DEV_VALID;
 		num_timers_used++;
@@ -709,10 +686,6 @@ static int hpet_cpuhp_notify(struct notifier_block *n,
 }
 #else
 
-static int hpet_setup_msi_irq(unsigned int irq)
-{
-	return 0;
-}
 static void hpet_msi_capability_lookup(unsigned int start_timer)
 {
 	return;

arch/x86/kernel/i8259.c

@@ -329,8 +329,8 @@ static void init_8259A(int auto_eoi)
 	 */
 	outb_pic(0x11, PIC_MASTER_CMD);	/* ICW1: select 8259A-1 init */
 
-	/* ICW2: 8259A-1 IR0-7 mapped to 0x30-0x37 */
-	outb_pic(IRQ0_VECTOR, PIC_MASTER_IMR);
+	/* ICW2: 8259A-1 IR0-7 mapped to ISA_IRQ_VECTOR(0) */
+	outb_pic(ISA_IRQ_VECTOR(0), PIC_MASTER_IMR);
 
 	/* 8259A-1 (the master) has a slave on IR2 */
 	outb_pic(1U << PIC_CASCADE_IR, PIC_MASTER_IMR);
@@ -342,8 +342,8 @@ static void init_8259A(int auto_eoi)
 
 	outb_pic(0x11, PIC_SLAVE_CMD);	/* ICW1: select 8259A-2 init */
 
-	/* ICW2: 8259A-2 IR0-7 mapped to IRQ8_VECTOR */
-	outb_pic(IRQ8_VECTOR, PIC_SLAVE_IMR);
+	/* ICW2: 8259A-2 IR0-7 mapped to ISA_IRQ_VECTOR(8) */
+	outb_pic(ISA_IRQ_VECTOR(8), PIC_SLAVE_IMR);
 	/* 8259A-2 is a slave on master's IR2 */
 	outb_pic(PIC_CASCADE_IR, PIC_SLAVE_IMR);
 	/* (slave's support for AEOI in flat mode is to be investigated) */

arch/x86/kernel/irq.c

@@ -22,6 +22,12 @@
 #define CREATE_TRACE_POINTS
 #include <asm/trace/irq_vectors.h>
 
+DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat);
+EXPORT_PER_CPU_SYMBOL(irq_stat);
+
+DEFINE_PER_CPU(struct pt_regs *, irq_regs);
+EXPORT_PER_CPU_SYMBOL(irq_regs);
+
 atomic_t irq_err_count;
 
 /* Function pointer for generic interrupt vector handling */
@@ -135,6 +141,18 @@ int arch_show_interrupts(struct seq_file *p, int prec)
 	seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count));
 #if defined(CONFIG_X86_IO_APIC)
 	seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read(&irq_mis_count));
+#endif
+#ifdef CONFIG_HAVE_KVM
+	seq_printf(p, "%*s: ", prec, "PIN");
+	for_each_online_cpu(j)
+		seq_printf(p, "%10u ", irq_stats(j)->kvm_posted_intr_ipis);
+	seq_puts(p, "  Posted-interrupt notification event\n");
+
+	seq_printf(p, "%*s: ", prec, "PIW");
+	for_each_online_cpu(j)
+		seq_printf(p, "%10u ",
+			   irq_stats(j)->kvm_posted_intr_wakeup_ipis);
+	seq_puts(p, "  Posted-interrupt wakeup event\n");
 #endif
 	return 0;
 }
@@ -192,8 +210,7 @@ __visible unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 	unsigned vector = ~regs->orig_ax;
 	unsigned irq;
 
-	irq_enter();
-	exit_idle();
+	entering_irq();
 
 	irq = __this_cpu_read(vector_irq[vector]);
 
@@ -209,7 +226,7 @@ __visible unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 		}
 	}
 
-	irq_exit();
+	exiting_irq();
 
 	set_irq_regs(old_regs);
 	return 1;
@@ -237,6 +254,18 @@ __visible void smp_x86_platform_ipi(struct pt_regs *regs)
 }
 
 #ifdef CONFIG_HAVE_KVM
+static void dummy_handler(void) {}
+static void (*kvm_posted_intr_wakeup_handler)(void) = dummy_handler;
+
+void kvm_set_posted_intr_wakeup_handler(void (*handler)(void))
+{
+	if (handler)
+		kvm_posted_intr_wakeup_handler = handler;
+	else
+		kvm_posted_intr_wakeup_handler = dummy_handler;
+}
+EXPORT_SYMBOL_GPL(kvm_set_posted_intr_wakeup_handler);
+
 /*
  * Handler for POSTED_INTERRUPT_VECTOR.
  */
@@ -244,16 +273,23 @@ __visible void smp_kvm_posted_intr_ipi(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
-	ack_APIC_irq();
-
-	irq_enter();
-
-	exit_idle();
-
+	entering_ack_irq();
 	inc_irq_stat(kvm_posted_intr_ipis);
+	exiting_irq();
+	set_irq_regs(old_regs);
+}
 
-	irq_exit();
+/*
+ * Handler for POSTED_INTERRUPT_WAKEUP_VECTOR.
+ */
+__visible void smp_kvm_posted_intr_wakeup_ipi(struct pt_regs *regs)
+{
+	struct pt_regs *old_regs = set_irq_regs(regs);
 
+	entering_ack_irq();
+	inc_irq_stat(kvm_posted_intr_wakeup_ipis);
+	kvm_posted_intr_wakeup_handler();
+	exiting_irq();
 	set_irq_regs(old_regs);
 }
 #endif

arch/x86/kernel/irq_32.c

@@ -21,12 +21,6 @@
 
 #include <asm/apic.h>
 
-DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat);
-EXPORT_PER_CPU_SYMBOL(irq_stat);
-
-DEFINE_PER_CPU(struct pt_regs *, irq_regs);
-EXPORT_PER_CPU_SYMBOL(irq_regs);
-
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 
 int sysctl_panic_on_stackoverflow __read_mostly;

arch/x86/kernel/irq_64.c

@@ -20,12 +20,6 @@
 #include <asm/idle.h>
 #include <asm/apic.h>
 
-DEFINE_PER_CPU_SHARED_ALIGNED(irq_cpustat_t, irq_stat);
-EXPORT_PER_CPU_SYMBOL(irq_stat);
-
-DEFINE_PER_CPU(struct pt_regs *, irq_regs);
-EXPORT_PER_CPU_SYMBOL(irq_regs);
-
 int sysctl_panic_on_stackoverflow;
 
 /*

arch/x86/kernel/irq_work.c

@@ -10,12 +10,6 @@
 #include <asm/apic.h>
 #include <asm/trace/irq_vectors.h>
 
-static inline void irq_work_entering_irq(void)
-{
-	irq_enter();
-	ack_APIC_irq();
-}
-
 static inline void __smp_irq_work_interrupt(void)
 {
 	inc_irq_stat(apic_irq_work_irqs);
@@ -24,14 +18,14 @@ static inline void __smp_irq_work_interrupt(void)
 
 __visible void smp_irq_work_interrupt(struct pt_regs *regs)
 {
-	irq_work_entering_irq();
+	ipi_entering_ack_irq();
 	__smp_irq_work_interrupt();
 	exiting_irq();
 }
 
 __visible void smp_trace_irq_work_interrupt(struct pt_regs *regs)
 {
-	irq_work_entering_irq();
+	ipi_entering_ack_irq();
 	trace_irq_work_entry(IRQ_WORK_VECTOR);
 	__smp_irq_work_interrupt();
 	trace_irq_work_exit(IRQ_WORK_VECTOR);

arch/x86/kernel/irqinit.c

@@ -86,7 +86,7 @@ void __init init_IRQ(void)
 	int i;
 
 	/*
-	 * On cpu 0, Assign IRQ0_VECTOR..IRQ15_VECTOR's to IRQ 0..15.
+	 * On cpu 0, Assign ISA_IRQ_VECTOR(irq) to IRQ 0..15.
 	 * If these IRQ's are handled by legacy interrupt-controllers like PIC,
 	 * then this configuration will likely be static after the boot. If
 	 * these IRQ's are handled by more mordern controllers like IO-APIC,
@@ -94,7 +94,7 @@ void __init init_IRQ(void)
 	 * irq's migrate etc.
 	 */
 	for (i = 0; i < nr_legacy_irqs(); i++)
-		per_cpu(vector_irq, 0)[IRQ0_VECTOR + i] = i;
+		per_cpu(vector_irq, 0)[ISA_IRQ_VECTOR(i)] = i;
 
 	x86_init.irqs.intr_init();
 }
@@ -144,6 +144,8 @@ static void __init apic_intr_init(void)
 #ifdef CONFIG_HAVE_KVM
 	/* IPI for KVM to deliver posted interrupt */
 	alloc_intr_gate(POSTED_INTR_VECTOR, kvm_posted_intr_ipi);
+	/* IPI for KVM to deliver interrupt to wake up tasks */
+	alloc_intr_gate(POSTED_INTR_WAKEUP_VECTOR, kvm_posted_intr_wakeup_ipi);
 #endif
 
 	/* IPI vectors for APIC spurious and error interrupts */

arch/x86/kernel/mpparse.c

@@ -19,8 +19,8 @@
 #include <linux/module.h>
 #include <linux/smp.h>
 #include <linux/pci.h>
-#include <linux/irqdomain.h>
 
+#include <asm/irqdomain.h>
 #include <asm/mtrr.h>
 #include <asm/mpspec.h>
 #include <asm/pgalloc.h>
@@ -113,11 +113,6 @@ static void __init MP_bus_info(struct mpc_bus *m)
 		pr_warn("Unknown bustype %s - ignoring\n", str);
 }
 
-static struct irq_domain_ops mp_ioapic_irqdomain_ops = {
-	.map = mp_irqdomain_map,
-	.unmap = mp_irqdomain_unmap,
-};
-
 static void __init MP_ioapic_info(struct mpc_ioapic *m)
 {
 	struct ioapic_domain_cfg cfg = {

arch/x86/kernel/paravirt.c

@@ -154,7 +154,9 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf,
 		ret = paravirt_patch_ident_64(insnbuf, len);
 
 	else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) ||
+#ifdef CONFIG_X86_32
 		 type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) ||
+#endif
 		 type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) ||
 		 type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret64))
 		/* If operation requires a jmp, then jmp */
@@ -371,7 +373,7 @@ __visible struct pv_cpu_ops pv_cpu_ops = {
 
 	.load_sp0 = native_load_sp0,
 
-#if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION)
+#if defined(CONFIG_X86_32)
 	.irq_enable_sysexit = native_irq_enable_sysexit,
 #endif
 #ifdef CONFIG_X86_64

arch/x86/kernel/paravirt_patch_64.c

@@ -49,7 +49,6 @@ unsigned native_patch(u8 type, u16 clobbers, void *ibuf,
 		PATCH_SITE(pv_irq_ops, save_fl);
 		PATCH_SITE(pv_irq_ops, irq_enable);
 		PATCH_SITE(pv_irq_ops, irq_disable);
-		PATCH_SITE(pv_cpu_ops, irq_enable_sysexit);
 		PATCH_SITE(pv_cpu_ops, usergs_sysret32);
 		PATCH_SITE(pv_cpu_ops, usergs_sysret64);
 		PATCH_SITE(pv_cpu_ops, swapgs);

arch/x86/kernel/process_32.c

@@ -302,13 +302,10 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	arch_end_context_switch(next_p);
 
 	/*
-	 * Reload esp0, kernel_stack, and current_top_of_stack.  This changes
+	 * Reload esp0 and cpu_current_top_of_stack.  This changes
 	 * current_thread_info().
 	 */
 	load_sp0(tss, next);
-	this_cpu_write(kernel_stack,
-		       (unsigned long)task_stack_page(next_p) +
-		       THREAD_SIZE);
 	this_cpu_write(cpu_current_top_of_stack,
 		       (unsigned long)task_stack_page(next_p) +
 		       THREAD_SIZE);

arch/x86/kernel/process_64.c

@@ -409,9 +409,6 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	/* Reload esp0 and ss1.  This changes current_thread_info(). */
 	load_sp0(tss, next);
 
-	this_cpu_write(kernel_stack,
-		(unsigned long)task_stack_page(next_p) + THREAD_SIZE);
-
 	/*
 	 * Now maybe reload the debug registers and handle I/O bitmaps
 	 */

arch/x86/kernel/setup.c

@@ -1222,8 +1222,7 @@ void __init setup_arch(char **cmdline_p)
 	init_cpu_to_node();
 
 	init_apic_mappings();
-	if (x86_io_apic_ops.init)
-		x86_io_apic_ops.init();
+	io_apic_init_mappings();
 
 	kvm_guest_init();
 

arch/x86/kernel/smp.c

@@ -170,8 +170,7 @@ static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
 
 asmlinkage __visible void smp_reboot_interrupt(void)
 {
-	ack_APIC_irq();
-	irq_enter();
+	ipi_entering_ack_irq();
 	stop_this_cpu(NULL);
 	irq_exit();
 }
@@ -265,12 +264,6 @@ __visible void smp_reschedule_interrupt(struct pt_regs *regs)
 	 */
 }
 
-static inline void smp_entering_irq(void)
-{
-	ack_APIC_irq();
-	irq_enter();
-}
-
 __visible void smp_trace_reschedule_interrupt(struct pt_regs *regs)
 {
 	/*
@@ -279,7 +272,7 @@ __visible void smp_trace_reschedule_interrupt(struct pt_regs *regs)
 	 * scheduler_ipi(). This is OK, since those functions are allowed
 	 * to nest.
 	 */
-	smp_entering_irq();
+	ipi_entering_ack_irq();
 	trace_reschedule_entry(RESCHEDULE_VECTOR);
 	__smp_reschedule_interrupt();
 	trace_reschedule_exit(RESCHEDULE_VECTOR);
@@ -297,14 +290,14 @@ static inline void __smp_call_function_interrupt(void)
 
 __visible void smp_call_function_interrupt(struct pt_regs *regs)
 {
-	smp_entering_irq();
+	ipi_entering_ack_irq();
 	__smp_call_function_interrupt();
 	exiting_irq();
 }
 
 __visible void smp_trace_call_function_interrupt(struct pt_regs *regs)
 {
-	smp_entering_irq();
+	ipi_entering_ack_irq();
 	trace_call_function_entry(CALL_FUNCTION_VECTOR);
 	__smp_call_function_interrupt();
 	trace_call_function_exit(CALL_FUNCTION_VECTOR);
@@ -319,14 +312,14 @@ static inline void __smp_call_function_single_interrupt(void)
 
 __visible void smp_call_function_single_interrupt(struct pt_regs *regs)
 {
-	smp_entering_irq();
+	ipi_entering_ack_irq();
 	__smp_call_function_single_interrupt();
 	exiting_irq();
 }
 
 __visible void smp_trace_call_function_single_interrupt(struct pt_regs *regs)
 {
-	smp_entering_irq();
+	ipi_entering_ack_irq();
 	trace_call_function_single_entry(CALL_FUNCTION_SINGLE_VECTOR);
 	__smp_call_function_single_interrupt();
 	trace_call_function_single_exit(CALL_FUNCTION_SINGLE_VECTOR);

arch/x86/kernel/smpboot.c

@@ -513,6 +513,40 @@ void __inquire_remote_apic(int apicid)
 	}
 }
 
+/*
+ * The Multiprocessor Specification 1.4 (1997) example code suggests
+ * that there should be a 10ms delay between the BSP asserting INIT
+ * and de-asserting INIT, when starting a remote processor.
+ * But that slows boot and resume on modern processors, which include
+ * many cores and don't require that delay.
+ *
+ * Cmdline "init_cpu_udelay=" is available to over-ride this delay.
+ * Modern processor families are quirked to remove the delay entirely.
+ */
+#define UDELAY_10MS_DEFAULT 10000
+
+static unsigned int init_udelay = UDELAY_10MS_DEFAULT;
+
+static int __init cpu_init_udelay(char *str)
+{
+	get_option(&str, &init_udelay);
+
+	return 0;
+}
+early_param("cpu_init_udelay", cpu_init_udelay);
+
+static void __init smp_quirk_init_udelay(void)
+{
+	/* if cmdline changed it from default, leave it alone */
+	if (init_udelay != UDELAY_10MS_DEFAULT)
+		return;
+
+	/* if modern processor, use no delay */
+	if (((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 == 6)) ||
+	    ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && (boot_cpu_data.x86 >= 0xF)))
+		init_udelay = 0;
+}
+
 /*
  * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
  * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
@@ -555,7 +589,7 @@ wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip)
 static int
 wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
 {
-	unsigned long send_status, accept_status = 0;
+	unsigned long send_status = 0, accept_status = 0;
 	int maxlvt, num_starts, j;
 
 	maxlvt = lapic_get_maxlvt();
@@ -583,7 +617,7 @@ wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
 	pr_debug("Waiting for send to finish...\n");
 	send_status = safe_apic_wait_icr_idle();
 
-	mdelay(10);
+	udelay(init_udelay);
 
 	pr_debug("Deasserting INIT\n");
 
@@ -651,6 +685,7 @@ wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
 		 * Give the other CPU some time to accept the IPI.
 		 */
 		udelay(200);
+
 		if (maxlvt > 3)		/* Due to the Pentium erratum 3AP.  */
 			apic_write(APIC_ESR, 0);
 		accept_status = (apic_read(APIC_ESR) & 0xEF);
@@ -792,8 +827,6 @@ void common_cpu_up(unsigned int cpu, struct task_struct *idle)
 	clear_tsk_thread_flag(idle, TIF_FORK);
 	initial_gs = per_cpu_offset(cpu);
 #endif
-	per_cpu(kernel_stack, cpu) =
-		(unsigned long)task_stack_page(idle) + THREAD_SIZE;
 }
 
 /*
@@ -1176,6 +1209,8 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 		uv_system_init();
 
 	set_mtrr_aps_delayed_init();
+
+	smp_quirk_init_udelay();
 }
 
 void arch_enable_nonboot_cpus_begin(void)

arch/x86/kernel/traps.c

@@ -997,8 +997,8 @@ void __init trap_init(void)
 #endif
 
 #ifdef CONFIG_X86_32
-	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
-	set_bit(SYSCALL_VECTOR, used_vectors);
+	set_system_trap_gate(IA32_SYSCALL_VECTOR, &system_call);
+	set_bit(IA32_SYSCALL_VECTOR, used_vectors);
 #endif
 
 	/*

arch/x86/kernel/x86_init.c

@@ -111,11 +111,9 @@ EXPORT_SYMBOL_GPL(x86_platform);
 #if defined(CONFIG_PCI_MSI)
 struct x86_msi_ops x86_msi = {
 	.setup_msi_irqs		= native_setup_msi_irqs,
-	.compose_msi_msg	= native_compose_msi_msg,
 	.teardown_msi_irq	= native_teardown_msi_irq,
 	.teardown_msi_irqs	= default_teardown_msi_irqs,
 	.restore_msi_irqs	= default_restore_msi_irqs,
-	.setup_hpet_msi		= default_setup_hpet_msi,
 };
 
 /* MSI arch specific hooks */
@@ -141,13 +139,6 @@ void arch_restore_msi_irqs(struct pci_dev *dev)
 #endif
 
 struct x86_io_apic_ops x86_io_apic_ops = {
-	.init			= native_io_apic_init_mappings,
 	.read			= native_io_apic_read,
-	.write			= native_io_apic_write,
-	.modify			= native_io_apic_modify,
 	.disable		= native_disable_io_apic,
-	.print_entries		= native_io_apic_print_entries,
-	.set_affinity		= native_ioapic_set_affinity,
-	.setup_entry		= native_setup_ioapic_entry,
-	.eoi_ioapic_pin		= native_eoi_ioapic_pin,
 };

arch/x86/lguest/boot.c

@@ -90,7 +90,7 @@ struct lguest_data lguest_data = {
 	.noirq_iret = (u32)lguest_noirq_iret,
 	.kernel_address = PAGE_OFFSET,
 	.blocked_interrupts = { 1 }, /* Block timer interrupts */
-	.syscall_vec = SYSCALL_VECTOR,
+	.syscall_vec = IA32_SYSCALL_VECTOR,
 };
 
 /*G:037
@@ -866,7 +866,7 @@ static void __init lguest_init_IRQ(void)
 	for (i = FIRST_EXTERNAL_VECTOR; i < FIRST_SYSTEM_VECTOR; i++) {
 		/* Some systems map "vectors" to interrupts weirdly.  Not us! */
 		__this_cpu_write(vector_irq[i], i - FIRST_EXTERNAL_VECTOR);
-		if (i != SYSCALL_VECTOR)
+		if (i != IA32_SYSCALL_VECTOR)
 			set_intr_gate(i, irq_entries_start +
 					8 * (i - FIRST_EXTERNAL_VECTOR));
 	}

arch/x86/lib/Makefile

@@ -40,6 +40,6 @@ else
         lib-y += csum-partial_64.o csum-copy_64.o csum-wrappers_64.o
         lib-y += clear_page_64.o copy_page_64.o
         lib-y += memmove_64.o memset_64.o
-        lib-y += copy_user_64.o copy_user_nocache_64.o
+        lib-y += copy_user_64.o
 	lib-y += cmpxchg16b_emu.o
 endif

arch/x86/lib/copy_user_64.S

@@ -16,30 +16,6 @@
 #include <asm/asm.h>
 #include <asm/smap.h>
 
-	.macro ALIGN_DESTINATION
-	/* check for bad alignment of destination */
-	movl %edi,%ecx
-	andl $7,%ecx
-	jz 102f				/* already aligned */
-	subl $8,%ecx
-	negl %ecx
-	subl %ecx,%edx
-100:	movb (%rsi),%al
-101:	movb %al,(%rdi)
-	incq %rsi
-	incq %rdi
-	decl %ecx
-	jnz 100b
-102:
-	.section .fixup,"ax"
-103:	addl %ecx,%edx			/* ecx is zerorest also */
-	jmp copy_user_handle_tail
-	.previous
-
-	_ASM_EXTABLE(100b,103b)
-	_ASM_EXTABLE(101b,103b)
-	.endm
-
 /* Standard copy_to_user with segment limit checking */
 ENTRY(_copy_to_user)
 	CFI_STARTPROC
@@ -266,3 +242,95 @@ ENTRY(copy_user_enhanced_fast_string)
 	_ASM_EXTABLE(1b,12b)
 	CFI_ENDPROC
 ENDPROC(copy_user_enhanced_fast_string)
+
+/*
+ * copy_user_nocache - Uncached memory copy with exception handling
+ * This will force destination/source out of cache for more performance.
+ */
+ENTRY(__copy_user_nocache)
+	CFI_STARTPROC
+	ASM_STAC
+	cmpl $8,%edx
+	jb 20f		/* less then 8 bytes, go to byte copy loop */
+	ALIGN_DESTINATION
+	movl %edx,%ecx
+	andl $63,%edx
+	shrl $6,%ecx
+	jz 17f
+1:	movq (%rsi),%r8
+2:	movq 1*8(%rsi),%r9
+3:	movq 2*8(%rsi),%r10
+4:	movq 3*8(%rsi),%r11
+5:	movnti %r8,(%rdi)
+6:	movnti %r9,1*8(%rdi)
+7:	movnti %r10,2*8(%rdi)
+8:	movnti %r11,3*8(%rdi)
+9:	movq 4*8(%rsi),%r8
+10:	movq 5*8(%rsi),%r9
+11:	movq 6*8(%rsi),%r10
+12:	movq 7*8(%rsi),%r11
+13:	movnti %r8,4*8(%rdi)
+14:	movnti %r9,5*8(%rdi)
+15:	movnti %r10,6*8(%rdi)
+16:	movnti %r11,7*8(%rdi)
+	leaq 64(%rsi),%rsi
+	leaq 64(%rdi),%rdi
+	decl %ecx
+	jnz 1b
+17:	movl %edx,%ecx
+	andl $7,%edx
+	shrl $3,%ecx
+	jz 20f
+18:	movq (%rsi),%r8
+19:	movnti %r8,(%rdi)
+	leaq 8(%rsi),%rsi
+	leaq 8(%rdi),%rdi
+	decl %ecx
+	jnz 18b
+20:	andl %edx,%edx
+	jz 23f
+	movl %edx,%ecx
+21:	movb (%rsi),%al
+22:	movb %al,(%rdi)
+	incq %rsi
+	incq %rdi
+	decl %ecx
+	jnz 21b
+23:	xorl %eax,%eax
+	ASM_CLAC
+	sfence
+	ret
+
+	.section .fixup,"ax"
+30:	shll $6,%ecx
+	addl %ecx,%edx
+	jmp 60f
+40:	lea (%rdx,%rcx,8),%rdx
+	jmp 60f
+50:	movl %ecx,%edx
+60:	sfence
+	jmp copy_user_handle_tail
+	.previous
+
+	_ASM_EXTABLE(1b,30b)
+	_ASM_EXTABLE(2b,30b)
+	_ASM_EXTABLE(3b,30b)
+	_ASM_EXTABLE(4b,30b)
+	_ASM_EXTABLE(5b,30b)
+	_ASM_EXTABLE(6b,30b)
+	_ASM_EXTABLE(7b,30b)
+	_ASM_EXTABLE(8b,30b)
+	_ASM_EXTABLE(9b,30b)
+	_ASM_EXTABLE(10b,30b)
+	_ASM_EXTABLE(11b,30b)
+	_ASM_EXTABLE(12b,30b)
+	_ASM_EXTABLE(13b,30b)
+	_ASM_EXTABLE(14b,30b)
+	_ASM_EXTABLE(15b,30b)
+	_ASM_EXTABLE(16b,30b)
+	_ASM_EXTABLE(18b,40b)
+	_ASM_EXTABLE(19b,40b)
+	_ASM_EXTABLE(21b,50b)
+	_ASM_EXTABLE(22b,50b)
+	CFI_ENDPROC
+ENDPROC(__copy_user_nocache)

arch/x86/lib/copy_user_nocache_64.S

@@ -1,136 +0,0 @@
-/*
- * Copyright 2008 Vitaly Mayatskikh <vmayatsk@redhat.com>
- * Copyright 2002 Andi Kleen, SuSE Labs.
- * Subject to the GNU Public License v2.
- *
- * Functions to copy from and to user space.
- */
-
-#include <linux/linkage.h>
-#include <asm/dwarf2.h>
-
-#define FIX_ALIGNMENT 1
-
-#include <asm/current.h>
-#include <asm/asm-offsets.h>
-#include <asm/thread_info.h>
-#include <asm/asm.h>
-#include <asm/smap.h>
-
-	.macro ALIGN_DESTINATION
-#ifdef FIX_ALIGNMENT
-	/* check for bad alignment of destination */
-	movl %edi,%ecx
-	andl $7,%ecx
-	jz 102f				/* already aligned */
-	subl $8,%ecx
-	negl %ecx
-	subl %ecx,%edx
-100:	movb (%rsi),%al
-101:	movb %al,(%rdi)
-	incq %rsi
-	incq %rdi
-	decl %ecx
-	jnz 100b
-102:
-	.section .fixup,"ax"
-103:	addl %ecx,%edx			/* ecx is zerorest also */
-	jmp copy_user_handle_tail
-	.previous
-
-	_ASM_EXTABLE(100b,103b)
-	_ASM_EXTABLE(101b,103b)
-#endif
-	.endm
-
-/*
- * copy_user_nocache - Uncached memory copy with exception handling
- * This will force destination/source out of cache for more performance.
- */
-ENTRY(__copy_user_nocache)
-	CFI_STARTPROC
-	ASM_STAC
-	cmpl $8,%edx
-	jb 20f		/* less then 8 bytes, go to byte copy loop */
-	ALIGN_DESTINATION
-	movl %edx,%ecx
-	andl $63,%edx
-	shrl $6,%ecx
-	jz 17f
-1:	movq (%rsi),%r8
-2:	movq 1*8(%rsi),%r9
-3:	movq 2*8(%rsi),%r10
-4:	movq 3*8(%rsi),%r11
-5:	movnti %r8,(%rdi)
-6:	movnti %r9,1*8(%rdi)
-7:	movnti %r10,2*8(%rdi)
-8:	movnti %r11,3*8(%rdi)
-9:	movq 4*8(%rsi),%r8
-10:	movq 5*8(%rsi),%r9
-11:	movq 6*8(%rsi),%r10
-12:	movq 7*8(%rsi),%r11
-13:	movnti %r8,4*8(%rdi)
-14:	movnti %r9,5*8(%rdi)
-15:	movnti %r10,6*8(%rdi)
-16:	movnti %r11,7*8(%rdi)
-	leaq 64(%rsi),%rsi
-	leaq 64(%rdi),%rdi
-	decl %ecx
-	jnz 1b
-17:	movl %edx,%ecx
-	andl $7,%edx
-	shrl $3,%ecx
-	jz 20f
-18:	movq (%rsi),%r8
-19:	movnti %r8,(%rdi)
-	leaq 8(%rsi),%rsi
-	leaq 8(%rdi),%rdi
-	decl %ecx
-	jnz 18b
-20:	andl %edx,%edx
-	jz 23f
-	movl %edx,%ecx
-21:	movb (%rsi),%al
-22:	movb %al,(%rdi)
-	incq %rsi
-	incq %rdi
-	decl %ecx
-	jnz 21b
-23:	xorl %eax,%eax
-	ASM_CLAC
-	sfence
-	ret
-
-	.section .fixup,"ax"
-30:	shll $6,%ecx
-	addl %ecx,%edx
-	jmp 60f
-40:	lea (%rdx,%rcx,8),%rdx
-	jmp 60f
-50:	movl %ecx,%edx
-60:	sfence
-	jmp copy_user_handle_tail
-	.previous
-
-	_ASM_EXTABLE(1b,30b)
-	_ASM_EXTABLE(2b,30b)
-	_ASM_EXTABLE(3b,30b)
-	_ASM_EXTABLE(4b,30b)
-	_ASM_EXTABLE(5b,30b)
-	_ASM_EXTABLE(6b,30b)
-	_ASM_EXTABLE(7b,30b)
-	_ASM_EXTABLE(8b,30b)
-	_ASM_EXTABLE(9b,30b)
-	_ASM_EXTABLE(10b,30b)
-	_ASM_EXTABLE(11b,30b)
-	_ASM_EXTABLE(12b,30b)
-	_ASM_EXTABLE(13b,30b)
-	_ASM_EXTABLE(14b,30b)
-	_ASM_EXTABLE(15b,30b)
-	_ASM_EXTABLE(16b,30b)
-	_ASM_EXTABLE(18b,40b)
-	_ASM_EXTABLE(19b,40b)
-	_ASM_EXTABLE(21b,50b)
-	_ASM_EXTABLE(22b,50b)
-	CFI_ENDPROC
-ENDPROC(__copy_user_nocache)

arch/x86/mm/iomap_32.c

@@ -82,7 +82,7 @@ iomap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
 	 * MTRR is UC or WC.  UC_MINUS gets the real intention, of the
 	 * user, which is "WC if the MTRR is WC, UC if you can't do that."
 	 */
-	if (!pat_enabled && pgprot_val(prot) ==
+	if (!pat_enabled() && pgprot_val(prot) ==
 	    (__PAGE_KERNEL | cachemode2protval(_PAGE_CACHE_MODE_WC)))
 		prot = __pgprot(__PAGE_KERNEL |
 				cachemode2protval(_PAGE_CACHE_MODE_UC_MINUS));

arch/x86/mm/ioremap.c

@@ -234,10 +234,11 @@ void __iomem *ioremap_nocache(resource_size_t phys_addr, unsigned long size)
 {
 	/*
 	 * Ideally, this should be:
-	 *	pat_enabled ? _PAGE_CACHE_MODE_UC : _PAGE_CACHE_MODE_UC_MINUS;
+	 *	pat_enabled() ? _PAGE_CACHE_MODE_UC : _PAGE_CACHE_MODE_UC_MINUS;
 	 *
 	 * Till we fix all X drivers to use ioremap_wc(), we will use
-	 * UC MINUS.
+	 * UC MINUS. Drivers that are certain they need or can already
+	 * be converted over to strong UC can use ioremap_uc().
 	 */
 	enum page_cache_mode pcm = _PAGE_CACHE_MODE_UC_MINUS;
 
@@ -246,6 +247,39 @@ void __iomem *ioremap_nocache(resource_size_t phys_addr, unsigned long size)
 }
 EXPORT_SYMBOL(ioremap_nocache);
 
+/**
+ * ioremap_uc     -   map bus memory into CPU space as strongly uncachable
+ * @phys_addr:    bus address of the memory
+ * @size:      size of the resource to map
+ *
+ * ioremap_uc performs a platform specific sequence of operations to
+ * make bus memory CPU accessible via the readb/readw/readl/writeb/
+ * writew/writel functions and the other mmio helpers. The returned
+ * address is not guaranteed to be usable directly as a virtual
+ * address.
+ *
+ * This version of ioremap ensures that the memory is marked with a strong
+ * preference as completely uncachable on the CPU when possible. For non-PAT
+ * systems this ends up setting page-attribute flags PCD=1, PWT=1. For PAT
+ * systems this will set the PAT entry for the pages as strong UC.  This call
+ * will honor existing caching rules from things like the PCI bus. Note that
+ * there are other caches and buffers on many busses. In particular driver
+ * authors should read up on PCI writes.
+ *
+ * It's useful if some control registers are in such an area and
+ * write combining or read caching is not desirable:
+ *
+ * Must be freed with iounmap.
+ */
+void __iomem *ioremap_uc(resource_size_t phys_addr, unsigned long size)
+{
+	enum page_cache_mode pcm = _PAGE_CACHE_MODE_UC;
+
+	return __ioremap_caller(phys_addr, size, pcm,
+				__builtin_return_address(0));
+}
+EXPORT_SYMBOL_GPL(ioremap_uc);
+
 /**
  * ioremap_wc	-	map memory into CPU space write combined
  * @phys_addr:	bus address of the memory
@@ -258,7 +292,7 @@ EXPORT_SYMBOL(ioremap_nocache);
  */
 void __iomem *ioremap_wc(resource_size_t phys_addr, unsigned long size)
 {
-	if (pat_enabled)
+	if (pat_enabled())
 		return __ioremap_caller(phys_addr, size, _PAGE_CACHE_MODE_WC,
 					__builtin_return_address(0));
 	else
@@ -331,7 +365,7 @@ void iounmap(volatile void __iomem *addr)
 }
 EXPORT_SYMBOL(iounmap);
 
-int arch_ioremap_pud_supported(void)
+int __init arch_ioremap_pud_supported(void)
 {
 #ifdef CONFIG_X86_64
 	return cpu_has_gbpages;
@@ -340,7 +374,7 @@ int arch_ioremap_pud_supported(void)
 #endif
 }
 
-int arch_ioremap_pmd_supported(void)
+int __init arch_ioremap_pmd_supported(void)
 {
 	return cpu_has_pse;
 }

arch/x86/mm/pageattr.c

@@ -129,16 +129,15 @@ within(unsigned long addr, unsigned long start, unsigned long end)
  */
 void clflush_cache_range(void *vaddr, unsigned int size)
 {
-	void *vend = vaddr + size - 1;
+	unsigned long clflush_mask = boot_cpu_data.x86_clflush_size - 1;
+	void *vend = vaddr + size;
+	void *p;
 
 	mb();
 
-	for (; vaddr < vend; vaddr += boot_cpu_data.x86_clflush_size)
-		clflushopt(vaddr);
-	/*
-	 * Flush any possible final partial cacheline:
-	 */
-	clflushopt(vend);
+	for (p = (void *)((unsigned long)vaddr & ~clflush_mask);
+	     p < vend; p += boot_cpu_data.x86_clflush_size)
+		clflushopt(p);
 
 	mb();
 }
@@ -418,13 +417,11 @@ phys_addr_t slow_virt_to_phys(void *__virt_addr)
 	phys_addr_t phys_addr;
 	unsigned long offset;
 	enum pg_level level;
-	unsigned long psize;
 	unsigned long pmask;
 	pte_t *pte;
 
 	pte = lookup_address(virt_addr, &level);
 	BUG_ON(!pte);
-	psize = page_level_size(level);
 	pmask = page_level_mask(level);
 	offset = virt_addr & ~pmask;
 	phys_addr = (phys_addr_t)pte_pfn(*pte) << PAGE_SHIFT;
@@ -1468,6 +1465,9 @@ int _set_memory_uc(unsigned long addr, int numpages)
 {
 	/*
 	 * for now UC MINUS. see comments in ioremap_nocache()
+	 * If you really need strong UC use ioremap_uc(), but note
+	 * that you cannot override IO areas with set_memory_*() as
+	 * these helpers cannot work with IO memory.
 	 */
 	return change_page_attr_set(&addr, numpages,
 				    cachemode2pgprot(_PAGE_CACHE_MODE_UC_MINUS),
@@ -1571,7 +1571,7 @@ int set_memory_wc(unsigned long addr, int numpages)
 {
 	int ret;
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return set_memory_uc(addr, numpages);
 
 	ret = reserve_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE,

arch/x86/mm/pat.c

@@ -33,13 +33,15 @@
 #include "pat_internal.h"
 #include "mm_internal.h"
 
-#ifdef CONFIG_X86_PAT
-int __read_mostly pat_enabled = 1;
+#undef pr_fmt
+#define pr_fmt(fmt) "" fmt
+
+static int __read_mostly __pat_enabled = IS_ENABLED(CONFIG_X86_PAT);
 
 static inline void pat_disable(const char *reason)
 {
-	pat_enabled = 0;
-	printk(KERN_INFO "%s\n", reason);
+	__pat_enabled = 0;
+	pr_info("x86/PAT: %s\n", reason);
 }
 
 static int __init nopat(char *str)
@@ -48,13 +50,12 @@ static int __init nopat(char *str)
 	return 0;
 }
 early_param("nopat", nopat);
-#else
-static inline void pat_disable(const char *reason)
+
+bool pat_enabled(void)
 {
-	(void)reason;
+	return !!__pat_enabled;
 }
-#endif
-
+EXPORT_SYMBOL_GPL(pat_enabled);
 
 int pat_debug_enable;
 
@@ -188,7 +189,7 @@ void pat_init_cache_modes(void)
 					   pat_msg + 4 * i);
 		update_cache_mode_entry(i, cache);
 	}
-	pr_info("PAT configuration [0-7]: %s\n", pat_msg);
+	pr_info("x86/PAT: Configuration [0-7]: %s\n", pat_msg);
 }
 
 #define PAT(x, y)	((u64)PAT_ ## y << ((x)*8))
@@ -198,7 +199,7 @@ void pat_init(void)
 	u64 pat;
 	bool boot_cpu = !boot_pat_state;
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return;
 
 	if (!cpu_has_pat) {
@@ -211,8 +212,7 @@ void pat_init(void)
 			 * switched to PAT on the boot CPU. We have no way to
 			 * undo PAT.
 			 */
-			printk(KERN_ERR "PAT enabled, "
-			       "but not supported by secondary CPU\n");
+			pr_err("x86/PAT: PAT enabled, but not supported by secondary CPU\n");
 			BUG();
 		}
 	}
@@ -267,9 +267,9 @@ static unsigned long pat_x_mtrr_type(u64 start, u64 end,
 	 * request is for WB.
 	 */
 	if (req_type == _PAGE_CACHE_MODE_WB) {
-		u8 mtrr_type;
+		u8 mtrr_type, uniform;
 
-		mtrr_type = mtrr_type_lookup(start, end);
+		mtrr_type = mtrr_type_lookup(start, end, &uniform);
 		if (mtrr_type != MTRR_TYPE_WRBACK)
 			return _PAGE_CACHE_MODE_UC_MINUS;
 
@@ -347,7 +347,7 @@ static int reserve_ram_pages_type(u64 start, u64 end,
 		page = pfn_to_page(pfn);
 		type = get_page_memtype(page);
 		if (type != -1) {
-			pr_info("reserve_ram_pages_type failed [mem %#010Lx-%#010Lx], track 0x%x, req 0x%x\n",
+			pr_info("x86/PAT: reserve_ram_pages_type failed [mem %#010Lx-%#010Lx], track 0x%x, req 0x%x\n",
 				start, end - 1, type, req_type);
 			if (new_type)
 				*new_type = type;
@@ -400,7 +400,7 @@ int reserve_memtype(u64 start, u64 end, enum page_cache_mode req_type,
 
 	BUG_ON(start >= end); /* end is exclusive */
 
-	if (!pat_enabled) {
+	if (!pat_enabled()) {
 		/* This is identical to page table setting without PAT */
 		if (new_type) {
 			if (req_type == _PAGE_CACHE_MODE_WC)
@@ -451,9 +451,9 @@ int reserve_memtype(u64 start, u64 end, enum page_cache_mode req_type,
 
 	err = rbt_memtype_check_insert(new, new_type);
 	if (err) {
-		printk(KERN_INFO "reserve_memtype failed [mem %#010Lx-%#010Lx], track %s, req %s\n",
-		       start, end - 1,
-		       cattr_name(new->type), cattr_name(req_type));
+		pr_info("x86/PAT: reserve_memtype failed [mem %#010Lx-%#010Lx], track %s, req %s\n",
+			start, end - 1,
+			cattr_name(new->type), cattr_name(req_type));
 		kfree(new);
 		spin_unlock(&memtype_lock);
 
@@ -475,7 +475,7 @@ int free_memtype(u64 start, u64 end)
 	int is_range_ram;
 	struct memtype *entry;
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return 0;
 
 	/* Low ISA region is always mapped WB. No need to track */
@@ -497,8 +497,8 @@ int free_memtype(u64 start, u64 end)
 	spin_unlock(&memtype_lock);
 
 	if (!entry) {
-		printk(KERN_INFO "%s:%d freeing invalid memtype [mem %#010Lx-%#010Lx]\n",
-		       current->comm, current->pid, start, end - 1);
+		pr_info("x86/PAT: %s:%d freeing invalid memtype [mem %#010Lx-%#010Lx]\n",
+			current->comm, current->pid, start, end - 1);
 		return -EINVAL;
 	}
 
@@ -623,13 +623,13 @@ static inline int range_is_allowed(unsigned long pfn, unsigned long size)
 	u64 to = from + size;
 	u64 cursor = from;
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return 1;
 
 	while (cursor < to) {
 		if (!devmem_is_allowed(pfn)) {
-			printk(KERN_INFO "Program %s tried to access /dev/mem between [mem %#010Lx-%#010Lx], PAT prevents it\n",
-			       current->comm, from, to - 1);
+			pr_info("x86/PAT: Program %s tried to access /dev/mem between [mem %#010Lx-%#010Lx], PAT prevents it\n",
+				current->comm, from, to - 1);
 			return 0;
 		}
 		cursor += PAGE_SIZE;
@@ -659,7 +659,7 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn,
 	 * caching for the high addresses through the KEN pin, but
 	 * we maintain the tradition of paranoia in this code.
 	 */
-	if (!pat_enabled &&
+	if (!pat_enabled() &&
 	    !(boot_cpu_has(X86_FEATURE_MTRR) ||
 	      boot_cpu_has(X86_FEATURE_K6_MTRR) ||
 	      boot_cpu_has(X86_FEATURE_CYRIX_ARR) ||
@@ -698,8 +698,7 @@ int kernel_map_sync_memtype(u64 base, unsigned long size,
 				size;
 
 	if (ioremap_change_attr((unsigned long)__va(base), id_sz, pcm) < 0) {
-		printk(KERN_INFO "%s:%d ioremap_change_attr failed %s "
-			"for [mem %#010Lx-%#010Lx]\n",
+		pr_info("x86/PAT: %s:%d ioremap_change_attr failed %s for [mem %#010Lx-%#010Lx]\n",
 			current->comm, current->pid,
 			cattr_name(pcm),
 			base, (unsigned long long)(base + size-1));
@@ -729,12 +728,12 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot,
 	 * the type requested matches the type of first page in the range.
 	 */
 	if (is_ram) {
-		if (!pat_enabled)
+		if (!pat_enabled())
 			return 0;
 
 		pcm = lookup_memtype(paddr);
 		if (want_pcm != pcm) {
-			printk(KERN_WARNING "%s:%d map pfn RAM range req %s for [mem %#010Lx-%#010Lx], got %s\n",
+			pr_warn("x86/PAT: %s:%d map pfn RAM range req %s for [mem %#010Lx-%#010Lx], got %s\n",
 				current->comm, current->pid,
 				cattr_name(want_pcm),
 				(unsigned long long)paddr,
@@ -755,13 +754,12 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot,
 		if (strict_prot ||
 		    !is_new_memtype_allowed(paddr, size, want_pcm, pcm)) {
 			free_memtype(paddr, paddr + size);
-			printk(KERN_ERR "%s:%d map pfn expected mapping type %s"
-				" for [mem %#010Lx-%#010Lx], got %s\n",
-				current->comm, current->pid,
-				cattr_name(want_pcm),
-				(unsigned long long)paddr,
-				(unsigned long long)(paddr + size - 1),
-				cattr_name(pcm));
+			pr_err("x86/PAT: %s:%d map pfn expected mapping type %s for [mem %#010Lx-%#010Lx], got %s\n",
+			       current->comm, current->pid,
+			       cattr_name(want_pcm),
+			       (unsigned long long)paddr,
+			       (unsigned long long)(paddr + size - 1),
+			       cattr_name(pcm));
 			return -EINVAL;
 		}
 		/*
@@ -844,7 +842,7 @@ int track_pfn_remap(struct vm_area_struct *vma, pgprot_t *prot,
 		return ret;
 	}
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return 0;
 
 	/*
@@ -872,7 +870,7 @@ int track_pfn_insert(struct vm_area_struct *vma, pgprot_t *prot,
 {
 	enum page_cache_mode pcm;
 
-	if (!pat_enabled)
+	if (!pat_enabled())
 		return 0;
 
 	/* Set prot based on lookup */
@@ -913,7 +911,7 @@ void untrack_pfn(struct vm_area_struct *vma, unsigned long pfn,
 
 pgprot_t pgprot_writecombine(pgprot_t prot)
 {
-	if (pat_enabled)
+	if (pat_enabled())
 		return __pgprot(pgprot_val(prot) |
 				cachemode2protval(_PAGE_CACHE_MODE_WC));
 	else
@@ -996,7 +994,7 @@ static const struct file_operations memtype_fops = {
 
 static int __init pat_memtype_list_init(void)
 {
-	if (pat_enabled) {
+	if (pat_enabled()) {
 		debugfs_create_file("pat_memtype_list", S_IRUSR,
 				    arch_debugfs_dir, NULL, &memtype_fops);
 	}

arch/x86/mm/pat_internal.h

@@ -4,7 +4,7 @@
 extern int pat_debug_enable;
 
 #define dprintk(fmt, arg...) \
-	do { if (pat_debug_enable) printk(KERN_INFO fmt, ##arg); } while (0)
+	do { if (pat_debug_enable) pr_info("x86/PAT: " fmt, ##arg); } while (0)
 
 struct memtype {
 	u64			start;

arch/x86/mm/pat_rbtree.c

@@ -160,9 +160,9 @@ success:
 	return 0;
 
 failure:
-	printk(KERN_INFO "%s:%d conflicting memory types "
-		"%Lx-%Lx %s<->%s\n", current->comm, current->pid, start,
-		end, cattr_name(found_type), cattr_name(match->type));
+	pr_info("x86/PAT: %s:%d conflicting memory types %Lx-%Lx %s<->%s\n",
+		current->comm, current->pid, start, end,
+		cattr_name(found_type), cattr_name(match->type));
 	return -EBUSY;
 }
 

arch/x86/mm/pgtable.c

@@ -563,16 +563,31 @@ void native_set_fixmap(enum fixed_addresses idx, phys_addr_t phys,
 }
 
 #ifdef CONFIG_HAVE_ARCH_HUGE_VMAP
+/**
+ * pud_set_huge - setup kernel PUD mapping
+ *
+ * MTRRs can override PAT memory types with 4KiB granularity. Therefore, this
+ * function sets up a huge page only if any of the following conditions are met:
+ *
+ * - MTRRs are disabled, or
+ *
+ * - MTRRs are enabled and the range is completely covered by a single MTRR, or
+ *
+ * - MTRRs are enabled and the corresponding MTRR memory type is WB, which
+ *   has no effect on the requested PAT memory type.
+ *
+ * Callers should try to decrease page size (1GB -> 2MB -> 4K) if the bigger
+ * page mapping attempt fails.
+ *
+ * Returns 1 on success and 0 on failure.
+ */
 int pud_set_huge(pud_t *pud, phys_addr_t addr, pgprot_t prot)
 {
-	u8 mtrr;
+	u8 mtrr, uniform;
 
-	/*
-	 * Do not use a huge page when the range is covered by non-WB type
-	 * of MTRRs.
-	 */
-	mtrr = mtrr_type_lookup(addr, addr + PUD_SIZE);
-	if ((mtrr != MTRR_TYPE_WRBACK) && (mtrr != 0xFF))
+	mtrr = mtrr_type_lookup(addr, addr + PUD_SIZE, &uniform);
+	if ((mtrr != MTRR_TYPE_INVALID) && (!uniform) &&
+	    (mtrr != MTRR_TYPE_WRBACK))
 		return 0;
 
 	prot = pgprot_4k_2_large(prot);
@@ -584,17 +599,24 @@ int pud_set_huge(pud_t *pud, phys_addr_t addr, pgprot_t prot)
 	return 1;
 }
 
+/**
+ * pmd_set_huge - setup kernel PMD mapping
+ *
+ * See text over pud_set_huge() above.
+ *
+ * Returns 1 on success and 0 on failure.
+ */
 int pmd_set_huge(pmd_t *pmd, phys_addr_t addr, pgprot_t prot)
 {
-	u8 mtrr;
+	u8 mtrr, uniform;
 
-	/*
-	 * Do not use a huge page when the range is covered by non-WB type
-	 * of MTRRs.
-	 */
-	mtrr = mtrr_type_lookup(addr, addr + PMD_SIZE);
-	if ((mtrr != MTRR_TYPE_WRBACK) && (mtrr != 0xFF))
+	mtrr = mtrr_type_lookup(addr, addr + PMD_SIZE, &uniform);
+	if ((mtrr != MTRR_TYPE_INVALID) && (!uniform) &&
+	    (mtrr != MTRR_TYPE_WRBACK)) {
+		pr_warn_once("%s: Cannot satisfy [mem %#010llx-%#010llx] with a huge-page mapping due to MTRR override.\n",
+			     __func__, addr, addr + PMD_SIZE);
 		return 0;
+	}
 
 	prot = pgprot_4k_2_large(prot);
 
@@ -605,6 +627,11 @@ int pmd_set_huge(pmd_t *pmd, phys_addr_t addr, pgprot_t prot)
 	return 1;
 }
 
+/**
+ * pud_clear_huge - clear kernel PUD mapping when it is set
+ *
+ * Returns 1 on success and 0 on failure (no PUD map is found).
+ */
 int pud_clear_huge(pud_t *pud)
 {
 	if (pud_large(*pud)) {
@@ -615,6 +642,11 @@ int pud_clear_huge(pud_t *pud)
 	return 0;
 }
 
+/**
+ * pmd_clear_huge - clear kernel PMD mapping when it is set
+ *
+ * Returns 1 on success and 0 on failure (no PMD map is found).
+ */
 int pmd_clear_huge(pmd_t *pmd)
 {
 	if (pmd_large(*pmd)) {

arch/x86/pci/i386.c

@@ -429,12 +429,12 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
  	 * Caller can followup with UC MINUS request and add a WC mtrr if there
  	 * is a free mtrr slot.
  	 */
-	if (!pat_enabled && write_combine)
+	if (!pat_enabled() && write_combine)
 		return -EINVAL;
 
-	if (pat_enabled && write_combine)
+	if (pat_enabled() && write_combine)
 		prot |= cachemode2protval(_PAGE_CACHE_MODE_WC);
-	else if (pat_enabled || boot_cpu_data.x86 > 3)
+	else if (pat_enabled() || boot_cpu_data.x86 > 3)
 		/*
 		 * ioremap() and ioremap_nocache() defaults to UC MINUS for now.
 		 * To avoid attribute conflicts, request UC MINUS here

arch/x86/pci/intel_mid_pci.c

@@ -208,6 +208,7 @@ static int pci_write(struct pci_bus *bus, unsigned int devfn, int where,
 
 static int intel_mid_pci_irq_enable(struct pci_dev *dev)
 {
+	struct irq_alloc_info info;
 	int polarity;
 
 	if (dev->irq_managed && dev->irq > 0)
@@ -217,14 +218,13 @@ static int intel_mid_pci_irq_enable(struct pci_dev *dev)
 		polarity = 0; /* active high */
 	else
 		polarity = 1; /* active low */
+	ioapic_set_alloc_attr(&info, dev_to_node(&dev->dev), 1, polarity);
 
 	/*
 	 * MRST only have IOAPIC, the PCI irq lines are 1:1 mapped to
 	 * IOAPIC RTE entries, so we just enable RTE for the device.
 	 */
-	if (mp_set_gsi_attr(dev->irq, 1, polarity, dev_to_node(&dev->dev)))
-		return -EBUSY;
-	if (mp_map_gsi_to_irq(dev->irq, IOAPIC_MAP_ALLOC) < 0)
+	if (mp_map_gsi_to_irq(dev->irq, IOAPIC_MAP_ALLOC, &info) < 0)
 		return -EBUSY;
 
 	dev->irq_managed = 1;

arch/x86/pci/irq.c

@@ -146,19 +146,20 @@ static void __init pirq_peer_trick(void)
 
 /*
  *  Code for querying and setting of IRQ routes on various interrupt routers.
+ *  PIC Edge/Level Control Registers (ELCR) 0x4d0 & 0x4d1.
  */
 
-void eisa_set_level_irq(unsigned int irq)
+void elcr_set_level_irq(unsigned int irq)
 {
 	unsigned char mask = 1 << (irq & 7);
 	unsigned int port = 0x4d0 + (irq >> 3);
 	unsigned char val;
-	static u16 eisa_irq_mask;
+	static u16 elcr_irq_mask;
 
-	if (irq >= 16 || (1 << irq) & eisa_irq_mask)
+	if (irq >= 16 || (1 << irq) & elcr_irq_mask)
 		return;
 
-	eisa_irq_mask |= (1 << irq);
+	elcr_irq_mask |= (1 << irq);
 	printk(KERN_DEBUG "PCI: setting IRQ %u as level-triggered\n", irq);
 	val = inb(port);
 	if (!(val & mask)) {
@@ -965,11 +966,11 @@ static int pcibios_lookup_irq(struct pci_dev *dev, int assign)
 	} else if (r->get && (irq = r->get(pirq_router_dev, dev, pirq)) && \
 	((!(pci_probe & PCI_USE_PIRQ_MASK)) || ((1 << irq) & mask))) {
 		msg = "found";
-		eisa_set_level_irq(irq);
+		elcr_set_level_irq(irq);
 	} else if (newirq && r->set &&
 		(dev->class >> 8) != PCI_CLASS_DISPLAY_VGA) {
 		if (r->set(pirq_router_dev, dev, pirq, newirq)) {
-			eisa_set_level_irq(newirq);
+			elcr_set_level_irq(newirq);
 			msg = "assigned";
 			irq = newirq;
 		}

arch/x86/platform/Makefile

@@ -1,4 +1,5 @@
 # Platform specific code goes here
+obj-y	+= atom/
 obj-y	+= ce4100/
 obj-y	+= efi/
 obj-y	+= geode/

arch/x86/platform/atom/Makefile

@@ -0,0 +1 @@
+obj-$(CONFIG_PUNIT_ATOM_DEBUG) += punit_atom_debug.o

arch/x86/platform/atom/punit_atom_debug.c

@@ -0,0 +1,183 @@
+/*
+ * Intel SOC Punit device state debug driver
+ * Punit controls power management for North Complex devices (Graphics
+ * blocks, Image Signal Processing, video processing, display, DSP etc.)
+ *
+ * Copyright (c) 2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/io.h>
+#include <asm/cpu_device_id.h>
+#include <asm/iosf_mbi.h>
+
+/* Side band Interface port */
+#define PUNIT_PORT		0x04
+/* Power gate status reg */
+#define PWRGT_STATUS		0x61
+/* Subsystem config/status Video processor */
+#define VED_SS_PM0		0x32
+/* Subsystem config/status ISP (Image Signal Processor) */
+#define ISP_SS_PM0		0x39
+/* Subsystem config/status Input/output controller */
+#define MIO_SS_PM		0x3B
+/* Shift bits for getting status for video, isp and i/o */
+#define SSS_SHIFT		24
+/* Shift bits for getting status for graphics rendering */
+#define RENDER_POS		0
+/* Shift bits for getting status for media control */
+#define MEDIA_POS		2
+/* Shift bits for getting status for Valley View/Baytrail display */
+#define VLV_DISPLAY_POS		6
+/* Subsystem config/status display for Cherry Trail SOC */
+#define CHT_DSP_SSS		0x36
+/* Shift bits for getting status for display */
+#define CHT_DSP_SSS_POS		16
+
+struct punit_device {
+	char *name;
+	int reg;
+	int sss_pos;
+};
+
+static const struct punit_device punit_device_byt[] = {
+	{ "GFX RENDER",	PWRGT_STATUS,	RENDER_POS },
+	{ "GFX MEDIA",	PWRGT_STATUS,	MEDIA_POS },
+	{ "DISPLAY",	PWRGT_STATUS,	VLV_DISPLAY_POS },
+	{ "VED",	VED_SS_PM0,	SSS_SHIFT },
+	{ "ISP",	ISP_SS_PM0,	SSS_SHIFT },
+	{ "MIO",	MIO_SS_PM,	SSS_SHIFT },
+	{ NULL }
+};
+
+static const struct punit_device punit_device_cht[] = {
+	{ "GFX RENDER",	PWRGT_STATUS,	RENDER_POS },
+	{ "GFX MEDIA",	PWRGT_STATUS,	MEDIA_POS },
+	{ "DISPLAY",	CHT_DSP_SSS,	CHT_DSP_SSS_POS },
+	{ "VED",	VED_SS_PM0,	SSS_SHIFT },
+	{ "ISP",	ISP_SS_PM0,	SSS_SHIFT },
+	{ "MIO",	MIO_SS_PM,	SSS_SHIFT },
+	{ NULL }
+};
+
+static const char * const dstates[] = {"D0", "D0i1", "D0i2", "D0i3"};
+
+static int punit_dev_state_show(struct seq_file *seq_file, void *unused)
+{
+	u32 punit_pwr_status;
+	struct punit_device *punit_devp = seq_file->private;
+	int index;
+	int status;
+
+	seq_puts(seq_file, "\n\nPUNIT NORTH COMPLEX DEVICES :\n");
+	while (punit_devp->name) {
+		status = iosf_mbi_read(PUNIT_PORT, BT_MBI_PMC_READ,
+				       punit_devp->reg,
+				       &punit_pwr_status);
+		if (status) {
+			seq_printf(seq_file, "%9s : Read Failed\n",
+				   punit_devp->name);
+		} else  {
+			index = (punit_pwr_status >> punit_devp->sss_pos) & 3;
+			seq_printf(seq_file, "%9s : %s\n", punit_devp->name,
+				   dstates[index]);
+		}
+		punit_devp++;
+	}
+
+	return 0;
+}
+
+static int punit_dev_state_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, punit_dev_state_show, inode->i_private);
+}
+
+static const struct file_operations punit_dev_state_ops = {
+	.open		= punit_dev_state_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static struct dentry *punit_dbg_file;
+
+static int punit_dbgfs_register(struct punit_device *punit_device)
+{
+	static struct dentry *dev_state;
+
+	punit_dbg_file = debugfs_create_dir("punit_atom", NULL);
+	if (!punit_dbg_file)
+		return -ENXIO;
+
+	dev_state = debugfs_create_file("dev_power_state", S_IFREG | S_IRUGO,
+					punit_dbg_file, punit_device,
+					&punit_dev_state_ops);
+	if (!dev_state) {
+		pr_err("punit_dev_state register failed\n");
+		debugfs_remove(punit_dbg_file);
+		return -ENXIO;
+	}
+
+	return 0;
+}
+
+static void punit_dbgfs_unregister(void)
+{
+	debugfs_remove_recursive(punit_dbg_file);
+}
+
+#define ICPU(model, drv_data) \
+	{ X86_VENDOR_INTEL, 6, model, X86_FEATURE_MWAIT,\
+	  (kernel_ulong_t)&drv_data }
+
+static const struct x86_cpu_id intel_punit_cpu_ids[] = {
+	ICPU(55, punit_device_byt), /* Valleyview, Bay Trail */
+	ICPU(76, punit_device_cht), /* Braswell, Cherry Trail */
+	{}
+};
+
+MODULE_DEVICE_TABLE(x86cpu, intel_punit_cpu_ids);
+
+static int __init punit_atom_debug_init(void)
+{
+	const struct x86_cpu_id *id;
+	int ret;
+
+	id = x86_match_cpu(intel_punit_cpu_ids);
+	if (!id)
+		return -ENODEV;
+
+	ret = punit_dbgfs_register((struct punit_device *)id->driver_data);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static void __exit punit_atom_debug_exit(void)
+{
+	punit_dbgfs_unregister();
+}
+
+module_init(punit_atom_debug_init);
+module_exit(punit_atom_debug_exit);
+
+MODULE_AUTHOR("Kumar P, Mahesh <mahesh.kumar.p@intel.com>");
+MODULE_AUTHOR("Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>");
+MODULE_DESCRIPTION("Driver for Punit devices states debugging");
+MODULE_LICENSE("GPL v2");

arch/x86/platform/intel-mid/device_libs/platform_wdt.c

@@ -27,6 +27,7 @@ static struct platform_device wdt_dev = {
 static int tangier_probe(struct platform_device *pdev)
 {
 	int gsi;
+	struct irq_alloc_info info;
 	struct intel_mid_wdt_pdata *pdata = pdev->dev.platform_data;
 
 	if (!pdata)
@@ -34,8 +35,8 @@ static int tangier_probe(struct platform_device *pdev)
 
 	/* IOAPIC builds identity mapping between GSI and IRQ on MID */
 	gsi = pdata->irq;
-	if (mp_set_gsi_attr(gsi, 1, 0, cpu_to_node(0)) ||
-	    mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC) <= 0) {
+	ioapic_set_alloc_attr(&info, cpu_to_node(0), 1, 0);
+	if (mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC, &info) <= 0) {
 		dev_warn(&pdev->dev, "cannot find interrupt %d in ioapic\n",
 			 gsi);
 		return -EINVAL;

arch/x86/platform/intel-mid/intel-mid.c

@@ -81,26 +81,34 @@ static unsigned long __init intel_mid_calibrate_tsc(void)
 	return 0;
 }
 
+static void __init intel_mid_setup_bp_timer(void)
+{
+	apbt_time_init();
+	setup_boot_APIC_clock();
+}
+
 static void __init intel_mid_time_init(void)
 {
 	sfi_table_parse(SFI_SIG_MTMR, NULL, NULL, sfi_parse_mtmr);
+
 	switch (intel_mid_timer_options) {
 	case INTEL_MID_TIMER_APBT_ONLY:
 		break;
 	case INTEL_MID_TIMER_LAPIC_APBT:
-		x86_init.timers.setup_percpu_clockev = setup_boot_APIC_clock;
+		/* Use apbt and local apic */
+		x86_init.timers.setup_percpu_clockev = intel_mid_setup_bp_timer;
 		x86_cpuinit.setup_percpu_clockev = setup_secondary_APIC_clock;
-		break;
+		return;
 	default:
 		if (!boot_cpu_has(X86_FEATURE_ARAT))
 			break;
+		/* Lapic only, no apbt */
 		x86_init.timers.setup_percpu_clockev = setup_boot_APIC_clock;
 		x86_cpuinit.setup_percpu_clockev = setup_secondary_APIC_clock;
 		return;
 	}
-	/* we need at least one APB timer */
-	pre_init_apic_IRQ0();
-	apbt_time_init();
+
+	x86_init.timers.setup_percpu_clockev = apbt_time_init;
 }
 
 static void intel_mid_arch_setup(void)

arch/x86/platform/intel-mid/sfi.c

@@ -95,18 +95,16 @@ int __init sfi_parse_mtmr(struct sfi_table_header *table)
 		pr_debug("timer[%d]: paddr = 0x%08x, freq = %dHz, irq = %d\n",
 			totallen, (u32)pentry->phys_addr,
 			pentry->freq_hz, pentry->irq);
-			if (!pentry->irq)
-				continue;
-			mp_irq.type = MP_INTSRC;
-			mp_irq.irqtype = mp_INT;
-/* triggering mode edge bit 2-3, active high polarity bit 0-1 */
-			mp_irq.irqflag = 5;
-			mp_irq.srcbus = MP_BUS_ISA;
-			mp_irq.srcbusirq = pentry->irq;	/* IRQ */
-			mp_irq.dstapic = MP_APIC_ALL;
-			mp_irq.dstirq = pentry->irq;
-			mp_save_irq(&mp_irq);
-			mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC);
+		mp_irq.type = MP_INTSRC;
+		mp_irq.irqtype = mp_INT;
+		/* triggering mode edge bit 2-3, active high polarity bit 0-1 */
+		mp_irq.irqflag = 5;
+		mp_irq.srcbus = MP_BUS_ISA;
+		mp_irq.srcbusirq = pentry->irq;	/* IRQ */
+		mp_irq.dstapic = MP_APIC_ALL;
+		mp_irq.dstirq = pentry->irq;
+		mp_save_irq(&mp_irq);
+		mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC, NULL);
 	}
 
 	return 0;
@@ -177,7 +175,7 @@ int __init sfi_parse_mrtc(struct sfi_table_header *table)
 		mp_irq.dstapic = MP_APIC_ALL;
 		mp_irq.dstirq = pentry->irq;
 		mp_save_irq(&mp_irq);
-		mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC);
+		mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC, NULL);
 	}
 	return 0;
 }
@@ -436,6 +434,7 @@ static int __init sfi_parse_devs(struct sfi_table_header *table)
 	struct devs_id *dev = NULL;
 	int num, i, ret;
 	int polarity;
+	struct irq_alloc_info info;
 
 	sb = (struct sfi_table_simple *)table;
 	num = SFI_GET_NUM_ENTRIES(sb, struct sfi_device_table_entry);
@@ -469,9 +468,8 @@ static int __init sfi_parse_devs(struct sfi_table_header *table)
 				polarity = 1;
 			}
 
-			ret = mp_set_gsi_attr(irq, 1, polarity, NUMA_NO_NODE);
-			if (ret == 0)
-				ret = mp_map_gsi_to_irq(irq, IOAPIC_MAP_ALLOC);
+			ioapic_set_alloc_attr(&info, NUMA_NO_NODE, 1, polarity);
+			ret = mp_map_gsi_to_irq(irq, IOAPIC_MAP_ALLOC, &info);
 			WARN_ON(ret < 0);
 		}
 

arch/x86/platform/sfi/sfi.c

@@ -25,8 +25,8 @@
 #include <linux/init.h>
 #include <linux/sfi.h>
 #include <linux/io.h>
-#include <linux/irqdomain.h>
 
+#include <asm/irqdomain.h>
 #include <asm/io_apic.h>
 #include <asm/mpspec.h>
 #include <asm/setup.h>
@@ -71,9 +71,6 @@ static int __init sfi_parse_cpus(struct sfi_table_header *table)
 #endif /* CONFIG_X86_LOCAL_APIC */
 
 #ifdef CONFIG_X86_IO_APIC
-static struct irq_domain_ops sfi_ioapic_irqdomain_ops = {
-	.map = mp_irqdomain_map,
-};
 
 static int __init sfi_parse_ioapic(struct sfi_table_header *table)
 {
@@ -82,7 +79,7 @@ static int __init sfi_parse_ioapic(struct sfi_table_header *table)
 	int i, num;
 	struct ioapic_domain_cfg cfg = {
 		.type = IOAPIC_DOMAIN_STRICT,
-		.ops = &sfi_ioapic_irqdomain_ops,
+		.ops = &mp_ioapic_irqdomain_ops,
 	};
 
 	sb = (struct sfi_table_simple *)table;

arch/x86/platform/uv/uv_irq.c

@@ -13,22 +13,37 @@
 #include <linux/slab.h>
 #include <linux/irq.h>
 
+#include <asm/irqdomain.h>
 #include <asm/apic.h>
 #include <asm/uv/uv_irq.h>
 #include <asm/uv/uv_hub.h>
 
 /* MMR offset and pnode of hub sourcing interrupts for a given irq */
-struct uv_irq_2_mmr_pnode{
-	struct rb_node		list;
+struct uv_irq_2_mmr_pnode {
 	unsigned long		offset;
 	int			pnode;
-	int			irq;
 };
 
-static DEFINE_SPINLOCK(uv_irq_lock);
-static struct rb_root		uv_irq_root;
+static void uv_program_mmr(struct irq_cfg *cfg, struct uv_irq_2_mmr_pnode *info)
+{
+	unsigned long mmr_value;
+	struct uv_IO_APIC_route_entry *entry;
+
+	BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
+		     sizeof(unsigned long));
+
+	mmr_value = 0;
+	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
+	entry->vector		= cfg->vector;
+	entry->delivery_mode	= apic->irq_delivery_mode;
+	entry->dest_mode	= apic->irq_dest_mode;
+	entry->polarity		= 0;
+	entry->trigger		= 0;
+	entry->mask		= 0;
+	entry->dest		= cfg->dest_apicid;
 
-static int uv_set_irq_affinity(struct irq_data *, const struct cpumask *, bool);
+	uv_write_global_mmr64(info->pnode, info->offset, mmr_value);
+}
 
 static void uv_noop(struct irq_data *data) { }
 
@@ -37,6 +52,23 @@ static void uv_ack_apic(struct irq_data *data)
 	ack_APIC_irq();
 }
 
+static int
+uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
+		    bool force)
+{
+	struct irq_data *parent = data->parent_data;
+	struct irq_cfg *cfg = irqd_cfg(data);
+	int ret;
+
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret >= 0) {
+		uv_program_mmr(cfg, data->chip_data);
+		send_cleanup_vector(cfg);
+	}
+
+	return ret;
+}
+
 static struct irq_chip uv_irq_chip = {
 	.name			= "UV-CORE",
 	.irq_mask		= uv_noop,
@@ -45,189 +77,99 @@ static struct irq_chip uv_irq_chip = {
 	.irq_set_affinity	= uv_set_irq_affinity,
 };
 
-/*
- * Add offset and pnode information of the hub sourcing interrupts to the
- * rb tree for a specific irq.
- */
-static int uv_set_irq_2_mmr_info(int irq, unsigned long offset, unsigned blade)
+static int uv_domain_alloc(struct irq_domain *domain, unsigned int virq,
+			   unsigned int nr_irqs, void *arg)
 {
-	struct rb_node **link = &uv_irq_root.rb_node;
-	struct rb_node *parent = NULL;
-	struct uv_irq_2_mmr_pnode *n;
-	struct uv_irq_2_mmr_pnode *e;
-	unsigned long irqflags;
-
-	n = kmalloc_node(sizeof(struct uv_irq_2_mmr_pnode), GFP_KERNEL,
-				uv_blade_to_memory_nid(blade));
-	if (!n)
+	struct uv_irq_2_mmr_pnode *chip_data;
+	struct irq_alloc_info *info = arg;
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+	int ret;
+
+	if (nr_irqs > 1 || !info || info->type != X86_IRQ_ALLOC_TYPE_UV)
+		return -EINVAL;
+
+	chip_data = kmalloc_node(sizeof(*chip_data), GFP_KERNEL,
+				 irq_data->node);
+	if (!chip_data)
 		return -ENOMEM;
 
-	n->irq = irq;
-	n->offset = offset;
-	n->pnode = uv_blade_to_pnode(blade);
-	spin_lock_irqsave(&uv_irq_lock, irqflags);
-	/* Find the right place in the rbtree: */
-	while (*link) {
-		parent = *link;
-		e = rb_entry(parent, struct uv_irq_2_mmr_pnode, list);
-
-		if (unlikely(irq == e->irq)) {
-			/* irq entry exists */
-			e->pnode = uv_blade_to_pnode(blade);
-			e->offset = offset;
-			spin_unlock_irqrestore(&uv_irq_lock, irqflags);
-			kfree(n);
-			return 0;
-		}
-
-		if (irq < e->irq)
-			link = &(*link)->rb_left;
+	ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
+	if (ret >= 0) {
+		if (info->uv_limit == UV_AFFINITY_CPU)
+			irq_set_status_flags(virq, IRQ_NO_BALANCING);
 		else
-			link = &(*link)->rb_right;
+			irq_set_status_flags(virq, IRQ_MOVE_PCNTXT);
+
+		chip_data->pnode = uv_blade_to_pnode(info->uv_blade);
+		chip_data->offset = info->uv_offset;
+		irq_domain_set_info(domain, virq, virq, &uv_irq_chip, chip_data,
+				    handle_percpu_irq, NULL, info->uv_name);
+	} else {
+		kfree(chip_data);
 	}
 
-	/* Insert the node into the rbtree. */
-	rb_link_node(&n->list, parent, link);
-	rb_insert_color(&n->list, &uv_irq_root);
-
-	spin_unlock_irqrestore(&uv_irq_lock, irqflags);
-	return 0;
+	return ret;
 }
 
-/* Retrieve offset and pnode information from the rb tree for a specific irq */
-int uv_irq_2_mmr_info(int irq, unsigned long *offset, int *pnode)
+static void uv_domain_free(struct irq_domain *domain, unsigned int virq,
+			   unsigned int nr_irqs)
 {
-	struct uv_irq_2_mmr_pnode *e;
-	struct rb_node *n;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&uv_irq_lock, irqflags);
-	n = uv_irq_root.rb_node;
-	while (n) {
-		e = rb_entry(n, struct uv_irq_2_mmr_pnode, list);
-
-		if (e->irq == irq) {
-			*offset = e->offset;
-			*pnode = e->pnode;
-			spin_unlock_irqrestore(&uv_irq_lock, irqflags);
-			return 0;
-		}
-
-		if (irq < e->irq)
-			n = n->rb_left;
-		else
-			n = n->rb_right;
-	}
-	spin_unlock_irqrestore(&uv_irq_lock, irqflags);
-	return -1;
+	struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+
+	BUG_ON(nr_irqs != 1);
+	kfree(irq_data->chip_data);
+	irq_clear_status_flags(virq, IRQ_MOVE_PCNTXT);
+	irq_clear_status_flags(virq, IRQ_NO_BALANCING);
+	irq_domain_free_irqs_top(domain, virq, nr_irqs);
 }
 
 /*
  * Re-target the irq to the specified CPU and enable the specified MMR located
  * on the specified blade to allow the sending of MSIs to the specified CPU.
  */
-static int
-arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
-		       unsigned long mmr_offset, int limit)
+static void uv_domain_activate(struct irq_domain *domain,
+			       struct irq_data *irq_data)
 {
-	const struct cpumask *eligible_cpu = cpumask_of(cpu);
-	struct irq_cfg *cfg = irq_cfg(irq);
-	unsigned long mmr_value;
-	struct uv_IO_APIC_route_entry *entry;
-	int mmr_pnode, err;
-	unsigned int dest;
-
-	BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
-			sizeof(unsigned long));
-
-	err = assign_irq_vector(irq, cfg, eligible_cpu);
-	if (err != 0)
-		return err;
-
-	err = apic->cpu_mask_to_apicid_and(eligible_cpu, eligible_cpu, &dest);
-	if (err != 0)
-		return err;
-
-	if (limit == UV_AFFINITY_CPU)
-		irq_set_status_flags(irq, IRQ_NO_BALANCING);
-	else
-		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-
-	irq_set_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq,
-				      irq_name);
-
-	mmr_value = 0;
-	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
-	entry->vector		= cfg->vector;
-	entry->delivery_mode	= apic->irq_delivery_mode;
-	entry->dest_mode	= apic->irq_dest_mode;
-	entry->polarity		= 0;
-	entry->trigger		= 0;
-	entry->mask		= 0;
-	entry->dest		= dest;
-
-	mmr_pnode = uv_blade_to_pnode(mmr_blade);
-	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
-
-	if (cfg->move_in_progress)
-		send_cleanup_vector(cfg);
-
-	return irq;
+	uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
 }
 
 /*
  * Disable the specified MMR located on the specified blade so that MSIs are
  * longer allowed to be sent.
  */
-static void arch_disable_uv_irq(int mmr_pnode, unsigned long mmr_offset)
+static void uv_domain_deactivate(struct irq_domain *domain,
+				 struct irq_data *irq_data)
 {
 	unsigned long mmr_value;
 	struct uv_IO_APIC_route_entry *entry;
 
-	BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
-			sizeof(unsigned long));
-
 	mmr_value = 0;
 	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
 	entry->mask = 1;
-
-	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
+	uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
 }
 
-static int
-uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
-		    bool force)
-{
-	struct irq_cfg *cfg = irqd_cfg(data);
-	unsigned int dest;
-	unsigned long mmr_value, mmr_offset;
-	struct uv_IO_APIC_route_entry *entry;
-	int mmr_pnode;
-
-	if (apic_set_affinity(data, mask, &dest))
-		return -1;
-
-	mmr_value = 0;
-	entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
-
-	entry->vector		= cfg->vector;
-	entry->delivery_mode	= apic->irq_delivery_mode;
-	entry->dest_mode	= apic->irq_dest_mode;
-	entry->polarity		= 0;
-	entry->trigger		= 0;
-	entry->mask		= 0;
-	entry->dest		= dest;
-
-	/* Get previously stored MMR and pnode of hub sourcing interrupts */
-	if (uv_irq_2_mmr_info(data->irq, &mmr_offset, &mmr_pnode))
-		return -1;
-
-	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
+static const struct irq_domain_ops uv_domain_ops = {
+	.alloc		= uv_domain_alloc,
+	.free		= uv_domain_free,
+	.activate	= uv_domain_activate,
+	.deactivate	= uv_domain_deactivate,
+};
 
-	if (cfg->move_in_progress)
-		send_cleanup_vector(cfg);
+static struct irq_domain *uv_get_irq_domain(void)
+{
+	static struct irq_domain *uv_domain;
+	static DEFINE_MUTEX(uv_lock);
+
+	mutex_lock(&uv_lock);
+	if (uv_domain == NULL) {
+		uv_domain = irq_domain_add_tree(NULL, &uv_domain_ops, NULL);
+		if (uv_domain)
+			uv_domain->parent = x86_vector_domain;
+	}
+	mutex_unlock(&uv_lock);
 
-	return IRQ_SET_MASK_OK_NOCOPY;
+	return uv_domain;
 }
 
 /*
@@ -238,19 +180,21 @@ uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
 int uv_setup_irq(char *irq_name, int cpu, int mmr_blade,
 		 unsigned long mmr_offset, int limit)
 {
-	int ret, irq = irq_alloc_hwirq(uv_blade_to_memory_nid(mmr_blade));
+	struct irq_alloc_info info;
+	struct irq_domain *domain = uv_get_irq_domain();
 
-	if (!irq)
-		return -EBUSY;
+	if (!domain)
+		return -ENOMEM;
 
-	ret = arch_enable_uv_irq(irq_name, irq, cpu, mmr_blade, mmr_offset,
-		limit);
-	if (ret == irq)
-		uv_set_irq_2_mmr_info(irq, mmr_offset, mmr_blade);
-	else
-		irq_free_hwirq(irq);
+	init_irq_alloc_info(&info, cpumask_of(cpu));
+	info.type = X86_IRQ_ALLOC_TYPE_UV;
+	info.uv_limit = limit;
+	info.uv_blade = mmr_blade;
+	info.uv_offset = mmr_offset;
+	info.uv_name = irq_name;
 
-	return ret;
+	return irq_domain_alloc_irqs(domain, 1,
+				     uv_blade_to_memory_nid(mmr_blade), &info);
 }
 EXPORT_SYMBOL_GPL(uv_setup_irq);
 
@@ -263,26 +207,6 @@ EXPORT_SYMBOL_GPL(uv_setup_irq);
  */
 void uv_teardown_irq(unsigned int irq)
 {
-	struct uv_irq_2_mmr_pnode *e;
-	struct rb_node *n;
-	unsigned long irqflags;
-
-	spin_lock_irqsave(&uv_irq_lock, irqflags);
-	n = uv_irq_root.rb_node;
-	while (n) {
-		e = rb_entry(n, struct uv_irq_2_mmr_pnode, list);
-		if (e->irq == irq) {
-			arch_disable_uv_irq(e->pnode, e->offset);
-			rb_erase(n, &uv_irq_root);
-			kfree(e);
-			break;
-		}
-		if (irq < e->irq)
-			n = n->rb_left;
-		else
-			n = n->rb_right;
-	}
-	spin_unlock_irqrestore(&uv_irq_lock, irqflags);
-	irq_free_hwirq(irq);
+	irq_domain_free_irqs(irq, 1);
 }
 EXPORT_SYMBOL_GPL(uv_teardown_irq);

arch/x86/power/hibernate_asm_64.S

@@ -78,9 +78,9 @@ ENTRY(restore_image)
 
 	/* code below has been relocated to a safe page */
 ENTRY(core_restore_code)
-loop:
+.Lloop:
 	testq	%rdx, %rdx
-	jz	done
+	jz	.Ldone
 
 	/* get addresses from the pbe and copy the page */
 	movq	pbe_address(%rdx), %rsi
@@ -91,8 +91,8 @@ loop:
 
 	/* progress to the next pbe */
 	movq	pbe_next(%rdx), %rdx
-	jmp	loop
-done:
+	jmp	.Lloop
+.Ldone:
 	/* jump to the restore_registers address from the image header */
 	jmpq	*%rax
 	/*

arch/x86/xen/enlighten.c

@@ -1181,10 +1181,11 @@ static const struct pv_cpu_ops xen_cpu_ops __initconst = {
 	.read_tscp = native_read_tscp,
 
 	.iret = xen_iret,
-	.irq_enable_sysexit = xen_sysexit,
 #ifdef CONFIG_X86_64
 	.usergs_sysret32 = xen_sysret32,
 	.usergs_sysret64 = xen_sysret64,
+#else
+	.irq_enable_sysexit = xen_sysexit,
 #endif
 
 	.load_tr_desc = paravirt_nop,

arch/x86/xen/xen-asm_64.S

@@ -15,6 +15,8 @@
 #include <asm/percpu.h>
 #include <asm/processor-flags.h>
 #include <asm/segment.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
 
 #include <xen/interface/xen.h>
 
@@ -47,29 +49,13 @@ ENTRY(xen_iret)
 ENDPATCH(xen_iret)
 RELOC(xen_iret, 1b+1)
 
-/*
- * sysexit is not used for 64-bit processes, so it's only ever used to
- * return to 32-bit compat userspace.
- */
-ENTRY(xen_sysexit)
-	pushq $__USER32_DS
-	pushq %rcx
-	pushq $X86_EFLAGS_IF
-	pushq $__USER32_CS
-	pushq %rdx
-
-	pushq $0
-1:	jmp hypercall_iret
-ENDPATCH(xen_sysexit)
-RELOC(xen_sysexit, 1b+1)
-
 ENTRY(xen_sysret64)
 	/*
 	 * We're already on the usermode stack at this point, but
 	 * still with the kernel gs, so we can easily switch back
 	 */
 	movq %rsp, PER_CPU_VAR(rsp_scratch)
-	movq PER_CPU_VAR(kernel_stack), %rsp
+	movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp
 
 	pushq $__USER_DS
 	pushq PER_CPU_VAR(rsp_scratch)
@@ -88,7 +74,7 @@ ENTRY(xen_sysret32)
 	 * still with the kernel gs, so we can easily switch back
 	 */
 	movq %rsp, PER_CPU_VAR(rsp_scratch)
-	movq PER_CPU_VAR(kernel_stack), %rsp
+	movq PER_CPU_VAR(cpu_current_top_of_stack), %rsp
 
 	pushq $__USER32_DS
 	pushq PER_CPU_VAR(rsp_scratch)

arch/x86/xen/xen-ops.h

@@ -134,7 +134,9 @@ DECL_ASM(void, xen_restore_fl_direct, unsigned long);
 
 /* These are not functions, and cannot be called normally */
 __visible void xen_iret(void);
+#ifdef CONFIG_X86_32
 __visible void xen_sysexit(void);
+#endif
 __visible void xen_sysret32(void);
 __visible void xen_sysret64(void);
 __visible void xen_adjust_exception_frame(void);

drivers/gpu/drm/drm_ioctl.c

@@ -36,9 +36,6 @@
 
 #include <linux/pci.h>
 #include <linux/export.h>
-#ifdef CONFIG_X86
-#include <asm/mtrr.h>
-#endif
 
 static int drm_version(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
@@ -197,16 +194,7 @@ static int drm_getmap(struct drm_device *dev, void *data,
 	map->type = r_list->map->type;
 	map->flags = r_list->map->flags;
 	map->handle = (void *)(unsigned long) r_list->user_token;
-
-#ifdef CONFIG_X86
-	/*
-	 * There appears to be exactly one user of the mtrr index: dritest.
-	 * It's easy enough to keep it working on non-PAT systems.
-	 */
-	map->mtrr = phys_wc_to_mtrr_index(r_list->map->mtrr);
-#else
-	map->mtrr = -1;
-#endif
+	map->mtrr = arch_phys_wc_index(r_list->map->mtrr);
 
 	mutex_unlock(&dev->struct_mutex);