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 3a61ec387c9092dfc91a5959145d36835a72fc4c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Fri, 25 Jul 2008 13:07:50 +0200
Subject: x86, AMD IOMMU: include amd_iommu_last_bdf in device initialization

All the values read while searching for amd_iommu_last_bdf are defined as
inclusive. Let the code handle this value as such. Found by Wei Wang. Thanks
Wei.

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
Cc: Wei Wang <wei.wang2@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/amd_iommu.c      | 4 ++--
 arch/x86/kernel/amd_iommu_init.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index c25210e6ac8..74697408576 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -667,7 +667,7 @@ static int get_device_resources(struct device *dev,
 	_bdf = calc_devid(pcidev->bus->number, pcidev->devfn);
 
 	/* device not translated by any IOMMU in the system? */
-	if (_bdf >= amd_iommu_last_bdf) {
+	if (_bdf > amd_iommu_last_bdf) {
 		*iommu = NULL;
 		*domain = NULL;
 		*bdf = 0xffff;
@@ -1085,7 +1085,7 @@ void prealloc_protection_domains(void)
 
 	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
 		devid = (dev->bus->number << 8) | dev->devfn;
-		if (devid >= amd_iommu_last_bdf)
+		if (devid > amd_iommu_last_bdf)
 			continue;
 		devid = amd_iommu_alias_table[devid];
 		if (domain_for_device(devid))
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c9d8ff2eb13..d9a9da597e7 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -732,7 +732,7 @@ static int __init init_exclusion_range(struct ivmd_header *m)
 		set_device_exclusion_range(m->devid, m);
 		break;
 	case ACPI_IVMD_TYPE_ALL:
-		for (i = 0; i < amd_iommu_last_bdf; ++i)
+		for (i = 0; i <= amd_iommu_last_bdf; ++i)
 			set_device_exclusion_range(i, m);
 		break;
 	case ACPI_IVMD_TYPE_RANGE:
@@ -934,7 +934,7 @@ int __init amd_iommu_init(void)
 	/*
 	 * let all alias entries point to itself
 	 */
-	for (i = 0; i < amd_iommu_last_bdf; ++i)
+	for (i = 0; i <= amd_iommu_last_bdf; ++i)
 		amd_iommu_alias_table[i] = i;
 
 	/*
-- 
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 8d8bb39b9eba32dd70e87fd5ad5c5dd4ba118e06 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 25 Jul 2008 19:44:49 -0700
Subject: dma-mapping: add the device argument to dma_mapping_error()

Add per-device dma_mapping_ops support for CONFIG_X86_64 as POWER
architecture does:

This enables us to cleanly fix the Calgary IOMMU issue that some devices
are not behind the IOMMU (http://lkml.org/lkml/2008/5/8/423).

I think that per-device dma_mapping_ops support would be also helpful for
KVM people to support PCI passthrough but Andi thinks that this makes it
difficult to support the PCI passthrough (see the above thread).  So I
CC'ed this to KVM camp.  Comments are appreciated.

A pointer to dma_mapping_ops to struct dev_archdata is added.  If the
pointer is non NULL, DMA operations in asm/dma-mapping.h use it.  If it's
NULL, the system-wide dma_ops pointer is used as before.

If it's useful for KVM people, I plan to implement a mechanism to register
a hook called when a new pci (or dma capable) device is created (it works
with hot plugging).  It enables IOMMUs to set up an appropriate
dma_mapping_ops per device.

The major obstacle is that dma_mapping_error doesn't take a pointer to the
device unlike other DMA operations.  So x86 can't have dma_mapping_ops per
device.  Note all the POWER IOMMUs use the same dma_mapping_error function
so this is not a problem for POWER but x86 IOMMUs use different
dma_mapping_error functions.

The first patch adds the device argument to dma_mapping_error.  The patch
is trivial but large since it touches lots of drivers and dma-mapping.h in
all the architecture.

This patch:

dma_mapping_error() doesn't take a pointer to the device unlike other DMA
operations.  So we can't have dma_mapping_ops per device.

Note that POWER already has dma_mapping_ops per device but all the POWER
IOMMUs use the same dma_mapping_error function.  x86 IOMMUs use device
argument.

[akpm@linux-foundation.org: fix sge]
[akpm@linux-foundation.org: fix svc_rdma]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix bnx2x]
[akpm@linux-foundation.org: fix s2io]
[akpm@linux-foundation.org: fix pasemi_mac]
[akpm@linux-foundation.org: fix sdhci]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix sparc]
[akpm@linux-foundation.org: fix ibmvscsi]
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Avi Kivity <avi@qumranet.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/pci-calgary_64.c |  2 +-
 arch/x86/kernel/pci-dma.c        | 27 ++++++++++++++++-----------
 arch/x86/kernel/pci-gart_64.c    |  3 +--
 arch/x86/kernel/pci-nommu.c      | 14 +-------------
 arch/x86/kernel/pci-swiotlb_64.c |  2 +-
 5 files changed, 20 insertions(+), 28 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 19e7fc7c2c4..1eb86be93d7 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -544,7 +544,7 @@ error:
 	return ret;
 }
 
-static const struct dma_mapping_ops calgary_dma_ops = {
+static struct dma_mapping_ops calgary_dma_ops = {
 	.alloc_coherent = calgary_alloc_coherent,
 	.map_single = calgary_map_single,
 	.unmap_single = calgary_unmap_single,
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index cbecb05551b..37544123896 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -11,7 +11,7 @@
 
 static int forbid_dac __read_mostly;
 
-const struct dma_mapping_ops *dma_ops;
+struct dma_mapping_ops *dma_ops;
 EXPORT_SYMBOL(dma_ops);
 
 static int iommu_sac_force __read_mostly;
@@ -312,6 +312,8 @@ static int dma_release_coherent(struct device *dev, int order, void *vaddr)
 
 int dma_supported(struct device *dev, u64 mask)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 #ifdef CONFIG_PCI
 	if (mask > 0xffffffff && forbid_dac > 0) {
 		dev_info(dev, "PCI: Disallowing DAC for device\n");
@@ -319,8 +321,8 @@ int dma_supported(struct device *dev, u64 mask)
 	}
 #endif
 
-	if (dma_ops->dma_supported)
-		return dma_ops->dma_supported(dev, mask);
+	if (ops->dma_supported)
+		return ops->dma_supported(dev, mask);
 
 	/* Copied from i386. Doesn't make much sense, because it will
 	   only work for pci_alloc_coherent.
@@ -367,6 +369,7 @@ void *
 dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		   gfp_t gfp)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
 	void *memory = NULL;
 	struct page *page;
 	unsigned long dma_mask = 0;
@@ -435,8 +438,8 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 			/* Let low level make its own zone decisions */
 			gfp &= ~(GFP_DMA32|GFP_DMA);
 
-			if (dma_ops->alloc_coherent)
-				return dma_ops->alloc_coherent(dev, size,
+			if (ops->alloc_coherent)
+				return ops->alloc_coherent(dev, size,
 							   dma_handle, gfp);
 			return NULL;
 		}
@@ -448,14 +451,14 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		}
 	}
 
-	if (dma_ops->alloc_coherent) {
+	if (ops->alloc_coherent) {
 		free_pages((unsigned long)memory, get_order(size));
 		gfp &= ~(GFP_DMA|GFP_DMA32);
-		return dma_ops->alloc_coherent(dev, size, dma_handle, gfp);
+		return ops->alloc_coherent(dev, size, dma_handle, gfp);
 	}
 
-	if (dma_ops->map_simple) {
-		*dma_handle = dma_ops->map_simple(dev, virt_to_phys(memory),
+	if (ops->map_simple) {
+		*dma_handle = ops->map_simple(dev, virt_to_phys(memory),
 					      size,
 					      PCI_DMA_BIDIRECTIONAL);
 		if (*dma_handle != bad_dma_address)
@@ -477,12 +480,14 @@ EXPORT_SYMBOL(dma_alloc_coherent);
 void dma_free_coherent(struct device *dev, size_t size,
 			 void *vaddr, dma_addr_t bus)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 	int order = get_order(size);
 	WARN_ON(irqs_disabled());	/* for portability */
 	if (dma_release_coherent(dev, order, vaddr))
 		return;
-	if (dma_ops->unmap_single)
-		dma_ops->unmap_single(dev, bus, size, 0);
+	if (ops->unmap_single)
+		ops->unmap_single(dev, bus, size, 0);
 	free_pages((unsigned long)vaddr, order);
 }
 EXPORT_SYMBOL(dma_free_coherent);
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index df5f142657d..744126e6495 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -692,8 +692,7 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 
 extern int agp_amd64_init(void);
 
-static const struct dma_mapping_ops gart_dma_ops = {
-	.mapping_error			= NULL,
+static struct dma_mapping_ops gart_dma_ops = {
 	.map_single			= gart_map_single,
 	.map_simple			= gart_map_simple,
 	.unmap_single			= gart_unmap_single,
diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index 792b9179eff..3f91f71cdc3 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -72,21 +72,9 @@ static int nommu_map_sg(struct device *hwdev, struct scatterlist *sg,
 	return nents;
 }
 
-/* Make sure we keep the same behaviour */
-static int nommu_mapping_error(dma_addr_t dma_addr)
-{
-#ifdef CONFIG_X86_32
-	return 0;
-#else
-	return (dma_addr == bad_dma_address);
-#endif
-}
-
-
-const struct dma_mapping_ops nommu_dma_ops = {
+struct dma_mapping_ops nommu_dma_ops = {
 	.map_single = nommu_map_single,
 	.map_sg = nommu_map_sg,
-	.mapping_error = nommu_mapping_error,
 	.is_phys = 1,
 };
 
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 20df839b9c2..c4ce0332759 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -18,7 +18,7 @@ swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
 	return swiotlb_map_single(hwdev, phys_to_virt(paddr), size, direction);
 }
 
-const struct dma_mapping_ops swiotlb_dma_ops = {
+struct dma_mapping_ops swiotlb_dma_ops = {
 	.mapping_error = swiotlb_dma_mapping_error,
 	.alloc_coherent = swiotlb_alloc_coherent,
 	.free_coherent = swiotlb_free_coherent,
-- 
cgit v1.2.3


From 1956a96de488feb05e95c08c9d5e80f63a4be2b1 Mon Sep 17 00:00:00 2001
From: Alexis Bruemmer <alexisb@us.ibm.com>
Date: Fri, 25 Jul 2008 19:44:51 -0700
Subject: x86 calgary: fix handling of devices that aren't behind the Calgary

The calgary code can give drivers addresses above 4GB which is very bad
for hardware that is only 32bit DMA addressable.

With this patch, the calgary code sets the global dma_ops to swiotlb or
nommu properly, and the dma_ops of devices behind the Calgary/CalIOC2
to calgary_dma_ops.  So the calgary code can handle devices safely that
aren't behind the Calgary/CalIOC2.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Alexis Bruemmer <alexisb@us.ibm.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/pci-calgary_64.c | 71 +++++++++++++++-------------------------
 1 file changed, 26 insertions(+), 45 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 1eb86be93d7..b67a4b1d4ea 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -37,6 +37,7 @@
 #include <linux/delay.h>
 #include <linux/scatterlist.h>
 #include <linux/iommu-helper.h>
+
 #include <asm/iommu.h>
 #include <asm/calgary.h>
 #include <asm/tce.h>
@@ -413,22 +414,6 @@ static void calgary_unmap_sg(struct device *dev,
 	}
 }
 
-static int calgary_nontranslate_map_sg(struct device* dev,
-	struct scatterlist *sg, int nelems, int direction)
-{
-	struct scatterlist *s;
-	int i;
-
-	for_each_sg(sg, s, nelems, i) {
-		struct page *p = sg_page(s);
-
-		BUG_ON(!p);
-		s->dma_address = virt_to_bus(sg_virt(s));
-		s->dma_length = s->length;
-	}
-	return nelems;
-}
-
 static int calgary_map_sg(struct device *dev, struct scatterlist *sg,
 	int nelems, int direction)
 {
@@ -439,9 +424,6 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg,
 	unsigned long entry;
 	int i;
 
-	if (!translation_enabled(tbl))
-		return calgary_nontranslate_map_sg(dev, sg, nelems, direction);
-
 	for_each_sg(sg, s, nelems, i) {
 		BUG_ON(!sg_page(s));
 
@@ -477,7 +459,6 @@ error:
 static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr,
 	size_t size, int direction)
 {
-	dma_addr_t dma_handle = bad_dma_address;
 	void *vaddr = phys_to_virt(paddr);
 	unsigned long uaddr;
 	unsigned int npages;
@@ -486,12 +467,7 @@ static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr,
 	uaddr = (unsigned long)vaddr;
 	npages = num_dma_pages(uaddr, size);
 
-	if (translation_enabled(tbl))
-		dma_handle = iommu_alloc(dev, tbl, vaddr, npages, direction);
-	else
-		dma_handle = virt_to_bus(vaddr);
-
-	return dma_handle;
+	return iommu_alloc(dev, tbl, vaddr, npages, direction);
 }
 
 static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle,
@@ -500,9 +476,6 @@ static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle,
 	struct iommu_table *tbl = find_iommu_table(dev);
 	unsigned int npages;
 
-	if (!translation_enabled(tbl))
-		return;
-
 	npages = num_dma_pages(dma_handle, size);
 	iommu_free(tbl, dma_handle, npages);
 }
@@ -525,18 +498,12 @@ static void* calgary_alloc_coherent(struct device *dev, size_t size,
 		goto error;
 	memset(ret, 0, size);
 
-	if (translation_enabled(tbl)) {
-		/* set up tces to cover the allocated range */
-		mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL);
-		if (mapping == bad_dma_address)
-			goto free;
-
-		*dma_handle = mapping;
-	} else /* non translated slot */
-		*dma_handle = virt_to_bus(ret);
-
+	/* set up tces to cover the allocated range */
+	mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL);
+	if (mapping == bad_dma_address)
+		goto free;
+	*dma_handle = mapping;
 	return ret;
-
 free:
 	free_pages((unsigned long)ret, get_order(size));
 	ret = NULL;
@@ -1241,6 +1208,16 @@ static int __init calgary_init(void)
 			goto error;
 	} while (1);
 
+	dev = NULL;
+	for_each_pci_dev(dev) {
+		struct iommu_table *tbl;
+
+		tbl = find_iommu_table(&dev->dev);
+
+		if (translation_enabled(tbl))
+			dev->dev.archdata.dma_ops = &calgary_dma_ops;
+	}
+
 	return ret;
 
 error:
@@ -1262,6 +1239,7 @@ error:
 		calgary_disable_translation(dev);
 		calgary_free_bus(dev);
 		pci_dev_put(dev); /* Undo calgary_init_one()'s pci_dev_get() */
+		dev->dev.archdata.dma_ops = NULL;
 	} while (1);
 
 	return ret;
@@ -1503,6 +1481,10 @@ void __init detect_calgary(void)
 		printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d, "
 		       "CONFIG_IOMMU_DEBUG is %s.\n", specified_table_size,
 		       debugging ? "enabled" : "disabled");
+
+		/* swiotlb for devices that aren't behind the Calgary. */
+		if (max_pfn > MAX_DMA32_PFN)
+			swiotlb = 1;
 	}
 	return;
 
@@ -1519,7 +1501,7 @@ int __init calgary_iommu_init(void)
 {
 	int ret;
 
-	if (no_iommu || swiotlb)
+	if (no_iommu || (swiotlb && !calgary_detected))
 		return -ENODEV;
 
 	if (!calgary_detected)
@@ -1532,15 +1514,14 @@ int __init calgary_iommu_init(void)
 	if (ret) {
 		printk(KERN_ERR "PCI-DMA: Calgary init failed %d, "
 		       "falling back to no_iommu\n", ret);
-		if (max_pfn > MAX_DMA32_PFN)
-			printk(KERN_ERR "WARNING more than 4GB of memory, "
-					"32bit PCI may malfunction.\n");
 		return ret;
 	}
 
 	force_iommu = 1;
 	bad_dma_address = 0x0;
-	dma_ops = &calgary_dma_ops;
+	/* dma_ops is set to swiotlb or nommu */
+	if (!dma_ops)
+		dma_ops = &nommu_dma_ops;
 
 	return 0;
 }
-- 
cgit v1.2.3


From 3ab83521378268044a448113c6aa9a9e245f4d2f Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 25 Jul 2008 19:45:07 -0700
Subject: kexec jump

This patch provides an enhancement to kexec/kdump.  It implements the
following features:

- Backup/restore memory used by the original kernel before/after
  kexec.

- Save/restore CPU state before/after kexec.

The features of this patch can be used as a general method to call program in
physical mode (paging turning off).  This can be used to call BIOS code under
Linux.

kexec-tools needs to be patched to support kexec jump. The patches and
the precompiled kexec can be download from the following URL:

       source: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-src_git_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-patches_git_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec_git_kh10

Usage example of calling some physical mode code and return:

1. Compile and install patched kernel with following options selected:

CONFIG_X86_32=y
CONFIG_KEXEC=y
CONFIG_PM=y
CONFIG_KEXEC_JUMP=y

2. Build patched kexec-tool or download the pre-built one.

3. Build some physical mode executable named such as "phy_mode"

4. Boot kernel compiled in step 1.

5. Load physical mode executable with /sbin/kexec. The shell command
   line can be as follow:

   /sbin/kexec --load-preserve-context --args-none phy_mode

6. Call physical mode executable with following shell command line:

   /sbin/kexec -e

Implementation point:

To support jumping without reserving memory.  One shadow backup page (source
page) is allocated for each page used by kexeced code image (destination
page).  When do kexec_load, the image of kexeced code is loaded into source
pages, and before executing, the destination pages and the source pages are
swapped, so the contents of destination pages are backupped.  Before jumping
to the kexeced code image and after jumping back to the original kernel, the
destination pages and the source pages are swapped too.

C ABI (calling convention) is used as communication protocol between
kernel and called code.

A flag named KEXEC_PRESERVE_CONTEXT for sys_kexec_load is added to
indicate that the loaded kernel image is used for jumping back.

Now, only the i386 architecture is supported.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/machine_kexec_32.c   |  27 ++++--
 arch/x86/kernel/machine_kexec_64.c   |   2 +-
 arch/x86/kernel/relocate_kernel_32.S | 174 ++++++++++++++++++++++++++++++-----
 3 files changed, 172 insertions(+), 31 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 8864230d55a..2b67609d0a1 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -22,6 +22,7 @@
 #include <asm/cpufeature.h>
 #include <asm/desc.h>
 #include <asm/system.h>
+#include <asm/cacheflush.h>
 
 #define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
 static u32 kexec_pgd[1024] PAGE_ALIGNED;
@@ -85,10 +86,12 @@ static void load_segments(void)
  * reboot code buffer to allow us to avoid allocations
  * later.
  *
- * Currently nothing.
+ * Make control page executable.
  */
 int machine_kexec_prepare(struct kimage *image)
 {
+	if (nx_enabled)
+		set_pages_x(image->control_code_page, 1);
 	return 0;
 }
 
@@ -98,16 +101,24 @@ int machine_kexec_prepare(struct kimage *image)
  */
 void machine_kexec_cleanup(struct kimage *image)
 {
+	if (nx_enabled)
+		set_pages_nx(image->control_code_page, 1);
 }
 
 /*
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
+	asmlinkage unsigned long
+		(*relocate_kernel_ptr)(unsigned long indirection_page,
+				       unsigned long control_page,
+				       unsigned long start_address,
+				       unsigned int has_pae,
+				       unsigned int preserve_context);
 
 	tracer_disable();
 
@@ -115,10 +126,11 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	local_irq_disable();
 
 	control_page = page_address(image->control_code_page);
-	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+	memcpy(control_page, relocate_kernel, PAGE_SIZE/2);
 
+	relocate_kernel_ptr = control_page;
 	page_list[PA_CONTROL_PAGE] = __pa(control_page);
-	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[VA_CONTROL_PAGE] = (unsigned long)control_page;
 	page_list[PA_PGD] = __pa(kexec_pgd);
 	page_list[VA_PGD] = (unsigned long)kexec_pgd;
 #ifdef CONFIG_X86_PAE
@@ -131,6 +143,7 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
 	page_list[PA_PTE_1] = __pa(kexec_pte1);
 	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
+	page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) << PAGE_SHIFT);
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -149,8 +162,10 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	set_idt(phys_to_virt(0),0);
 
 	/* now call it */
-	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
-			image->start, cpu_has_pae);
+	image->start = relocate_kernel_ptr((unsigned long)image->head,
+					   (unsigned long)page_list,
+					   image->start, cpu_has_pae,
+					   image->preserve_context);
 }
 
 void arch_crash_save_vmcoreinfo(void)
diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index 9dd9262693a..c43caa3a91f 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -181,7 +181,7 @@ void machine_kexec_cleanup(struct kimage *image)
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S
index c30fe25d470..703310a9902 100644
--- a/arch/x86/kernel/relocate_kernel_32.S
+++ b/arch/x86/kernel/relocate_kernel_32.S
@@ -20,11 +20,44 @@
 #define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define PAE_PGD_ATTR (_PAGE_PRESENT)
 
+/* control_page + PAGE_SIZE/2 ~ control_page + PAGE_SIZE * 3/4 are
+ * used to save some data for jumping back
+ */
+#define DATA(offset)		(PAGE_SIZE/2+(offset))
+
+/* Minimal CPU state */
+#define ESP			DATA(0x0)
+#define CR0			DATA(0x4)
+#define CR3			DATA(0x8)
+#define CR4			DATA(0xc)
+
+/* other data */
+#define CP_VA_CONTROL_PAGE	DATA(0x10)
+#define CP_PA_PGD		DATA(0x14)
+#define CP_PA_SWAP_PAGE		DATA(0x18)
+#define CP_PA_BACKUP_PAGES_MAP	DATA(0x1c)
+
 	.text
 	.align PAGE_SIZE
 	.globl relocate_kernel
 relocate_kernel:
-	movl	8(%esp), %ebp /* list of pages */
+	/* Save the CPU context, used for jumping back */
+
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	pushl	%ebp
+	pushf
+
+	movl	20+8(%esp), %ebp /* list of pages */
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %edi
+	movl	%esp, ESP(%edi)
+	movl	%cr0, %eax
+	movl	%eax, CR0(%edi)
+	movl	%cr3, %eax
+	movl	%eax, CR3(%edi)
+	movl	%cr4, %eax
+	movl	%eax, CR4(%edi)
 
 #ifdef CONFIG_X86_PAE
 	/* map the control page at its virtual address */
@@ -138,15 +171,25 @@ relocate_kernel:
 
 relocate_new_kernel:
 	/* read the arguments and say goodbye to the stack */
-	movl  4(%esp), %ebx /* page_list */
-	movl  8(%esp), %ebp /* list of pages */
-	movl  12(%esp), %edx /* start address */
-	movl  16(%esp), %ecx /* cpu_has_pae */
+	movl  20+4(%esp), %ebx /* page_list */
+	movl  20+8(%esp), %ebp /* list of pages */
+	movl  20+12(%esp), %edx /* start address */
+	movl  20+16(%esp), %ecx /* cpu_has_pae */
+	movl  20+20(%esp), %esi /* preserve_context */
 
 	/* zero out flags, and disable interrupts */
 	pushl $0
 	popfl
 
+	/* save some information for jumping back */
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %edi
+	movl	%edi, CP_VA_CONTROL_PAGE(%edi)
+	movl	PTR(PA_PGD)(%ebp), %eax
+	movl	%eax, CP_PA_PGD(%edi)
+	movl	PTR(PA_SWAP_PAGE)(%ebp), %eax
+	movl	%eax, CP_PA_SWAP_PAGE(%edi)
+	movl	%ebx, CP_PA_BACKUP_PAGES_MAP(%edi)
+
 	/* get physical address of control page now */
 	/* this is impossible after page table switch */
 	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edi
@@ -197,8 +240,90 @@ identity_mapped:
 	xorl	%eax, %eax
 	movl	%eax, %cr3
 
+	movl	CP_PA_SWAP_PAGE(%edi), %eax
+	pushl	%eax
+	pushl	%ebx
+	call	swap_pages
+	addl	$8, %esp
+
+	/* To be certain of avoiding problems with self-modifying code
+	 * I need to execute a serializing instruction here.
+	 * So I flush the TLB, it's handy, and not processor dependent.
+	 */
+	xorl	%eax, %eax
+	movl	%eax, %cr3
+
+	/* set all of the registers to known values */
+	/* leave %esp alone */
+
+	testl	%esi, %esi
+	jnz 1f
+	xorl	%edi, %edi
+	xorl	%eax, %eax
+	xorl	%ebx, %ebx
+	xorl    %ecx, %ecx
+	xorl    %edx, %edx
+	xorl    %esi, %esi
+	xorl    %ebp, %ebp
+	ret
+1:
+	popl	%edx
+	movl	CP_PA_SWAP_PAGE(%edi), %esp
+	addl	$PAGE_SIZE, %esp
+2:
+	call	*%edx
+
+	/* get the re-entry point of the peer system */
+	movl	0(%esp), %ebp
+	call	1f
+1:
+	popl	%ebx
+	subl	$(1b - relocate_kernel), %ebx
+	movl	CP_VA_CONTROL_PAGE(%ebx), %edi
+	lea	PAGE_SIZE(%ebx), %esp
+	movl	CP_PA_SWAP_PAGE(%ebx), %eax
+	movl	CP_PA_BACKUP_PAGES_MAP(%ebx), %edx
+	pushl	%eax
+	pushl	%edx
+	call	swap_pages
+	addl	$8, %esp
+	movl	CP_PA_PGD(%ebx), %eax
+	movl	%eax, %cr3
+	movl	%cr0, %eax
+	orl	$(1<<31), %eax
+	movl	%eax, %cr0
+	lea	PAGE_SIZE(%edi), %esp
+	movl	%edi, %eax
+	addl	$(virtual_mapped - relocate_kernel), %eax
+	pushl	%eax
+	ret
+
+virtual_mapped:
+	movl	CR4(%edi), %eax
+	movl	%eax, %cr4
+	movl	CR3(%edi), %eax
+	movl	%eax, %cr3
+	movl	CR0(%edi), %eax
+	movl	%eax, %cr0
+	movl	ESP(%edi), %esp
+	movl	%ebp, %eax
+
+	popf
+	popl	%ebp
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	ret
+
 	/* Do the copies */
-	movl	%ebx, %ecx
+swap_pages:
+	movl	8(%esp), %edx
+	movl	4(%esp), %ecx
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%edi
+	pushl	%esi
+	movl	%ecx, %ebx
 	jmp	1f
 
 0:	/* top, read another word from the indirection page */
@@ -226,27 +351,28 @@ identity_mapped:
 	movl    %ecx,   %esi /* For every source page do a copy */
 	andl    $0xfffff000, %esi
 
+	movl	%edi, %eax
+	movl	%esi, %ebp
+
+	movl	%edx, %edi
 	movl    $1024, %ecx
 	rep ; movsl
-	jmp     0b
 
-3:
-
-	/* To be certain of avoiding problems with self-modifying code
-	 * I need to execute a serializing instruction here.
-	 * So I flush the TLB, it's handy, and not processor dependent.
-	 */
-	xorl	%eax, %eax
-	movl	%eax, %cr3
+	movl	%ebp, %edi
+	movl	%eax, %esi
+	movl	$1024, %ecx
+	rep ; movsl
 
-	/* set all of the registers to known values */
-	/* leave %esp alone */
+	movl	%eax, %edi
+	movl	%edx, %esi
+	movl	$1024, %ecx
+	rep ; movsl
 
-	xorl	%eax, %eax
-	xorl	%ebx, %ebx
-	xorl    %ecx, %ecx
-	xorl    %edx, %edx
-	xorl    %esi, %esi
-	xorl    %edi, %edi
-	xorl    %ebp, %ebp
+	lea	PAGE_SIZE(%ebp), %esi
+	jmp     0b
+3:
+	popl	%esi
+	popl	%edi
+	popl	%ebx
+	popl	%ebp
 	ret
-- 
cgit v1.2.3


From 89081d17f7bb81d89fa1aa9b70f821c5cf4d39e9 Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 25 Jul 2008 19:45:10 -0700
Subject: kexec jump: save/restore device state

This patch implements devices state save/restore before after kexec.

This patch together with features in kexec_jump patch can be used for
following:

- A simple hibernation implementation without ACPI support.  You can kexec a
  hibernating kernel, save the memory image of original system and shutdown
  the system.  When resuming, you restore the memory image of original system
  via ordinary kexec load then jump back.

- Kernel/system debug through making system snapshot.  You can make system
  snapshot, jump back, do some thing and make another system snapshot.

- Cooperative multi-kernel/system.  With kexec jump, you can switch between
  several kernels/systems quickly without boot process except the first time.
  This appears like swap a whole kernel/system out/in.

- A general method to call program in physical mode (paging turning
  off). This can be used to invoke BIOS code under Linux.

The following user-space tools can be used with kexec jump:

- kexec-tools needs to be patched to support kexec jump. The patches
  and the precompiled kexec can be download from the following URL:
       source: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-src_git_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-patches_git_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec_git_kh10

- makedumpfile with patches are used as memory image saving tool, it
  can exclude free pages from original kernel memory image file. The
  patches and the precompiled makedumpfile can be download from the
  following URL:
       source: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-src_cvs_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-patches_cvs_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile_cvs_kh10

- An initramfs image can be used as the root file system of kexeced
  kernel. An initramfs image built with "BuildRoot" can be downloaded
  from the following URL:
       initramfs image: http://khibernation.sourceforge.net/download/release_v10/initramfs/rootfs_cvs_kh10.gz
  All user space tools above are included in the initramfs image.

Usage example of simple hibernation:

1. Compile and install patched kernel with following options selected:

CONFIG_X86_32=y
CONFIG_RELOCATABLE=y
CONFIG_KEXEC=y
CONFIG_CRASH_DUMP=y
CONFIG_PM=y
CONFIG_HIBERNATION=y
CONFIG_KEXEC_JUMP=y

2. Build an initramfs image contains kexec-tool and makedumpfile, or
   download the pre-built initramfs image, called rootfs.gz in
   following text.

3. Prepare a partition to save memory image of original kernel, called
   hibernating partition in following text.

4. Boot kernel compiled in step 1 (kernel A).

5. In the kernel A, load kernel compiled in step 1 (kernel B) with
   /sbin/kexec. The shell command line can be as follow:

   /sbin/kexec --load-preserve-context /boot/bzImage --mem-min=0x100000
     --mem-max=0xffffff --initrd=rootfs.gz

6. Boot the kernel B with following shell command line:

   /sbin/kexec -e

7. The kernel B will boot as normal kexec. In kernel B the memory
   image of kernel A can be saved into hibernating partition as
   follow:

   jump_back_entry=`cat /proc/cmdline | tr ' ' '\n' | grep kexec_jump_back_entry | cut -d '='`
   echo $jump_back_entry > kexec_jump_back_entry
   cp /proc/vmcore dump.elf

   Then you can shutdown the machine as normal.

8. Boot kernel compiled in step 1 (kernel C). Use the rootfs.gz as
   root file system.

9. In kernel C, load the memory image of kernel A as follow:

   /sbin/kexec -l --args-none --entry=`cat kexec_jump_back_entry` dump.elf

10. Jump back to the kernel A as follow:

   /sbin/kexec -e

   Then, kernel A is resumed.

Implementation point:

To support jumping between two kernels, before jumping to (executing)
the new kernel and jumping back to the original kernel, the devices
are put into quiescent state, and the state of devices and CPU is
saved. After jumping back from kexeced kernel and jumping to the new
kernel, the state of devices and CPU are restored accordingly. The
devices/CPU state save/restore code of software suspend is called to
implement corresponding function.

Known issues:

- Because the segment number supported by sys_kexec_load is limited,
  hibernation image with many segments may not be load. This is
  planned to be eliminated by adding a new flag to sys_kexec_load to
  make a image can be loaded with multiple sys_kexec_load invoking.

Now, only the i386 architecture is supported.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/machine_kexec_32.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 2b67609d0a1..9fe478d9840 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -125,6 +125,18 @@ void machine_kexec(struct kimage *image)
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
+	if (image->preserve_context) {
+#ifdef CONFIG_X86_IO_APIC
+		/* We need to put APICs in legacy mode so that we can
+		 * get timer interrupts in second kernel. kexec/kdump
+		 * paths already have calls to disable_IO_APIC() in
+		 * one form or other. kexec jump path also need
+		 * one.
+		 */
+		disable_IO_APIC();
+#endif
+	}
+
 	control_page = page_address(image->control_code_page);
 	memcpy(control_page, relocate_kernel, PAGE_SIZE/2);
 
-- 
cgit v1.2.3


From 583323b9d2f624884a8c9563fb5a4d795f39ab07 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 27 Jul 2008 21:43:11 +0200
Subject: x86: fix cpu hotplug on 32bit

commit 3e9704739daf46a8ba6593d749c67b5f7cd633d2 ("x86: boot secondary
cpus through initial_code") causes the kernel to crash when a CPU is
brought online after the read only sections have been write
protected. The write to initial_code in do_boot_cpu() fails.

Move inital_code to .cpuinit.data section.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/head_32.S | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'arch/x86/kernel')

diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index f67e93441ca..a7010c3a377 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -456,9 +456,6 @@ is386:	movl $2,%ecx		# set MP
 1:
 #endif /* CONFIG_SMP */
 	jmp *(initial_code)
-.align 4
-ENTRY(initial_code)
-	.long i386_start_kernel
 
 /*
  * We depend on ET to be correct. This checks for 287/387.
@@ -601,6 +598,11 @@ ignore_int:
 #endif
 	iret
 
+.section .cpuinit.data,"wa"
+.align 4
+ENTRY(initial_code)
+	.long i386_start_kernel
+
 .section .text
 /*
  * Real beginning of normal "text" segment
-- 
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 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