All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ingo Molnar <mingo@kernel.org>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org,
	Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Suresh Siddha <suresh.b.siddha@intel.com>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: [GIT PULL] x86/platform changes for v3.6
Date: Sun, 22 Jul 2012 16:05:27 +0200	[thread overview]
Message-ID: <20120722140527.GA8313@gmail.com> (raw)

Linus,

Please pull the latest x86-platform-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-platform-for-linus

   HEAD: 36d93d88a5396baa135f8bcde7b8501dfe3b8e53 Revert "x86/early_printk: Replace obsolete simple_strtoul() usage with kstrtoint()"

This tree mostly involves various APIC driver 
cleanups/robustization, and vSMP motivated platform callback 
improvements/cleanups.

out-of-topic modifications in x86-platform-for-linus:
-----------------------------------------------------
drivers/iommu/intel_irq_remapping.c# 7eb9ae0: irq/apic: Use config_enabled(CONF
                                   # ed88bed: x86/apic/irq_remap: Silence a bog
                                   # ff16432: x86/apic: Make cpu_mask_to_apicid
drivers/iommu/irq_remapping.c      # 7eb9ae0: irq/apic: Use config_enabled(CONF
drivers/iommu/irq_remapping.h      # 7eb9ae0: irq/apic: Use config_enabled(CONF
include/linux/irq.h                # 7eb9ae0: irq/apic: Use config_enabled(CONF

 Thanks,

	Ingo

------------------>
Alexander Gordeev (14):
      x86/apic: Trivial whitespace fixes
      x86/apic: Factor out default target_cpus() operation
      x86/apic: Factor out default cpu_mask_to_apicid() operations
      x86/apic: Factor out default vector_allocation_domain() operation
      x86/apic: Try to spread IRQ vectors to different priority levels
      x86/apic: Avoid useless scanning thru a cpumask in assign_irq_vector()
      x86/apic: Make cpu_mask_to_apicid() operations return error code
      x86/apic: Make cpu_mask_to_apicid() operations check cpu_online_mask
      x86/x2apic/cluster: Vector_allocation_domain() should return a value
      x86/apic: Eliminate cpu_mask_to_apicid() operation
      x86/apic: Fix ugly casting and branching in cpu_mask_to_apicid_and()
      x86/apic/es7000+summit: Fix compile warning in cpu_mask_to_apicid()
      x86/apic/es7000+summit: Always make valid apicid from a cpumask
      x86/apic/es7000: Make apicid of a cluster (not CPU) from a cpumask

Dan Carpenter (1):
      x86/apic/irq_remap: Silence a bogus pr_err()

Feng Tang (1):
      Revert "x86/platform: Add a wallclock_init func to x86_platforms ops"

Ido Yariv (3):
      x86/platform: Introduce APIC post-initialization callback
      x86/vsmp: Fix linker error when CONFIG_PROC_FS is not set
      x86/vsmp: Fix vector_allocation_domain's return value

Ingo Molnar (1):
      Revert "x86/early_printk: Replace obsolete simple_strtoul() usage with kstrtoint()"

Jiang Liu (1):
      x86: Return IRQ_SET_MASK_OK_NOCOPY from irq affinity functions

Ravikiran Thirumalai (1):
      x86/vsmp: Ignore IOAPIC IRQ affinity if possible

Shuah Khan (1):
      x86/early_printk: Replace obsolete simple_strtoul() usage with kstrtoint()

Suresh Siddha (6):
      x86/irq: Update irq_cfg domain unless the new affinity is a subset of the current domain
      x86/x2apic/cluster: Use all the members of one cluster specified in the smp_affinity mask for the interrupt destination
      irq/apic: Use config_enabled(CONFIG_SMP) checks to clean up irq_set_affinity() for UP
      x86/apic: Optimize cpu traversal in __assign_irq_vector() using domain membership
      x86/apic/x2apic: Limit the vector reservation to the user specified mask
      x86/apic/x2apic: Use multiple cluster members for the irq destination only with the explicit affinity


 arch/x86/include/asm/apic.h           |   61 +++++--
 arch/x86/include/asm/x2apic.h         |   18 --
 arch/x86/include/asm/x86_init.h       |    4 +-
 arch/x86/kernel/apic/apic.c           |   19 ++
 arch/x86/kernel/apic/apic_flat_64.c   |   76 +-------
 arch/x86/kernel/apic/apic_noop.c      |    9 +-
 arch/x86/kernel/apic/apic_numachip.c  |   50 +-----
 arch/x86/kernel/apic/bigsmp_32.c      |   48 +----
 arch/x86/kernel/apic/es7000_32.c      |   51 ++----
 arch/x86/kernel/apic/io_apic.c        |  312 +++++++++++++++++----------------
 arch/x86/kernel/apic/numaq_32.c       |   30 +---
 arch/x86/kernel/apic/probe_32.c       |   23 +--
 arch/x86/kernel/apic/probe_64.c       |   11 +-
 arch/x86/kernel/apic/summit_32.c      |   46 ++---
 arch/x86/kernel/apic/x2apic_cluster.c |   82 ++++++---
 arch/x86/kernel/apic/x2apic_phys.c    |   39 +----
 arch/x86/kernel/apic/x2apic_uv_x.c    |   45 ++---
 arch/x86/kernel/setup.c               |    2 -
 arch/x86/kernel/vsmp_64.c             |   44 ++++-
 arch/x86/kernel/x86_init.c            |    2 -
 arch/x86/platform/uv/uv_irq.c         |    9 +-
 drivers/iommu/intel_irq_remapping.c   |   20 ++-
 drivers/iommu/irq_remapping.c         |    5 +-
 drivers/iommu/irq_remapping.h         |    2 -
 include/linux/irq.h                   |    2 -
 25 files changed, 429 insertions(+), 581 deletions(-)

diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index eaff479..88093c1 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -306,7 +306,8 @@ struct apic {
 	unsigned long (*check_apicid_used)(physid_mask_t *map, int apicid);
 	unsigned long (*check_apicid_present)(int apicid);
 
-	void (*vector_allocation_domain)(int cpu, struct cpumask *retmask);
+	void (*vector_allocation_domain)(int cpu, struct cpumask *retmask,
+					 const struct cpumask *mask);
 	void (*init_apic_ldr)(void);
 
 	void (*ioapic_phys_id_map)(physid_mask_t *phys_map, physid_mask_t *retmap);
@@ -331,9 +332,9 @@ struct apic {
 	unsigned long (*set_apic_id)(unsigned int id);
 	unsigned long apic_id_mask;
 
-	unsigned int (*cpu_mask_to_apicid)(const struct cpumask *cpumask);
-	unsigned int (*cpu_mask_to_apicid_and)(const struct cpumask *cpumask,
-					       const struct cpumask *andmask);
+	int (*cpu_mask_to_apicid_and)(const struct cpumask *cpumask,
+				      const struct cpumask *andmask,
+				      unsigned int *apicid);
 
 	/* ipi */
 	void (*send_IPI_mask)(const struct cpumask *mask, int vector);
@@ -537,6 +538,11 @@ static inline const struct cpumask *default_target_cpus(void)
 #endif
 }
 
+static inline const struct cpumask *online_target_cpus(void)
+{
+	return cpu_online_mask;
+}
+
 DECLARE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid);
 
 
@@ -586,21 +592,50 @@ static inline int default_phys_pkg_id(int cpuid_apic, int index_msb)
 
 #endif
 
-static inline unsigned int
-default_cpu_mask_to_apicid(const struct cpumask *cpumask)
+static inline int
+flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+			    const struct cpumask *andmask,
+			    unsigned int *apicid)
 {
-	return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS;
+	unsigned long cpu_mask = cpumask_bits(cpumask)[0] &
+				 cpumask_bits(andmask)[0] &
+				 cpumask_bits(cpu_online_mask)[0] &
+				 APIC_ALL_CPUS;
+
+	if (likely(cpu_mask)) {
+		*apicid = (unsigned int)cpu_mask;
+		return 0;
+	} else {
+		return -EINVAL;
+	}
 }
 
-static inline unsigned int
+extern int
 default_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			       const struct cpumask *andmask)
+			       const struct cpumask *andmask,
+			       unsigned int *apicid);
+
+static inline void
+flat_vector_allocation_domain(int cpu, struct cpumask *retmask,
+			      const struct cpumask *mask)
 {
-	unsigned long mask1 = cpumask_bits(cpumask)[0];
-	unsigned long mask2 = cpumask_bits(andmask)[0];
-	unsigned long mask3 = cpumask_bits(cpu_online_mask)[0];
+	/* Careful. Some cpus do not strictly honor the set of cpus
+	 * specified in the interrupt destination when using lowest
+	 * priority interrupt delivery mode.
+	 *
+	 * In particular there was a hyperthreading cpu observed to
+	 * deliver interrupts to the wrong hyperthread when only one
+	 * hyperthread was specified in the interrupt desitination.
+	 */
+	cpumask_clear(retmask);
+	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
+}
 
-	return (unsigned int)(mask1 & mask2 & mask3);
+static inline void
+default_vector_allocation_domain(int cpu, struct cpumask *retmask,
+				 const struct cpumask *mask)
+{
+	cpumask_copy(retmask, cpumask_of(cpu));
 }
 
 static inline unsigned long default_check_apicid_used(physid_mask_t *map, int apicid)
diff --git a/arch/x86/include/asm/x2apic.h b/arch/x86/include/asm/x2apic.h
index 92e54ab..f90f0a5 100644
--- a/arch/x86/include/asm/x2apic.h
+++ b/arch/x86/include/asm/x2apic.h
@@ -9,15 +9,6 @@
 #include <asm/ipi.h>
 #include <linux/cpumask.h>
 
-/*
- * Need to use more than cpu 0, because we need more vectors
- * when MSI-X are used.
- */
-static const struct cpumask *x2apic_target_cpus(void)
-{
-	return cpu_online_mask;
-}
-
 static int x2apic_apic_id_valid(int apicid)
 {
 	return 1;
@@ -28,15 +19,6 @@ static int x2apic_apic_id_registered(void)
 	return 1;
 }
 
-/*
- * For now each logical cpu is in its own vector allocation domain.
- */
-static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
-}
-
 static void
 __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest)
 {
diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
index c090af1..38155f6 100644
--- a/arch/x86/include/asm/x86_init.h
+++ b/arch/x86/include/asm/x86_init.h
@@ -156,7 +156,6 @@ struct x86_cpuinit_ops {
 /**
  * struct x86_platform_ops - platform specific runtime functions
  * @calibrate_tsc:		calibrate TSC
- * @wallclock_init:		init the wallclock device
  * @get_wallclock:		get time from HW clock like RTC etc.
  * @set_wallclock:		set time back to HW clock
  * @is_untracked_pat_range	exclude from PAT logic
@@ -164,10 +163,10 @@ struct x86_cpuinit_ops {
  * @i8042_detect		pre-detect if i8042 controller exists
  * @save_sched_clock_state:	save state for sched_clock() on suspend
  * @restore_sched_clock_state:	restore state for sched_clock() on resume
+ * @apic_post_init:		adjust apic if neeeded
  */
 struct x86_platform_ops {
 	unsigned long (*calibrate_tsc)(void);
-	void (*wallclock_init)(void);
 	unsigned long (*get_wallclock)(void);
 	int (*set_wallclock)(unsigned long nowtime);
 	void (*iommu_shutdown)(void);
@@ -177,6 +176,7 @@ struct x86_platform_ops {
 	int (*i8042_detect)(void);
 	void (*save_sched_clock_state)(void);
 	void (*restore_sched_clock_state)(void);
+	void (*apic_post_init)(void);
 };
 
 struct pci_dev;
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index 39a222e..c421512 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -2123,6 +2123,25 @@ void default_init_apic_ldr(void)
 	apic_write(APIC_LDR, val);
 }
 
+int default_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+				   const struct cpumask *andmask,
+				   unsigned int *apicid)
+{
+	unsigned int cpu;
+
+	for_each_cpu_and(cpu, cpumask, andmask) {
+		if (cpumask_test_cpu(cpu, cpu_online_mask))
+			break;
+	}
+
+	if (likely(cpu < nr_cpu_ids)) {
+		*apicid = per_cpu(x86_cpu_to_apicid, cpu);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
 /*
  * Power management
  */
diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c
index 0e881c4..00c77cf 100644
--- a/arch/x86/kernel/apic/apic_flat_64.c
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -36,25 +36,6 @@ static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 1;
 }
 
-static const struct cpumask *flat_target_cpus(void)
-{
-	return cpu_online_mask;
-}
-
-static void flat_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	/* Careful. Some cpus do not strictly honor the set of cpus
-	 * specified in the interrupt destination when using lowest
-	 * priority interrupt delivery mode.
-	 *
-	 * In particular there was a hyperthreading cpu observed to
-	 * deliver interrupts to the wrong hyperthread when only one
-	 * hyperthread was specified in the interrupt desitination.
-	 */
-	cpumask_clear(retmask);
-	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
-}
-
 /*
  * Set up the logical destination ID.
  *
@@ -92,7 +73,7 @@ static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector)
 }
 
 static void
- flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector)
+flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector)
 {
 	unsigned long mask = cpumask_bits(cpumask)[0];
 	int cpu = smp_processor_id();
@@ -186,7 +167,7 @@ static struct apic apic_flat =  {
 	.irq_delivery_mode		= dest_LowestPrio,
 	.irq_dest_mode			= 1, /* logical */
 
-	.target_cpus			= flat_target_cpus,
+	.target_cpus			= online_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= APIC_DEST_LOGICAL,
 	.check_apicid_used		= NULL,
@@ -210,8 +191,7 @@ static struct apic apic_flat =  {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xFFu << 24,
 
-	.cpu_mask_to_apicid		= default_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= flat_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= flat_send_IPI_mask,
 	.send_IPI_mask_allbutself	= flat_send_IPI_mask_allbutself,
@@ -262,17 +242,6 @@ static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;
 }
 
-static const struct cpumask *physflat_target_cpus(void)
-{
-	return cpu_online_mask;
-}
-
-static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
-}
-
 static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector)
 {
 	default_send_IPI_mask_sequence_phys(cpumask, vector);
@@ -294,38 +263,6 @@ static void physflat_send_IPI_all(int vector)
 	physflat_send_IPI_mask(cpu_online_mask, vector);
 }
 
-static unsigned int physflat_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	cpu = cpumask_first(cpumask);
-	if ((unsigned)cpu < nr_cpu_ids)
-		return per_cpu(x86_cpu_to_apicid, cpu);
-	else
-		return BAD_APICID;
-}
-
-static unsigned int
-physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-				const struct cpumask *andmask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	for_each_cpu_and(cpu, cpumask, andmask) {
-		if (cpumask_test_cpu(cpu, cpu_online_mask))
-			break;
-	}
-	return per_cpu(x86_cpu_to_apicid, cpu);
-}
-
 static int physflat_probe(void)
 {
 	if (apic == &apic_physflat || num_possible_cpus() > 8)
@@ -345,13 +282,13 @@ static struct apic apic_physflat =  {
 	.irq_delivery_mode		= dest_Fixed,
 	.irq_dest_mode			= 0, /* physical */
 
-	.target_cpus			= physflat_target_cpus,
+	.target_cpus			= online_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= 0,
 	.check_apicid_used		= NULL,
 	.check_apicid_present		= NULL,
 
-	.vector_allocation_domain	= physflat_vector_allocation_domain,
+	.vector_allocation_domain	= default_vector_allocation_domain,
 	/* not needed, but shouldn't hurt: */
 	.init_apic_ldr			= flat_init_apic_ldr,
 
@@ -370,8 +307,7 @@ static struct apic apic_physflat =  {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xFFu << 24,
 
-	.cpu_mask_to_apicid		= physflat_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= physflat_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= physflat_send_IPI_mask,
 	.send_IPI_mask_allbutself	= physflat_send_IPI_mask_allbutself,
diff --git a/arch/x86/kernel/apic/apic_noop.c b/arch/x86/kernel/apic/apic_noop.c
index a6e4c6e..e145f28 100644
--- a/arch/x86/kernel/apic/apic_noop.c
+++ b/arch/x86/kernel/apic/apic_noop.c
@@ -100,12 +100,12 @@ static unsigned long noop_check_apicid_present(int bit)
 	return physid_isset(bit, phys_cpu_present_map);
 }
 
-static void noop_vector_allocation_domain(int cpu, struct cpumask *retmask)
+static void noop_vector_allocation_domain(int cpu, struct cpumask *retmask,
+					  const struct cpumask *mask)
 {
 	if (cpu != 0)
 		pr_warning("APIC: Vector allocated for non-BSP cpu\n");
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
+	cpumask_copy(retmask, cpumask_of(cpu));
 }
 
 static u32 noop_apic_read(u32 reg)
@@ -159,8 +159,7 @@ struct apic apic_noop = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0x0F << 24,
 
-	.cpu_mask_to_apicid		= default_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= flat_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= noop_send_IPI_mask,
 	.send_IPI_mask_allbutself	= noop_send_IPI_mask_allbutself,
diff --git a/arch/x86/kernel/apic/apic_numachip.c b/arch/x86/kernel/apic/apic_numachip.c
index 6ec6d5d..bc552cf 100644
--- a/arch/x86/kernel/apic/apic_numachip.c
+++ b/arch/x86/kernel/apic/apic_numachip.c
@@ -72,17 +72,6 @@ static int numachip_phys_pkg_id(int initial_apic_id, int index_msb)
 	return initial_apic_id >> index_msb;
 }
 
-static const struct cpumask *numachip_target_cpus(void)
-{
-	return cpu_online_mask;
-}
-
-static void numachip_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
-}
-
 static int __cpuinit numachip_wakeup_secondary(int phys_apicid, unsigned long start_rip)
 {
 	union numachip_csr_g3_ext_irq_gen int_gen;
@@ -157,38 +146,6 @@ static void numachip_send_IPI_self(int vector)
 	__default_send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL);
 }
 
-static unsigned int numachip_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	cpu = cpumask_first(cpumask);
-	if (likely((unsigned)cpu < nr_cpu_ids))
-		return per_cpu(x86_cpu_to_apicid, cpu);
-
-	return BAD_APICID;
-}
-
-static unsigned int
-numachip_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-				const struct cpumask *andmask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	for_each_cpu_and(cpu, cpumask, andmask) {
-		if (cpumask_test_cpu(cpu, cpu_online_mask))
-			break;
-	}
-	return per_cpu(x86_cpu_to_apicid, cpu);
-}
-
 static int __init numachip_probe(void)
 {
 	return apic == &apic_numachip;
@@ -253,13 +210,13 @@ static struct apic apic_numachip __refconst = {
 	.irq_delivery_mode		= dest_Fixed,
 	.irq_dest_mode			= 0, /* physical */
 
-	.target_cpus			= numachip_target_cpus,
+	.target_cpus			= online_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= 0,
 	.check_apicid_used		= NULL,
 	.check_apicid_present		= NULL,
 
-	.vector_allocation_domain	= numachip_vector_allocation_domain,
+	.vector_allocation_domain	= default_vector_allocation_domain,
 	.init_apic_ldr			= flat_init_apic_ldr,
 
 	.ioapic_phys_id_map		= NULL,
@@ -277,8 +234,7 @@ static struct apic apic_numachip __refconst = {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xffU << 24,
 
-	.cpu_mask_to_apicid		= numachip_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= numachip_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= numachip_send_IPI_mask,
 	.send_IPI_mask_allbutself	= numachip_send_IPI_mask_allbutself,
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
index 31fbdbf..d50e364 100644
--- a/arch/x86/kernel/apic/bigsmp_32.c
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -26,15 +26,6 @@ static int bigsmp_apic_id_registered(void)
 	return 1;
 }
 
-static const struct cpumask *bigsmp_target_cpus(void)
-{
-#ifdef CONFIG_SMP
-	return cpu_online_mask;
-#else
-	return cpumask_of(0);
-#endif
-}
-
 static unsigned long bigsmp_check_apicid_used(physid_mask_t *map, int apicid)
 {
 	return 0;
@@ -105,32 +96,6 @@ static int bigsmp_check_phys_apicid_present(int phys_apicid)
 	return 1;
 }
 
-/* As we are using single CPU as destination, pick only one CPU here */
-static unsigned int bigsmp_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	int cpu = cpumask_first(cpumask);
-
-	if (cpu < nr_cpu_ids)
-		return cpu_physical_id(cpu);
-	return BAD_APICID;
-}
-
-static unsigned int bigsmp_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			      const struct cpumask *andmask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	for_each_cpu_and(cpu, cpumask, andmask) {
-		if (cpumask_test_cpu(cpu, cpu_online_mask))
-			return cpu_physical_id(cpu);
-	}
-	return BAD_APICID;
-}
-
 static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
 {
 	return cpuid_apic >> index_msb;
@@ -177,12 +142,6 @@ static const struct dmi_system_id bigsmp_dmi_table[] = {
 	{ } /* NULL entry stops DMI scanning */
 };
 
-static void bigsmp_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
-}
-
 static int probe_bigsmp(void)
 {
 	if (def_to_bigsmp)
@@ -205,13 +164,13 @@ static struct apic apic_bigsmp = {
 	/* phys delivery to target CPU: */
 	.irq_dest_mode			= 0,
 
-	.target_cpus			= bigsmp_target_cpus,
+	.target_cpus			= default_target_cpus,
 	.disable_esr			= 1,
 	.dest_logical			= 0,
 	.check_apicid_used		= bigsmp_check_apicid_used,
 	.check_apicid_present		= bigsmp_check_apicid_present,
 
-	.vector_allocation_domain	= bigsmp_vector_allocation_domain,
+	.vector_allocation_domain	= default_vector_allocation_domain,
 	.init_apic_ldr			= bigsmp_init_apic_ldr,
 
 	.ioapic_phys_id_map		= bigsmp_ioapic_phys_id_map,
@@ -229,8 +188,7 @@ static struct apic apic_bigsmp = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0xFF << 24,
 
-	.cpu_mask_to_apicid		= bigsmp_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= bigsmp_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= bigsmp_send_IPI_mask,
 	.send_IPI_mask_allbutself	= NULL,
diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c
index db4ab1b..0874799 100644
--- a/arch/x86/kernel/apic/es7000_32.c
+++ b/arch/x86/kernel/apic/es7000_32.c
@@ -394,21 +394,6 @@ static void es7000_enable_apic_mode(void)
 		WARN(1, "Command failed, status = %x\n", mip_status);
 }
 
-static void es7000_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	/* Careful. Some cpus do not strictly honor the set of cpus
-	 * specified in the interrupt destination when using lowest
-	 * priority interrupt delivery mode.
-	 *
-	 * In particular there was a hyperthreading cpu observed to
-	 * deliver interrupts to the wrong hyperthread when only one
-	 * hyperthread was specified in the interrupt desitination.
-	 */
-	cpumask_clear(retmask);
-	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
-}
-
-
 static void es7000_wait_for_init_deassert(atomic_t *deassert)
 {
 	while (!atomic_read(deassert))
@@ -540,45 +525,49 @@ static int es7000_check_phys_apicid_present(int cpu_physical_apicid)
 	return 1;
 }
 
-static unsigned int es7000_cpu_mask_to_apicid(const struct cpumask *cpumask)
+static inline int
+es7000_cpu_mask_to_apicid(const struct cpumask *cpumask, unsigned int *dest_id)
 {
 	unsigned int round = 0;
-	int cpu, uninitialized_var(apicid);
+	unsigned int cpu, uninitialized_var(apicid);
 
 	/*
 	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	for_each_cpu(cpu, cpumask) {
+	for_each_cpu_and(cpu, cpumask, cpu_online_mask) {
 		int new_apicid = early_per_cpu(x86_cpu_to_logical_apicid, cpu);
 
 		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
 			WARN(1, "Not a valid mask!");
 
-			return BAD_APICID;
+			return -EINVAL;
 		}
-		apicid = new_apicid;
+		apicid |= new_apicid;
 		round++;
 	}
-	return apicid;
+	if (!round)
+		return -EINVAL;
+	*dest_id = apicid;
+	return 0;
 }
 
-static unsigned int
+static int
 es7000_cpu_mask_to_apicid_and(const struct cpumask *inmask,
-			      const struct cpumask *andmask)
+			      const struct cpumask *andmask,
+			      unsigned int *apicid)
 {
-	int apicid = early_per_cpu(x86_cpu_to_logical_apicid, 0);
 	cpumask_var_t cpumask;
+	*apicid = early_per_cpu(x86_cpu_to_logical_apicid, 0);
 
 	if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC))
-		return apicid;
+		return 0;
 
 	cpumask_and(cpumask, inmask, andmask);
-	cpumask_and(cpumask, cpumask, cpu_online_mask);
-	apicid = es7000_cpu_mask_to_apicid(cpumask);
+	es7000_cpu_mask_to_apicid(cpumask, apicid);
 
 	free_cpumask_var(cpumask);
 
-	return apicid;
+	return 0;
 }
 
 static int es7000_phys_pkg_id(int cpuid_apic, int index_msb)
@@ -638,7 +627,7 @@ static struct apic __refdata apic_es7000_cluster = {
 	.check_apicid_used		= es7000_check_apicid_used,
 	.check_apicid_present		= es7000_check_apicid_present,
 
-	.vector_allocation_domain	= es7000_vector_allocation_domain,
+	.vector_allocation_domain	= flat_vector_allocation_domain,
 	.init_apic_ldr			= es7000_init_apic_ldr_cluster,
 
 	.ioapic_phys_id_map		= es7000_ioapic_phys_id_map,
@@ -656,7 +645,6 @@ static struct apic __refdata apic_es7000_cluster = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0xFF << 24,
 
-	.cpu_mask_to_apicid		= es7000_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= es7000_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= es7000_send_IPI_mask,
@@ -705,7 +693,7 @@ static struct apic __refdata apic_es7000 = {
 	.check_apicid_used		= es7000_check_apicid_used,
 	.check_apicid_present		= es7000_check_apicid_present,
 
-	.vector_allocation_domain	= es7000_vector_allocation_domain,
+	.vector_allocation_domain	= flat_vector_allocation_domain,
 	.init_apic_ldr			= es7000_init_apic_ldr,
 
 	.ioapic_phys_id_map		= es7000_ioapic_phys_id_map,
@@ -723,7 +711,6 @@ static struct apic __refdata apic_es7000 = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0xFF << 24,
 
-	.cpu_mask_to_apicid		= es7000_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= es7000_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= es7000_send_IPI_mask,
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index ac96561..9684f96 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1112,8 +1112,7 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
 	 */
 	static int current_vector = FIRST_EXTERNAL_VECTOR + VECTOR_OFFSET_START;
-	static int current_offset = VECTOR_OFFSET_START % 8;
-	unsigned int old_vector;
+	static int current_offset = VECTOR_OFFSET_START % 16;
 	int cpu, err;
 	cpumask_var_t tmp_mask;
 
@@ -1123,35 +1122,45 @@ __assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask)
 	if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC))
 		return -ENOMEM;
 
-	old_vector = cfg->vector;
-	if (old_vector) {
-		cpumask_and(tmp_mask, mask, cpu_online_mask);
-		cpumask_and(tmp_mask, cfg->domain, tmp_mask);
-		if (!cpumask_empty(tmp_mask)) {
-			free_cpumask_var(tmp_mask);
-			return 0;
-		}
-	}
-
 	/* Only try and allocate irqs on cpus that are present */
 	err = -ENOSPC;
-	for_each_cpu_and(cpu, mask, cpu_online_mask) {
-		int new_cpu;
-		int vector, offset;
+	cpumask_clear(cfg->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);
+		apic->vector_allocation_domain(cpu, tmp_mask, mask);
+
+		if (cpumask_subset(tmp_mask, cfg->domain)) {
+			err = 0;
+			if (cpumask_equal(tmp_mask, cfg->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 = 1;
+			cpumask_and(cfg->domain, cfg->domain, tmp_mask);
+			break;
+		}
 
 		vector = current_vector;
 		offset = current_offset;
 next:
-		vector += 8;
+		vector += 16;
 		if (vector >= first_system_vector) {
-			/* If out of vectors on large boxen, must share them. */
-			offset = (offset + 1) % 8;
+			offset = (offset + 1) % 16;
 			vector = FIRST_EXTERNAL_VECTOR + offset;
 		}
-		if (unlikely(current_vector == vector))
+
+		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);
 			continue;
+		}
 
 		if (test_bit(vector, used_vectors))
 			goto next;
@@ -1162,7 +1171,7 @@ next:
 		/* Found one! */
 		current_vector = vector;
 		current_offset = offset;
-		if (old_vector) {
+		if (cfg->vector) {
 			cfg->move_in_progress = 1;
 			cpumask_copy(cfg->old_domain, cfg->domain);
 		}
@@ -1346,18 +1355,18 @@ static void setup_ioapic_irq(unsigned int irq, struct irq_cfg *cfg,
 
 	if (!IO_APIC_IRQ(irq))
 		return;
-	/*
-	 * For legacy irqs, cfg->domain starts with cpu 0 for legacy
-	 * controllers like 8259. Now that IO-APIC can handle this irq, update
-	 * the cfg->domain.
-	 */
-	if (irq < legacy_pic->nr_legacy_irqs && cpumask_test_cpu(0, cfg->domain))
-		apic->vector_allocation_domain(0, cfg->domain);
 
 	if (assign_irq_vector(irq, cfg, apic->target_cpus()))
 		return;
 
-	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
+	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 -> "
@@ -1366,7 +1375,7 @@ static void setup_ioapic_irq(unsigned int irq, struct irq_cfg *cfg,
 		    cfg->vector, irq, attr->trigger, attr->polarity, dest);
 
 	if (setup_ioapic_entry(irq, &entry, dest, cfg->vector, attr)) {
-		pr_warn("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
+		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);
 
@@ -1469,9 +1478,10 @@ void setup_IO_APIC_irq_extra(u32 gsi)
  * 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)
+					unsigned int pin, int vector)
 {
 	struct IO_APIC_route_entry entry;
+	unsigned int dest;
 
 	if (irq_remapping_enabled)
 		return;
@@ -1482,9 +1492,13 @@ static void __init setup_timer_IRQ0_pin(unsigned int ioapic_idx,
 	 * 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 = apic->cpu_mask_to_apicid(apic->target_cpus());
+	entry.dest = dest;
 	entry.delivery_mode = apic->irq_delivery_mode;
 	entry.polarity = 0;
 	entry.trigger = 0;
@@ -2210,71 +2224,6 @@ void send_cleanup_vector(struct irq_cfg *cfg)
 	cfg->move_in_progress = 0;
 }
 
-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;
-		/*
-		 * With interrupt-remapping, destination information comes
-		 * from interrupt-remapping table entry.
-		 */
-		if (!irq_remapped(cfg))
-			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);
-	}
-}
-
-/*
- * 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 __ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
-			  unsigned int *dest_id)
-{
-	struct irq_cfg *cfg = data->chip_data;
-
-	if (!cpumask_intersects(mask, cpu_online_mask))
-		return -1;
-
-	if (assign_irq_vector(data->irq, data->chip_data, mask))
-		return -1;
-
-	cpumask_copy(data->affinity, mask);
-
-	*dest_id = apic->cpu_mask_to_apicid_and(mask, cfg->domain);
-	return 0;
-}
-
-static int
-ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
-		    bool force)
-{
-	unsigned int dest, irq = data->irq;
-	unsigned long flags;
-	int ret;
-
-	raw_spin_lock_irqsave(&ioapic_lock, flags);
-	ret = __ioapic_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, data->chip_data);
-	}
-	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
-	return ret;
-}
-
 asmlinkage void smp_irq_move_cleanup_interrupt(void)
 {
 	unsigned vector, me;
@@ -2362,6 +2311,87 @@ void irq_force_complete_move(int irq)
 static inline void irq_complete_move(struct irq_cfg *cfg) { }
 #endif
 
+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;
+		/*
+		 * With interrupt-remapping, destination information comes
+		 * from interrupt-remapping table entry.
+		 */
+		if (!irq_remapped(cfg))
+			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);
+	}
+}
+
+/*
+ * 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 __ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
+			  unsigned int *dest_id)
+{
+	struct irq_cfg *cfg = data->chip_data;
+	unsigned int irq = data->irq;
+	int err;
+
+	if (!config_enabled(CONFIG_SMP))
+		return -1;
+
+	if (!cpumask_intersects(mask, 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);
+	if (err) {
+		if (assign_irq_vector(irq, cfg, data->affinity))
+			pr_err("Failed to recover vector for irq %d\n", irq);
+		return err;
+	}
+
+	cpumask_copy(data->affinity, mask);
+
+	return 0;
+}
+
+static int
+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 -1;
+
+	raw_spin_lock_irqsave(&ioapic_lock, flags);
+	ret = __ioapic_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, data->chip_data);
+		ret = IRQ_SET_MASK_OK_NOCOPY;
+	}
+	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+	return ret;
+}
+
 static void ack_apic_edge(struct irq_data *data)
 {
 	irq_complete_move(data->chip_data);
@@ -2541,9 +2571,7 @@ static void irq_remap_modify_chip_defaults(struct irq_chip *chip)
 	chip->irq_ack = ir_ack_apic_edge;
 	chip->irq_eoi = ir_ack_apic_level;
 
-#ifdef CONFIG_SMP
 	chip->irq_set_affinity = set_remapped_irq_affinity;
-#endif
 }
 #endif /* CONFIG_IRQ_REMAP */
 
@@ -2554,9 +2582,7 @@ static struct irq_chip ioapic_chip __read_mostly = {
 	.irq_unmask		= unmask_ioapic_irq,
 	.irq_ack		= ack_apic_edge,
 	.irq_eoi		= ack_apic_level,
-#ifdef CONFIG_SMP
 	.irq_set_affinity	= ioapic_set_affinity,
-#endif
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
@@ -3038,7 +3064,10 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq,
 	if (err)
 		return err;
 
-	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());
+	err = apic->cpu_mask_to_apicid_and(cfg->domain,
+					   apic->target_cpus(), &dest);
+	if (err)
+		return err;
 
 	if (irq_remapped(cfg)) {
 		compose_remapped_msi_msg(pdev, irq, dest, msg, hpet_id);
@@ -3072,7 +3101,6 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq,
 	return err;
 }
 
-#ifdef CONFIG_SMP
 static int
 msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 {
@@ -3092,9 +3120,8 @@ msi_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 
 	__write_msi_msg(data->msi_desc, &msg);
 
-	return 0;
+	return IRQ_SET_MASK_OK_NOCOPY;
 }
-#endif /* CONFIG_SMP */
 
 /*
  * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices,
@@ -3105,9 +3132,7 @@ static struct irq_chip msi_chip = {
 	.irq_unmask		= unmask_msi_irq,
 	.irq_mask		= mask_msi_irq,
 	.irq_ack		= ack_apic_edge,
-#ifdef CONFIG_SMP
 	.irq_set_affinity	= msi_set_affinity,
-#endif
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
@@ -3192,7 +3217,6 @@ void native_teardown_msi_irq(unsigned int irq)
 }
 
 #ifdef CONFIG_DMAR_TABLE
-#ifdef CONFIG_SMP
 static int
 dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
 		      bool force)
@@ -3214,19 +3238,15 @@ dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
 
 	dmar_msi_write(irq, &msg);
 
-	return 0;
+	return IRQ_SET_MASK_OK_NOCOPY;
 }
 
-#endif /* CONFIG_SMP */
-
 static struct irq_chip dmar_msi_type = {
 	.name			= "DMAR_MSI",
 	.irq_unmask		= dmar_msi_unmask,
 	.irq_mask		= dmar_msi_mask,
 	.irq_ack		= ack_apic_edge,
-#ifdef CONFIG_SMP
 	.irq_set_affinity	= dmar_msi_set_affinity,
-#endif
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
@@ -3247,7 +3267,6 @@ int arch_setup_dmar_msi(unsigned int irq)
 
 #ifdef CONFIG_HPET_TIMER
 
-#ifdef CONFIG_SMP
 static int hpet_msi_set_affinity(struct irq_data *data,
 				 const struct cpumask *mask, bool force)
 {
@@ -3267,19 +3286,15 @@ static int hpet_msi_set_affinity(struct irq_data *data,
 
 	hpet_msi_write(data->handler_data, &msg);
 
-	return 0;
+	return IRQ_SET_MASK_OK_NOCOPY;
 }
 
-#endif /* CONFIG_SMP */
-
 static struct irq_chip hpet_msi_type = {
 	.name = "HPET_MSI",
 	.irq_unmask = hpet_msi_unmask,
 	.irq_mask = hpet_msi_mask,
 	.irq_ack = ack_apic_edge,
-#ifdef CONFIG_SMP
 	.irq_set_affinity = hpet_msi_set_affinity,
-#endif
 	.irq_retrigger = ioapic_retrigger_irq,
 };
 
@@ -3314,8 +3329,6 @@ int arch_setup_hpet_msi(unsigned int irq, unsigned int id)
  */
 #ifdef CONFIG_HT_IRQ
 
-#ifdef CONFIG_SMP
-
 static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
 {
 	struct ht_irq_msg msg;
@@ -3340,25 +3353,23 @@ ht_set_affinity(struct irq_data *data, const struct cpumask *mask, bool force)
 		return -1;
 
 	target_ht_irq(data->irq, dest, cfg->vector);
-	return 0;
+	return IRQ_SET_MASK_OK_NOCOPY;
 }
 
-#endif
-
 static struct irq_chip ht_irq_chip = {
 	.name			= "PCI-HT",
 	.irq_mask		= mask_ht_irq,
 	.irq_unmask		= unmask_ht_irq,
 	.irq_ack		= ack_apic_edge,
-#ifdef CONFIG_SMP
 	.irq_set_affinity	= ht_set_affinity,
-#endif
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
 int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 {
 	struct irq_cfg *cfg;
+	struct ht_irq_msg msg;
+	unsigned dest;
 	int err;
 
 	if (disable_apic)
@@ -3366,36 +3377,37 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
 
 	cfg = irq_cfg(irq);
 	err = assign_irq_vector(irq, cfg, apic->target_cpus());
-	if (!err) {
-		struct ht_irq_msg msg;
-		unsigned dest;
+	if (err)
+		return err;
+
+	err = apic->cpu_mask_to_apicid_and(cfg->domain,
+					   apic->target_cpus(), &dest);
+	if (err)
+		return err;
 
-		dest = apic->cpu_mask_to_apicid_and(cfg->domain,
-						    apic->target_cpus());
+	msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
 
-		msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);
+	msg.address_lo =
+		HT_IRQ_LOW_BASE |
+		HT_IRQ_LOW_DEST_ID(dest) |
+		HT_IRQ_LOW_VECTOR(cfg->vector) |
+		((apic->irq_dest_mode == 0) ?
+			HT_IRQ_LOW_DM_PHYSICAL :
+			HT_IRQ_LOW_DM_LOGICAL) |
+		HT_IRQ_LOW_RQEOI_EDGE |
+		((apic->irq_delivery_mode != dest_LowestPrio) ?
+			HT_IRQ_LOW_MT_FIXED :
+			HT_IRQ_LOW_MT_ARBITRATED) |
+		HT_IRQ_LOW_IRQ_MASKED;
 
-		msg.address_lo =
-			HT_IRQ_LOW_BASE |
-			HT_IRQ_LOW_DEST_ID(dest) |
-			HT_IRQ_LOW_VECTOR(cfg->vector) |
-			((apic->irq_dest_mode == 0) ?
-				HT_IRQ_LOW_DM_PHYSICAL :
-				HT_IRQ_LOW_DM_LOGICAL) |
-			HT_IRQ_LOW_RQEOI_EDGE |
-			((apic->irq_delivery_mode != dest_LowestPrio) ?
-				HT_IRQ_LOW_MT_FIXED :
-				HT_IRQ_LOW_MT_ARBITRATED) |
-			HT_IRQ_LOW_IRQ_MASKED;
+	write_ht_irq_msg(irq, &msg);
 
-		write_ht_irq_msg(irq, &msg);
+	irq_set_chip_and_handler_name(irq, &ht_irq_chip,
+				      handle_edge_irq, "edge");
 
-		irq_set_chip_and_handler_name(irq, &ht_irq_chip,
-					      handle_edge_irq, "edge");
+	dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq);
 
-		dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq);
-	}
-	return err;
+	return 0;
 }
 #endif /* CONFIG_HT_IRQ */
 
diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c
index f00a68c..d661ee9 100644
--- a/arch/x86/kernel/apic/numaq_32.c
+++ b/arch/x86/kernel/apic/numaq_32.c
@@ -406,16 +406,13 @@ static inline int numaq_check_phys_apicid_present(int phys_apicid)
  * We use physical apicids here, not logical, so just return the default
  * physical broadcast to stop people from breaking us
  */
-static unsigned int numaq_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	return 0x0F;
-}
-
-static inline unsigned int
+static int
 numaq_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			     const struct cpumask *andmask)
+			     const struct cpumask *andmask,
+			     unsigned int *apicid)
 {
-	return 0x0F;
+	*apicid = 0x0F;
+	return 0;
 }
 
 /* No NUMA-Q box has a HT CPU, but it can't hurt to use the default code. */
@@ -441,20 +438,6 @@ static int probe_numaq(void)
 	return found_numaq;
 }
 
-static void numaq_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	/* Careful. Some cpus do not strictly honor the set of cpus
-	 * specified in the interrupt destination when using lowest
-	 * priority interrupt delivery mode.
-	 *
-	 * In particular there was a hyperthreading cpu observed to
-	 * deliver interrupts to the wrong hyperthread when only one
-	 * hyperthread was specified in the interrupt desitination.
-	 */
-	cpumask_clear(retmask);
-	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
-}
-
 static void numaq_setup_portio_remap(void)
 {
 	int num_quads = num_online_nodes();
@@ -491,7 +474,7 @@ static struct apic __refdata apic_numaq = {
 	.check_apicid_used		= numaq_check_apicid_used,
 	.check_apicid_present		= numaq_check_apicid_present,
 
-	.vector_allocation_domain	= numaq_vector_allocation_domain,
+	.vector_allocation_domain	= flat_vector_allocation_domain,
 	.init_apic_ldr			= numaq_init_apic_ldr,
 
 	.ioapic_phys_id_map		= numaq_ioapic_phys_id_map,
@@ -509,7 +492,6 @@ static struct apic __refdata apic_numaq = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0x0F << 24,
 
-	.cpu_mask_to_apicid		= numaq_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= numaq_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= numaq_send_IPI_mask,
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
index 1b291da..eb35ef9 100644
--- a/arch/x86/kernel/apic/probe_32.c
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -66,21 +66,6 @@ static void setup_apic_flat_routing(void)
 #endif
 }
 
-static void default_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	/*
-	 * Careful. Some cpus do not strictly honor the set of cpus
-	 * specified in the interrupt destination when using lowest
-	 * priority interrupt delivery mode.
-	 *
-	 * In particular there was a hyperthreading cpu observed to
-	 * deliver interrupts to the wrong hyperthread when only one
-	 * hyperthread was specified in the interrupt desitination.
-	 */
-	cpumask_clear(retmask);
-	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
-}
-
 /* should be called last. */
 static int probe_default(void)
 {
@@ -105,7 +90,7 @@ static struct apic apic_default = {
 	.check_apicid_used		= default_check_apicid_used,
 	.check_apicid_present		= default_check_apicid_present,
 
-	.vector_allocation_domain	= default_vector_allocation_domain,
+	.vector_allocation_domain	= flat_vector_allocation_domain,
 	.init_apic_ldr			= default_init_apic_ldr,
 
 	.ioapic_phys_id_map		= default_ioapic_phys_id_map,
@@ -123,8 +108,7 @@ static struct apic apic_default = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0x0F << 24,
 
-	.cpu_mask_to_apicid		= default_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= flat_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= default_send_IPI_mask_logical,
 	.send_IPI_mask_allbutself	= default_send_IPI_mask_allbutself_logical,
@@ -208,6 +192,9 @@ void __init default_setup_apic_routing(void)
 
 	if (apic->setup_apic_routing)
 		apic->setup_apic_routing();
+
+	if (x86_platform.apic_post_init)
+		x86_platform.apic_post_init();
 }
 
 void __init generic_apic_probe(void)
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index 3fe9866..1793dba 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -23,11 +23,6 @@
 #include <asm/ipi.h>
 #include <asm/setup.h>
 
-static int apicid_phys_pkg_id(int initial_apic_id, int index_msb)
-{
-	return hard_smp_processor_id() >> index_msb;
-}
-
 /*
  * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
  */
@@ -48,10 +43,8 @@ void __init default_setup_apic_routing(void)
 		}
 	}
 
-	if (is_vsmp_box()) {
-		/* need to update phys_pkg_id */
-		apic->phys_pkg_id = apicid_phys_pkg_id;
-	}
+	if (x86_platform.apic_post_init)
+		x86_platform.apic_post_init();
 }
 
 /* Same for both flat and physical. */
diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c
index 659897c..b6e6185 100644
--- a/arch/x86/kernel/apic/summit_32.c
+++ b/arch/x86/kernel/apic/summit_32.c
@@ -263,43 +263,48 @@ static int summit_check_phys_apicid_present(int physical_apicid)
 	return 1;
 }
 
-static unsigned int summit_cpu_mask_to_apicid(const struct cpumask *cpumask)
+static inline int
+summit_cpu_mask_to_apicid(const struct cpumask *cpumask, unsigned int *dest_id)
 {
 	unsigned int round = 0;
-	int cpu, apicid = 0;
+	unsigned int cpu, apicid = 0;
 
 	/*
 	 * The cpus in the mask must all be on the apic cluster.
 	 */
-	for_each_cpu(cpu, cpumask) {
+	for_each_cpu_and(cpu, cpumask, cpu_online_mask) {
 		int new_apicid = early_per_cpu(x86_cpu_to_logical_apicid, cpu);
 
 		if (round && APIC_CLUSTER(apicid) != APIC_CLUSTER(new_apicid)) {
 			printk("%s: Not a valid mask!\n", __func__);
-			return BAD_APICID;
+			return -EINVAL;
 		}
 		apicid |= new_apicid;
 		round++;
 	}
-	return apicid;
+	if (!round)
+		return -EINVAL;
+	*dest_id = apicid;
+	return 0;
 }
 
-static unsigned int summit_cpu_mask_to_apicid_and(const struct cpumask *inmask,
-			      const struct cpumask *andmask)
+static int
+summit_cpu_mask_to_apicid_and(const struct cpumask *inmask,
+			      const struct cpumask *andmask,
+			      unsigned int *apicid)
 {
-	int apicid = early_per_cpu(x86_cpu_to_logical_apicid, 0);
 	cpumask_var_t cpumask;
+	*apicid = early_per_cpu(x86_cpu_to_logical_apicid, 0);
 
 	if (!alloc_cpumask_var(&cpumask, GFP_ATOMIC))
-		return apicid;
+		return 0;
 
 	cpumask_and(cpumask, inmask, andmask);
-	cpumask_and(cpumask, cpumask, cpu_online_mask);
-	apicid = summit_cpu_mask_to_apicid(cpumask);
+	summit_cpu_mask_to_apicid(cpumask, apicid);
 
 	free_cpumask_var(cpumask);
 
-	return apicid;
+	return 0;
 }
 
 /*
@@ -320,20 +325,6 @@ static int probe_summit(void)
 	return 0;
 }
 
-static void summit_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	/* Careful. Some cpus do not strictly honor the set of cpus
-	 * specified in the interrupt destination when using lowest
-	 * priority interrupt delivery mode.
-	 *
-	 * In particular there was a hyperthreading cpu observed to
-	 * deliver interrupts to the wrong hyperthread when only one
-	 * hyperthread was specified in the interrupt desitination.
-	 */
-	cpumask_clear(retmask);
-	cpumask_bits(retmask)[0] = APIC_ALL_CPUS;
-}
-
 #ifdef CONFIG_X86_SUMMIT_NUMA
 static struct rio_table_hdr *rio_table_hdr;
 static struct scal_detail   *scal_devs[MAX_NUMNODES];
@@ -509,7 +500,7 @@ static struct apic apic_summit = {
 	.check_apicid_used		= summit_check_apicid_used,
 	.check_apicid_present		= summit_check_apicid_present,
 
-	.vector_allocation_domain	= summit_vector_allocation_domain,
+	.vector_allocation_domain	= flat_vector_allocation_domain,
 	.init_apic_ldr			= summit_init_apic_ldr,
 
 	.ioapic_phys_id_map		= summit_ioapic_phys_id_map,
@@ -527,7 +518,6 @@ static struct apic apic_summit = {
 	.set_apic_id			= NULL,
 	.apic_id_mask			= 0xFF << 24,
 
-	.cpu_mask_to_apicid		= summit_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= summit_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= summit_send_IPI_mask,
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
index ff35cff..c88baa4 100644
--- a/arch/x86/kernel/apic/x2apic_cluster.c
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -81,7 +81,7 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
 }
 
 static void
- x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
 {
 	__x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLBUT);
 }
@@ -96,36 +96,37 @@ static void x2apic_send_IPI_all(int vector)
 	__x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLINC);
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
+static int
+x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
+			      const struct cpumask *andmask,
+			      unsigned int *apicid)
 {
-	/*
-	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
-	 * May as well be the first.
-	 */
-	int cpu = cpumask_first(cpumask);
+	u32 dest = 0;
+	u16 cluster;
+	int i;
 
-	if ((unsigned)cpu < nr_cpu_ids)
-		return per_cpu(x86_cpu_to_logical_apicid, cpu);
-	else
-		return BAD_APICID;
-}
+	for_each_cpu_and(i, cpumask, andmask) {
+		if (!cpumask_test_cpu(i, cpu_online_mask))
+			continue;
+		dest = per_cpu(x86_cpu_to_logical_apicid, i);
+		cluster = x2apic_cluster(i);
+		break;
+	}
 
-static unsigned int
-x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			      const struct cpumask *andmask)
-{
-	int cpu;
+	if (!dest)
+		return -EINVAL;
 
-	/*
-	 * We're using fixed IRQ delivery, can only return one logical APIC ID.
-	 * May as well be the first.
-	 */
-	for_each_cpu_and(cpu, cpumask, andmask) {
-		if (cpumask_test_cpu(cpu, cpu_online_mask))
-			break;
+	for_each_cpu_and(i, cpumask, andmask) {
+		if (!cpumask_test_cpu(i, cpu_online_mask))
+			continue;
+		if (cluster != x2apic_cluster(i))
+			continue;
+		dest |= per_cpu(x86_cpu_to_logical_apicid, i);
 	}
 
-	return per_cpu(x86_cpu_to_logical_apicid, cpu);
+	*apicid = dest;
+
+	return 0;
 }
 
 static void init_x2apic_ldr(void)
@@ -208,6 +209,32 @@ static int x2apic_cluster_probe(void)
 		return 0;
 }
 
+static const struct cpumask *x2apic_cluster_target_cpus(void)
+{
+	return cpu_all_mask;
+}
+
+/*
+ * Each x2apic cluster is an allocation domain.
+ */
+static void cluster_vector_allocation_domain(int cpu, struct cpumask *retmask,
+					     const struct cpumask *mask)
+{
+	/*
+	 * To minimize vector pressure, default case of boot, device bringup
+	 * etc will use a single cpu for the interrupt destination.
+	 *
+	 * On explicit migration requests coming from irqbalance etc,
+	 * interrupts will be routed to the x2apic cluster (cluster-id
+	 * derived from the first cpu in the mask) members specified
+	 * in the mask.
+	 */
+	if (mask == x2apic_cluster_target_cpus())
+		cpumask_copy(retmask, cpumask_of(cpu));
+	else
+		cpumask_and(retmask, mask, per_cpu(cpus_in_cluster, cpu));
+}
+
 static struct apic apic_x2apic_cluster = {
 
 	.name				= "cluster x2apic",
@@ -219,13 +246,13 @@ static struct apic apic_x2apic_cluster = {
 	.irq_delivery_mode		= dest_LowestPrio,
 	.irq_dest_mode			= 1, /* logical */
 
-	.target_cpus			= x2apic_target_cpus,
+	.target_cpus			= x2apic_cluster_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= APIC_DEST_LOGICAL,
 	.check_apicid_used		= NULL,
 	.check_apicid_present		= NULL,
 
-	.vector_allocation_domain	= x2apic_vector_allocation_domain,
+	.vector_allocation_domain	= cluster_vector_allocation_domain,
 	.init_apic_ldr			= init_x2apic_ldr,
 
 	.ioapic_phys_id_map		= NULL,
@@ -243,7 +270,6 @@ static struct apic apic_x2apic_cluster = {
 	.set_apic_id			= x2apic_set_apic_id,
 	.apic_id_mask			= 0xFFFFFFFFu,
 
-	.cpu_mask_to_apicid		= x2apic_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= x2apic_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= x2apic_send_IPI_mask,
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
index c17e982..e03a1e1 100644
--- a/arch/x86/kernel/apic/x2apic_phys.c
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -76,38 +76,6 @@ static void x2apic_send_IPI_all(int vector)
 	__x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLINC);
 }
 
-static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	int cpu = cpumask_first(cpumask);
-
-	if ((unsigned)cpu < nr_cpu_ids)
-		return per_cpu(x86_cpu_to_apicid, cpu);
-	else
-		return BAD_APICID;
-}
-
-static unsigned int
-x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			      const struct cpumask *andmask)
-{
-	int cpu;
-
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	for_each_cpu_and(cpu, cpumask, andmask) {
-		if (cpumask_test_cpu(cpu, cpu_online_mask))
-			break;
-	}
-
-	return per_cpu(x86_cpu_to_apicid, cpu);
-}
-
 static void init_x2apic_ldr(void)
 {
 }
@@ -131,13 +99,13 @@ static struct apic apic_x2apic_phys = {
 	.irq_delivery_mode		= dest_Fixed,
 	.irq_dest_mode			= 0, /* physical */
 
-	.target_cpus			= x2apic_target_cpus,
+	.target_cpus			= online_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= 0,
 	.check_apicid_used		= NULL,
 	.check_apicid_present		= NULL,
 
-	.vector_allocation_domain	= x2apic_vector_allocation_domain,
+	.vector_allocation_domain	= default_vector_allocation_domain,
 	.init_apic_ldr			= init_x2apic_ldr,
 
 	.ioapic_phys_id_map		= NULL,
@@ -155,8 +123,7 @@ static struct apic apic_x2apic_phys = {
 	.set_apic_id			= x2apic_set_apic_id,
 	.apic_id_mask			= 0xFFFFFFFFu,
 
-	.cpu_mask_to_apicid		= x2apic_cpu_mask_to_apicid,
-	.cpu_mask_to_apicid_and		= x2apic_cpu_mask_to_apicid_and,
+	.cpu_mask_to_apicid_and		= default_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= x2apic_send_IPI_mask,
 	.send_IPI_mask_allbutself	= x2apic_send_IPI_mask_allbutself,
diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index c6d03f7..8cfade9 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -185,17 +185,6 @@ EXPORT_SYMBOL_GPL(uv_possible_blades);
 unsigned long sn_rtc_cycles_per_second;
 EXPORT_SYMBOL(sn_rtc_cycles_per_second);
 
-static const struct cpumask *uv_target_cpus(void)
-{
-	return cpu_online_mask;
-}
-
-static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask)
-{
-	cpumask_clear(retmask);
-	cpumask_set_cpu(cpu, retmask);
-}
-
 static int __cpuinit uv_wakeup_secondary(int phys_apicid, unsigned long start_rip)
 {
 #ifdef CONFIG_SMP
@@ -280,25 +269,12 @@ static void uv_init_apic_ldr(void)
 {
 }
 
-static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask)
-{
-	/*
-	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
-	 * May as well be the first.
-	 */
-	int cpu = cpumask_first(cpumask);
-
-	if ((unsigned)cpu < nr_cpu_ids)
-		return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits;
-	else
-		return BAD_APICID;
-}
-
-static unsigned int
+static int
 uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
-			  const struct cpumask *andmask)
+			  const struct cpumask *andmask,
+			  unsigned int *apicid)
 {
-	int cpu;
+	int unsigned cpu;
 
 	/*
 	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
@@ -308,7 +284,13 @@ uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask,
 		if (cpumask_test_cpu(cpu, cpu_online_mask))
 			break;
 	}
-	return per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits;
+
+	if (likely(cpu < nr_cpu_ids)) {
+		*apicid = per_cpu(x86_cpu_to_apicid, cpu) | uv_apicid_hibits;
+		return 0;
+	}
+
+	return -EINVAL;
 }
 
 static unsigned int x2apic_get_apic_id(unsigned long x)
@@ -362,13 +344,13 @@ static struct apic __refdata apic_x2apic_uv_x = {
 	.irq_delivery_mode		= dest_Fixed,
 	.irq_dest_mode			= 0, /* physical */
 
-	.target_cpus			= uv_target_cpus,
+	.target_cpus			= online_target_cpus,
 	.disable_esr			= 0,
 	.dest_logical			= APIC_DEST_LOGICAL,
 	.check_apicid_used		= NULL,
 	.check_apicid_present		= NULL,
 
-	.vector_allocation_domain	= uv_vector_allocation_domain,
+	.vector_allocation_domain	= default_vector_allocation_domain,
 	.init_apic_ldr			= uv_init_apic_ldr,
 
 	.ioapic_phys_id_map		= NULL,
@@ -386,7 +368,6 @@ static struct apic __refdata apic_x2apic_uv_x = {
 	.set_apic_id			= set_apic_id,
 	.apic_id_mask			= 0xFFFFFFFFu,
 
-	.cpu_mask_to_apicid		= uv_cpu_mask_to_apicid,
 	.cpu_mask_to_apicid_and		= uv_cpu_mask_to_apicid_and,
 
 	.send_IPI_mask			= uv_send_IPI_mask,
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 16be6dc..f4b9b80 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -1031,8 +1031,6 @@ void __init setup_arch(char **cmdline_p)
 
 	x86_init.timers.wallclock_init();
 
-	x86_platform.wallclock_init();
-
 	mcheck_init();
 
 	arch_init_ideal_nops();
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index 8eeb55a..992f890 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -16,6 +16,7 @@
 #include <linux/pci_ids.h>
 #include <linux/pci_regs.h>
 #include <linux/smp.h>
+#include <linux/irq.h>
 
 #include <asm/apic.h>
 #include <asm/pci-direct.h>
@@ -95,6 +96,18 @@ static void __init set_vsmp_pv_ops(void)
 	ctl = readl(address + 4);
 	printk(KERN_INFO "vSMP CTL: capabilities:0x%08x  control:0x%08x\n",
 	       cap, ctl);
+
+	/* If possible, let the vSMP foundation route the interrupt optimally */
+#ifdef CONFIG_SMP
+	if (cap & ctl & BIT(8)) {
+		ctl &= ~BIT(8);
+#ifdef CONFIG_PROC_FS
+		/* Don't let users change irq affinity via procfs */
+		no_irq_affinity = 1;
+#endif
+	}
+#endif
+
 	if (cap & ctl & (1 << 4)) {
 		/* Setup irq ops and turn on vSMP  IRQ fastpath handling */
 		pv_irq_ops.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable);
@@ -102,12 +115,11 @@ static void __init set_vsmp_pv_ops(void)
 		pv_irq_ops.save_fl  = PV_CALLEE_SAVE(vsmp_save_fl);
 		pv_irq_ops.restore_fl  = PV_CALLEE_SAVE(vsmp_restore_fl);
 		pv_init_ops.patch = vsmp_patch;
-
 		ctl &= ~(1 << 4);
-		writel(ctl, address + 4);
-		ctl = readl(address + 4);
-		printk(KERN_INFO "vSMP CTL: control set to:0x%08x\n", ctl);
 	}
+	writel(ctl, address + 4);
+	ctl = readl(address + 4);
+	pr_info("vSMP CTL: control set to:0x%08x\n", ctl);
 
 	early_iounmap(address, 8);
 }
@@ -187,12 +199,36 @@ static void __init vsmp_cap_cpus(void)
 #endif
 }
 
+static int apicid_phys_pkg_id(int initial_apic_id, int index_msb)
+{
+	return hard_smp_processor_id() >> index_msb;
+}
+
+/*
+ * In vSMP, all cpus should be capable of handling interrupts, regardless of
+ * the APIC used.
+ */
+static void fill_vector_allocation_domain(int cpu, struct cpumask *retmask,
+					  const struct cpumask *mask)
+{
+	cpumask_setall(retmask);
+}
+
+static void vsmp_apic_post_init(void)
+{
+	/* need to update phys_pkg_id */
+	apic->phys_pkg_id = apicid_phys_pkg_id;
+	apic->vector_allocation_domain = fill_vector_allocation_domain;
+}
+
 void __init vsmp_init(void)
 {
 	detect_vsmp_box();
 	if (!is_vsmp_box())
 		return;
 
+	x86_platform.apic_post_init = vsmp_apic_post_init;
+
 	vsmp_cap_cpus();
 
 	set_vsmp_pv_ops();
diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c
index 35c5e54..9f3167e 100644
--- a/arch/x86/kernel/x86_init.c
+++ b/arch/x86/kernel/x86_init.c
@@ -29,7 +29,6 @@ void __init x86_init_uint_noop(unsigned int unused) { }
 void __init x86_init_pgd_noop(pgd_t *unused) { }
 int __init iommu_init_noop(void) { return 0; }
 void iommu_shutdown_noop(void) { }
-void wallclock_init_noop(void) { }
 
 /*
  * The platform setup functions are preset with the default functions
@@ -101,7 +100,6 @@ static int default_i8042_detect(void) { return 1; };
 
 struct x86_platform_ops x86_platform = {
 	.calibrate_tsc			= native_calibrate_tsc,
-	.wallclock_init			= wallclock_init_noop,
 	.get_wallclock			= mach_get_cmos_time,
 	.set_wallclock			= mach_set_rtc_mmss,
 	.iommu_shutdown			= iommu_shutdown_noop,
diff --git a/arch/x86/platform/uv/uv_irq.c b/arch/x86/platform/uv/uv_irq.c
index f25c276..acf7752 100644
--- a/arch/x86/platform/uv/uv_irq.c
+++ b/arch/x86/platform/uv/uv_irq.c
@@ -135,6 +135,7 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	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));
@@ -143,6 +144,10 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	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
@@ -159,7 +164,7 @@ arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
 	entry->polarity		= 0;
 	entry->trigger		= 0;
 	entry->mask		= 0;
-	entry->dest		= apic->cpu_mask_to_apicid(eligible_cpu);
+	entry->dest		= dest;
 
 	mmr_pnode = uv_blade_to_pnode(mmr_blade);
 	uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value);
@@ -222,7 +227,7 @@ uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
 	if (cfg->move_in_progress)
 		send_cleanup_vector(cfg);
 
-	return 0;
+	return IRQ_SET_MASK_OK_NOCOPY;
 }
 
 /*
diff --git a/drivers/iommu/intel_irq_remapping.c b/drivers/iommu/intel_irq_remapping.c
index 6d34706..e0b18f3 100644
--- a/drivers/iommu/intel_irq_remapping.c
+++ b/drivers/iommu/intel_irq_remapping.c
@@ -902,7 +902,6 @@ static int intel_setup_ioapic_entry(int irq,
 	return 0;
 }
 
-#ifdef CONFIG_SMP
 /*
  * Migrate the IO-APIC irq in the presence of intr-remapping.
  *
@@ -924,6 +923,10 @@ intel_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
 	struct irq_cfg *cfg = data->chip_data;
 	unsigned int dest, irq = data->irq;
 	struct irte irte;
+	int err;
+
+	if (!config_enabled(CONFIG_SMP))
+		return -EINVAL;
 
 	if (!cpumask_intersects(mask, cpu_online_mask))
 		return -EINVAL;
@@ -931,10 +934,16 @@ intel_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
 	if (get_irte(irq, &irte))
 		return -EBUSY;
 
-	if (assign_irq_vector(irq, cfg, mask))
-		return -EBUSY;
+	err = assign_irq_vector(irq, cfg, mask);
+	if (err)
+		return err;
 
-	dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask);
+	err = apic->cpu_mask_to_apicid_and(cfg->domain, mask, &dest);
+	if (err) {
+		if (assign_irq_vector(irq, cfg, data->affinity))
+			pr_err("Failed to recover vector for irq %d\n", irq);
+		return err;
+	}
 
 	irte.vector = cfg->vector;
 	irte.dest_id = IRTE_DEST(dest);
@@ -956,7 +965,6 @@ intel_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
 	cpumask_copy(data->affinity, mask);
 	return 0;
 }
-#endif
 
 static void intel_compose_msi_msg(struct pci_dev *pdev,
 				  unsigned int irq, unsigned int dest,
@@ -1058,9 +1066,7 @@ struct irq_remap_ops intel_irq_remap_ops = {
 	.reenable		= reenable_irq_remapping,
 	.enable_faulting	= enable_drhd_fault_handling,
 	.setup_ioapic_entry	= intel_setup_ioapic_entry,
-#ifdef CONFIG_SMP
 	.set_affinity		= intel_ioapic_set_affinity,
-#endif
 	.free_irq		= free_irte,
 	.compose_msi_msg	= intel_compose_msi_msg,
 	.msi_alloc_irq		= intel_msi_alloc_irq,
diff --git a/drivers/iommu/irq_remapping.c b/drivers/iommu/irq_remapping.c
index 40cda8e..1d29b1c 100644
--- a/drivers/iommu/irq_remapping.c
+++ b/drivers/iommu/irq_remapping.c
@@ -111,16 +111,15 @@ int setup_ioapic_remapped_entry(int irq,
 					     vector, attr);
 }
 
-#ifdef CONFIG_SMP
 int set_remapped_irq_affinity(struct irq_data *data, const struct cpumask *mask,
 			      bool force)
 {
-	if (!remap_ops || !remap_ops->set_affinity)
+	if (!config_enabled(CONFIG_SMP) || !remap_ops ||
+	    !remap_ops->set_affinity)
 		return 0;
 
 	return remap_ops->set_affinity(data, mask, force);
 }
-#endif
 
 void free_remapped_irq(int irq)
 {
diff --git a/drivers/iommu/irq_remapping.h b/drivers/iommu/irq_remapping.h
index be9d729..b12974c 100644
--- a/drivers/iommu/irq_remapping.h
+++ b/drivers/iommu/irq_remapping.h
@@ -59,11 +59,9 @@ struct irq_remap_ops {
 				  unsigned int, int,
 				  struct io_apic_irq_attr *);
 
-#ifdef CONFIG_SMP
 	/* Set the CPU affinity of a remapped interrupt */
 	int (*set_affinity)(struct irq_data *data, const struct cpumask *mask,
 			    bool force);
-#endif
 
 	/* Free an IRQ */
 	int (*free_irq)(int);
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 61f5cec..47a937c 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -150,9 +150,7 @@ struct irq_data {
 	void			*handler_data;
 	void			*chip_data;
 	struct msi_desc		*msi_desc;
-#ifdef CONFIG_SMP
 	cpumask_var_t		affinity;
-#endif
 };
 
 /*

                 reply	other threads:[~2012-07-22 14:05 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120722140527.GA8313@gmail.com \
    --to=mingo@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.