From 323ec001c6bb98eeabb5abbdbb8c8055d9496554 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Sun, 29 Jun 2008 14:19:31 +0400
Subject: x86: use generic per-device dma coherent allocator

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-dma.c | 122 +---------------------------------------------
 1 file changed, 2 insertions(+), 120 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index dc00a1331ac..b75c81a8d3e 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -197,124 +197,6 @@ static __init int iommu_setup(char *p)
 }
 early_param("iommu", iommu_setup);
 
-#ifdef CONFIG_X86_32
-int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
-				dma_addr_t device_addr, size_t size, int flags)
-{
-	void __iomem *mem_base = NULL;
-	int pages = size >> PAGE_SHIFT;
-	int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
-
-	if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
-		goto out;
-	if (!size)
-		goto out;
-	if (dev->dma_mem)
-		goto out;
-
-	/* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
-
-	mem_base = ioremap(bus_addr, size);
-	if (!mem_base)
-		goto out;
-
-	dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
-	if (!dev->dma_mem)
-		goto out;
-	dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL);
-	if (!dev->dma_mem->bitmap)
-		goto free1_out;
-
-	dev->dma_mem->virt_base = mem_base;
-	dev->dma_mem->device_base = device_addr;
-	dev->dma_mem->size = pages;
-	dev->dma_mem->flags = flags;
-
-	if (flags & DMA_MEMORY_MAP)
-		return DMA_MEMORY_MAP;
-
-	return DMA_MEMORY_IO;
-
- free1_out:
-	kfree(dev->dma_mem);
- out:
-	if (mem_base)
-		iounmap(mem_base);
-	return 0;
-}
-EXPORT_SYMBOL(dma_declare_coherent_memory);
-
-void dma_release_declared_memory(struct device *dev)
-{
-	struct dma_coherent_mem *mem = dev->dma_mem;
-
-	if (!mem)
-		return;
-	dev->dma_mem = NULL;
-	iounmap(mem->virt_base);
-	kfree(mem->bitmap);
-	kfree(mem);
-}
-EXPORT_SYMBOL(dma_release_declared_memory);
-
-void *dma_mark_declared_memory_occupied(struct device *dev,
-					dma_addr_t device_addr, size_t size)
-{
-	struct dma_coherent_mem *mem = dev->dma_mem;
-	int pos, err;
-	int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1);
-
-	pages >>= PAGE_SHIFT;
-
-	if (!mem)
-		return ERR_PTR(-EINVAL);
-
-	pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
-	err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
-	if (err != 0)
-		return ERR_PTR(err);
-	return mem->virt_base + (pos << PAGE_SHIFT);
-}
-EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
-
-static int dma_alloc_from_coherent_mem(struct device *dev, ssize_t size,
-				       dma_addr_t *dma_handle, void **ret)
-{
-	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-	int order = get_order(size);
-
-	if (mem) {
-		int page = bitmap_find_free_region(mem->bitmap, mem->size,
-						     order);
-		if (page >= 0) {
-			*dma_handle = mem->device_base + (page << PAGE_SHIFT);
-			*ret = mem->virt_base + (page << PAGE_SHIFT);
-			memset(*ret, 0, size);
-		}
-		if (mem->flags & DMA_MEMORY_EXCLUSIVE)
-			*ret = NULL;
-	}
-	return (mem != NULL);
-}
-
-static int dma_release_coherent(struct device *dev, int order, void *vaddr)
-{
-	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-
-	if (mem && vaddr >= mem->virt_base && vaddr <
-		   (mem->virt_base + (mem->size << PAGE_SHIFT))) {
-		int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
-
-		bitmap_release_region(mem->bitmap, page, order);
-		return 1;
-	}
-	return 0;
-}
-#else
-#define dma_alloc_from_coherent_mem(dev, size, handle, ret) (0)
-#define dma_release_coherent(dev, order, vaddr) (0)
-#endif /* CONFIG_X86_32 */
-
 int dma_supported(struct device *dev, u64 mask)
 {
 #ifdef CONFIG_PCI
@@ -383,7 +265,7 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 	/* ignore region specifiers */
 	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
 
-	if (dma_alloc_from_coherent_mem(dev, size, dma_handle, &memory))
+	if (dma_alloc_from_coherent(dev, size, dma_handle, &memory))
 		return memory;
 
 	if (!dev) {
@@ -486,7 +368,7 @@ void dma_free_coherent(struct device *dev, size_t size,
 {
 	int order = get_order(size);
 	WARN_ON(irqs_disabled());	/* for portability */
-	if (dma_release_coherent(dev, order, vaddr))
+	if (dma_release_from_coherent(dev, order, vaddr))
 		return;
 	if (dma_ops->unmap_single)
 		dma_ops->unmap_single(dev, bus, size, 0);
-- 
cgit v1.2.3


From 91467bdf6e53058af13fd255375d6634ba0c70e0 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bwalle@suse.de>
Date: Fri, 18 Jul 2008 19:07:53 +0200
Subject: x86: move dma32_reserve_bootmem() after reserve_crashkernel()

On a x86-64 machine (nothing special I could encounter) I had the problem that
crashkernel reservation with the usual "64M@16M" failed. While debugging that,
I encountered that dma32_reserve_bootmem() reserves a memory region which is in
that area.

Because dma32_reserve_bootmem() does not rely on a specific offset but
crashkernel does, it makes sense to move the dma32_reserve_bootmem()
reservation down a bit. I tested that patch and it works without problems. I
don't see any negative effects of that move, but maybe I oversaw something ...

While we strictly don't need that patch in 2.6.27 because we have the
automatic, dynamic offset detection, it makes sense to also include it here
because:

  - it's easier to get it in -stable then,
  - many people are still used to the 'crashkernel=...@16M' syntax,
  - not everybody may be using a reloatable kernel.

Signed-off-by: Bernhard Walle <bwalle@suse.de>
Cc: kexec@lists.infradead.org
Cc: vgoyal@redhat.com
Cc: akpm@linux-foundation.org
Cc: Bernhard Walle <bwalle@suse.de>
Cc: yhlu.kernel@gmail.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 531b55b8e81..74d110ef269 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -792,10 +792,6 @@ void __init setup_arch(char **cmdline_p)
 
 	initmem_init(0, max_pfn);
 
-#ifdef CONFIG_X86_64
-	dma32_reserve_bootmem();
-#endif
-
 #ifdef CONFIG_ACPI_SLEEP
 	/*
 	 * Reserve low memory region for sleep support.
@@ -810,6 +806,15 @@ void __init setup_arch(char **cmdline_p)
 #endif
 	reserve_crashkernel();
 
+#ifdef CONFIG_X86_64
+	/*
+	 * dma32_reserve_bootmem() allocates bootmem which may conflict
+	 * with the crashkernel command line, so do that after
+	 * reserve_crashkernel()
+	 */
+	dma32_reserve_bootmem();
+#endif
+
 	reserve_ibft_region();
 
 #ifdef CONFIG_KVM_CLOCK
-- 
cgit v1.2.3


From a8132e5fe2c4f3f780b8bd3cce7851640f79f1c7 Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Fri, 25 Jul 2008 14:57:59 +0200
Subject: x86, AMD IOMMU: replace to_pages macro with iommu_num_pages

This patch removes the to_pages macro from AMD IOMMU code and calls the generic
iommu_num_pages function instead.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Cc: iommu@lists.linux-foundation.org
Cc: bhavna.sarathy@amd.com
Cc: robert.richter@amd.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/amd_iommu.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index c25210e6ac8..79eff735fde 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -29,9 +29,6 @@
 
 #define CMD_SET_TYPE(cmd, t) ((cmd)->data[1] |= ((t) << 28))
 
-#define to_pages(addr, size) \
-	 (round_up(((addr) & ~PAGE_MASK) + (size), PAGE_SIZE) >> PAGE_SHIFT)
-
 #define EXIT_LOOP_COUNT 10000000
 
 static DEFINE_RWLOCK(amd_iommu_devtable_lock);
@@ -185,7 +182,7 @@ static int iommu_flush_pages(struct amd_iommu *iommu, u16 domid,
 		u64 address, size_t size)
 {
 	int s = 0;
-	unsigned pages = to_pages(address, size);
+	unsigned pages = iommu_num_pages(address, size);
 
 	address &= PAGE_MASK;
 
@@ -557,8 +554,8 @@ static struct dma_ops_domain *dma_ops_domain_alloc(struct amd_iommu *iommu,
 	if (iommu->exclusion_start &&
 	    iommu->exclusion_start < dma_dom->aperture_size) {
 		unsigned long startpage = iommu->exclusion_start >> PAGE_SHIFT;
-		int pages = to_pages(iommu->exclusion_start,
-				iommu->exclusion_length);
+		int pages = iommu_num_pages(iommu->exclusion_start,
+					    iommu->exclusion_length);
 		dma_ops_reserve_addresses(dma_dom, startpage, pages);
 	}
 
@@ -767,7 +764,7 @@ static dma_addr_t __map_single(struct device *dev,
 	unsigned int pages;
 	int i;
 
-	pages = to_pages(paddr, size);
+	pages = iommu_num_pages(paddr, size);
 	paddr &= PAGE_MASK;
 
 	address = dma_ops_alloc_addresses(dev, dma_dom, pages);
@@ -802,7 +799,7 @@ static void __unmap_single(struct amd_iommu *iommu,
 	if ((dma_addr == 0) || (dma_addr + size > dma_dom->aperture_size))
 		return;
 
-	pages = to_pages(dma_addr, size);
+	pages = iommu_num_pages(dma_addr, size);
 	dma_addr &= PAGE_MASK;
 	start = dma_addr;
 
-- 
cgit v1.2.3


From 87e39ea5714dd59ba31e36c25833d2b20255a29d Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Fri, 25 Jul 2008 14:58:00 +0200
Subject: x86 gart: replace to_pages macro with iommu_num_pages

This patch removes the to_pages macro from x86 GART code and calls the generic
iommu_num_pages function instead.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Cc: iommu@lists.linux-foundation.org
Cc: bhavna.sarathy@amd.com
Cc: robert.richter@amd.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-gart_64.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index df5f142657d..c9824043156 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -67,9 +67,6 @@ static u32 gart_unmapped_entry;
 	(((x) & 0xfffff000) | (((x) >> 32) << 4) | GPTE_VALID | GPTE_COHERENT)
 #define GPTE_DECODE(x) (((x) & 0xfffff000) | (((u64)(x) & 0xff0) << 28))
 
-#define to_pages(addr, size) \
-	(round_up(((addr) & ~PAGE_MASK) + (size), PAGE_SIZE) >> PAGE_SHIFT)
-
 #define EMERGENCY_PAGES 32 /* = 128KB */
 
 #ifdef CONFIG_AGP
@@ -241,7 +238,7 @@ nonforced_iommu(struct device *dev, unsigned long addr, size_t size)
 static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem,
 				size_t size, int dir)
 {
-	unsigned long npages = to_pages(phys_mem, size);
+	unsigned long npages = iommu_num_pages(phys_mem, size);
 	unsigned long iommu_page = alloc_iommu(dev, npages);
 	int i;
 
@@ -304,7 +301,7 @@ static void gart_unmap_single(struct device *dev, dma_addr_t dma_addr,
 		return;
 
 	iommu_page = (dma_addr - iommu_bus_base)>>PAGE_SHIFT;
-	npages = to_pages(dma_addr, size);
+	npages = iommu_num_pages(dma_addr, size);
 	for (i = 0; i < npages; i++) {
 		iommu_gatt_base[iommu_page + i] = gart_unmapped_entry;
 		CLEAR_LEAK(iommu_page + i);
@@ -387,7 +384,7 @@ static int __dma_map_cont(struct device *dev, struct scatterlist *start,
 		}
 
 		addr = phys_addr;
-		pages = to_pages(s->offset, s->length);
+		pages = iommu_num_pages(s->offset, s->length);
 		while (pages--) {
 			iommu_gatt_base[iommu_page] = GPTE_ENCODE(addr);
 			SET_LEAK(iommu_page);
@@ -470,7 +467,7 @@ gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir)
 
 		seg_size += s->length;
 		need = nextneed;
-		pages += to_pages(s->offset, s->length);
+		pages += iommu_num_pages(s->offset, s->length);
 		ps = s;
 	}
 	if (dma_map_cont(dev, start_sg, i - start, sgmap, pages, need) < 0)
-- 
cgit v1.2.3


From 6524d938b3360504b43a1278b5a8403e85383d1a Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Thu, 24 Jul 2008 18:21:30 -0700
Subject: cpumask: put cpumask_of_cpu_map in the initdata section

  * Create the cpumask_of_cpu_map statically in the init data section
    using NR_CPUS but replace it during boot up with one sized by
    nr_cpu_ids (num possible cpus).

Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_percpu.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index f7745f94c00..1cd53dfcd30 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -81,10 +81,12 @@ static void __init setup_per_cpu_maps(void)
 }
 
 #ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
-cpumask_t *cpumask_of_cpu_map __read_mostly;
-EXPORT_SYMBOL(cpumask_of_cpu_map);
-
-/* requires nr_cpu_ids to be initialized */
+/*
+ * Replace static cpumask_of_cpu_map in the initdata section,
+ * with one that's allocated sized by the possible number of cpus.
+ *
+ * (requires nr_cpu_ids to be initialized)
+ */
 static void __init setup_cpumask_of_cpu(void)
 {
 	int i;
-- 
cgit v1.2.3


From 0bc3cc03fa6e1c20aecb5a33356bcaae410640b9 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Thu, 24 Jul 2008 18:21:31 -0700
Subject: cpumask: change cpumask_of_cpu_ptr to use new cpumask_of_cpu

  * Replace previous instances of the cpumask_of_cpu_ptr* macros
    with a the new (lvalue capable) generic cpumask_of_cpu().

Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/acpi/cstate.c                    |  3 +--
 arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c       | 10 +++-------
 arch/x86/kernel/cpu/cpufreq/powernow-k8.c        | 15 +++++----------
 arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c | 12 ++++--------
 arch/x86/kernel/cpu/cpufreq/speedstep-ich.c      |  3 +--
 arch/x86/kernel/cpu/intel_cacheinfo.c            |  3 +--
 arch/x86/kernel/ldt.c                            |  6 ++----
 arch/x86/kernel/microcode.c                      | 17 +++++------------
 arch/x86/kernel/reboot.c                         | 11 +++--------
 9 files changed, 25 insertions(+), 55 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index 9220cf46aa1..c2502eb9aa8 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -73,7 +73,6 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 
 	cpumask_t saved_mask;
-	cpumask_of_cpu_ptr(new_mask, cpu);
 	int retval;
 	unsigned int eax, ebx, ecx, edx;
 	unsigned int edx_part;
@@ -92,7 +91,7 @@ int acpi_processor_ffh_cstate_probe(unsigned int cpu,
 
 	/* Make sure we are running on right CPU */
 	saved_mask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, new_mask);
+	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	if (retval)
 		return -1;
 
diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index ff2fff56f0a..dd097b83583 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -200,12 +200,10 @@ static void drv_read(struct drv_cmd *cmd)
 static void drv_write(struct drv_cmd *cmd)
 {
 	cpumask_t saved_mask = current->cpus_allowed;
-	cpumask_of_cpu_ptr_declare(cpu_mask);
 	unsigned int i;
 
 	for_each_cpu_mask_nr(i, cmd->mask) {
-		cpumask_of_cpu_ptr_next(cpu_mask, i);
-		set_cpus_allowed_ptr(current, cpu_mask);
+		set_cpus_allowed_ptr(current, &cpumask_of_cpu(i));
 		do_drv_write(cmd);
 	}
 
@@ -269,12 +267,11 @@ static unsigned int get_measured_perf(unsigned int cpu)
 	} aperf_cur, mperf_cur;
 
 	cpumask_t saved_mask;
-	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	unsigned int perf_percent;
 	unsigned int retval;
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, cpu_mask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	if (get_cpu() != cpu) {
 		/* We were not able to run on requested processor */
 		put_cpu();
@@ -340,7 +337,6 @@ static unsigned int get_measured_perf(unsigned int cpu)
 
 static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 {
-	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	struct acpi_cpufreq_data *data = per_cpu(drv_data, cpu);
 	unsigned int freq;
 	unsigned int cached_freq;
@@ -353,7 +349,7 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
 	}
 
 	cached_freq = data->freq_table[data->acpi_data->state].frequency;
-	freq = extract_freq(get_cur_val(cpu_mask), data);
+	freq = extract_freq(get_cur_val(&cpumask_of_cpu(cpu)), data);
 	if (freq != cached_freq) {
 		/*
 		 * The dreaded BIOS frequency change behind our back.
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index 53c7b693697..c45ca6d4dce 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -479,12 +479,11 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvi
 static int check_supported_cpu(unsigned int cpu)
 {
 	cpumask_t oldmask;
-	cpumask_of_cpu_ptr(cpu_mask, cpu);
 	u32 eax, ebx, ecx, edx;
 	unsigned int rc = 0;
 
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, cpu_mask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", cpu);
@@ -1017,7 +1016,6 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i
 static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsigned relation)
 {
 	cpumask_t oldmask;
-	cpumask_of_cpu_ptr(cpu_mask, pol->cpu);
 	struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu);
 	u32 checkfid;
 	u32 checkvid;
@@ -1032,7 +1030,7 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, cpu_mask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1107,7 +1105,6 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask;
-	cpumask_of_cpu_ptr_declare(newmask);
 	int rc;
 
 	if (!cpu_online(pol->cpu))
@@ -1159,8 +1156,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 
 	/* only run on specific CPU from here on */
 	oldmask = current->cpus_allowed;
-	cpumask_of_cpu_ptr_next(newmask, pol->cpu);
-	set_cpus_allowed_ptr(current, newmask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
 
 	if (smp_processor_id() != pol->cpu) {
 		printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu);
@@ -1182,7 +1178,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	set_cpus_allowed_ptr(current, &oldmask);
 
 	if (cpu_family == CPU_HW_PSTATE)
-		pol->cpus = *newmask;
+		pol->cpus = cpumask_of_cpu(pol->cpu);
 	else
 		pol->cpus = per_cpu(cpu_core_map, pol->cpu);
 	data->available_cores = &(pol->cpus);
@@ -1248,7 +1244,6 @@ static unsigned int powernowk8_get (unsigned int cpu)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask = current->cpus_allowed;
-	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int khz = 0;
 	unsigned int first;
 
@@ -1258,7 +1253,7 @@ static unsigned int powernowk8_get (unsigned int cpu)
 	if (!data)
 		return -EINVAL;
 
-	set_cpus_allowed_ptr(current, newmask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	if (smp_processor_id() != cpu) {
 		printk(KERN_ERR PFX
 			"limiting to CPU %d failed in powernowk8_get\n", cpu);
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
index ca2ac13b7af..15e13c01cc3 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c
@@ -324,10 +324,9 @@ static unsigned int get_cur_freq(unsigned int cpu)
 	unsigned l, h;
 	unsigned clock_freq;
 	cpumask_t saved_mask;
-	cpumask_of_cpu_ptr(new_mask, cpu);
 
 	saved_mask = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, new_mask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	if (smp_processor_id() != cpu)
 		return 0;
 
@@ -585,15 +584,12 @@ static int centrino_target (struct cpufreq_policy *policy,
 		 * Best effort undo..
 		 */
 
-		if (!cpus_empty(*covered_cpus)) {
-			cpumask_of_cpu_ptr_declare(new_mask);
-
+		if (!cpus_empty(*covered_cpus))
 			for_each_cpu_mask_nr(j, *covered_cpus) {
-				cpumask_of_cpu_ptr_next(new_mask, j);
-				set_cpus_allowed_ptr(current, new_mask);
+				set_cpus_allowed_ptr(current,
+						     &cpumask_of_cpu(j));
 				wrmsr(MSR_IA32_PERF_CTL, oldmsr, h);
 			}
-		}
 
 		tmp = freqs.new;
 		freqs.new = freqs.old;
diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
index 2f3728dc24f..191f7263c61 100644
--- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
+++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c
@@ -244,8 +244,7 @@ static unsigned int _speedstep_get(const cpumask_t *cpus)
 
 static unsigned int speedstep_get(unsigned int cpu)
 {
-	cpumask_of_cpu_ptr(newmask, cpu);
-	return _speedstep_get(newmask);
+	return _speedstep_get(&cpumask_of_cpu(cpu));
 }
 
 /**
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 650d40f7912..6b0a10b002f 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -516,7 +516,6 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 	unsigned long		j;
 	int			retval;
 	cpumask_t		oldmask;
-	cpumask_of_cpu_ptr(newmask, cpu);
 
 	if (num_cache_leaves == 0)
 		return -ENOENT;
@@ -527,7 +526,7 @@ static int __cpuinit detect_cache_attributes(unsigned int cpu)
 		return -ENOMEM;
 
 	oldmask = current->cpus_allowed;
-	retval = set_cpus_allowed_ptr(current, newmask);
+	retval = set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	if (retval)
 		goto out;
 
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 3fee2aa50f3..b68e21f06f4 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -62,12 +62,10 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 
 	if (reload) {
 #ifdef CONFIG_SMP
-		cpumask_of_cpu_ptr_declare(mask);
-
 		preempt_disable();
 		load_LDT(pc);
-		cpumask_of_cpu_ptr_next(mask, smp_processor_id());
-		if (!cpus_equal(current->mm->cpu_vm_mask, *mask))
+		if (!cpus_equal(current->mm->cpu_vm_mask,
+				cpumask_of_cpu(smp_processor_id())))
 			smp_call_function(flush_ldt, current->mm, 1);
 		preempt_enable();
 #else
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
index 6994c751590..652fa5c38eb 100644
--- a/arch/x86/kernel/microcode.c
+++ b/arch/x86/kernel/microcode.c
@@ -388,7 +388,6 @@ static int do_microcode_update (void)
 	void *new_mc = NULL;
 	int cpu;
 	cpumask_t old;
-	cpumask_of_cpu_ptr_declare(newmask);
 
 	old = current->cpus_allowed;
 
@@ -405,8 +404,7 @@ static int do_microcode_update (void)
 
 			if (!uci->valid)
 				continue;
-			cpumask_of_cpu_ptr_next(newmask, cpu);
-			set_cpus_allowed_ptr(current, newmask);
+			set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 			error = get_maching_microcode(new_mc, cpu);
 			if (error < 0)
 				goto out;
@@ -576,7 +574,6 @@ static int apply_microcode_check_cpu(int cpu)
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 	cpumask_t old;
-	cpumask_of_cpu_ptr(newmask, cpu);
 	unsigned int val[2];
 	int err = 0;
 
@@ -585,7 +582,7 @@ static int apply_microcode_check_cpu(int cpu)
 		return 0;
 
 	old = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, newmask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 
 	/* Check if the microcode we have in memory matches the CPU */
 	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
@@ -623,12 +620,11 @@ static int apply_microcode_check_cpu(int cpu)
 static void microcode_init_cpu(int cpu, int resume)
 {
 	cpumask_t old;
-	cpumask_of_cpu_ptr(newmask, cpu);
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 
 	old = current->cpus_allowed;
 
-	set_cpus_allowed_ptr(current, newmask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 	mutex_lock(&microcode_mutex);
 	collect_cpu_info(cpu);
 	if (uci->valid && system_state == SYSTEM_RUNNING && !resume)
@@ -661,13 +657,10 @@ static ssize_t reload_store(struct sys_device *dev,
 	if (end == buf)
 		return -EINVAL;
 	if (val == 1) {
-		cpumask_t old;
-		cpumask_of_cpu_ptr(newmask, cpu);
-
-		old = current->cpus_allowed;
+		cpumask_t old = current->cpus_allowed;
 
 		get_online_cpus();
-		set_cpus_allowed_ptr(current, newmask);
+		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
 
 		mutex_lock(&microcode_mutex);
 		if (uci->valid)
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 06a9f643817..724adfc63cb 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -414,25 +414,20 @@ void native_machine_shutdown(void)
 
 	/* The boot cpu is always logical cpu 0 */
 	int reboot_cpu_id = 0;
-	cpumask_of_cpu_ptr(newmask, reboot_cpu_id);
 
 #ifdef CONFIG_X86_32
 	/* See if there has been given a command line override */
 	if ((reboot_cpu != -1) && (reboot_cpu < NR_CPUS) &&
-		cpu_online(reboot_cpu)) {
+		cpu_online(reboot_cpu))
 		reboot_cpu_id = reboot_cpu;
-		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
-	}
 #endif
 
 	/* Make certain the cpu I'm about to reboot on is online */
-	if (!cpu_online(reboot_cpu_id)) {
+	if (!cpu_online(reboot_cpu_id))
 		reboot_cpu_id = smp_processor_id();
-		cpumask_of_cpu_ptr_next(newmask, reboot_cpu_id);
-	}
 
 	/* Make certain I only run on the appropriate processor */
-	set_cpus_allowed_ptr(current, newmask);
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(reboot_cpu_id));
 
 	/* O.K Now that I'm on the appropriate processor,
 	 * stop all of the others.
-- 
cgit v1.2.3


From e56b3bc7942982ac2589c942fb345e38bc7a341a Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 28 Jul 2008 11:32:33 -0700
Subject: cpu masks: optimize and clean up cpumask_of_cpu()

Clean up and optimize cpumask_of_cpu(), by sharing all the zero words.

Instead of stupidly generating all possible i=0...NR_CPUS 2^i patterns
creating a huge array of constant bitmasks, realize that the zero words
can be shared.

In other words, on a 64-bit architecture, we only ever need 64 of these
arrays - with a different bit set in one single world (with enough zero
words around it so that we can create any bitmask by just offsetting in
that big array). And then we just put enough zeroes around it that we
can point every single cpumask to be one of those things.

So when we have 4k CPU's, instead of having 4k arrays (of 4k bits each,
with one bit set in each array - 2MB memory total), we have exactly 64
arrays instead, each 8k bits in size (64kB total).

And then we just point cpumask(n) to the right position (which we can
calculate dynamically). Once we have the right arrays, getting
"cpumask(n)" ends up being:

  static inline const cpumask_t *get_cpu_mask(unsigned int cpu)
  {
          const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
          p -= cpu / BITS_PER_LONG;
          return (const cpumask_t *)p;
  }

This brings other advantages and simplifications as well:

 - we are not wasting memory that is just filled with a single bit in
   various different places

 - we don't need all those games to re-create the arrays in some dense
   format, because they're already going to be dense enough.

if we compile a kernel for up to 4k CPU's, "wasting" that 64kB of memory
is a non-issue (especially since by doing this "overlapping" trick we
probably get better cache behaviour anyway).

[ mingo@elte.hu:

  Converted Linus's mails into a commit. See:

     http://lkml.org/lkml/2008/7/27/156
     http://lkml.org/lkml/2008/7/28/320

  Also applied a family filter - which also has the side-effect of leaving
  out the bits where Linus calls me an idio... Oh, never mind ;-)
]

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup_percpu.c | 23 -----------------------
 1 file changed, 23 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index 1cd53dfcd30..76e305e064f 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -80,26 +80,6 @@ static void __init setup_per_cpu_maps(void)
 #endif
 }
 
-#ifdef CONFIG_HAVE_CPUMASK_OF_CPU_MAP
-/*
- * Replace static cpumask_of_cpu_map in the initdata section,
- * with one that's allocated sized by the possible number of cpus.
- *
- * (requires nr_cpu_ids to be initialized)
- */
-static void __init setup_cpumask_of_cpu(void)
-{
-	int i;
-
-	/* alloc_bootmem zeroes memory */
-	cpumask_of_cpu_map = alloc_bootmem_low(sizeof(cpumask_t) * nr_cpu_ids);
-	for (i = 0; i < nr_cpu_ids; i++)
-		cpu_set(i, cpumask_of_cpu_map[i]);
-}
-#else
-static inline void setup_cpumask_of_cpu(void) { }
-#endif
-
 #ifdef CONFIG_X86_32
 /*
  * Great future not-so-futuristic plan: make i386 and x86_64 do it
@@ -199,9 +179,6 @@ void __init setup_per_cpu_areas(void)
 
 	/* Setup node to cpumask map */
 	setup_node_to_cpumask_map();
-
-	/* Setup cpumask_of_cpu map */
-	setup_cpumask_of_cpu();
 }
 
 #endif
-- 
cgit v1.2.3


From 8978b74253280d59e97cf49a3ec2c0cbccd5b801 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Tue, 29 Jul 2008 13:38:53 +0900
Subject: generic, x86: fix add iommu_num_pages helper function

This IOMMU helper function doesn't work for some architectures:

  http://marc.info/?l=linux-kernel&m=121699304403202&w=2

It also breaks POWER and SPARC builds:

  http://marc.info/?l=linux-kernel&m=121730388001890&w=2

Currently, only x86 IOMMUs use this so let's move it to x86 for
now.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-dma.c | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 8dbffb846de..87d4d6964ec 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -123,6 +123,14 @@ void __init pci_iommu_alloc(void)
 
 	pci_swiotlb_init();
 }
+
+unsigned long iommu_num_pages(unsigned long addr, unsigned long len)
+{
+	unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE);
+
+	return size >> PAGE_SHIFT;
+}
+EXPORT_SYMBOL(iommu_num_pages);
 #endif
 
 /*
-- 
cgit v1.2.3


From 0d39741a27d86d305cc75ba626392be410dcbab9 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Tue, 29 Jul 2008 22:34:01 -0700
Subject: GRU Driver: export is_uv_system(), zap_page_range() & follow_page()

Exports needed by the GRU driver.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/genapic_64.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index 1fa8be5bd21..eaff0bbb144 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -99,3 +99,4 @@ int is_uv_system(void)
 {
 	return uv_system_type != UV_NONE;
 }
+EXPORT_SYMBOL_GPL(is_uv_system);
-- 
cgit v1.2.3


From e0d22d03c06c4e2c194d7010bc1e4a972199f156 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Thu, 31 Jul 2008 23:43:44 +0200
Subject: x86: fdiv bug detection fix

The fdiv detection code writes s32 integer into
the boot_cpu_data.fdiv_bug.
However, the boot_cpu_data.fdiv_bug is only char (s8)
field so the detection overwrites already set fields for
other bugs, e.g. the f00f bug field.

Use local s32 variable to receive result.

This is a partial fix to Bugzilla #9928  - fixes wrong
information about the f00f bug (tested) and probably
for coma bug (I have no cpu to test this).

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/bugs.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index c9b58a806e8..c8e315f1aa8 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -50,6 +50,8 @@ static double __initdata y = 3145727.0;
  */
 static void __init check_fpu(void)
 {
+	s32 fdiv_bug;
+
 	if (!boot_cpu_data.hard_math) {
 #ifndef CONFIG_MATH_EMULATION
 		printk(KERN_EMERG "No coprocessor found and no math emulation present.\n");
@@ -74,8 +76,10 @@ static void __init check_fpu(void)
 		"fistpl %0\n\t"
 		"fwait\n\t"
 		"fninit"
-		: "=m" (*&boot_cpu_data.fdiv_bug)
+		: "=m" (*&fdiv_bug)
 		: "m" (*&x), "m" (*&y));
+
+	boot_cpu_data.fdiv_bug = fdiv_bug;
 	if (boot_cpu_data.fdiv_bug)
 		printk("Hmm, FPU with FDIV bug.\n");
 }
-- 
cgit v1.2.3


From ec983f7060cd73e14cdd3edd910339127a8a4e96 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Wed, 30 Jul 2008 12:05:03 -0400
Subject: [CPUFREQ] Remove EXPERIMENTAL annotation from VIA C7 powersaver
 kconfig.

This has been pretty solid, and doesn't see much change at all.

Noticed by Harald Welte.

Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig
index cb7a5715596..efae3b22a0f 100644
--- a/arch/x86/kernel/cpu/cpufreq/Kconfig
+++ b/arch/x86/kernel/cpu/cpufreq/Kconfig
@@ -235,9 +235,9 @@ config X86_LONGHAUL
 	  If in doubt, say N.
 
 config X86_E_POWERSAVER
-	tristate "VIA C7 Enhanced PowerSaver (EXPERIMENTAL)"
+	tristate "VIA C7 Enhanced PowerSaver"
 	select CPU_FREQ_TABLE
-	depends on X86_32 && EXPERIMENTAL
+	depends on X86_32
 	help
 	  This adds the CPUFreq driver for VIA C7 processors.
 
-- 
cgit v1.2.3


From 460f5ef2835dbc33825f611f408eb09c29be4b85 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Wed, 30 Jul 2008 13:01:42 -0400
Subject: [CPUFREQ] Fix warning in elanfreq

arch/x86/kernel/cpu/cpufreq/elanfreq.c:47:26: warning: symbol 'elan_multiplier' was not declared. Should it be static?

Yes, yes it should.

Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/elanfreq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
index 94619c22f56..e4a4bf870e9 100644
--- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/elanfreq.c
@@ -44,7 +44,7 @@ struct s_elan_multiplier {
  * It is important that the frequencies
  * are listed in ascending order here!
  */
-struct s_elan_multiplier elan_multiplier[] = {
+static struct s_elan_multiplier elan_multiplier[] = {
 	{1000,	0x02,	0x18},
 	{2000,	0x02,	0x10},
 	{4000,	0x02,	0x08},
-- 
cgit v1.2.3


From 23431b495fcbbde6d09aba30019b6a224d543cf9 Mon Sep 17 00:00:00 2001
From: Mark Langsdorf <mark.langsdorf@amd.com>
Date: Thu, 31 Jul 2008 12:39:05 -0500
Subject: [CPUFREQ][1/2] whitespace fix for powernow-k8

Trivial whitespace fix for powernow-k8.

Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 3 ---
 1 file changed, 3 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index c45ca6d4dce..84bb395038d 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -66,7 +66,6 @@ static u32 find_freq_from_fid(u32 fid)
 	return 800 + (fid * 100);
 }
 
-
 /* Return a frequency in KHz, given an input fid */
 static u32 find_khz_freq_from_fid(u32 fid)
 {
@@ -78,7 +77,6 @@ static u32 find_khz_freq_from_pstate(struct cpufreq_frequency_table *data, u32 p
 	return data[pstate].frequency;
 }
 
-
 /* Return the vco fid for an input fid
  *
  * Each "low" fid has corresponding "high" fid, and you can get to "low" fids
@@ -166,7 +164,6 @@ static void fidvid_msr_init(void)
 	wrmsr(MSR_FIDVID_CTL, lo, hi);
 }
 
-
 /* write the new fid value along with the other control fields to the msr */
 static int write_new_fid(struct powernow_k8_data *data, u32 fid)
 {
-- 
cgit v1.2.3


From 34ae7f35a21694aa5cb8829dc5142c39d73d6ba0 Mon Sep 17 00:00:00 2001
From: Mark Langsdorf <mark.langsdorf@amd.com>
Date: Thu, 31 Jul 2008 12:39:12 -0500
Subject: [CPUFREQ][2/2] preregister support for powernow-k8

This patch provides support for the _PSD ACPI object in the Powernow-k8
driver.  Although it looks like an invasive patch, most of it is
simply the consequence of turning the static acpi_performance_data
structure into a pointer.

AMD has tested it on several machines over the past few days without issue.

[trivial checkpatch warnings fixed up by davej]
[X86_POWERNOW_K8_ACPI=n buildfix from Randy Dunlap]

Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Tested-by: Frank Arnold <frank.arnold@amd.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 109 ++++++++++++++++++++----------
 arch/x86/kernel/cpu/cpufreq/powernow-k8.h |   3 +-
 2 files changed, 75 insertions(+), 37 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
index 84bb395038d..4e7271999a7 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -737,44 +737,63 @@ static int find_psb_table(struct powernow_k8_data *data)
 #ifdef CONFIG_X86_POWERNOW_K8_ACPI
 static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index)
 {
-	if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE))
+	if (!data->acpi_data->state_count || (cpu_family == CPU_HW_PSTATE))
 		return;
 
-	data->irt = (data->acpi_data.states[index].control >> IRT_SHIFT) & IRT_MASK;
-	data->rvo = (data->acpi_data.states[index].control >> RVO_SHIFT) & RVO_MASK;
-	data->exttype = (data->acpi_data.states[index].control >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK;
-	data->plllock = (data->acpi_data.states[index].control >> PLL_L_SHIFT) & PLL_L_MASK;
-	data->vidmvs = 1 << ((data->acpi_data.states[index].control >> MVS_SHIFT) & MVS_MASK);
-	data->vstable = (data->acpi_data.states[index].control >> VST_SHIFT) & VST_MASK;
+	data->irt = (data->acpi_data->states[index].control >> IRT_SHIFT) & IRT_MASK;
+	data->rvo = (data->acpi_data->states[index].control >> RVO_SHIFT) & RVO_MASK;
+	data->exttype = (data->acpi_data->states[index].control >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK;
+	data->plllock = (data->acpi_data->states[index].control >> PLL_L_SHIFT) & PLL_L_MASK;
+	data->vidmvs = 1 << ((data->acpi_data->states[index].control >> MVS_SHIFT) & MVS_MASK);
+	data->vstable = (data->acpi_data->states[index].control >> VST_SHIFT) & VST_MASK;
+}
+
+
+static struct acpi_processor_performance *acpi_perf_data;
+static int preregister_valid;
+
+static int powernow_k8_cpu_preinit_acpi(void)
+{
+	acpi_perf_data = alloc_percpu(struct acpi_processor_performance);
+	if (!acpi_perf_data)
+		return -ENODEV;
+
+	if (acpi_processor_preregister_performance(acpi_perf_data))
+		return -ENODEV;
+	else
+		preregister_valid = 1;
+	return 0;
 }
 
 static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 {
 	struct cpufreq_frequency_table *powernow_table;
 	int ret_val;
+	int cpu = 0;
 
-	if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) {
+	data->acpi_data = percpu_ptr(acpi_perf_data, cpu);
+	if (acpi_processor_register_performance(data->acpi_data, data->cpu)) {
 		dprintk("register performance failed: bad ACPI data\n");
 		return -EIO;
 	}
 
 	/* verify the data contained in the ACPI structures */
-	if (data->acpi_data.state_count <= 1) {
+	if (data->acpi_data->state_count <= 1) {
 		dprintk("No ACPI P-States\n");
 		goto err_out;
 	}
 
-	if ((data->acpi_data.control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) ||
-		(data->acpi_data.status_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) {
+	if ((data->acpi_data->control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) ||
+		(data->acpi_data->status_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) {
 		dprintk("Invalid control/status registers (%x - %x)\n",
-			data->acpi_data.control_register.space_id,
-			data->acpi_data.status_register.space_id);
+			data->acpi_data->control_register.space_id,
+			data->acpi_data->status_register.space_id);
 		goto err_out;
 	}
 
 	/* fill in data->powernow_table */
 	powernow_table = kmalloc((sizeof(struct cpufreq_frequency_table)
-		* (data->acpi_data.state_count + 1)), GFP_KERNEL);
+		* (data->acpi_data->state_count + 1)), GFP_KERNEL);
 	if (!powernow_table) {
 		dprintk("powernow_table memory alloc failure\n");
 		goto err_out;
@@ -787,12 +806,12 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 	if (ret_val)
 		goto err_out_mem;
 
-	powernow_table[data->acpi_data.state_count].frequency = CPUFREQ_TABLE_END;
-	powernow_table[data->acpi_data.state_count].index = 0;
+	powernow_table[data->acpi_data->state_count].frequency = CPUFREQ_TABLE_END;
+	powernow_table[data->acpi_data->state_count].index = 0;
 	data->powernow_table = powernow_table;
 
 	/* fill in data */
-	data->numps = data->acpi_data.state_count;
+	data->numps = data->acpi_data->state_count;
 	if (first_cpu(per_cpu(cpu_core_map, data->cpu)) == data->cpu)
 		print_basics(data);
 	powernow_k8_acpi_pst_values(data, 0);
@@ -800,16 +819,31 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
 	/* notify BIOS that we exist */
 	acpi_processor_notify_smm(THIS_MODULE);
 
+	/* determine affinity, from ACPI if available */
+	if (preregister_valid) {
+		if ((data->acpi_data->shared_type == CPUFREQ_SHARED_TYPE_ALL) ||
+			(data->acpi_data->shared_type == CPUFREQ_SHARED_TYPE_ANY))
+			data->starting_core_affinity = data->acpi_data->shared_cpu_map;
+		else
+			data->starting_core_affinity = cpumask_of_cpu(data->cpu);
+	} else {
+	/* best guess from family if not */
+		if (cpu_family == CPU_HW_PSTATE)
+			data->starting_core_affinity = cpumask_of_cpu(data->cpu);
+		else
+			data->starting_core_affinity = per_cpu(cpu_core_map, data->cpu);
+	}
+
 	return 0;
 
 err_out_mem:
 	kfree(powernow_table);
 
 err_out:
-	acpi_processor_unregister_performance(&data->acpi_data, data->cpu);
+	acpi_processor_unregister_performance(data->acpi_data, data->cpu);
 
 	/* data->acpi_data.state_count informs us at ->exit() whether ACPI was used */
-	data->acpi_data.state_count = 0;
+	data->acpi_data->state_count = 0;
 
 	return -ENODEV;
 }
@@ -821,10 +855,10 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpuf
 	rdmsr(MSR_PSTATE_CUR_LIMIT, hi, lo);
 	data->max_hw_pstate = (hi & HW_PSTATE_MAX_MASK) >> HW_PSTATE_MAX_SHIFT;
 
-	for (i = 0; i < data->acpi_data.state_count; i++) {
+	for (i = 0; i < data->acpi_data->state_count; i++) {
 		u32 index;
 
-		index = data->acpi_data.states[i].control & HW_PSTATE_MASK;
+		index = data->acpi_data->states[i].control & HW_PSTATE_MASK;
 		if (index > data->max_hw_pstate) {
 			printk(KERN_ERR PFX "invalid pstate %d - bad value %d.\n", i, index);
 			printk(KERN_ERR PFX "Please report to BIOS manufacturer\n");
@@ -840,7 +874,7 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpuf
 
 		powernow_table[i].index = index;
 
-		powernow_table[i].frequency = data->acpi_data.states[i].core_frequency * 1000;
+		powernow_table[i].frequency = data->acpi_data->states[i].core_frequency * 1000;
 	}
 	return 0;
 }
@@ -849,16 +883,16 @@ static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpuf
 {
 	int i;
 	int cntlofreq = 0;
-	for (i = 0; i < data->acpi_data.state_count; i++) {
+	for (i = 0; i < data->acpi_data->state_count; i++) {
 		u32 fid;
 		u32 vid;
 
 		if (data->exttype) {
-			fid = data->acpi_data.states[i].status & EXT_FID_MASK;
-			vid = (data->acpi_data.states[i].status >> VID_SHIFT) & EXT_VID_MASK;
+			fid = data->acpi_data->states[i].status & EXT_FID_MASK;
+			vid = (data->acpi_data->states[i].status >> VID_SHIFT) & EXT_VID_MASK;
 		} else {
-			fid = data->acpi_data.states[i].control & FID_MASK;
-			vid = (data->acpi_data.states[i].control >> VID_SHIFT) & VID_MASK;
+			fid = data->acpi_data->states[i].control & FID_MASK;
+			vid = (data->acpi_data->states[i].control >> VID_SHIFT) & VID_MASK;
 		}
 
 		dprintk("   %d : fid 0x%x, vid 0x%x\n", i, fid, vid);
@@ -899,10 +933,10 @@ static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpuf
 				cntlofreq = i;
 		}
 
-		if (powernow_table[i].frequency != (data->acpi_data.states[i].core_frequency * 1000)) {
+		if (powernow_table[i].frequency != (data->acpi_data->states[i].core_frequency * 1000)) {
 			printk(KERN_INFO PFX "invalid freq entries %u kHz vs. %u kHz\n",
 				powernow_table[i].frequency,
-				(unsigned int) (data->acpi_data.states[i].core_frequency * 1000));
+				(unsigned int) (data->acpi_data->states[i].core_frequency * 1000));
 			powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID;
 			continue;
 		}
@@ -912,11 +946,12 @@ static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpuf
 
 static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data)
 {
-	if (data->acpi_data.state_count)
-		acpi_processor_unregister_performance(&data->acpi_data, data->cpu);
+	if (data->acpi_data->state_count)
+		acpi_processor_unregister_performance(data->acpi_data, data->cpu);
 }
 
 #else
+static int powernow_k8_cpu_preinit_acpi(void) { return -ENODEV; }
 static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { return -ENODEV; }
 static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) { return; }
 static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) { return; }
@@ -1101,7 +1136,7 @@ static int powernowk8_verify(struct cpufreq_policy *pol)
 static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 {
 	struct powernow_k8_data *data;
-	cpumask_t oldmask;
+	cpumask_t oldmask = CPU_MASK_ALL;
 	int rc;
 
 	if (!cpu_online(pol->cpu))
@@ -1174,10 +1209,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 	/* run on any CPU again */
 	set_cpus_allowed_ptr(current, &oldmask);
 
-	if (cpu_family == CPU_HW_PSTATE)
-		pol->cpus = cpumask_of_cpu(pol->cpu);
-	else
-		pol->cpus = per_cpu(cpu_core_map, pol->cpu);
+	pol->cpus = data->starting_core_affinity;
 	data->available_cores = &(pol->cpus);
 
 	/* Take a crude guess here.
@@ -1300,6 +1332,7 @@ static int __cpuinit powernowk8_init(void)
 	}
 
 	if (supported_cpus == num_online_cpus()) {
+		powernow_k8_cpu_preinit_acpi();
 		printk(KERN_INFO PFX "Found %d %s "
 			"processors (%d cpu cores) (" VERSION ")\n",
 			num_online_nodes(),
@@ -1316,6 +1349,10 @@ static void __exit powernowk8_exit(void)
 	dprintk("exit\n");
 
 	cpufreq_unregister_driver(&cpufreq_amd64_driver);
+
+#ifdef CONFIG_X86_POWERNOW_K8_ACPI
+	free_percpu(acpi_perf_data);
+#endif
 }
 
 MODULE_AUTHOR("Paul Devriendt <paul.devriendt@amd.com> and Mark Langsdorf <mark.langsdorf@amd.com>");
diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
index ab48cfed4d9..a62612cd4be 100644
--- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
+++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h
@@ -33,12 +33,13 @@ struct powernow_k8_data {
 #ifdef CONFIG_X86_POWERNOW_K8_ACPI
 	/* the acpi table needs to be kept. it's only available if ACPI was
 	 * used to determine valid frequency/vid/fid states */
-	struct acpi_processor_performance acpi_data;
+	struct acpi_processor_performance *acpi_data;
 #endif
 	/* we need to keep track of associated cores, but let cpufreq
 	 * handle hotplug events - so just point at cpufreq pol->cpus
 	 * structure */
 	cpumask_t *available_cores;
+	cpumask_t starting_core_affinity;
 };
 
 
-- 
cgit v1.2.3


From 31343d8a5079cda57ffd539fcf4f00cea344fe98 Mon Sep 17 00:00:00 2001
From: Alok Kataria <akataria@vmware.com>
Date: Fri, 8 Aug 2008 12:15:57 -0700
Subject: x86: Fix broken VMI in 2.6.27-rc..

The lowmem mapping table created by VMI need not depend on max_low_pfn
at all.  Instead we now create an extra large mapping which covers all
possible lowmem instead of the physical ram that is actually available.

This allows the vmi initialization to be done before max_low_pfn could
be computed. We also move the vmi_init code very early in the boot process
so that nobody accidentally breaks the fixmap dependancy.

Signed-off-by: Alok N Kataria <akataria@vmware.com>
Acked-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/setup.c  | 16 ++++++++--------
 arch/x86/kernel/vmi_32.c |  3 ++-
 2 files changed, 10 insertions(+), 9 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 2d888586385..6e5823b9e76 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -604,6 +604,14 @@ void __init setup_arch(char **cmdline_p)
 	early_cpu_init();
 	early_ioremap_init();
 
+#if defined(CONFIG_VMI) && defined(CONFIG_X86_32)
+	/*
+	 * Must be before kernel pagetables are setup
+	 * or fixmap area is touched.
+	 */
+	vmi_init();
+#endif
+
 	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
 	screen_info = boot_params.screen_info;
 	edid_info = boot_params.edid_info;
@@ -817,14 +825,6 @@ void __init setup_arch(char **cmdline_p)
 	kvmclock_init();
 #endif
 
-#if defined(CONFIG_VMI) && defined(CONFIG_X86_32)
-	/*
-	 * Must be after max_low_pfn is determined, and before kernel
-	 * pagetables are setup.
-	 */
-	vmi_init();
-#endif
-
 	paravirt_pagetable_setup_start(swapper_pg_dir);
 	paging_init();
 	paravirt_pagetable_setup_done(swapper_pg_dir);
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 0a1b1a9d922..6ca515d6db5 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -37,6 +37,7 @@
 #include <asm/timer.h>
 #include <asm/vmi_time.h>
 #include <asm/kmap_types.h>
+#include <asm/setup.h>
 
 /* Convenient for calling VMI functions indirectly in the ROM */
 typedef u32 __attribute__((regparm(1))) (VROMFUNC)(void);
@@ -683,7 +684,7 @@ void vmi_bringup(void)
 {
  	/* We must establish the lowmem mapping for MMU ops to work */
 	if (vmi_ops.set_linear_mapping)
-		vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, max_low_pfn, 0);
+		vmi_ops.set_linear_mapping(0, (void *)__PAGE_OFFSET, MAXMEM_PFN, 0);
 }
 
 /*
-- 
cgit v1.2.3


From d388e5fdc461344d04307a3fa83862b9ed429647 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Sat, 9 Aug 2008 15:09:02 -0700
Subject: x86: Restore proper vector locking during cpu hotplug

Having cpu_online_map change during assign_irq_vector can result
in some really nasty and weird things happening.  The one that
bit me last time was accessing non existent per cpu memory for non
existent cpus.

This locking was removed in a sloppy x86_64 and x86_32 merge patch.

Guys can we please try and avoid subtly breaking x86 when we are
merging files together?

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/io_apic_32.c |  6 +-----
 arch/x86/kernel/io_apic_64.c | 25 +++++++++++++++----------
 arch/x86/kernel/smpboot.c    | 12 +++++++++---
 3 files changed, 25 insertions(+), 18 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index de9aa0e3a9c..09cddb57bec 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -57,7 +57,7 @@ atomic_t irq_mis_count;
 static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
 
 static DEFINE_SPINLOCK(ioapic_lock);
-static DEFINE_SPINLOCK(vector_lock);
+DEFINE_SPINLOCK(vector_lock);
 
 int timer_through_8259 __initdata;
 
@@ -1209,10 +1209,6 @@ static int assign_irq_vector(int irq)
 	return vector;
 }
 
-void setup_vector_irq(int cpu)
-{
-}
-
 static struct irq_chip ioapic_chip;
 
 #define IOAPIC_AUTO	-1
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 8269434d170..61a83b70c18 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -101,7 +101,7 @@ int timer_through_8259 __initdata;
 static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
 
 static DEFINE_SPINLOCK(ioapic_lock);
-DEFINE_SPINLOCK(vector_lock);
+static DEFINE_SPINLOCK(vector_lock);
 
 /*
  * # of IRQ routing registers
@@ -697,6 +697,19 @@ static int pin_2_irq(int idx, int apic, int pin)
 	return irq;
 }
 
+void lock_vector_lock(void)
+{
+	/* Used to the online set of cpus does not change
+	 * during assign_irq_vector.
+	 */
+	spin_lock(&vector_lock);
+}
+
+void unlock_vector_lock(void)
+{
+	spin_unlock(&vector_lock);
+}
+
 static int __assign_irq_vector(int irq, cpumask_t mask)
 {
 	/*
@@ -802,7 +815,7 @@ static void __clear_irq_vector(int irq)
 	cpus_clear(cfg->domain);
 }
 
-static void __setup_vector_irq(int cpu)
+void __setup_vector_irq(int cpu)
 {
 	/* Initialize vector_irq on a new cpu */
 	/* This function must be called with vector_lock held */
@@ -825,14 +838,6 @@ static void __setup_vector_irq(int cpu)
 	}
 }
 
-void setup_vector_irq(int cpu)
-{
-	spin_lock(&vector_lock);
-	__setup_vector_irq(smp_processor_id());
-	spin_unlock(&vector_lock);
-}
-
-
 static struct irq_chip ioapic_chip;
 
 static void ioapic_register_intr(int irq, unsigned long trigger)
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 332512767f4..da10f07fc59 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -326,12 +326,16 @@ static void __cpuinit start_secondary(void *unused)
 	 * for which cpus receive the IPI. Holding this
 	 * lock helps us to not include this cpu in a currently in progress
 	 * smp_call_function().
+	 *
+	 * We need to hold vector_lock so there the set of online cpus
+	 * does not change while we are assigning vectors to cpus.  Holding
+	 * this lock ensures we don't half assign or remove an irq from a cpu.
 	 */
 	ipi_call_lock_irq();
-#ifdef CONFIG_X86_IO_APIC
-	setup_vector_irq(smp_processor_id());
-#endif
+	lock_vector_lock();
+	__setup_vector_irq(smp_processor_id());
 	cpu_set(smp_processor_id(), cpu_online_map);
+	unlock_vector_lock();
 	ipi_call_unlock_irq();
 	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 
@@ -1336,7 +1340,9 @@ int __cpu_disable(void)
 	remove_siblinginfo(cpu);
 
 	/* It's now safe to remove this processor from the online map */
+	lock_vector_lock();
 	remove_cpu_from_maps(cpu);
+	unlock_vector_lock();
 	fixup_irqs(cpu_online_map);
 	return 0;
 }
-- 
cgit v1.2.3


From eeb0d7d113895556db473ff1e638803d7d49bff9 Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Mon, 11 Aug 2008 17:44:57 +0200
Subject: x86, debug: tone down arch/x86/kernel/mpparse.c debugging printk

commit 11a62a056093a7f25f1595fbd8bd5f93559572b6 turns some formerly
nopped debugging printks in arch/x86/kernel/mppparse.c into regular
ones. The one at the top of smp_scan_config() in particular also
prints on !CONFIG_SMP/CONFIG_X86_LOCAL_APIC kernels and UP machines
without anything resembling MP tables which makes their lowly UP
owners wonder...

Turn the former Dprintk()s into apic_printk()s instead meaning that
their printing is dependent on passing the apic=verbose (or =debug)
command line param.

On 32-bit, "apic" is a __setup() param which isn't early enough
for this code and therefore needs a followup changing it into an
early_param(). On 64-bit, it already is.

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/mpparse.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index 6ae005ccaed..678090508a6 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -83,7 +83,7 @@ static void __init MP_bus_info(struct mpc_config_bus *m)
 	if (x86_quirks->mpc_oem_bus_info)
 		x86_quirks->mpc_oem_bus_info(m, str);
 	else
-		printk(KERN_INFO "Bus #%d is %s\n", m->mpc_busid, str);
+		apic_printk(APIC_VERBOSE, "Bus #%d is %s\n", m->mpc_busid, str);
 
 #if MAX_MP_BUSSES < 256
 	if (m->mpc_busid >= MAX_MP_BUSSES) {
@@ -154,7 +154,7 @@ static void __init MP_ioapic_info(struct mpc_config_ioapic *m)
 
 static void print_MP_intsrc_info(struct mpc_config_intsrc *m)
 {
-	printk(KERN_CONT "Int: type %d, pol %d, trig %d, bus %02x,"
+	apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x,"
 		" IRQ %02x, APIC ID %x, APIC INT %02x\n",
 		m->mpc_irqtype, m->mpc_irqflag & 3,
 		(m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
@@ -163,7 +163,7 @@ static void print_MP_intsrc_info(struct mpc_config_intsrc *m)
 
 static void __init print_mp_irq_info(struct mp_config_intsrc *mp_irq)
 {
-	printk(KERN_CONT "Int: type %d, pol %d, trig %d, bus %02x,"
+	apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x,"
 		" IRQ %02x, APIC ID %x, APIC INT %02x\n",
 		mp_irq->mp_irqtype, mp_irq->mp_irqflag & 3,
 		(mp_irq->mp_irqflag >> 2) & 3, mp_irq->mp_srcbus,
@@ -235,7 +235,7 @@ static void __init MP_intsrc_info(struct mpc_config_intsrc *m)
 
 static void __init MP_lintsrc_info(struct mpc_config_lintsrc *m)
 {
-	printk(KERN_INFO "Lint: type %d, pol %d, trig %d, bus %02x,"
+	apic_printk(APIC_VERBOSE, "Lint: type %d, pol %d, trig %d, bus %02x,"
 		" IRQ %02x, APIC ID %x, APIC LINT %02x\n",
 		m->mpc_irqtype, m->mpc_irqflag & 3,
 		(m->mpc_irqflag >> 2) & 3, m->mpc_srcbusid,
@@ -695,7 +695,8 @@ static int __init smp_scan_config(unsigned long base, unsigned long length,
 	unsigned int *bp = phys_to_virt(base);
 	struct intel_mp_floating *mpf;
 
-	printk(KERN_DEBUG "Scan SMP from %p for %ld bytes.\n", bp, length);
+	apic_printk(APIC_VERBOSE, "Scan SMP from %p for %ld bytes.\n",
+			bp, length);
 	BUILD_BUG_ON(sizeof(*mpf) != 16);
 
 	while (length > 0) {
-- 
cgit v1.2.3


From fb6bef8002d54a9c3062abc281893ec7c896d3ce Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Mon, 11 Aug 2008 17:45:53 +0200
Subject: x86: make "apic" an early_param() on 32-bit

On 32-bit, "apic" is a __setup() param meaning it is parsed rather
late in the game. Make it an early_param() for apic_printk() use
by arch/x86/kernel/mpparse.c.

On 64-bit, it already is an early_param().

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index d6c89835837..f432d4833a1 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1726,9 +1726,9 @@ static int __init apic_set_verbosity(char *str)
 		apic_verbosity = APIC_DEBUG;
 	else if (strcmp("verbose", str) == 0)
 		apic_verbosity = APIC_VERBOSE;
-	return 1;
+	return 0;
 }
-__setup("apic=", apic_set_verbosity);
+early_param("apic", apic_set_verbosity);
 
 static int __init lapic_insert_resource(void)
 {
-- 
cgit v1.2.3


From 9b0094f7f289d752868aae1a3984517d712cbb22 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 7 Aug 2008 15:14:55 -0700
Subject: x86, pci-calgary: fix function declaration

Fix function declaration:

 linux-next-20080807/arch/x86/kernel/pci-calgary_64.c:1353:36: warning: non-ANSI function declaration of function 'get_tce_space_from_tar'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Acked-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-calgary_64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index b67a4b1d4ea..02d19328525 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -1350,7 +1350,7 @@ static void calgary_init_bitmap_from_tce_table(struct iommu_table *tbl)
  * Function for kdump case. Get the tce tables from first kernel
  * by reading the contents of the base adress register of calgary iommu
  */
-static void get_tce_space_from_tar()
+static void get_tce_space_from_tar(void)
 {
 	int bus;
 	void __iomem *target;
-- 
cgit v1.2.3


From 48d97cb65e62a5f1122ac2cf1149800d4f4693e8 Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Mon, 11 Aug 2008 19:20:17 +0200
Subject: x86: make "apic" an early_param() on 32-bit, NULL check

Cyrill Gorcunov observed:

> you turned it into early_param so now it's NULL injecting vulnerabled.
> Could you please add checking for NULL str param?

fix that.

Also, change the name of 'str' into 'arg', to make it more apparent
that this is an optional argument that can be NULL, not a string
parameter that is empty when unset.

Reported-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic_32.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index f432d4833a1..039a8d4aaf6 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -1720,12 +1720,16 @@ static int __init parse_lapic_timer_c2_ok(char *arg)
 }
 early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok);
 
-static int __init apic_set_verbosity(char *str)
+static int __init apic_set_verbosity(char *arg)
 {
-	if (strcmp("debug", str) == 0)
+	if (!arg)
+		return -EINVAL;
+
+	if (strcmp(arg, "debug") == 0)
 		apic_verbosity = APIC_DEBUG;
-	else if (strcmp("verbose", str) == 0)
+	else if (strcmp(arg, "verbose") == 0)
 		apic_verbosity = APIC_VERBOSE;
+
 	return 0;
 }
 early_param("apic", apic_set_verbosity);
-- 
cgit v1.2.3


From b74548e76a0eab1f29546e7c5a589429c069a680 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yhlu.kernel@gmail.com>
Date: Mon, 11 Aug 2008 13:36:04 -0700
Subject: x86: fix 2.6.27rc1 cannot boot more than 8CPUs

Jeff Chua reported that booting a !bigsmp kernel on a 16-way box
hangs silently.

this is a long-standing issue, smp start AP cpu could check the
apic id >=8 etc before trying to start it.

achieve this by moving the def_to_bigsmp check later and skip the
apicid id > 8

[ mingo@elte.hu: clean up the message that is printed. ]

Reported-by: "Jeff Chua" <jeff.chua.linux@gmail.com>
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

 arch/x86/kernel/setup.c   |    6 ------
 arch/x86/kernel/smpboot.c |   10 ++++++++++
 2 files changed, 10 insertions(+), 6 deletions(-)
---
 arch/x86/kernel/setup.c   |  6 ------
 arch/x86/kernel/smpboot.c | 10 ++++++++++
 2 files changed, 10 insertions(+), 6 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 6e5823b9e76..68b48e3fbcb 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -861,12 +861,6 @@ void __init setup_arch(char **cmdline_p)
 	init_apic_mappings();
 	ioapic_init_mappings();
 
-#if defined(CONFIG_SMP) && defined(CONFIG_X86_PC) && defined(CONFIG_X86_32)
-	if (def_to_bigsmp)
-		printk(KERN_WARNING "More than 8 CPUs detected and "
-			"CONFIG_X86_PC cannot handle it.\nUse "
-			"CONFIG_X86_GENERICARCH or CONFIG_X86_BIGSMP.\n");
-#endif
 	kvm_guest_init();
 
 	e820_reserve_resources();
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index da10f07fc59..91055d7fc1b 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -994,7 +994,17 @@ int __cpuinit native_cpu_up(unsigned int cpu)
 	flush_tlb_all();
 	low_mappings = 1;
 
+#ifdef CONFIG_X86_PC
+	if (def_to_bigsmp && apicid > 8) {
+		printk(KERN_WARNING
+			"More than 8 CPUs detected - skipping them.\n"
+			"Use CONFIG_X86_GENERICARCH and CONFIG_X86_BIGSMP.\n");
+		err = -1;
+	} else
+		err = do_boot_cpu(apicid, cpu);
+#else
 	err = do_boot_cpu(apicid, cpu);
+#endif
 
 	zap_low_mappings();
 	low_mappings = 0;
-- 
cgit v1.2.3