Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86: (77 commits)
  x86: UV startup of slave cpus
  x86: integrate pci-dma.c
  x86: don't do dma if mask is NULL.
  x86: return conditional to mmu
  x86: remove kludge from x86_64
  x86: unify gfp masks
  x86: retry allocation if failed
  x86: don't try to allocate from DMA zone at first
  x86: use a fallback dev for i386
  x86: use numa allocation function in i386
  x86: remove virt_to_bus in pci-dma_64.c
  x86: adjust dma_free_coherent for i386
  x86: move bad_dma_address
  x86: isolate coherent mapping functions
  x86: move dma_coherent functions to pci-dma.c
  x86: merge iommu initialization parameters
  x86: merge dma_supported
  x86: move pci fixup to pci-dma.c
  x86: move x86_64-specific to common code.
  x86: move initialization functions to pci-dma.c
  ...

Documentation/feature-removal-schedule.txt

@@ -282,6 +282,13 @@ Why:	Not used in-tree. The current out-of-tree users used it to
 	out-of-tree driver.
 Who:	Thomas Gleixner <tglx@linutronix.de>
 
+----------------------------
+
+What:	usedac i386 kernel parameter
+When:	2.6.27
+Why:	replaced by allowdac and no dac combination
+Who:	Glauber Costa <gcosta@redhat.com>
+
 ---------------------------
 
 What:	/sys/o2cb symlink

Documentation/kernel-parameters.txt

@@ -1280,8 +1280,16 @@ and is between 256 and 4096 characters. It is defined in the file
 	noexec		[IA-64]
 
 	noexec		[X86-32,X86-64]
+			On X86-32 available only on PAE configured kernels.
 			noexec=on: enable non-executable mappings (default)
-			noexec=off: disable nn-executable mappings
+			noexec=off: disable non-executable mappings
+
+	noexec32	[X86-64]
+			This affects only 32-bit executables.
+			noexec32=on: enable non-executable mappings (default)
+				read doesn't imply executable mappings
+			noexec32=off: disable non-executable mappings
+				read implies executable mappings
 
 	nofxsr		[BUGS=X86-32] Disables x86 floating point extended
 			register save and restore. The kernel will only save

Documentation/prctl/disable-tsc-ctxt-sw-stress-test.c

@@ -0,0 +1,96 @@
+/*
+ * Tests for prctl(PR_GET_TSC, ...) / prctl(PR_SET_TSC, ...)
+ *
+ * Tests if the control register is updated correctly
+ * at context switches
+ *
+ * Warning: this test will cause a very high load for a few seconds
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <signal.h>
+#include <inttypes.h>
+#include <wait.h>
+
+
+#include <sys/prctl.h>
+#include <linux/prctl.h>
+
+/* Get/set the process' ability to use the timestamp counter instruction */
+#ifndef PR_GET_TSC
+#define PR_GET_TSC 25
+#define PR_SET_TSC 26
+# define PR_TSC_ENABLE		1   /* allow the use of the timestamp counter */
+# define PR_TSC_SIGSEGV		2   /* throw a SIGSEGV instead of reading the TSC */
+#endif
+
+uint64_t rdtsc() {
+uint32_t lo, hi;
+/* We cannot use "=A", since this would use %rax on x86_64 */
+__asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi));
+return (uint64_t)hi << 32 | lo;
+}
+
+void sigsegv_expect(int sig)
+{
+	/* */
+}
+
+void segvtask(void)
+{
+	if (prctl(PR_SET_TSC, PR_TSC_SIGSEGV) < 0)
+	{
+		perror("prctl");
+		exit(0);
+	}
+	signal(SIGSEGV, sigsegv_expect);
+	alarm(10);
+	rdtsc();
+	fprintf(stderr, "FATAL ERROR, rdtsc() succeeded while disabled\n");
+	exit(0);
+}
+
+
+void sigsegv_fail(int sig)
+{
+	fprintf(stderr, "FATAL ERROR, rdtsc() failed while enabled\n");
+	exit(0);
+}
+
+void rdtsctask(void)
+{
+	if (prctl(PR_SET_TSC, PR_TSC_ENABLE) < 0)
+	{
+		perror("prctl");
+		exit(0);
+	}
+	signal(SIGSEGV, sigsegv_fail);
+	alarm(10);
+	for(;;) rdtsc();
+}
+
+
+int main(int argc, char **argv)
+{
+	int n_tasks = 100, i;
+
+	fprintf(stderr, "[No further output means we're allright]\n");
+
+	for (i=0; i<n_tasks; i++)
+		if (fork() == 0)
+		{
+			if (i & 1)
+				segvtask();
+			else
+				rdtsctask();
+		}
+
+	for (i=0; i<n_tasks; i++)
+		wait(NULL);
+
+	exit(0);
+}
+

Documentation/prctl/disable-tsc-on-off-stress-test.c

@@ -0,0 +1,95 @@
+/*
+ * Tests for prctl(PR_GET_TSC, ...) / prctl(PR_SET_TSC, ...)
+ *
+ * Tests if the control register is updated correctly
+ * when set with prctl()
+ *
+ * Warning: this test will cause a very high load for a few seconds
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <signal.h>
+#include <inttypes.h>
+#include <wait.h>
+
+
+#include <sys/prctl.h>
+#include <linux/prctl.h>
+
+/* Get/set the process' ability to use the timestamp counter instruction */
+#ifndef PR_GET_TSC
+#define PR_GET_TSC 25
+#define PR_SET_TSC 26
+# define PR_TSC_ENABLE		1   /* allow the use of the timestamp counter */
+# define PR_TSC_SIGSEGV		2   /* throw a SIGSEGV instead of reading the TSC */
+#endif
+
+/* snippet from wikipedia :-) */
+
+uint64_t rdtsc() {
+uint32_t lo, hi;
+/* We cannot use "=A", since this would use %rax on x86_64 */
+__asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi));
+return (uint64_t)hi << 32 | lo;
+}
+
+int should_segv = 0;
+
+void sigsegv_cb(int sig)
+{
+	if (!should_segv)
+	{
+		fprintf(stderr, "FATAL ERROR, rdtsc() failed while enabled\n");
+		exit(0);
+	}
+	if (prctl(PR_SET_TSC, PR_TSC_ENABLE) < 0)
+	{
+		perror("prctl");
+		exit(0);
+	}
+	should_segv = 0;
+
+	rdtsc();
+}
+
+void task(void)
+{
+	signal(SIGSEGV, sigsegv_cb);
+	alarm(10);
+	for(;;)
+	{
+		rdtsc();
+		if (should_segv)
+		{
+			fprintf(stderr, "FATAL ERROR, rdtsc() succeeded while disabled\n");
+			exit(0);
+		}
+		if (prctl(PR_SET_TSC, PR_TSC_SIGSEGV) < 0)
+		{
+			perror("prctl");
+			exit(0);
+		}
+		should_segv = 1;
+	}
+}
+
+
+int main(int argc, char **argv)
+{
+	int n_tasks = 100, i;
+
+	fprintf(stderr, "[No further output means we're allright]\n");
+
+	for (i=0; i<n_tasks; i++)
+		if (fork() == 0)
+			task();
+
+	for (i=0; i<n_tasks; i++)
+		wait(NULL);
+
+	exit(0);
+}
+

Documentation/prctl/disable-tsc-test.c

@@ -0,0 +1,94 @@
+/*
+ * Tests for prctl(PR_GET_TSC, ...) / prctl(PR_SET_TSC, ...)
+ *
+ * Basic test to test behaviour of PR_GET_TSC and PR_SET_TSC
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <signal.h>
+#include <inttypes.h>
+
+
+#include <sys/prctl.h>
+#include <linux/prctl.h>
+
+/* Get/set the process' ability to use the timestamp counter instruction */
+#ifndef PR_GET_TSC
+#define PR_GET_TSC 25
+#define PR_SET_TSC 26
+# define PR_TSC_ENABLE		1   /* allow the use of the timestamp counter */
+# define PR_TSC_SIGSEGV		2   /* throw a SIGSEGV instead of reading the TSC */
+#endif
+
+const char *tsc_names[] =
+{
+	[0] = "[not set]",
+	[PR_TSC_ENABLE] = "PR_TSC_ENABLE",
+	[PR_TSC_SIGSEGV] = "PR_TSC_SIGSEGV",
+};
+
+uint64_t rdtsc() {
+uint32_t lo, hi;
+/* We cannot use "=A", since this would use %rax on x86_64 */
+__asm__ __volatile__ ("rdtsc" : "=a" (lo), "=d" (hi));
+return (uint64_t)hi << 32 | lo;
+}
+
+void sigsegv_cb(int sig)
+{
+	int tsc_val = 0;
+
+	printf("[ SIG_SEGV ]\n");
+	printf("prctl(PR_GET_TSC, &tsc_val); ");
+	fflush(stdout);
+
+	if ( prctl(PR_GET_TSC, &tsc_val) == -1)
+		perror("prctl");
+
+	printf("tsc_val == %s\n", tsc_names[tsc_val]);
+	printf("prctl(PR_SET_TSC, PR_TSC_ENABLE)\n");
+	fflush(stdout);
+	if ( prctl(PR_SET_TSC, PR_TSC_ENABLE) == -1)
+		perror("prctl");
+
+	printf("rdtsc() == ");
+}
+
+int main(int argc, char **argv)
+{
+	int tsc_val = 0;
+
+	signal(SIGSEGV, sigsegv_cb);
+
+	printf("rdtsc() == %llu\n", (unsigned long long)rdtsc());
+	printf("prctl(PR_GET_TSC, &tsc_val); ");
+	fflush(stdout);
+
+	if ( prctl(PR_GET_TSC, &tsc_val) == -1)
+		perror("prctl");
+
+	printf("tsc_val == %s\n", tsc_names[tsc_val]);
+	printf("rdtsc() == %llu\n", (unsigned long long)rdtsc());
+	printf("prctl(PR_SET_TSC, PR_TSC_ENABLE)\n");
+	fflush(stdout);
+
+	if ( prctl(PR_SET_TSC, PR_TSC_ENABLE) == -1)
+		perror("prctl");
+
+	printf("rdtsc() == %llu\n", (unsigned long long)rdtsc());
+	printf("prctl(PR_SET_TSC, PR_TSC_SIGSEGV)\n");
+	fflush(stdout);
+
+	if ( prctl(PR_SET_TSC, PR_TSC_SIGSEGV) == -1)
+		perror("prctl");
+
+	printf("rdtsc() == ");
+	fflush(stdout);
+	printf("%llu\n", (unsigned long long)rdtsc());
+	fflush(stdout);
+
+	exit(EXIT_SUCCESS);
+}
+

arch/x86/Kconfig

@@ -903,6 +903,15 @@ config X86_64_ACPI_NUMA
 	help
 	  Enable ACPI SRAT based node topology detection.
 
+# Some NUMA nodes have memory ranges that span
+# other nodes.  Even though a pfn is valid and
+# between a node's start and end pfns, it may not
+# reside on that node.  See memmap_init_zone()
+# for details.
+config NODES_SPAN_OTHER_NODES
+	def_bool y
+	depends on X86_64_ACPI_NUMA
+
 config NUMA_EMU
 	bool "NUMA emulation"
 	depends on X86_64 && NUMA

arch/x86/boot/a20.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/a20.c
- *
  * Enable A20 gate (return -1 on failure)
  */
 

arch/x86/boot/apm.c

@@ -12,8 +12,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/apm.c
- *
  * Get APM BIOS information
  */
 

arch/x86/boot/bitops.h

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/bitops.h
- *
  * Very simple bitops for the boot code.
  */
 

arch/x86/boot/boot.h

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/boot.h
- *
  * Header file for the real-mode kernel code
  */
 

arch/x86/boot/cmdline.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/cmdline.c
- *
  * Simple command-line parser for early boot.
  */
 

arch/x86/boot/compressed/head_32.S

@@ -130,7 +130,7 @@ relocated:
 /*
  * Setup the stack for the decompressor
  */
-	leal stack_end(%ebx), %esp
+	leal boot_stack_end(%ebx), %esp
 
 /*
  * Do the decompression, and jump to the new kernel..
@@ -142,8 +142,8 @@ relocated:
 	pushl %eax	# input_len
 	leal input_data(%ebx), %eax
 	pushl %eax	# input_data
-	leal _end(%ebx), %eax
+	leal boot_heap(%ebx), %eax
-	pushl %eax	# end of the image as third argument
+	pushl %eax	# heap area as third argument
 	pushl %esi	# real mode pointer as second arg
 	call decompress_kernel
 	addl $20, %esp
@@ -181,7 +181,10 @@ relocated:
 	jmp *%ebp
 
 .bss
+/* Stack and heap for uncompression */
 .balign 4
-stack:
+boot_heap:
-	.fill 4096, 1, 0
+	.fill BOOT_HEAP_SIZE, 1, 0
-stack_end:
+boot_stack:
+	.fill BOOT_STACK_SIZE, 1, 0
+boot_stack_end:

arch/x86/boot/compressed/head_64.S

@@ -28,6 +28,7 @@
 #include <asm/segment.h>
 #include <asm/pgtable.h>
 #include <asm/page.h>
+#include <asm/boot.h>
 #include <asm/msr.h>
 #include <asm/asm-offsets.h>
 
@@ -62,7 +63,7 @@ startup_32:
 	subl	$1b, %ebp
 
 /* setup a stack and make sure cpu supports long mode. */
-	movl	$user_stack_end, %eax
+	movl	$boot_stack_end, %eax
 	addl	%ebp, %eax
 	movl	%eax, %esp
 
@@ -243,9 +244,9 @@ ENTRY(startup_64)
 /* Copy the compressed kernel to the end of our buffer
  * where decompression in place becomes safe.
  */
-	leaq	_end(%rip), %r8
+	leaq	_end_before_pgt(%rip), %r8
-	leaq	_end(%rbx), %r9
+	leaq	_end_before_pgt(%rbx), %r9
-	movq	$_end /* - $startup_32 */, %rcx
+	movq	$_end_before_pgt /* - $startup_32 */, %rcx
 1:	subq	$8, %r8
 	subq	$8, %r9
 	movq	0(%r8), %rax
@@ -267,14 +268,14 @@ relocated:
  */
 	xorq	%rax, %rax
 	leaq    _edata(%rbx), %rdi
-	leaq    _end(%rbx), %rcx
+	leaq    _end_before_pgt(%rbx), %rcx
 	subq	%rdi, %rcx
 	cld
 	rep
 	stosb
 
 	/* Setup the stack */
-	leaq	user_stack_end(%rip), %rsp
+	leaq	boot_stack_end(%rip), %rsp
 
 	/* zero EFLAGS after setting rsp */
 	pushq	$0
@@ -285,7 +286,7 @@ relocated:
  */
 	pushq	%rsi			# Save the real mode argument
 	movq	%rsi, %rdi		# real mode address
-	leaq	_heap(%rip), %rsi	# _heap
+	leaq	boot_heap(%rip), %rsi	# malloc area for uncompression
 	leaq	input_data(%rip), %rdx  # input_data
 	movl	input_len(%rip), %eax
 	movq	%rax, %rcx		# input_len
@@ -310,9 +311,12 @@ gdt:
 	.quad	0x0080890000000000	/* TS descriptor */
 	.quad   0x0000000000000000	/* TS continued */
 gdt_end:
-	.bss
+
-/* Stack for uncompression */
+.bss
-	.balign 4
+/* Stack and heap for uncompression */
-user_stack:
+.balign 4
-	.fill 4096,4,0
+boot_heap:
-user_stack_end:
+	.fill BOOT_HEAP_SIZE, 1, 0
+boot_stack:
+	.fill BOOT_STACK_SIZE, 1, 0
+boot_stack_end:

arch/x86/boot/compressed/misc.c

@@ -217,12 +217,6 @@ static void putstr(const char *);
 static memptr free_mem_ptr;
 static memptr free_mem_end_ptr;
 
-#ifdef CONFIG_X86_64
-#define HEAP_SIZE             0x7000
-#else
-#define HEAP_SIZE             0x4000
-#endif
-
 static char *vidmem;
 static int vidport;
 static int lines, cols;
@@ -449,7 +443,7 @@ asmlinkage void decompress_kernel(void *rmode, memptr heap,
 
 	window = output;		/* Output buffer (Normally at 1M) */
 	free_mem_ptr     = heap;	/* Heap */
-	free_mem_end_ptr = heap + HEAP_SIZE;
+	free_mem_end_ptr = heap + BOOT_HEAP_SIZE;
 	inbuf  = input_data;		/* Input buffer */
 	insize = input_len;
 	inptr  = 0;

arch/x86/boot/compressed/vmlinux_64.lds

@@ -39,10 +39,10 @@ SECTIONS
 		*(.bss.*)
 		*(COMMON)
 		. = ALIGN(8);
-		_end = . ;
+		_end_before_pgt = . ;
 		. = ALIGN(4096);
 		pgtable = . ;
 		. = . + 4096 * 6;
-		_heap = .;
+		_ebss = .;
 	}
 }

arch/x86/boot/copy.S

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/copy.S
- *
  * Memory copy routines
  */
 

arch/x86/boot/cpucheck.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/cpucheck.c
- *
  * Check for obligatory CPU features and abort if the features are not
  * present.  This code should be compilable as 16-, 32- or 64-bit
  * code, so be very careful with types and inline assembly.

arch/x86/boot/edd.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/edd.c
- *
  * Get EDD BIOS disk information
  */
 

arch/x86/boot/install.sh

@@ -1,7 +1,5 @@
 #!/bin/sh
 #
-# arch/i386/boot/install.sh
-#
 # This file is subject to the terms and conditions of the GNU General Public
 # License.  See the file "COPYING" in the main directory of this archive
 # for more details.

arch/x86/boot/main.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/main.c
- *
  * Main module for the real-mode kernel code
  */
 

arch/x86/boot/mca.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/mca.c
- *
  * Get the MCA system description table
  */
 

arch/x86/boot/memory.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/memory.c
- *
  * Memory detection code
  */
 

arch/x86/boot/pm.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/pm.c
- *
  * Prepare the machine for transition to protected mode.
  */
 

arch/x86/boot/pmjump.S

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/pmjump.S
- *
  * The actual transition into protected mode
  */
 

arch/x86/boot/printf.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/printf.c
- *
  * Oh, it's a waste of space, but oh-so-yummy for debugging.  This
  * version of printf() does not include 64-bit support.  "Live with
  * it."

arch/x86/boot/string.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/string.c
- *
  * Very basic string functions
  */
 

arch/x86/boot/tty.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/tty.c
- *
  * Very simple screen I/O
  * XXX: Probably should add very simple serial I/O?
  */

arch/x86/boot/version.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/version.c
- *
  * Kernel version string
  */
 

arch/x86/boot/video-bios.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/video-bios.c
- *
  * Standard video BIOS modes
  *
  * We have two options for this; silent and scanned.

arch/x86/boot/video-vesa.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/video-vesa.c
- *
  * VESA text modes
  */
 

arch/x86/boot/video-vga.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/video-vga.c
- *
  * Common all-VGA modes
  */
 

arch/x86/boot/video.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/video.c
- *
  * Select video mode
  */
 

arch/x86/boot/video.h

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/video.h
- *
  * Header file for the real-mode video probing code
  */
 

arch/x86/boot/voyager.c

@@ -9,8 +9,6 @@
  * ----------------------------------------------------------------------- */
 
 /*
- * arch/i386/boot/voyager.c
- *
  * Get the Voyager config information
  */
 

arch/x86/kernel/Makefile

@@ -22,13 +22,14 @@ obj-y			+= setup_$(BITS).o i8259_$(BITS).o setup.o
 obj-$(CONFIG_X86_32)	+= sys_i386_32.o i386_ksyms_32.o
 obj-$(CONFIG_X86_64)	+= sys_x86_64.o x8664_ksyms_64.o
 obj-$(CONFIG_X86_64)	+= syscall_64.o vsyscall_64.o setup64.o
-obj-y			+= pci-dma_$(BITS).o  bootflag.o e820_$(BITS).o
+obj-y			+= bootflag.o e820_$(BITS).o
-obj-y			+= quirks.o i8237.o topology.o kdebugfs.o
+obj-y			+= pci-dma.o quirks.o i8237.o topology.o kdebugfs.o
-obj-y			+= alternative.o i8253.o
+obj-y			+= alternative.o i8253.o pci-nommu.o
-obj-$(CONFIG_X86_64)	+= pci-nommu_64.o bugs_64.o
+obj-$(CONFIG_X86_64)	+= bugs_64.o
 obj-y			+= tsc_$(BITS).o io_delay.o rtc.o
 
 obj-$(CONFIG_X86_TRAMPOLINE)	+= trampoline.o
+obj-y				+= process.o
 obj-y				+= i387.o
 obj-y				+= ptrace.o
 obj-y				+= ds.o

arch/x86/kernel/acpi/cstate.c

@@ -1,6 +1,4 @@
 /*
- * arch/i386/kernel/acpi/cstate.c
- *
  * Copyright (C) 2005 Intel Corporation
  * 	Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
  * 	- Added _PDC for SMP C-states on Intel CPUs

arch/x86/kernel/acpi/processor.c

@@ -1,6 +1,4 @@
 /*
- * arch/i386/kernel/acpi/processor.c
- *
  * Copyright (C) 2005 Intel Corporation
  * 	Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
  * 	- Added _PDC for platforms with Intel CPUs

arch/x86/kernel/cpu/cpufreq/p4-clockmod.c

@@ -289,8 +289,8 @@ static int __init cpufreq_p4_init(void)
 	if (c->x86_vendor != X86_VENDOR_INTEL)
 		return -ENODEV;
 
-	if (!test_bit(X86_FEATURE_ACPI, c->x86_capability) ||
+	if (!test_cpu_cap(c, X86_FEATURE_ACPI) ||
-		!test_bit(X86_FEATURE_ACC, c->x86_capability))
+				!test_cpu_cap(c, X86_FEATURE_ACC))
 		return -ENODEV;
 
 	ret = cpufreq_register_driver(&p4clockmod_driver);

arch/x86/kernel/cpu/mcheck/therm_throt.c

@@ -1,5 +1,4 @@
 /*
- * linux/arch/i386/kernel/cpu/mcheck/therm_throt.c
  *
  * Thermal throttle event support code (such as syslog messaging and rate
  * limiting) that was factored out from x86_64 (mce_intel.c) and i386 (p4.c).

arch/x86/kernel/e820_32.c

@@ -475,7 +475,7 @@ int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
 /*
  * Find the highest page frame number we have available
  */
-void __init find_max_pfn(void)
+void __init propagate_e820_map(void)
 {
 	int i;
 
@@ -704,7 +704,7 @@ static int __init parse_memmap(char *arg)
 		 * size before original memory map is
 		 * reset.
 		 */
-		find_max_pfn();
+		propagate_e820_map();
 		saved_max_pfn = max_pfn;
 #endif
 		e820.nr_map = 0;

arch/x86/kernel/e820_64.c

@@ -96,7 +96,7 @@ void __init early_res_to_bootmem(void)
 }
 
 /* Check for already reserved areas */
-static inline int
+static inline int __init
 bad_addr(unsigned long *addrp, unsigned long size, unsigned long align)
 {
 	int i;
@@ -116,7 +116,7 @@ again:
 }
 
 /* Check for already reserved areas */
-static inline int
+static inline int __init
 bad_addr_size(unsigned long *addrp, unsigned long *sizep, unsigned long align)
 {
 	int i;

arch/x86/kernel/efi.c

@@ -383,6 +383,7 @@ static void __init runtime_code_page_mkexec(void)
 {
 	efi_memory_desc_t *md;
 	void *p;
+	u64 addr, npages;
 
 	/* Make EFI runtime service code area executable */
 	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
@@ -391,7 +392,10 @@ static void __init runtime_code_page_mkexec(void)
 		if (md->type != EFI_RUNTIME_SERVICES_CODE)
 			continue;
 
-		set_memory_x(md->virt_addr, md->num_pages);
+		addr = md->virt_addr;
+		npages = md->num_pages;
+		memrange_efi_to_native(&addr, &npages);
+		set_memory_x(addr, npages);
 	}
 }
 
@@ -408,7 +412,7 @@ void __init efi_enter_virtual_mode(void)
 	efi_memory_desc_t *md;
 	efi_status_t status;
 	unsigned long size;
-	u64 end, systab;
+	u64 end, systab, addr, npages;
 	void *p, *va;
 
 	efi.systab = NULL;
@@ -420,7 +424,7 @@ void __init efi_enter_virtual_mode(void)
 		size = md->num_pages << EFI_PAGE_SHIFT;
 		end = md->phys_addr + size;
 
-		if ((end >> PAGE_SHIFT) <= max_pfn_mapped)
+		if (PFN_UP(end) <= max_pfn_mapped)
 			va = __va(md->phys_addr);
 		else
 			va = efi_ioremap(md->phys_addr, size);
@@ -433,8 +437,12 @@ void __init efi_enter_virtual_mode(void)
 			continue;
 		}
 
-		if (!(md->attribute & EFI_MEMORY_WB))
+		if (!(md->attribute & EFI_MEMORY_WB)) {
-			set_memory_uc(md->virt_addr, md->num_pages);
+			addr = md->virt_addr;
+			npages = md->num_pages;
+			memrange_efi_to_native(&addr, &npages);
+			set_memory_uc(addr, npages);
+		}
 
 		systab = (u64) (unsigned long) efi_phys.systab;
 		if (md->phys_addr <= systab && systab < end) {

arch/x86/kernel/efi_64.c

@@ -105,14 +105,14 @@ void __init efi_reserve_bootmem(void)
 
 void __iomem * __init efi_ioremap(unsigned long phys_addr, unsigned long size)
 {
-	static unsigned pages_mapped;
+	static unsigned pages_mapped __initdata;
 	unsigned i, pages;
+	unsigned long offset;
 
-	/* phys_addr and size must be page aligned */
+	pages = PFN_UP(phys_addr + size) - PFN_DOWN(phys_addr);
-	if ((phys_addr & ~PAGE_MASK) || (size & ~PAGE_MASK))
+	offset = phys_addr & ~PAGE_MASK;
-		return NULL;
+	phys_addr &= PAGE_MASK;
 
-	pages = size >> PAGE_SHIFT;
 	if (pages_mapped + pages > MAX_EFI_IO_PAGES)
 		return NULL;
 
@@ -124,5 +124,5 @@ void __iomem * __init efi_ioremap(unsigned long phys_addr, unsigned long size)
 	}
 
 	return (void __iomem *)__fix_to_virt(FIX_EFI_IO_MAP_FIRST_PAGE - \
-					     (pages_mapped - pages));
+					     (pages_mapped - pages)) + offset;
 }

arch/x86/kernel/entry_32.S

@@ -1,5 +1,4 @@
 /*
- *  linux/arch/i386/entry.S
  *
  *  Copyright (C) 1991, 1992  Linus Torvalds
  */

arch/x86/kernel/genx2apic_uv_x.c

@@ -61,26 +61,31 @@ int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip)
 	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
 	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
 	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
-	    (6 << UVH_IPI_INT_DELIVERY_MODE_SHFT);
+	    APIC_DM_INIT;
+	uv_write_global_mmr64(nasid, UVH_IPI_INT, val);
+	mdelay(10);
+
+	val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+	    (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+	    (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+	    APIC_DM_STARTUP;
 	uv_write_global_mmr64(nasid, UVH_IPI_INT, val);
 	return 0;
 }
 
 static void uv_send_IPI_one(int cpu, int vector)
 {
-	unsigned long val, apicid;
+	unsigned long val, apicid, lapicid;
 	int nasid;
 
 	apicid = per_cpu(x86_cpu_to_apicid, cpu); /* ZZZ - cache node-local ? */
+	lapicid = apicid & 0x3f;		/* ZZZ macro needed */
 	nasid = uv_apicid_to_nasid(apicid);
 	val =
-	    (1UL << UVH_IPI_INT_SEND_SHFT) | (apicid <<
+	    (1UL << UVH_IPI_INT_SEND_SHFT) | (lapicid <<
 					      UVH_IPI_INT_APIC_ID_SHFT) |
 	    (vector << UVH_IPI_INT_VECTOR_SHFT);
 	uv_write_global_mmr64(nasid, UVH_IPI_INT, val);
-	printk(KERN_DEBUG
-	     "UV: IPI to cpu %d, apicid 0x%lx, vec %d, nasid%d, val 0x%lx\n",
-	     cpu, apicid, vector, nasid, val);
 }
 
 static void uv_send_IPI_mask(cpumask_t mask, int vector)

arch/x86/kernel/head64.c

@@ -146,6 +146,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
 
+#ifdef CONFIG_BLK_DEV_INITRD
 	/* Reserve INITRD */
 	if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
 		unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
@@ -153,6 +154,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 		unsigned long ramdisk_end   = ramdisk_image + ramdisk_size;
 		reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
 	}
+#endif
 
 	reserve_ebda_region();
 

arch/x86/kernel/head_32.S

@@ -1,5 +1,4 @@
 /*
- *  linux/arch/i386/kernel/head.S -- the 32-bit startup code.
  *
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *

arch/x86/kernel/i387.c

@@ -35,17 +35,18 @@
 #endif
 
 static unsigned int		mxcsr_feature_mask __read_mostly = 0xffffffffu;
+unsigned int xstate_size;
+static struct i387_fxsave_struct fx_scratch __cpuinitdata;
 
-void mxcsr_feature_mask_init(void)
+void __cpuinit mxcsr_feature_mask_init(void)
 {
 	unsigned long mask = 0;
 
 	clts();
 	if (cpu_has_fxsr) {
-		memset(&current->thread.i387.fxsave, 0,
+		memset(&fx_scratch, 0, sizeof(struct i387_fxsave_struct));
-		       sizeof(struct i387_fxsave_struct));
+		asm volatile("fxsave %0" : : "m" (fx_scratch));
-		asm volatile("fxsave %0" : : "m" (current->thread.i387.fxsave));
+		mask = fx_scratch.mxcsr_mask;
-		mask = current->thread.i387.fxsave.mxcsr_mask;
 		if (mask == 0)
 			mask = 0x0000ffbf;
 	}
@@ -53,6 +54,16 @@ void mxcsr_feature_mask_init(void)
 	stts();
 }
 
+void __init init_thread_xstate(void)
+{
+	if (cpu_has_fxsr)
+		xstate_size = sizeof(struct i387_fxsave_struct);
+#ifdef CONFIG_X86_32
+	else
+		xstate_size = sizeof(struct i387_fsave_struct);
+#endif
+}
+
 #ifdef CONFIG_X86_64
 /*
  * Called at bootup to set up the initial FPU state that is later cloned
@@ -61,10 +72,6 @@ void mxcsr_feature_mask_init(void)
 void __cpuinit fpu_init(void)
 {
 	unsigned long oldcr0 = read_cr0();
-	extern void __bad_fxsave_alignment(void);
-
-	if (offsetof(struct task_struct, thread.i387.fxsave) & 15)
-		__bad_fxsave_alignment();
 
 	set_in_cr4(X86_CR4_OSFXSR);
 	set_in_cr4(X86_CR4_OSXMMEXCPT);
@@ -84,32 +91,44 @@ void __cpuinit fpu_init(void)
  * value at reset if we support XMM instructions and then
  * remeber the current task has used the FPU.
  */
-void init_fpu(struct task_struct *tsk)
+int init_fpu(struct task_struct *tsk)
 {
 	if (tsk_used_math(tsk)) {
 		if (tsk == current)
 			unlazy_fpu(tsk);
-		return;
+		return 0;
+	}
+
+	/*
+	 * Memory allocation at the first usage of the FPU and other state.
+	 */
+	if (!tsk->thread.xstate) {
+		tsk->thread.xstate = kmem_cache_alloc(task_xstate_cachep,
+						      GFP_KERNEL);
+		if (!tsk->thread.xstate)
+			return -ENOMEM;
 	}
 
 	if (cpu_has_fxsr) {
-		memset(&tsk->thread.i387.fxsave, 0,
+		struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave;
-		       sizeof(struct i387_fxsave_struct));
+
-		tsk->thread.i387.fxsave.cwd = 0x37f;
+		memset(fx, 0, xstate_size);
+		fx->cwd = 0x37f;
 		if (cpu_has_xmm)
-			tsk->thread.i387.fxsave.mxcsr = MXCSR_DEFAULT;
+			fx->mxcsr = MXCSR_DEFAULT;
 	} else {
-		memset(&tsk->thread.i387.fsave, 0,
+		struct i387_fsave_struct *fp = &tsk->thread.xstate->fsave;
-		       sizeof(struct i387_fsave_struct));
+		memset(fp, 0, xstate_size);
-		tsk->thread.i387.fsave.cwd = 0xffff037fu;
+		fp->cwd = 0xffff037fu;
-		tsk->thread.i387.fsave.swd = 0xffff0000u;
+		fp->swd = 0xffff0000u;
-		tsk->thread.i387.fsave.twd = 0xffffffffu;
+		fp->twd = 0xffffffffu;
-		tsk->thread.i387.fsave.fos = 0xffff0000u;
+		fp->fos = 0xffff0000u;
 	}
 	/*
 	 * Only the device not available exception or ptrace can call init_fpu.
 	 */
 	set_stopped_child_used_math(tsk);
+	return 0;
 }
 
 int fpregs_active(struct task_struct *target, const struct user_regset *regset)
@@ -126,13 +145,17 @@ int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
 		unsigned int pos, unsigned int count,
 		void *kbuf, void __user *ubuf)
 {
+	int ret;
+
 	if (!cpu_has_fxsr)
 		return -ENODEV;
 
-	init_fpu(target);
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
 
 	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
-				   &target->thread.i387.fxsave, 0, -1);
+				   &target->thread.xstate->fxsave, 0, -1);
 }
 
 int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
@@ -144,16 +167,19 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
 	if (!cpu_has_fxsr)
 		return -ENODEV;
 
-	init_fpu(target);
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
+
 	set_stopped_child_used_math(target);
 
 	ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
-				 &target->thread.i387.fxsave, 0, -1);
+				 &target->thread.xstate->fxsave, 0, -1);
 
 	/*
 	 * mxcsr reserved bits must be masked to zero for security reasons.
 	 */
-	target->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
+	target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
 
 	return ret;
 }
@@ -233,7 +259,7 @@ static inline u32 twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave)
 static void
 convert_from_fxsr(struct user_i387_ia32_struct *env, struct task_struct *tsk)
 {
-	struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave;
+	struct i387_fxsave_struct *fxsave = &tsk->thread.xstate->fxsave;
 	struct _fpreg *to = (struct _fpreg *) &env->st_space[0];
 	struct _fpxreg *from = (struct _fpxreg *) &fxsave->st_space[0];
 	int i;
@@ -273,7 +299,7 @@ static void convert_to_fxsr(struct task_struct *tsk,
 			    const struct user_i387_ia32_struct *env)
 
 {
-	struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave;
+	struct i387_fxsave_struct *fxsave = &tsk->thread.xstate->fxsave;
 	struct _fpreg *from = (struct _fpreg *) &env->st_space[0];
 	struct _fpxreg *to = (struct _fpxreg *) &fxsave->st_space[0];
 	int i;
@@ -302,15 +328,19 @@ int fpregs_get(struct task_struct *target, const struct user_regset *regset,
 	       void *kbuf, void __user *ubuf)
 {
 	struct user_i387_ia32_struct env;
+	int ret;
 
 	if (!HAVE_HWFP)
 		return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
 
-	init_fpu(target);
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
 
 	if (!cpu_has_fxsr) {
 		return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
-					   &target->thread.i387.fsave, 0, -1);
+					   &target->thread.xstate->fsave, 0,
+					   -1);
 	}
 
 	if (kbuf && pos == 0 && count == sizeof(env)) {
@@ -333,12 +363,15 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
 	if (!HAVE_HWFP)
 		return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
 
-	init_fpu(target);
+	ret = init_fpu(target);
+	if (ret)
+		return ret;
+
 	set_stopped_child_used_math(target);
 
 	if (!cpu_has_fxsr) {
 		return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
-					  &target->thread.i387.fsave, 0, -1);
+					  &target->thread.xstate->fsave, 0, -1);
 	}
 
 	if (pos > 0 || count < sizeof(env))
@@ -358,11 +391,11 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
 static inline int save_i387_fsave(struct _fpstate_ia32 __user *buf)
 {
 	struct task_struct *tsk = current;
+	struct i387_fsave_struct *fp = &tsk->thread.xstate->fsave;
 
 	unlazy_fpu(tsk);
-	tsk->thread.i387.fsave.status = tsk->thread.i387.fsave.swd;
+	fp->status = fp->swd;
-	if (__copy_to_user(buf, &tsk->thread.i387.fsave,
+	if (__copy_to_user(buf, fp, sizeof(struct i387_fsave_struct)))
-			   sizeof(struct i387_fsave_struct)))
 		return -1;
 	return 1;
 }
@@ -370,6 +403,7 @@ static inline int save_i387_fsave(struct _fpstate_ia32 __user *buf)
 static int save_i387_fxsave(struct _fpstate_ia32 __user *buf)
 {
 	struct task_struct *tsk = current;
+	struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave;
 	struct user_i387_ia32_struct env;
 	int err = 0;
 
@@ -379,12 +413,12 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf)
 	if (__copy_to_user(buf, &env, sizeof(env)))
 		return -1;
 
-	err |= __put_user(tsk->thread.i387.fxsave.swd, &buf->status);
+	err |= __put_user(fx->swd, &buf->status);
 	err |= __put_user(X86_FXSR_MAGIC, &buf->magic);
 	if (err)
 		return -1;
 
-	if (__copy_to_user(&buf->_fxsr_env[0], &tsk->thread.i387.fxsave,
+	if (__copy_to_user(&buf->_fxsr_env[0], fx,
 			   sizeof(struct i387_fxsave_struct)))
 		return -1;
 	return 1;
@@ -417,7 +451,7 @@ static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
 	struct task_struct *tsk = current;
 
 	clear_fpu(tsk);
-	return __copy_from_user(&tsk->thread.i387.fsave, buf,
+	return __copy_from_user(&tsk->thread.xstate->fsave, buf,
 				sizeof(struct i387_fsave_struct));
 }
 
@@ -428,10 +462,10 @@ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf)
 	int err;
 
 	clear_fpu(tsk);
-	err = __copy_from_user(&tsk->thread.i387.fxsave, &buf->_fxsr_env[0],
+	err = __copy_from_user(&tsk->thread.xstate->fxsave, &buf->_fxsr_env[0],
 			       sizeof(struct i387_fxsave_struct));
 	/* mxcsr reserved bits must be masked to zero for security reasons */
-	tsk->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
+	tsk->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
 	if (err || __copy_from_user(&env, buf, sizeof(env)))
 		return 1;
 	convert_to_fxsr(tsk, &env);

arch/x86/kernel/kgdb.c

@@ -46,11 +46,7 @@
 #include <asm/apicdef.h>
 #include <asm/system.h>
 
-#ifdef CONFIG_X86_32
+#include <mach_ipi.h>
-# include <mach_ipi.h>
-#else
-# include <asm/mach_apic.h>
-#endif
 
 /*
  * Put the error code here just in case the user cares:

arch/x86/kernel/nmi_32.c

@@ -321,7 +321,8 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-__kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+notrace __kprobes int
+nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 {
 
 	/*

arch/x86/kernel/nmi_64.c

@@ -313,7 +313,8 @@ void touch_nmi_watchdog(void)
 }
 EXPORT_SYMBOL(touch_nmi_watchdog);
 
-int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+notrace __kprobes int
+nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
@@ -384,7 +385,8 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 
 static unsigned ignore_nmis;
 
-asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
+asmlinkage notrace __kprobes void
+do_nmi(struct pt_regs *regs, long error_code)
 {
 	nmi_enter();
 	add_pda(__nmi_count,1);

arch/x86/kernel/pci-calgary_64.c

@@ -470,10 +470,11 @@ error:
 	return 0;
 }
 
-static dma_addr_t calgary_map_single(struct device *dev, void *vaddr,
+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;
 	struct iommu_table *tbl = find_iommu_table(dev);

arch/x86/kernel/pci-dma.c

@@ -1,225 +1,133 @@
-/*
+#include <linux/dma-mapping.h>
- * Dynamic DMA mapping support.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <linux/module.h>
 #include <linux/dmar.h>
-#include <asm/io.h>
+#include <linux/bootmem.h>
+#include <linux/pci.h>
+
+#include <asm/proto.h>
+#include <asm/dma.h>
 #include <asm/gart.h>
 #include <asm/calgary.h>
 
+int forbid_dac __read_mostly;
+EXPORT_SYMBOL(forbid_dac);
+
+const struct dma_mapping_ops *dma_ops;
+EXPORT_SYMBOL(dma_ops);
+
+int iommu_sac_force __read_mostly = 0;
+
+#ifdef CONFIG_IOMMU_DEBUG
+int panic_on_overflow __read_mostly = 1;
+int force_iommu __read_mostly = 1;
+#else
+int panic_on_overflow __read_mostly = 0;
+int force_iommu __read_mostly = 0;
+#endif
+
 int iommu_merge __read_mostly = 0;
 
-dma_addr_t bad_dma_address __read_mostly;
+int no_iommu __read_mostly;
-EXPORT_SYMBOL(bad_dma_address);
+/* Set this to 1 if there is a HW IOMMU in the system */
+int iommu_detected __read_mostly = 0;
 
 /* This tells the BIO block layer to assume merging. Default to off
    because we cannot guarantee merging later. */
 int iommu_bio_merge __read_mostly = 0;
 EXPORT_SYMBOL(iommu_bio_merge);
 
-static int iommu_sac_force __read_mostly = 0;
+dma_addr_t bad_dma_address __read_mostly = 0;
-
+EXPORT_SYMBOL(bad_dma_address);
-int no_iommu __read_mostly;
-#ifdef CONFIG_IOMMU_DEBUG
-int panic_on_overflow __read_mostly = 1;
-int force_iommu __read_mostly = 1;
-#else
-int panic_on_overflow __read_mostly = 0;
-int force_iommu __read_mostly= 0;
-#endif
-
-/* Set this to 1 if there is a HW IOMMU in the system */
-int iommu_detected __read_mostly = 0;
 
 /* Dummy device used for NULL arguments (normally ISA). Better would
    be probably a smaller DMA mask, but this is bug-to-bug compatible
-   to i386. */
+   to older i386. */
 struct device fallback_dev = {
 	.bus_id = "fallback device",
 	.coherent_dma_mask = DMA_32BIT_MASK,
 	.dma_mask = &fallback_dev.coherent_dma_mask,
 };
 
-/* Allocate DMA memory on node near device */
-noinline static void *
-dma_alloc_pages(struct device *dev, gfp_t gfp, unsigned order)
-{
-	struct page *page;
-	int node;
-
-	node = dev_to_node(dev);
-
-	page = alloc_pages_node(node, gfp, order);
-	return page ? page_address(page) : NULL;
-}
-
-/*
- * Allocate memory for a coherent mapping.
- */
-void *
-dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
-		   gfp_t gfp)
-{
-	void *memory;
-	unsigned long dma_mask = 0;
-	u64 bus;
-
-	if (!dev)
-		dev = &fallback_dev;
-	dma_mask = dev->coherent_dma_mask;
-	if (dma_mask == 0)
-		dma_mask = DMA_32BIT_MASK;
-
-	/* Device not DMA able */
-	if (dev->dma_mask == NULL)
-		return NULL;
-
-	/* Don't invoke OOM killer */
-	gfp |= __GFP_NORETRY;
-
-	/* Kludge to make it bug-to-bug compatible with i386. i386
-	   uses the normal dma_mask for alloc_coherent. */
-	dma_mask &= *dev->dma_mask;
-
-	/* Why <=? Even when the mask is smaller than 4GB it is often
-	   larger than 16MB and in this case we have a chance of
-	   finding fitting memory in the next higher zone first. If
-	   not retry with true GFP_DMA. -AK */
-	if (dma_mask <= DMA_32BIT_MASK)
-		gfp |= GFP_DMA32;
-
- again:
-	memory = dma_alloc_pages(dev, gfp, get_order(size));
-	if (memory == NULL)
-		return NULL;
-
-	{
-		int high, mmu;
-		bus = virt_to_bus(memory);
-	        high = (bus + size) >= dma_mask;
-		mmu = high;
-		if (force_iommu && !(gfp & GFP_DMA))
-			mmu = 1;
-		else if (high) {
-			free_pages((unsigned long)memory,
-				   get_order(size));
-
-			/* Don't use the 16MB ZONE_DMA unless absolutely
-			   needed. It's better to use remapping first. */
-			if (dma_mask < DMA_32BIT_MASK && !(gfp & GFP_DMA)) {
-				gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
-				goto again;
-			}
-
-			/* 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,
-							   dma_handle, gfp);
-			return NULL;
-		}
-
-		memset(memory, 0, size);
-		if (!mmu) {
-			*dma_handle = virt_to_bus(memory);
-			return memory;
-		}
-	}
-
-	if (dma_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);
-	}
-
-	if (dma_ops->map_simple) {
-		*dma_handle = dma_ops->map_simple(dev, memory,
-					      size,
-					      PCI_DMA_BIDIRECTIONAL);
-		if (*dma_handle != bad_dma_address)
-			return memory;
-	}
-
-	if (panic_on_overflow)
-		panic("dma_alloc_coherent: IOMMU overflow by %lu bytes\n",size);
-	free_pages((unsigned long)memory, get_order(size));
-	return NULL;
-}
-EXPORT_SYMBOL(dma_alloc_coherent);
-
-/*
- * Unmap coherent memory.
- * The caller must ensure that the device has finished accessing the mapping.
- */
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t bus)
-{
-	WARN_ON(irqs_disabled());	/* for portability */
-	if (dma_ops->unmap_single)
-		dma_ops->unmap_single(dev, bus, size, 0);
-	free_pages((unsigned long)vaddr, get_order(size));
-}
-EXPORT_SYMBOL(dma_free_coherent);
-
-static int forbid_dac __read_mostly;
-
-int dma_supported(struct device *dev, u64 mask)
-{
-#ifdef CONFIG_PCI
-	if (mask > 0xffffffff && forbid_dac > 0) {
-
-
-
-		printk(KERN_INFO "PCI: Disallowing DAC for device %s\n", dev->bus_id);
-		return 0;
-	}
-#endif
-
-	if (dma_ops->dma_supported)
-		return dma_ops->dma_supported(dev, mask);
-
-	/* Copied from i386. Doesn't make much sense, because it will
-	   only work for pci_alloc_coherent.
-	   The caller just has to use GFP_DMA in this case. */
-        if (mask < DMA_24BIT_MASK)
-                return 0;
-
-	/* Tell the device to use SAC when IOMMU force is on.  This
-	   allows the driver to use cheaper accesses in some cases.
-
-	   Problem with this is that if we overflow the IOMMU area and
-	   return DAC as fallback address the device may not handle it
-	   correctly.
-
-	   As a special case some controllers have a 39bit address
-	   mode that is as efficient as 32bit (aic79xx). Don't force
-	   SAC for these.  Assume all masks <= 40 bits are of this
-	   type. Normally this doesn't make any difference, but gives
-	   more gentle handling of IOMMU overflow. */
-	if (iommu_sac_force && (mask >= DMA_40BIT_MASK)) {
-		printk(KERN_INFO "%s: Force SAC with mask %Lx\n", dev->bus_id,mask);
-		return 0;
-	}
-
-	return 1;
-}
-EXPORT_SYMBOL(dma_supported);
-
 int dma_set_mask(struct device *dev, u64 mask)
 {
 	if (!dev->dma_mask || !dma_supported(dev, mask))
 		return -EIO;
+
 	*dev->dma_mask = mask;
+
 	return 0;
 }
 EXPORT_SYMBOL(dma_set_mask);
 
+#ifdef CONFIG_X86_64
+static __initdata void *dma32_bootmem_ptr;
+static unsigned long dma32_bootmem_size __initdata = (128ULL<<20);
+
+static int __init parse_dma32_size_opt(char *p)
+{
+	if (!p)
+		return -EINVAL;
+	dma32_bootmem_size = memparse(p, &p);
+	return 0;
+}
+early_param("dma32_size", parse_dma32_size_opt);
+
+void __init dma32_reserve_bootmem(void)
+{
+	unsigned long size, align;
+	if (end_pfn <= MAX_DMA32_PFN)
+		return;
+
+	align = 64ULL<<20;
+	size = round_up(dma32_bootmem_size, align);
+	dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align,
+				 __pa(MAX_DMA_ADDRESS));
+	if (dma32_bootmem_ptr)
+		dma32_bootmem_size = size;
+	else
+		dma32_bootmem_size = 0;
+}
+static void __init dma32_free_bootmem(void)
+{
+	int node;
+
+	if (end_pfn <= MAX_DMA32_PFN)
+		return;
+
+	if (!dma32_bootmem_ptr)
+		return;
+
+	for_each_online_node(node)
+		free_bootmem_node(NODE_DATA(node), __pa(dma32_bootmem_ptr),
+				  dma32_bootmem_size);
+
+	dma32_bootmem_ptr = NULL;
+	dma32_bootmem_size = 0;
+}
+
+void __init pci_iommu_alloc(void)
+{
+	/* free the range so iommu could get some range less than 4G */
+	dma32_free_bootmem();
+	/*
+	 * The order of these functions is important for
+	 * fall-back/fail-over reasons
+	 */
+#ifdef CONFIG_GART_IOMMU
+	gart_iommu_hole_init();
+#endif
+
+#ifdef CONFIG_CALGARY_IOMMU
+	detect_calgary();
+#endif
+
+	detect_intel_iommu();
+
+#ifdef CONFIG_SWIOTLB
+	pci_swiotlb_init();
+#endif
+}
+#endif
+
 /*
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
@@ -263,7 +171,10 @@ static __init int iommu_setup(char *p)
 			forbid_dac = 0;
 		if (!strncmp(p, "nodac", 5))
 			forbid_dac = -1;
-
+		if (!strncmp(p, "usedac", 6)) {
+			forbid_dac = -1;
+			return 1;
+		}
 #ifdef CONFIG_SWIOTLB
 		if (!strncmp(p, "soft", 4))
 			swiotlb = 1;
@@ -286,26 +197,294 @@ static __init int iommu_setup(char *p)
 }
 early_param("iommu", iommu_setup);
 
-void __init pci_iommu_alloc(void)
+#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;
-	 * The order of these functions is important for
+	int pages = size >> PAGE_SHIFT;
-	 * fall-back/fail-over reasons
+	int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long);
-	 */
-#ifdef CONFIG_GART_IOMMU
-	gart_iommu_hole_init();
-#endif
 
-#ifdef CONFIG_CALGARY_IOMMU
+	if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
-	detect_calgary();
+		goto out;
-#endif
+	if (!size)
+		goto out;
+	if (dev->dma_mem)
+		goto out;
 
-	detect_intel_iommu();
+	/* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
 
-#ifdef CONFIG_SWIOTLB
+	mem_base = ioremap(bus_addr, size);
-	pci_swiotlb_init();
+	if (!mem_base)
-#endif
+		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
+	if (mask > 0xffffffff && forbid_dac > 0) {
+		printk(KERN_INFO "PCI: Disallowing DAC for device %s\n",
+				 dev->bus_id);
+		return 0;
+	}
+#endif
+
+	if (dma_ops->dma_supported)
+		return dma_ops->dma_supported(dev, mask);
+
+	/* Copied from i386. Doesn't make much sense, because it will
+	   only work for pci_alloc_coherent.
+	   The caller just has to use GFP_DMA in this case. */
+	if (mask < DMA_24BIT_MASK)
+		return 0;
+
+	/* Tell the device to use SAC when IOMMU force is on.  This
+	   allows the driver to use cheaper accesses in some cases.
+
+	   Problem with this is that if we overflow the IOMMU area and
+	   return DAC as fallback address the device may not handle it
+	   correctly.
+
+	   As a special case some controllers have a 39bit address
+	   mode that is as efficient as 32bit (aic79xx). Don't force
+	   SAC for these.  Assume all masks <= 40 bits are of this
+	   type. Normally this doesn't make any difference, but gives
+	   more gentle handling of IOMMU overflow. */
+	if (iommu_sac_force && (mask >= DMA_40BIT_MASK)) {
+		printk(KERN_INFO "%s: Force SAC with mask %Lx\n",
+				 dev->bus_id, mask);
+		return 0;
+	}
+
+	return 1;
+}
+EXPORT_SYMBOL(dma_supported);
+
+/* Allocate DMA memory on node near device */
+noinline struct page *
+dma_alloc_pages(struct device *dev, gfp_t gfp, unsigned order)
+{
+	int node;
+
+	node = dev_to_node(dev);
+
+	return alloc_pages_node(node, gfp, order);
+}
+
+/*
+ * Allocate memory for a coherent mapping.
+ */
+void *
+dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
+		   gfp_t gfp)
+{
+	void *memory = NULL;
+	struct page *page;
+	unsigned long dma_mask = 0;
+	dma_addr_t bus;
+
+	/* ignore region specifiers */
+	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
+
+	if (dma_alloc_from_coherent_mem(dev, size, dma_handle, &memory))
+		return memory;
+
+	if (!dev)
+		dev = &fallback_dev;
+	dma_mask = dev->coherent_dma_mask;
+	if (dma_mask == 0)
+		dma_mask = DMA_32BIT_MASK;
+
+	/* Device not DMA able */
+	if (dev->dma_mask == NULL)
+		return NULL;
+
+	/* Don't invoke OOM killer */
+	gfp |= __GFP_NORETRY;
+
+#ifdef CONFIG_X86_64
+	/* Why <=? Even when the mask is smaller than 4GB it is often
+	   larger than 16MB and in this case we have a chance of
+	   finding fitting memory in the next higher zone first. If
+	   not retry with true GFP_DMA. -AK */
+	if (dma_mask <= DMA_32BIT_MASK)
+		gfp |= GFP_DMA32;
+#endif
+
+ again:
+	page = dma_alloc_pages(dev, gfp, get_order(size));
+	if (page == NULL)
+		return NULL;
+
+	{
+		int high, mmu;
+		bus = page_to_phys(page);
+		memory = page_address(page);
+		high = (bus + size) >= dma_mask;
+		mmu = high;
+		if (force_iommu && !(gfp & GFP_DMA))
+			mmu = 1;
+		else if (high) {
+			free_pages((unsigned long)memory,
+				   get_order(size));
+
+			/* Don't use the 16MB ZONE_DMA unless absolutely
+			   needed. It's better to use remapping first. */
+			if (dma_mask < DMA_32BIT_MASK && !(gfp & GFP_DMA)) {
+				gfp = (gfp & ~GFP_DMA32) | GFP_DMA;
+				goto again;
+			}
+
+			/* 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,
+							   dma_handle, gfp);
+			return NULL;
+		}
+
+		memset(memory, 0, size);
+		if (!mmu) {
+			*dma_handle = bus;
+			return memory;
+		}
+	}
+
+	if (dma_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);
+	}
+
+	if (dma_ops->map_simple) {
+		*dma_handle = dma_ops->map_simple(dev, virt_to_phys(memory),
+					      size,
+					      PCI_DMA_BIDIRECTIONAL);
+		if (*dma_handle != bad_dma_address)
+			return memory;
+	}
+
+	if (panic_on_overflow)
+		panic("dma_alloc_coherent: IOMMU overflow by %lu bytes\n",
+		      (unsigned long)size);
+	free_pages((unsigned long)memory, get_order(size));
+	return NULL;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+/*
+ * Unmap coherent memory.
+ * The caller must ensure that the device has finished accessing the mapping.
+ */
+void dma_free_coherent(struct device *dev, size_t size,
+			 void *vaddr, dma_addr_t bus)
+{
+	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);
+	free_pages((unsigned long)vaddr, order);
+}
+EXPORT_SYMBOL(dma_free_coherent);
 
 static int __init pci_iommu_init(void)
 {
@@ -327,6 +506,8 @@ void pci_iommu_shutdown(void)
 {
 	gart_iommu_shutdown();
 }
+/* Must execute after PCI subsystem */
+fs_initcall(pci_iommu_init);
 
 #ifdef CONFIG_PCI
 /* Many VIA bridges seem to corrupt data for DAC. Disable it here */
@@ -334,11 +515,10 @@ void pci_iommu_shutdown(void)
 static __devinit void via_no_dac(struct pci_dev *dev)
 {
 	if ((dev->class >> 8) == PCI_CLASS_BRIDGE_PCI && forbid_dac == 0) {
-		printk(KERN_INFO "PCI: VIA PCI bridge detected. Disabling DAC.\n");
+		printk(KERN_INFO "PCI: VIA PCI bridge detected."
+				 "Disabling DAC.\n");
 		forbid_dac = 1;
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_ANY_ID, via_no_dac);
 #endif
-/* Must execute after PCI subsystem */
-fs_initcall(pci_iommu_init);

arch/x86/kernel/pci-dma_32.c

@@ -1,177 +0,0 @@
-/*
- * Dynamic DMA mapping support.
- *
- * On i386 there is no hardware dynamic DMA address translation,
- * so consistent alloc/free are merely page allocation/freeing.
- * The rest of the dynamic DMA mapping interface is implemented
- * in asm/pci.h.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <linux/module.h>
-#include <asm/io.h>
-
-struct dma_coherent_mem {
-	void		*virt_base;
-	u32		device_base;
-	int		size;
-	int		flags;
-	unsigned long	*bitmap;
-};
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t gfp)
-{
-	void *ret;
-	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-	int order = get_order(size);
-	/* ignore region specifiers */
-	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
-
-	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);
-			return ret;
-		}
-		if (mem->flags & DMA_MEMORY_EXCLUSIVE)
-			return NULL;
-	}
-
-	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
-		gfp |= GFP_DMA;
-
-	ret = (void *)__get_free_pages(gfp, order);
-
-	if (ret != NULL) {
-		memset(ret, 0, size);
-		*dma_handle = virt_to_phys(ret);
-	}
-	return ret;
-}
-EXPORT_SYMBOL(dma_alloc_coherent);
-
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
-{
-	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
-	int order = get_order(size);
-
-	WARN_ON(irqs_disabled());	/* for portability */
-	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);
-	} else
-		free_pages((unsigned long)vaddr, order);
-}
-EXPORT_SYMBOL(dma_free_coherent);
-
-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 pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	int pos, err;
-
-	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);
-
-#ifdef CONFIG_PCI
-/* Many VIA bridges seem to corrupt data for DAC. Disable it here */
-
-int forbid_dac;
-EXPORT_SYMBOL(forbid_dac);
-
-static __devinit void via_no_dac(struct pci_dev *dev)
-{
-	if ((dev->class >> 8) == PCI_CLASS_BRIDGE_PCI && forbid_dac == 0) {
-		printk(KERN_INFO "PCI: VIA PCI bridge detected. Disabling DAC.\n");
-		forbid_dac = 1;
-	}
-}
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_ANY_ID, via_no_dac);
-
-static int check_iommu(char *s)
-{
-	if (!strcmp(s, "usedac")) {
-		forbid_dac = -1;
-		return 1;
-	}
-	return 0;
-}
-__setup("iommu=", check_iommu);
-#endif

arch/x86/kernel/pci-gart_64.c

@@ -264,9 +264,9 @@ static dma_addr_t dma_map_area(struct device *dev, dma_addr_t phys_mem,
 }
 
 static dma_addr_t
-gart_map_simple(struct device *dev, char *buf, size_t size, int dir)
+gart_map_simple(struct device *dev, phys_addr_t paddr, size_t size, int dir)
 {
-	dma_addr_t map = dma_map_area(dev, virt_to_bus(buf), size, dir);
+	dma_addr_t map = dma_map_area(dev, paddr, size, dir);
 
 	flush_gart();
 
@@ -275,18 +275,17 @@ gart_map_simple(struct device *dev, char *buf, size_t size, int dir)
 
 /* Map a single area into the IOMMU */
 static dma_addr_t
-gart_map_single(struct device *dev, void *addr, size_t size, int dir)
+gart_map_single(struct device *dev, phys_addr_t paddr, size_t size, int dir)
 {
-	unsigned long phys_mem, bus;
+	unsigned long bus;
 
 	if (!dev)
 		dev = &fallback_dev;
 
-	phys_mem = virt_to_phys(addr);
+	if (!need_iommu(dev, paddr, size))
-	if (!need_iommu(dev, phys_mem, size))
+		return paddr;
-		return phys_mem;
 
-	bus = gart_map_simple(dev, addr, size, dir);
+	bus = gart_map_simple(dev, paddr, size, dir);
 
 	return bus;
 }

arch/x86/kernel/pci-nommu.c

@@ -14,7 +14,7 @@
 static int
 check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size)
 {
-        if (hwdev && bus + size > *hwdev->dma_mask) {
+	if (hwdev && bus + size > *hwdev->dma_mask) {
 		if (*hwdev->dma_mask >= DMA_32BIT_MASK)
 			printk(KERN_ERR
 			    "nommu_%s: overflow %Lx+%zu of device mask %Lx\n",
@@ -26,19 +26,17 @@ check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size)
 }
 
 static dma_addr_t
-nommu_map_single(struct device *hwdev, void *ptr, size_t size,
+nommu_map_single(struct device *hwdev, phys_addr_t paddr, size_t size,
 	       int direction)
 {
-	dma_addr_t bus = virt_to_bus(ptr);
+	dma_addr_t bus = paddr;
+	WARN_ON(size == 0);
 	if (!check_addr("map_single", hwdev, bus, size))
 				return bad_dma_address;
+	flush_write_buffers();
 	return bus;
 }
 
-static void nommu_unmap_single(struct device *dev, dma_addr_t addr,size_t size,
-			int direction)
-{
-}
 
 /* Map a set of buffers described by scatterlist in streaming
  * mode for DMA.  This is the scatter-gather version of the
@@ -61,30 +59,34 @@ static int nommu_map_sg(struct device *hwdev, struct scatterlist *sg,
 	struct scatterlist *s;
 	int i;
 
+	WARN_ON(nents == 0 || sg[0].length == 0);
+
 	for_each_sg(sg, s, nents, i) {
 		BUG_ON(!sg_page(s));
-		s->dma_address = virt_to_bus(sg_virt(s));
+		s->dma_address = sg_phys(s);
 		if (!check_addr("map_sg", hwdev, s->dma_address, s->length))
 			return 0;
 		s->dma_length = s->length;
 	}
+	flush_write_buffers();
 	return nents;
 }
 
-/* Unmap a set of streaming mode DMA translations.
+/* Make sure we keep the same behaviour */
- * Again, cpu read rules concerning calls here are the same as for
+static int nommu_mapping_error(dma_addr_t dma_addr)
- * pci_unmap_single() above.
- */
-static void nommu_unmap_sg(struct device *dev, struct scatterlist *sg,
-		  int nents, int dir)
 {
+#ifdef CONFIG_X86_32
+	return 0;
+#else
+	return (dma_addr == bad_dma_address);
+#endif
 }
 
+
 const struct dma_mapping_ops nommu_dma_ops = {
 	.map_single = nommu_map_single,
-	.unmap_single = nommu_unmap_single,
 	.map_sg = nommu_map_sg,
-	.unmap_sg = nommu_unmap_sg,
+	.mapping_error = nommu_mapping_error,
 	.is_phys = 1,
 };
 

arch/x86/kernel/pci-swiotlb_64.c

@@ -11,11 +11,18 @@
 
 int swiotlb __read_mostly;
 
+static dma_addr_t
+swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
+			int direction)
+{
+	return swiotlb_map_single(hwdev, phys_to_virt(paddr), size, direction);
+}
+
 const struct dma_mapping_ops swiotlb_dma_ops = {
 	.mapping_error = swiotlb_dma_mapping_error,
 	.alloc_coherent = swiotlb_alloc_coherent,
 	.free_coherent = swiotlb_free_coherent,
-	.map_single = swiotlb_map_single,
+	.map_single = swiotlb_map_single_phys,
 	.unmap_single = swiotlb_unmap_single,
 	.sync_single_for_cpu = swiotlb_sync_single_for_cpu,
 	.sync_single_for_device = swiotlb_sync_single_for_device,

arch/x86/kernel/process.c

@@ -0,0 +1,44 @@
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+
+struct kmem_cache *task_xstate_cachep;
+
+int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
+{
+	*dst = *src;
+	if (src->thread.xstate) {
+		dst->thread.xstate = kmem_cache_alloc(task_xstate_cachep,
+						      GFP_KERNEL);
+		if (!dst->thread.xstate)
+			return -ENOMEM;
+		WARN_ON((unsigned long)dst->thread.xstate & 15);
+		memcpy(dst->thread.xstate, src->thread.xstate, xstate_size);
+	}
+	return 0;
+}
+
+void free_thread_xstate(struct task_struct *tsk)
+{
+	if (tsk->thread.xstate) {
+		kmem_cache_free(task_xstate_cachep, tsk->thread.xstate);
+		tsk->thread.xstate = NULL;
+	}
+}
+
+void free_thread_info(struct thread_info *ti)
+{
+	free_thread_xstate(ti->task);
+	free_pages((unsigned long)ti, get_order(THREAD_SIZE));
+}
+
+void arch_task_cache_init(void)
+{
+        task_xstate_cachep =
+        	kmem_cache_create("task_xstate", xstate_size,
+				  __alignof__(union thread_xstate),
+				  SLAB_PANIC, NULL);
+}

arch/x86/kernel/process_32.c

@@ -36,6 +36,7 @@
 #include <linux/personality.h>
 #include <linux/tick.h>
 #include <linux/percpu.h>
+#include <linux/prctl.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -45,7 +46,6 @@
 #include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/desc.h>
-#include <asm/vm86.h>
 #ifdef CONFIG_MATH_EMULATION
 #include <asm/math_emu.h>
 #endif
@@ -521,14 +521,18 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
 	regs->cs		= __USER_CS;
 	regs->ip		= new_ip;
 	regs->sp		= new_sp;
+	/*
+	 * Free the old FP and other extended state
+	 */
+	free_thread_xstate(current);
 }
 EXPORT_SYMBOL_GPL(start_thread);
 
-#ifdef CONFIG_SECCOMP
 static void hard_disable_TSC(void)
 {
 	write_cr4(read_cr4() | X86_CR4_TSD);
 }
+
 void disable_TSC(void)
 {
 	preempt_disable();
@@ -540,11 +544,47 @@ void disable_TSC(void)
 		hard_disable_TSC();
 	preempt_enable();
 }
+
 static void hard_enable_TSC(void)
 {
 	write_cr4(read_cr4() & ~X86_CR4_TSD);
 }
-#endif /* CONFIG_SECCOMP */
+
+void enable_TSC(void)
+{
+	preempt_disable();
+	if (test_and_clear_thread_flag(TIF_NOTSC))
+		/*
+		 * Must flip the CPU state synchronously with
+		 * TIF_NOTSC in the current running context.
+		 */
+		hard_enable_TSC();
+	preempt_enable();
+}
+
+int get_tsc_mode(unsigned long adr)
+{
+	unsigned int val;
+
+	if (test_thread_flag(TIF_NOTSC))
+		val = PR_TSC_SIGSEGV;
+	else
+		val = PR_TSC_ENABLE;
+
+	return put_user(val, (unsigned int __user *)adr);
+}
+
+int set_tsc_mode(unsigned int val)
+{
+	if (val == PR_TSC_SIGSEGV)
+		disable_TSC();
+	else if (val == PR_TSC_ENABLE)
+		enable_TSC();
+	else
+		return -EINVAL;
+
+	return 0;
+}
 
 static noinline void
 __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
@@ -578,7 +618,6 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		set_debugreg(next->debugreg7, 7);
 	}
 
-#ifdef CONFIG_SECCOMP
 	if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
 	    test_tsk_thread_flag(next_p, TIF_NOTSC)) {
 		/* prev and next are different */
@@ -587,7 +626,6 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		else
 			hard_enable_TSC();
 	}
-#endif
 
 #ifdef X86_BTS
 	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
@@ -669,7 +707,7 @@ struct task_struct * __switch_to(struct task_struct *prev_p, struct task_struct
 
 	/* we're going to use this soon, after a few expensive things */
 	if (next_p->fpu_counter > 5)
-		prefetch(&next->i387.fxsave);
+		prefetch(next->xstate);
 
 	/*
 	 * Reload esp0.

arch/x86/kernel/process_64.c

@@ -36,6 +36,7 @@
 #include <linux/kprobes.h>
 #include <linux/kdebug.h>
 #include <linux/tick.h>
+#include <linux/prctl.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -532,9 +533,71 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
 	regs->ss		= __USER_DS;
 	regs->flags		= 0x200;
 	set_fs(USER_DS);
+	/*
+	 * Free the old FP and other extended state
+	 */
+	free_thread_xstate(current);
 }
 EXPORT_SYMBOL_GPL(start_thread);
 
+static void hard_disable_TSC(void)
+{
+	write_cr4(read_cr4() | X86_CR4_TSD);
+}
+
+void disable_TSC(void)
+{
+	preempt_disable();
+	if (!test_and_set_thread_flag(TIF_NOTSC))
+		/*
+		 * Must flip the CPU state synchronously with
+		 * TIF_NOTSC in the current running context.
+		 */
+		hard_disable_TSC();
+	preempt_enable();
+}
+
+static void hard_enable_TSC(void)
+{
+	write_cr4(read_cr4() & ~X86_CR4_TSD);
+}
+
+void enable_TSC(void)
+{
+	preempt_disable();
+	if (test_and_clear_thread_flag(TIF_NOTSC))
+		/*
+		 * Must flip the CPU state synchronously with
+		 * TIF_NOTSC in the current running context.
+		 */
+		hard_enable_TSC();
+	preempt_enable();
+}
+
+int get_tsc_mode(unsigned long adr)
+{
+	unsigned int val;
+
+	if (test_thread_flag(TIF_NOTSC))
+		val = PR_TSC_SIGSEGV;
+	else
+		val = PR_TSC_ENABLE;
+
+	return put_user(val, (unsigned int __user *)adr);
+}
+
+int set_tsc_mode(unsigned int val)
+{
+	if (val == PR_TSC_SIGSEGV)
+		disable_TSC();
+	else if (val == PR_TSC_ENABLE)
+		enable_TSC();
+	else
+		return -EINVAL;
+
+	return 0;
+}
+
 /*
  * This special macro can be used to load a debugging register
  */
@@ -572,6 +635,15 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 		loaddebug(next, 7);
 	}
 
+	if (test_tsk_thread_flag(prev_p, TIF_NOTSC) ^
+	    test_tsk_thread_flag(next_p, TIF_NOTSC)) {
+		/* prev and next are different */
+		if (test_tsk_thread_flag(next_p, TIF_NOTSC))
+			hard_disable_TSC();
+		else
+			hard_enable_TSC();
+	}
+
 	if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
 		/*
 		 * Copy the relevant range of the IO bitmap.
@@ -614,7 +686,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 
 	/* we're going to use this soon, after a few expensive things */
 	if (next_p->fpu_counter>5)
-		prefetch(&next->i387.fxsave);
+		prefetch(next->xstate);
 
 	/*
 	 * Reload esp0, LDT and the page table pointer:

arch/x86/kernel/setup64.c

@@ -74,8 +74,8 @@ int force_personality32 = 0;
 Control non executable heap for 32bit processes.
 To control the stack too use noexec=off
 
-on	PROT_READ does not imply PROT_EXEC for 32bit processes
+on	PROT_READ does not imply PROT_EXEC for 32bit processes (default)
-off	PROT_READ implies PROT_EXEC (default)
+off	PROT_READ implies PROT_EXEC
 */
 static int __init nonx32_setup(char *str)
 {

arch/x86/kernel/setup_32.c

@@ -812,10 +812,10 @@ void __init setup_arch(char **cmdline_p)
 		efi_init();
 
 	/* update e820 for memory not covered by WB MTRRs */
-	find_max_pfn();
+	propagate_e820_map();
 	mtrr_bp_init();
 	if (mtrr_trim_uncached_memory(max_pfn))
-		find_max_pfn();
+		propagate_e820_map();
 
 	max_low_pfn = setup_memory();
 

arch/x86/kernel/setup_64.c

@@ -398,6 +398,8 @@ void __init setup_arch(char **cmdline_p)
 
 	early_res_to_bootmem();
 
+	dma32_reserve_bootmem();
+
 #ifdef CONFIG_ACPI_SLEEP
 	/*
 	 * Reserve low memory region for sleep support.
@@ -420,11 +422,14 @@ void __init setup_arch(char **cmdline_p)
 		unsigned long end_of_mem    = end_pfn << PAGE_SHIFT;
 
 		if (ramdisk_end <= end_of_mem) {
-			reserve_bootmem_generic(ramdisk_image, ramdisk_size);
+			/*
+			 * don't need to reserve again, already reserved early
+			 * in x86_64_start_kernel, and early_res_to_bootmem
+			 * convert that to reserved in bootmem
+			 */
 			initrd_start = ramdisk_image + PAGE_OFFSET;
 			initrd_end = initrd_start+ramdisk_size;
 		} else {
-			/* Assumes everything on node 0 */
 			free_bootmem(ramdisk_image, ramdisk_size);
 			printk(KERN_ERR "initrd extends beyond end of memory "
 			       "(0x%08lx > 0x%08lx)\ndisabling initrd\n",

arch/x86/kernel/smpboot.c

@@ -61,6 +61,7 @@
 #include <asm/mtrr.h>
 #include <asm/nmi.h>
 #include <asm/vmi.h>
+#include <asm/genapic.h>
 #include <linux/mc146818rtc.h>
 
 #include <mach_apic.h>
@@ -677,6 +678,12 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 	unsigned long send_status, accept_status = 0;
 	int maxlvt, num_starts, j;
 
+	if (get_uv_system_type() == UV_NON_UNIQUE_APIC) {
+		send_status = uv_wakeup_secondary(phys_apicid, start_eip);
+		atomic_set(&init_deasserted, 1);
+		return send_status;
+	}
+
 	/*
 	 * Be paranoid about clearing APIC errors.
 	 */
@@ -918,16 +925,19 @@ do_rest:
 
 	atomic_set(&init_deasserted, 0);
 
-	Dprintk("Setting warm reset code and vector.\n");
+	if (get_uv_system_type() != UV_NON_UNIQUE_APIC) {
 
-	store_NMI_vector(&nmi_high, &nmi_low);
+		Dprintk("Setting warm reset code and vector.\n");
 
-	smpboot_setup_warm_reset_vector(start_ip);
+		store_NMI_vector(&nmi_high, &nmi_low);
-	/*
+
-	 * Be paranoid about clearing APIC errors.
+		smpboot_setup_warm_reset_vector(start_ip);
-	 */
+		/*
-	apic_write(APIC_ESR, 0);
+		 * Be paranoid about clearing APIC errors.
-	apic_read(APIC_ESR);
+	 	*/
+		apic_write(APIC_ESR, 0);
+		apic_read(APIC_ESR);
+	}
 
 	/*
 	 * Starting actual IPI sequence...
@@ -966,7 +976,8 @@ do_rest:
 			else
 				/* trampoline code not run */
 				printk(KERN_ERR "Not responding.\n");
-			inquire_remote_apic(apicid);
+			if (get_uv_system_type() != UV_NON_UNIQUE_APIC)
+				inquire_remote_apic(apicid);
 		}
 	}
 

arch/x86/kernel/traps_32.c

@@ -681,7 +681,7 @@ gp_in_kernel:
 	}
 }
 
-static __kprobes void
+static notrace __kprobes void
 mem_parity_error(unsigned char reason, struct pt_regs *regs)
 {
 	printk(KERN_EMERG
@@ -707,7 +707,7 @@ mem_parity_error(unsigned char reason, struct pt_regs *regs)
 	clear_mem_error(reason);
 }
 
-static __kprobes void
+static notrace __kprobes void
 io_check_error(unsigned char reason, struct pt_regs *regs)
 {
 	unsigned long i;
@@ -727,7 +727,7 @@ io_check_error(unsigned char reason, struct pt_regs *regs)
 	outb(reason, 0x61);
 }
 
-static __kprobes void
+static notrace __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 {
 	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -755,7 +755,7 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 
 static DEFINE_SPINLOCK(nmi_print_lock);
 
-void __kprobes die_nmi(struct pt_regs *regs, const char *msg)
+void notrace __kprobes die_nmi(struct pt_regs *regs, const char *msg)
 {
 	if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) == NOTIFY_STOP)
 		return;
@@ -786,7 +786,7 @@ void __kprobes die_nmi(struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
-static __kprobes void default_do_nmi(struct pt_regs *regs)
+static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;
 
@@ -828,7 +828,7 @@ static __kprobes void default_do_nmi(struct pt_regs *regs)
 
 static int ignore_nmis;
 
-__kprobes void do_nmi(struct pt_regs *regs, long error_code)
+notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code)
 {
 	int cpu;
 
@@ -1148,9 +1148,22 @@ asmlinkage void math_state_restore(void)
 	struct thread_info *thread = current_thread_info();
 	struct task_struct *tsk = thread->task;
 
+	if (!tsk_used_math(tsk)) {
+		local_irq_enable();
+		/*
+		 * does a slab alloc which can sleep
+		 */
+		if (init_fpu(tsk)) {
+			/*
+			 * ran out of memory!
+			 */
+			do_group_exit(SIGKILL);
+			return;
+		}
+		local_irq_disable();
+	}
+
 	clts();				/* Allow maths ops (or we recurse) */
-	if (!tsk_used_math(tsk))
-		init_fpu(tsk);
 	restore_fpu(tsk);
 	thread->status |= TS_USEDFPU;	/* So we fnsave on switch_to() */
 	tsk->fpu_counter++;
@@ -1208,11 +1221,6 @@ void __init trap_init(void)
 #endif
 	set_trap_gate(19, &simd_coprocessor_error);
 
-	/*
-	 * Verify that the FXSAVE/FXRSTOR data will be 16-byte aligned.
-	 * Generate a build-time error if the alignment is wrong.
-	 */
-	BUILD_BUG_ON(offsetof(struct task_struct, thread.i387.fxsave) & 15);
 	if (cpu_has_fxsr) {
 		printk(KERN_INFO "Enabling fast FPU save and restore... ");
 		set_in_cr4(X86_CR4_OSFXSR);
@@ -1233,6 +1241,7 @@ void __init trap_init(void)
 
 	set_bit(SYSCALL_VECTOR, used_vectors);
 
+	init_thread_xstate();
 	/*
 	 * Should be a barrier for any external CPU state:
 	 */

arch/x86/kernel/traps_64.c

@@ -600,7 +600,8 @@ void die(const char * str, struct pt_regs * regs, long err)
 	oops_end(flags, regs, SIGSEGV);
 }
 
-void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
+notrace __kprobes void
+die_nmi(char *str, struct pt_regs *regs, int do_panic)
 {
 	unsigned long flags;
 
@@ -772,7 +773,7 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 	die("general protection fault", regs, error_code);
 }
 
-static __kprobes void
+static notrace __kprobes void
 mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
@@ -796,7 +797,7 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs)
 	outb(reason, 0x61);
 }
 
-static __kprobes void
+static notrace __kprobes void
 io_check_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk("NMI: IOCK error (debug interrupt?)\n");
@@ -810,7 +811,7 @@ io_check_error(unsigned char reason, struct pt_regs * regs)
 	outb(reason, 0x61);
 }
 
-static __kprobes void
+static notrace __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {
 	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -827,7 +828,7 @@ unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
    Nested NMIs are prevented by the CPU. */
-asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
+asmlinkage notrace  __kprobes void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;
 	int cpu;
@@ -1123,11 +1124,24 @@ asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void)
 asmlinkage void math_state_restore(void)
 {
 	struct task_struct *me = current;
-	clts();			/* Allow maths ops (or we recurse) */
 
-	if (!used_math())
+	if (!used_math()) {
-		init_fpu(me);
+		local_irq_enable();
-	restore_fpu_checking(&me->thread.i387.fxsave);
+		/*
+		 * does a slab alloc which can sleep
+		 */
+		if (init_fpu(me)) {
+			/*
+			 * ran out of memory!
+			 */
+			do_group_exit(SIGKILL);
+			return;
+		}
+		local_irq_disable();
+	}
+
+	clts();			/* Allow maths ops (or we recurse) */
+	restore_fpu_checking(&me->thread.xstate->fxsave);
 	task_thread_info(me)->status |= TS_USEDFPU;
 	me->fpu_counter++;
 }
@@ -1162,6 +1176,10 @@ void __init trap_init(void)
 	set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall);
 #endif
        
+	/*
+	 * initialize the per thread extended state:
+	 */
+        init_thread_xstate();
 	/*
 	 * Should be a barrier for any external CPU state.
 	 */

arch/x86/kernel/tsc_32.c

@@ -221,9 +221,9 @@ EXPORT_SYMBOL(recalibrate_cpu_khz);
  * if the CPU frequency is scaled, TSC-based delays will need a different
  * loops_per_jiffy value to function properly.
  */
-static unsigned int ref_freq = 0;
+static unsigned int ref_freq;
-static unsigned long loops_per_jiffy_ref = 0;
+static unsigned long loops_per_jiffy_ref;
-static unsigned long cpu_khz_ref = 0;
+static unsigned long cpu_khz_ref;
 
 static int
 time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data)
@@ -283,15 +283,28 @@ core_initcall(cpufreq_tsc);
 
 /* clock source code */
 
-static unsigned long current_tsc_khz = 0;
+static unsigned long current_tsc_khz;
+static struct clocksource clocksource_tsc;
 
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp issue. This can be observed in
+ * a very small window right after one CPU updated cycle_last under
+ * xtime lock and the other CPU reads a TSC value which is smaller
+ * than the cycle_last reference value due to a TSC which is slighty
+ * behind. This delta is nowhere else observable, but in that case it
+ * results in a forward time jump in the range of hours due to the
+ * unsigned delta calculation of the time keeping core code, which is
+ * necessary to support wrapping clocksources like pm timer.
+ */
 static cycle_t read_tsc(void)
 {
 	cycle_t ret;
 
 	rdtscll(ret);
 
-	return ret;
+	return ret >= clocksource_tsc.cycle_last ?
+		ret : clocksource_tsc.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {

arch/x86/kernel/tsc_64.c

@@ -11,6 +11,7 @@
 #include <asm/hpet.h>
 #include <asm/timex.h>
 #include <asm/timer.h>
+#include <asm/vgtod.h>
 
 static int notsc __initdata = 0;
 
@@ -287,18 +288,34 @@ int __init notsc_setup(char *s)
 
 __setup("notsc", notsc_setup);
 
+static struct clocksource clocksource_tsc;
 
-/* clock source code: */
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp. This can be observed in a
+ * very small window right after one CPU updated cycle_last under
+ * xtime/vsyscall_gtod lock and the other CPU reads a TSC value which
+ * is smaller than the cycle_last reference value due to a TSC which
+ * is slighty behind. This delta is nowhere else observable, but in
+ * that case it results in a forward time jump in the range of hours
+ * due to the unsigned delta calculation of the time keeping core
+ * code, which is necessary to support wrapping clocksources like pm
+ * timer.
+ */
 static cycle_t read_tsc(void)
 {
 	cycle_t ret = (cycle_t)get_cycles();
-	return ret;
+
+	return ret >= clocksource_tsc.cycle_last ?
+		ret : clocksource_tsc.cycle_last;
 }
 
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
 	cycle_t ret = (cycle_t)vget_cycles();
-	return ret;
+
+	return ret >= __vsyscall_gtod_data.clock.cycle_last ?
+		ret : __vsyscall_gtod_data.clock.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {

arch/x86/mach-visws/visws_apic.c

@@ -1,6 +1,4 @@
 /*
- *	linux/arch/i386/mach-visws/visws_apic.c
- *
  *	Copyright (C) 1999 Bent Hagemark, Ingo Molnar
  *
  *  SGI Visual Workstation interrupt controller

arch/x86/mach-voyager/voyager_basic.c

@@ -2,8 +2,6 @@
  *
  * Author: J.E.J.Bottomley@HansenPartnership.com
  *
- * linux/arch/i386/kernel/voyager.c
- *
  * This file contains all the voyager specific routines for getting
  * initialisation of the architecture to function.  For additional
  * features see:

arch/x86/mach-voyager/voyager_cat.c

@@ -4,8 +4,6 @@
  *
  * Author: J.E.J.Bottomley@HansenPartnership.com
  *
- * linux/arch/i386/kernel/voyager_cat.c
- *
  * This file contains all the logic for manipulating the CAT bus
  * in a level 5 machine.
  *

arch/x86/mach-voyager/voyager_smp.c

@@ -4,8 +4,6 @@
  *
  * Author: J.E.J.Bottomley@HansenPartnership.com
  *
- * linux/arch/i386/kernel/voyager_smp.c
- *
  * This file provides all the same external entries as smp.c but uses
  * the voyager hal to provide the functionality
  */

arch/x86/mach-voyager/voyager_thread.c

@@ -4,8 +4,6 @@
  *
  * Author: J.E.J.Bottomley@HansenPartnership.com
  *
- * linux/arch/i386/kernel/voyager_thread.c
- *
  * This module provides the machine status monitor thread for the
  * voyager architecture.  This allows us to monitor the machine
  * environment (temp, voltage, fan function) and the front panel and

arch/x86/math-emu/fpu_entry.c

@@ -678,7 +678,7 @@ int fpregs_soft_set(struct task_struct *target,
 		    unsigned int pos, unsigned int count,
 		    const void *kbuf, const void __user *ubuf)
 {
-	struct i387_soft_struct *s387 = &target->thread.i387.soft;
+	struct i387_soft_struct *s387 = &target->thread.xstate->soft;
 	void *space = s387->st_space;
 	int ret;
 	int offset, other, i, tags, regnr, tag, newtop;
@@ -730,7 +730,7 @@ int fpregs_soft_get(struct task_struct *target,
 		    unsigned int pos, unsigned int count,
 		    void *kbuf, void __user *ubuf)
 {
-	struct i387_soft_struct *s387 = &target->thread.i387.soft;
+	struct i387_soft_struct *s387 = &target->thread.xstate->soft;
 	const void *space = s387->st_space;
 	int ret;
 	int offset = (S387->ftop & 7) * 10, other = 80 - offset;

arch/x86/math-emu/fpu_system.h

@@ -35,8 +35,8 @@
 #define SEG_EXPAND_DOWN(s)	(((s).b & ((1 << 11) | (1 << 10))) \
 				 == (1 << 10))
 
-#define I387			(current->thread.i387)
+#define I387			(current->thread.xstate)
-#define FPU_info		(I387.soft.info)
+#define FPU_info		(I387->soft.info)
 
 #define FPU_CS			(*(unsigned short *) &(FPU_info->___cs))
 #define FPU_SS			(*(unsigned short *) &(FPU_info->___ss))
@@ -46,25 +46,25 @@
 #define FPU_EIP			(FPU_info->___eip)
 #define FPU_ORIG_EIP		(FPU_info->___orig_eip)
 
-#define FPU_lookahead           (I387.soft.lookahead)
+#define FPU_lookahead           (I387->soft.lookahead)
 
 /* nz if ip_offset and cs_selector are not to be set for the current
    instruction. */
-#define no_ip_update		(*(u_char *)&(I387.soft.no_update))
+#define no_ip_update		(*(u_char *)&(I387->soft.no_update))
-#define FPU_rm			(*(u_char *)&(I387.soft.rm))
+#define FPU_rm			(*(u_char *)&(I387->soft.rm))
 
 /* Number of bytes of data which can be legally accessed by the current
    instruction. This only needs to hold a number <= 108, so a byte will do. */
-#define access_limit		(*(u_char *)&(I387.soft.alimit))
+#define access_limit		(*(u_char *)&(I387->soft.alimit))
 
-#define partial_status		(I387.soft.swd)
+#define partial_status		(I387->soft.swd)
-#define control_word		(I387.soft.cwd)
+#define control_word		(I387->soft.cwd)
-#define fpu_tag_word		(I387.soft.twd)
+#define fpu_tag_word		(I387->soft.twd)
-#define registers		(I387.soft.st_space)
+#define registers		(I387->soft.st_space)
-#define top			(I387.soft.ftop)
+#define top			(I387->soft.ftop)
 
-#define instruction_address	(*(struct address *)&I387.soft.fip)
+#define instruction_address	(*(struct address *)&I387->soft.fip)
-#define operand_address		(*(struct address *)&I387.soft.foo)
+#define operand_address		(*(struct address *)&I387->soft.foo)
 
 #define FPU_access_ok(x,y,z)	if ( !access_ok(x,y,z) ) \
 				math_abort(FPU_info,SIGSEGV)

arch/x86/math-emu/reg_ld_str.c

@@ -1180,8 +1180,8 @@ u_char __user *fstenv(fpu_addr_modes addr_modes, u_char __user *d)
 		control_word |= 0xffff0040;
 		partial_status = status_word() | 0xffff0000;
 		fpu_tag_word |= 0xffff0000;
-		I387.soft.fcs &= ~0xf8000000;
+		I387->soft.fcs &= ~0xf8000000;
-		I387.soft.fos |= 0xffff0000;
+		I387->soft.fos |= 0xffff0000;
 #endif /* PECULIAR_486 */
 		if (__copy_to_user(d, &control_word, 7 * 4))
 			FPU_abort;

arch/x86/mm/discontig_32.c

@@ -120,7 +120,7 @@ int __init get_memcfg_numa_flat(void)
 	printk("NUMA - single node, flat memory mode\n");
 
 	/* Run the memory configuration and find the top of memory. */
-	find_max_pfn();
+	propagate_e820_map();
 	node_start_pfn[0] = 0;
 	node_end_pfn[0] = max_pfn;
 	memory_present(0, 0, max_pfn);
@@ -134,7 +134,7 @@ int __init get_memcfg_numa_flat(void)
 /*
  * Find the highest page frame number we have available for the node
  */
-static void __init find_max_pfn_node(int nid)
+static void __init propagate_e820_map_node(int nid)
 {
 	if (node_end_pfn[nid] > max_pfn)
 		node_end_pfn[nid] = max_pfn;
@@ -379,7 +379,7 @@ unsigned long __init setup_memory(void)
 	printk("High memory starts at vaddr %08lx\n",
 			(ulong) pfn_to_kaddr(highstart_pfn));
 	for_each_online_node(nid)
-		find_max_pfn_node(nid);
+		propagate_e820_map_node(nid);
 
 	memset(NODE_DATA(0), 0, sizeof(struct pglist_data));
 	NODE_DATA(0)->bdata = &node0_bdata;

arch/x86/mm/init_32.c

@@ -1,5 +1,4 @@
 /*
- *  linux/arch/i386/mm/init.c
  *
  *  Copyright (C) 1995  Linus Torvalds
  *

arch/x86/mm/init_64.c

@@ -47,9 +47,6 @@
 #include <asm/numa.h>
 #include <asm/cacheflush.h>
 
-const struct dma_mapping_ops *dma_ops;
-EXPORT_SYMBOL(dma_ops);
-
 static unsigned long dma_reserve __initdata;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);

arch/x86/mm/ioremap.c

@@ -134,7 +134,7 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 
 	if (!phys_addr_valid(phys_addr)) {
 		printk(KERN_WARNING "ioremap: invalid physical address %llx\n",
-		       phys_addr);
+		       (unsigned long long)phys_addr);
 		WARN_ON_ONCE(1);
 		return NULL;
 	}
@@ -187,7 +187,8 @@ static void __iomem *__ioremap(resource_size_t phys_addr, unsigned long size,
 		     new_prot_val == _PAGE_CACHE_WB)) {
 			pr_debug(
 		"ioremap error for 0x%llx-0x%llx, requested 0x%lx, got 0x%lx\n",
-				phys_addr, phys_addr + size,
+				(unsigned long long)phys_addr,
+				(unsigned long long)(phys_addr + size),
 				prot_val, new_prot_val);
 			free_memtype(phys_addr, phys_addr + size);
 			return NULL;

arch/x86/mm/k8topology_64.c

@@ -164,7 +164,7 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 	if (!found)
 		return -1;
 
-	memnode_shift = compute_hash_shift(nodes, 8);
+	memnode_shift = compute_hash_shift(nodes, 8, NULL);
 	if (memnode_shift < 0) {
 		printk(KERN_ERR "No NUMA node hash function found. Contact maintainer\n");
 		return -1;

arch/x86/mm/numa_64.c

@@ -60,7 +60,7 @@ unsigned long __initdata nodemap_size;
  * -1 if node overlap or lost ram (shift too big)
  */
 static int __init populate_memnodemap(const struct bootnode *nodes,
-				      int numnodes, int shift)
+				      int numnodes, int shift, int *nodeids)
 {
 	unsigned long addr, end;
 	int i, res = -1;
@@ -76,7 +76,12 @@ static int __init populate_memnodemap(const struct bootnode *nodes,
 		do {
 			if (memnodemap[addr >> shift] != NUMA_NO_NODE)
 				return -1;
-			memnodemap[addr >> shift] = i;
+
+			if (!nodeids)
+				memnodemap[addr >> shift] = i;
+			else
+				memnodemap[addr >> shift] = nodeids[i];
+
 			addr += (1UL << shift);
 		} while (addr < end);
 		res = 1;
@@ -139,7 +144,8 @@ static int __init extract_lsb_from_nodes(const struct bootnode *nodes,
 	return i;
 }
 
-int __init compute_hash_shift(struct bootnode *nodes, int numnodes)
+int __init compute_hash_shift(struct bootnode *nodes, int numnodes,
+			      int *nodeids)
 {
 	int shift;
 
@@ -149,7 +155,7 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes)
 	printk(KERN_DEBUG "NUMA: Using %d for the hash shift.\n",
 		shift);
 
-	if (populate_memnodemap(nodes, numnodes, shift) != 1) {
+	if (populate_memnodemap(nodes, numnodes, shift, nodeids) != 1) {
 		printk(KERN_INFO "Your memory is not aligned you need to "
 		       "rebuild your kernel with a bigger NODEMAPSIZE "
 		       "shift=%d\n", shift);
@@ -462,7 +468,7 @@ done:
 		}
 	}
 out:
-	memnode_shift = compute_hash_shift(nodes, num_nodes);
+	memnode_shift = compute_hash_shift(nodes, num_nodes, NULL);
 	if (memnode_shift < 0) {
 		memnode_shift = 0;
 		printk(KERN_ERR "No NUMA hash function found.  NUMA emulation "

arch/x86/mm/pgtable_32.c

@@ -1,7 +1,3 @@
-/*
- *  linux/arch/i386/mm/pgtable.c
- */
-
 #include <linux/sched.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>

arch/x86/mm/srat_64.c

@@ -32,6 +32,10 @@ static struct bootnode nodes_add[MAX_NUMNODES];
 static int found_add_area __initdata;
 int hotadd_percent __initdata = 0;
 
+static int num_node_memblks __initdata;
+static struct bootnode node_memblk_range[NR_NODE_MEMBLKS] __initdata;
+static int memblk_nodeid[NR_NODE_MEMBLKS] __initdata;
+
 /* Too small nodes confuse the VM badly. Usually they result
    from BIOS bugs. */
 #define NODE_MIN_SIZE (4*1024*1024)
@@ -41,17 +45,17 @@ static __init int setup_node(int pxm)
 	return acpi_map_pxm_to_node(pxm);
 }
 
-static __init int conflicting_nodes(unsigned long start, unsigned long end)
+static __init int conflicting_memblks(unsigned long start, unsigned long end)
 {
 	int i;
-	for_each_node_mask(i, nodes_parsed) {
+	for (i = 0; i < num_node_memblks; i++) {
-		struct bootnode *nd = &nodes[i];
+		struct bootnode *nd = &node_memblk_range[i];
 		if (nd->start == nd->end)
 			continue;
 		if (nd->end > start && nd->start < end)
-			return i;
+			return memblk_nodeid[i];
 		if (nd->end == end && nd->start == start)
-			return i;
+			return memblk_nodeid[i];
 	}
 	return -1;
 }
@@ -258,7 +262,7 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
 		bad_srat();
 		return;
 	}
-	i = conflicting_nodes(start, end);
+	i = conflicting_memblks(start, end);
 	if (i == node) {
 		printk(KERN_WARNING
 		"SRAT: Warning: PXM %d (%lx-%lx) overlaps with itself (%Lx-%Lx)\n",
@@ -283,10 +287,10 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
 			nd->end = end;
 	}
 
-	printk(KERN_INFO "SRAT: Node %u PXM %u %Lx-%Lx\n", node, pxm,
+	printk(KERN_INFO "SRAT: Node %u PXM %u %lx-%lx\n", node, pxm,
-	       nd->start, nd->end);
+	       start, end);
-	e820_register_active_regions(node, nd->start >> PAGE_SHIFT,
+	e820_register_active_regions(node, start >> PAGE_SHIFT,
-						nd->end >> PAGE_SHIFT);
+				     end >> PAGE_SHIFT);
 	push_node_boundaries(node, nd->start >> PAGE_SHIFT,
 						nd->end >> PAGE_SHIFT);
 
@@ -298,6 +302,11 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
 		if ((nd->start | nd->end) == 0)
 			node_clear(node, nodes_parsed);
 	}
+
+	node_memblk_range[num_node_memblks].start = start;
+	node_memblk_range[num_node_memblks].end = end;
+	memblk_nodeid[num_node_memblks] = node;
+	num_node_memblks++;
 }
 
 /* Sanity check to catch more bad SRATs (they are amazingly common).
@@ -368,7 +377,8 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 		return -1;
 	}
 
-	memnode_shift = compute_hash_shift(nodes, MAX_NUMNODES);
+	memnode_shift = compute_hash_shift(node_memblk_range, num_node_memblks,
+					   memblk_nodeid);
 	if (memnode_shift < 0) {
 		printk(KERN_ERR
 		     "SRAT: No NUMA node hash function found. Contact maintainer\n");

arch/x86/vdso/Makefile

@@ -37,7 +37,8 @@ $(obj)/%.so: OBJCOPYFLAGS := -S
 $(obj)/%.so: $(obj)/%.so.dbg FORCE
 	$(call if_changed,objcopy)
 
-CFL := $(PROFILING) -mcmodel=small -fPIC -g0 -O2 -fasynchronous-unwind-tables -m64
+CFL := $(PROFILING) -mcmodel=small -fPIC -O2 -fasynchronous-unwind-tables -m64 \
+       $(filter -g%,$(KBUILD_CFLAGS))
 
 $(vobjs): KBUILD_CFLAGS += $(CFL)
 

arch/x86/video/fbdev.c

@@ -1,5 +1,4 @@
 /*
- * arch/i386/video/fbdev.c - i386 Framebuffer
  *
  * Copyright (C) 2007 Antonino Daplas <adaplas@gmail.com>
  *

include/asm-x86/boot.h

@@ -17,4 +17,12 @@
 				+ (CONFIG_PHYSICAL_ALIGN - 1)) \
 				& ~(CONFIG_PHYSICAL_ALIGN - 1))
 
+#ifdef CONFIG_X86_64
+#define BOOT_HEAP_SIZE	0x7000
+#define BOOT_STACK_SIZE	0x4000
+#else
+#define BOOT_HEAP_SIZE	0x4000
+#define BOOT_STACK_SIZE	0x1000
+#endif
+
 #endif /* _ASM_BOOT_H */

include/asm-x86/dma-mapping.h

@@ -1,5 +1,237 @@
+#ifndef _ASM_DMA_MAPPING_H_
+#define _ASM_DMA_MAPPING_H_
+
+/*
+ * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for
+ * documentation.
+ */
+
+#include <linux/scatterlist.h>
+#include <asm/io.h>
+#include <asm/swiotlb.h>
+
+extern dma_addr_t bad_dma_address;
+extern int iommu_merge;
+extern struct device fallback_dev;
+extern int panic_on_overflow;
+extern int forbid_dac;
+extern int force_iommu;
+
+struct dma_mapping_ops {
+	int             (*mapping_error)(dma_addr_t dma_addr);
+	void*           (*alloc_coherent)(struct device *dev, size_t size,
+				dma_addr_t *dma_handle, gfp_t gfp);
+	void            (*free_coherent)(struct device *dev, size_t size,
+				void *vaddr, dma_addr_t dma_handle);
+	dma_addr_t      (*map_single)(struct device *hwdev, phys_addr_t ptr,
+				size_t size, int direction);
+	/* like map_single, but doesn't check the device mask */
+	dma_addr_t      (*map_simple)(struct device *hwdev, phys_addr_t ptr,
+				size_t size, int direction);
+	void            (*unmap_single)(struct device *dev, dma_addr_t addr,
+				size_t size, int direction);
+	void            (*sync_single_for_cpu)(struct device *hwdev,
+				dma_addr_t dma_handle, size_t size,
+				int direction);
+	void            (*sync_single_for_device)(struct device *hwdev,
+				dma_addr_t dma_handle, size_t size,
+				int direction);
+	void            (*sync_single_range_for_cpu)(struct device *hwdev,
+				dma_addr_t dma_handle, unsigned long offset,
+				size_t size, int direction);
+	void            (*sync_single_range_for_device)(struct device *hwdev,
+				dma_addr_t dma_handle, unsigned long offset,
+				size_t size, int direction);
+	void            (*sync_sg_for_cpu)(struct device *hwdev,
+				struct scatterlist *sg, int nelems,
+				int direction);
+	void            (*sync_sg_for_device)(struct device *hwdev,
+				struct scatterlist *sg, int nelems,
+				int direction);
+	int             (*map_sg)(struct device *hwdev, struct scatterlist *sg,
+				int nents, int direction);
+	void            (*unmap_sg)(struct device *hwdev,
+				struct scatterlist *sg, int nents,
+				int direction);
+	int             (*dma_supported)(struct device *hwdev, u64 mask);
+	int		is_phys;
+};
+
+extern const struct dma_mapping_ops *dma_ops;
+
+static inline int dma_mapping_error(dma_addr_t dma_addr)
+{
+	if (dma_ops->mapping_error)
+		return dma_ops->mapping_error(dma_addr);
+
+	return (dma_addr == bad_dma_address);
+}
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			   dma_addr_t *dma_handle, gfp_t flag);
+
+void dma_free_coherent(struct device *dev, size_t size,
+			 void *vaddr, dma_addr_t dma_handle);
+
+
+extern int dma_supported(struct device *hwdev, u64 mask);
+extern int dma_set_mask(struct device *dev, u64 mask);
+
+static inline dma_addr_t
+dma_map_single(struct device *hwdev, void *ptr, size_t size,
+	       int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	return dma_ops->map_single(hwdev, virt_to_phys(ptr), size, direction);
+}
+
+static inline void
+dma_unmap_single(struct device *dev, dma_addr_t addr, size_t size,
+		 int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->unmap_single)
+		dma_ops->unmap_single(dev, addr, size, direction);
+}
+
+static inline int
+dma_map_sg(struct device *hwdev, struct scatterlist *sg,
+	   int nents, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	return dma_ops->map_sg(hwdev, sg, nents, direction);
+}
+
+static inline void
+dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
+	     int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->unmap_sg)
+		dma_ops->unmap_sg(hwdev, sg, nents, direction);
+}
+
+static inline void
+dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
+			size_t size, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_single_for_cpu)
+		dma_ops->sync_single_for_cpu(hwdev, dma_handle, size,
+					     direction);
+	flush_write_buffers();
+}
+
+static inline void
+dma_sync_single_for_device(struct device *hwdev, dma_addr_t dma_handle,
+			   size_t size, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_single_for_device)
+		dma_ops->sync_single_for_device(hwdev, dma_handle, size,
+						direction);
+	flush_write_buffers();
+}
+
+static inline void
+dma_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
+			      unsigned long offset, size_t size, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_single_range_for_cpu)
+		dma_ops->sync_single_range_for_cpu(hwdev, dma_handle, offset,
+						   size, direction);
+
+	flush_write_buffers();
+}
+
+static inline void
+dma_sync_single_range_for_device(struct device *hwdev, dma_addr_t dma_handle,
+				 unsigned long offset, size_t size,
+				 int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_single_range_for_device)
+		dma_ops->sync_single_range_for_device(hwdev, dma_handle,
+						      offset, size, direction);
+
+	flush_write_buffers();
+}
+
+static inline void
+dma_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
+		    int nelems, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_sg_for_cpu)
+		dma_ops->sync_sg_for_cpu(hwdev, sg, nelems, direction);
+	flush_write_buffers();
+}
+
+static inline void
+dma_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
+		       int nelems, int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	if (dma_ops->sync_sg_for_device)
+		dma_ops->sync_sg_for_device(hwdev, sg, nelems, direction);
+
+	flush_write_buffers();
+}
+
+static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
+				      size_t offset, size_t size,
+				      int direction)
+{
+	BUG_ON(!valid_dma_direction(direction));
+	return dma_ops->map_single(dev, page_to_phys(page)+offset,
+				   size, direction);
+}
+
+static inline void dma_unmap_page(struct device *dev, dma_addr_t addr,
+				  size_t size, int direction)
+{
+	dma_unmap_single(dev, addr, size, direction);
+}
+
+static inline void
+dma_cache_sync(struct device *dev, void *vaddr, size_t size,
+	enum dma_data_direction dir)
+{
+	flush_write_buffers();
+}
+
+static inline int dma_get_cache_alignment(void)
+{
+	/* no easy way to get cache size on all x86, so return the
+	 * maximum possible, to be safe */
+	return boot_cpu_data.x86_clflush_size;
+}
+
+#define dma_is_consistent(d, h)	(1)
+
 #ifdef CONFIG_X86_32
-# include "dma-mapping_32.h"
+#  define ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
-#else
+struct dma_coherent_mem {
-# include "dma-mapping_64.h"
+	void		*virt_base;
+	u32		device_base;
+	int		size;
+	int		flags;
+	unsigned long	*bitmap;
+};
+
+extern int
+dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+			    dma_addr_t device_addr, size_t size, int flags);
+
+extern void
+dma_release_declared_memory(struct device *dev);
+
+extern void *
+dma_mark_declared_memory_occupied(struct device *dev,
+				  dma_addr_t device_addr, size_t size);
+#endif /* CONFIG_X86_32 */
 #endif

include/asm-x86/dma-mapping_32.h

@@ -1,187 +0,0 @@
-#ifndef _ASM_I386_DMA_MAPPING_H
-#define _ASM_I386_DMA_MAPPING_H
-
-#include <linux/mm.h>
-#include <linux/scatterlist.h>
-
-#include <asm/cache.h>
-#include <asm/io.h>
-#include <asm/bug.h>
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
-void *dma_alloc_coherent(struct device *dev, size_t size,
-			   dma_addr_t *dma_handle, gfp_t flag);
-
-void dma_free_coherent(struct device *dev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle);
-
-static inline dma_addr_t
-dma_map_single(struct device *dev, void *ptr, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(size == 0);
-	flush_write_buffers();
-	return virt_to_phys(ptr);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
-		 enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-}
-
-static inline int
-dma_map_sg(struct device *dev, struct scatterlist *sglist, int nents,
-	   enum dma_data_direction direction)
-{
-	struct scatterlist *sg;
-	int i;
-
-	BUG_ON(!valid_dma_direction(direction));
-	WARN_ON(nents == 0 || sglist[0].length == 0);
-
-	for_each_sg(sglist, sg, nents, i) {
-		BUG_ON(!sg_page(sg));
-
-		sg->dma_address = sg_phys(sg);
-	}
-
-	flush_write_buffers();
-	return nents;
-}
-
-static inline dma_addr_t
-dma_map_page(struct device *dev, struct page *page, unsigned long offset,
-	     size_t size, enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	return page_to_phys(page) + offset;
-}
-
-static inline void
-dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
-	       enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-}
-
-
-static inline void
-dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
-	     enum dma_data_direction direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-}
-
-static inline void
-dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
-			enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
-			enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
-			      unsigned long offset, size_t size,
-			      enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size,
-				 enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
-		    enum dma_data_direction direction)
-{
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
-		    enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-static inline int
-dma_mapping_error(dma_addr_t dma_addr)
-{
-	return 0;
-}
-
-extern int forbid_dac;
-
-static inline int
-dma_supported(struct device *dev, u64 mask)
-{
-        /*
-         * we fall back to GFP_DMA when the mask isn't all 1s,
-         * so we can't guarantee allocations that must be
-         * within a tighter range than GFP_DMA..
-         */
-        if(mask < 0x00ffffff)
-                return 0;
-
-	/* Work around chipset bugs */
-	if (forbid_dac > 0 && mask > 0xffffffffULL)
-		return 0;
-
-	return 1;
-}
-
-static inline int
-dma_set_mask(struct device *dev, u64 mask)
-{
-	if(!dev->dma_mask || !dma_supported(dev, mask))
-		return -EIO;
-
-	*dev->dma_mask = mask;
-
-	return 0;
-}
-
-static inline int
-dma_get_cache_alignment(void)
-{
-	/* no easy way to get cache size on all x86, so return the
-	 * maximum possible, to be safe */
-	return (1 << INTERNODE_CACHE_SHIFT);
-}
-
-#define dma_is_consistent(d, h)	(1)
-
-static inline void
-dma_cache_sync(struct device *dev, void *vaddr, size_t size,
-	       enum dma_data_direction direction)
-{
-	flush_write_buffers();
-}
-
-#define ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
-extern int
-dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
-			    dma_addr_t device_addr, size_t size, int flags);
-
-extern void
-dma_release_declared_memory(struct device *dev);
-
-extern void *
-dma_mark_declared_memory_occupied(struct device *dev,
-				  dma_addr_t device_addr, size_t size);
-
-#endif

include/asm-x86/dma-mapping_64.h

@@ -1,202 +0,0 @@
-#ifndef _X8664_DMA_MAPPING_H
-#define _X8664_DMA_MAPPING_H 1
-
-/*
- * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for
- * documentation.
- */
-
-#include <linux/scatterlist.h>
-#include <asm/io.h>
-#include <asm/swiotlb.h>
-
-struct dma_mapping_ops {
-	int             (*mapping_error)(dma_addr_t dma_addr);
-	void*           (*alloc_coherent)(struct device *dev, size_t size,
-                                dma_addr_t *dma_handle, gfp_t gfp);
-	void            (*free_coherent)(struct device *dev, size_t size,
-                                void *vaddr, dma_addr_t dma_handle);
-	dma_addr_t      (*map_single)(struct device *hwdev, void *ptr,
-                                size_t size, int direction);
-	/* like map_single, but doesn't check the device mask */
-	dma_addr_t      (*map_simple)(struct device *hwdev, char *ptr,
-                                size_t size, int direction);
-	void            (*unmap_single)(struct device *dev, dma_addr_t addr,
-		                size_t size, int direction);
-	void            (*sync_single_for_cpu)(struct device *hwdev,
-		                dma_addr_t dma_handle, size_t size,
-				int direction);
-	void            (*sync_single_for_device)(struct device *hwdev,
-                                dma_addr_t dma_handle, size_t size,
-				int direction);
-	void            (*sync_single_range_for_cpu)(struct device *hwdev,
-                                dma_addr_t dma_handle, unsigned long offset,
-		                size_t size, int direction);
-	void            (*sync_single_range_for_device)(struct device *hwdev,
-				dma_addr_t dma_handle, unsigned long offset,
-		                size_t size, int direction);
-	void            (*sync_sg_for_cpu)(struct device *hwdev,
-                                struct scatterlist *sg, int nelems,
-				int direction);
-	void            (*sync_sg_for_device)(struct device *hwdev,
-				struct scatterlist *sg, int nelems,
-				int direction);
-	int             (*map_sg)(struct device *hwdev, struct scatterlist *sg,
-		                int nents, int direction);
-	void            (*unmap_sg)(struct device *hwdev,
-				struct scatterlist *sg, int nents,
-				int direction);
-	int             (*dma_supported)(struct device *hwdev, u64 mask);
-	int		is_phys;
-};
-
-extern dma_addr_t bad_dma_address;
-extern const struct dma_mapping_ops* dma_ops;
-extern int iommu_merge;
-
-static inline int dma_mapping_error(dma_addr_t dma_addr)
-{
-	if (dma_ops->mapping_error)
-		return dma_ops->mapping_error(dma_addr);
-
-	return (dma_addr == bad_dma_address);
-}
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
-#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
-#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
-
-extern void *dma_alloc_coherent(struct device *dev, size_t size,
-				dma_addr_t *dma_handle, gfp_t gfp);
-extern void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
-			      dma_addr_t dma_handle);
-
-static inline dma_addr_t
-dma_map_single(struct device *hwdev, void *ptr, size_t size,
-	       int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	return dma_ops->map_single(hwdev, ptr, size, direction);
-}
-
-static inline void
-dma_unmap_single(struct device *dev, dma_addr_t addr,size_t size,
-		 int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	dma_ops->unmap_single(dev, addr, size, direction);
-}
-
-#define dma_map_page(dev,page,offset,size,dir) \
-	dma_map_single((dev), page_address(page)+(offset), (size), (dir))
-
-#define dma_unmap_page dma_unmap_single
-
-static inline void
-dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
-			size_t size, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_for_cpu)
-		dma_ops->sync_single_for_cpu(hwdev, dma_handle, size,
-					     direction);
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_single_for_device(struct device *hwdev, dma_addr_t dma_handle,
-			   size_t size, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_for_device)
-		dma_ops->sync_single_for_device(hwdev, dma_handle, size,
-						direction);
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
-			      unsigned long offset, size_t size, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_range_for_cpu) {
-		dma_ops->sync_single_range_for_cpu(hwdev, dma_handle, offset, size, direction);
-	}
-
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_single_range_for_device(struct device *hwdev, dma_addr_t dma_handle,
-				 unsigned long offset, size_t size, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_range_for_device)
-		dma_ops->sync_single_range_for_device(hwdev, dma_handle,
-						      offset, size, direction);
-
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
-		    int nelems, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_sg_for_cpu)
-		dma_ops->sync_sg_for_cpu(hwdev, sg, nelems, direction);
-	flush_write_buffers();
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
-		       int nelems, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_sg_for_device) {
-		dma_ops->sync_sg_for_device(hwdev, sg, nelems, direction);
-	}
-
-	flush_write_buffers();
-}
-
-static inline int
-dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents, int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	return dma_ops->map_sg(hwdev, sg, nents, direction);
-}
-
-static inline void
-dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
-	     int direction)
-{
-	BUG_ON(!valid_dma_direction(direction));
-	dma_ops->unmap_sg(hwdev, sg, nents, direction);
-}
-
-extern int dma_supported(struct device *hwdev, u64 mask);
-
-/* same for gart, swiotlb, and nommu */
-static inline int dma_get_cache_alignment(void)
-{
-	return boot_cpu_data.x86_clflush_size;
-}
-
-#define dma_is_consistent(d, h) 1
-
-extern int dma_set_mask(struct device *dev, u64 mask);
-
-static inline void
-dma_cache_sync(struct device *dev, void *vaddr, size_t size,
-	enum dma_data_direction dir)
-{
-	flush_write_buffers();
-}
-
-extern struct device fallback_dev;
-extern int panic_on_overflow;
-
-#endif /* _X8664_DMA_MAPPING_H */

include/asm-x86/e820_32.h

@@ -24,7 +24,7 @@ extern void update_e820(void);
 extern int e820_all_mapped(unsigned long start, unsigned long end,
 			   unsigned type);
 extern int e820_any_mapped(u64 start, u64 end, unsigned type);
-extern void find_max_pfn(void);
+extern void propagate_e820_map(void);
 extern void register_bootmem_low_pages(unsigned long max_low_pfn);
 extern void add_memory_region(unsigned long long start,
 			      unsigned long long size, int type);

include/asm-x86/genapic_32.h

@@ -117,6 +117,7 @@ extern struct genapic *genapic;
 enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC};
 #define get_uv_system_type()		UV_NONE
 #define is_uv_system()			0
+#define uv_wakeup_secondary(a, b)	1
 
 
 #endif

include/asm-x86/i387.h

@@ -21,8 +21,9 @@
 
 extern void fpu_init(void);
 extern void mxcsr_feature_mask_init(void);
-extern void init_fpu(struct task_struct *child);
+extern int init_fpu(struct task_struct *child);
 extern asmlinkage void math_state_restore(void);
+extern void init_thread_xstate(void);
 
 extern user_regset_active_fn fpregs_active, xfpregs_active;
 extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
@@ -117,24 +118,22 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 	/* Using "fxsaveq %0" would be the ideal choice, but is only supported
 	   starting with gas 2.16. */
 	__asm__ __volatile__("fxsaveq %0"
-			     : "=m" (tsk->thread.i387.fxsave));
+			     : "=m" (tsk->thread.xstate->fxsave));
 #elif 0
 	/* Using, as a workaround, the properly prefixed form below isn't
 	   accepted by any binutils version so far released, complaining that
 	   the same type of prefix is used twice if an extended register is
 	   needed for addressing (fix submitted to mainline 2005-11-21). */
 	__asm__ __volatile__("rex64/fxsave %0"
-			     : "=m" (tsk->thread.i387.fxsave));
+			     : "=m" (tsk->thread.xstate->fxsave));
 #else
 	/* This, however, we can work around by forcing the compiler to select
 	   an addressing mode that doesn't require extended registers. */
-	__asm__ __volatile__("rex64/fxsave %P2(%1)"
+	__asm__ __volatile__("rex64/fxsave (%1)"
-			     : "=m" (tsk->thread.i387.fxsave)
+			     : "=m" (tsk->thread.xstate->fxsave)
-			     : "cdaSDb" (tsk),
+			     : "cdaSDb" (&tsk->thread.xstate->fxsave));
-				"i" (offsetof(__typeof__(*tsk),
-					      thread.i387.fxsave)));
 #endif
-	clear_fpu_state(&tsk->thread.i387.fxsave);
+	clear_fpu_state(&tsk->thread.xstate->fxsave);
 	task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
@@ -148,7 +147,7 @@ static inline int save_i387(struct _fpstate __user *buf)
 	int err = 0;
 
 	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
-			sizeof(tsk->thread.i387.fxsave));
+			sizeof(tsk->thread.xstate->fxsave));
 
 	if ((unsigned long)buf % 16)
 		printk("save_i387: bad fpstate %p\n", buf);
@@ -164,7 +163,7 @@ static inline int save_i387(struct _fpstate __user *buf)
 		task_thread_info(tsk)->status &= ~TS_USEDFPU;
 		stts();
 	} else {
-		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,
+		if (__copy_to_user(buf, &tsk->thread.xstate->fxsave,
 				   sizeof(struct i387_fxsave_struct)))
 			return -1;
 	}
@@ -201,7 +200,7 @@ static inline void restore_fpu(struct task_struct *tsk)
 		"nop ; frstor %1",
 		"fxrstor %1",
 		X86_FEATURE_FXSR,
-		"m" ((tsk)->thread.i387.fxsave));
+		"m" (tsk->thread.xstate->fxsave));
 }
 
 /* We need a safe address that is cheap to find and that is already
@@ -225,8 +224,8 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 		"fxsave %[fx]\n"
 		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",
 		X86_FEATURE_FXSR,
-		[fx] "m" (tsk->thread.i387.fxsave),
+		[fx] "m" (tsk->thread.xstate->fxsave),
-		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");
+		[fsw] "m" (tsk->thread.xstate->fxsave.swd) : "memory");
 	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
 	   is pending.  Clear the x87 state here by setting it to fixed
 	   values. safe_address is a random variable that should be in L1 */
@@ -327,25 +326,25 @@ static inline void clear_fpu(struct task_struct *tsk)
 static inline unsigned short get_fpu_cwd(struct task_struct *tsk)
 {
 	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.cwd;
+		return tsk->thread.xstate->fxsave.cwd;
 	} else {
-		return (unsigned short)tsk->thread.i387.fsave.cwd;
+		return (unsigned short)tsk->thread.xstate->fsave.cwd;
 	}
 }
 
 static inline unsigned short get_fpu_swd(struct task_struct *tsk)
 {
 	if (cpu_has_fxsr) {
-		return tsk->thread.i387.fxsave.swd;
+		return tsk->thread.xstate->fxsave.swd;
 	} else {
-		return (unsigned short)tsk->thread.i387.fsave.swd;
+		return (unsigned short)tsk->thread.xstate->fsave.swd;
 	}
 }
 
 static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)
 {
 	if (cpu_has_xmm) {
-		return tsk->thread.i387.fxsave.mxcsr;
+		return tsk->thread.xstate->fxsave.mxcsr;
 	} else {
 		return MXCSR_DEFAULT;
 	}

include/asm-x86/numa_64.h

@@ -9,7 +9,8 @@ struct bootnode {
 	u64 end;
 };
 
-extern int compute_hash_shift(struct bootnode *nodes, int numnodes);
+extern int compute_hash_shift(struct bootnode *nodes, int numblks,
+			      int *nodeids);
 
 #define ZONE_ALIGN (1UL << (MAX_ORDER+PAGE_SHIFT))
 

include/asm-x86/pci_64.h

@@ -22,6 +22,7 @@ extern int (*pci_config_read)(int seg, int bus, int dev, int fn,
 extern int (*pci_config_write)(int seg, int bus, int dev, int fn,
 			       int reg, int len, u32 value);
 
+extern void dma32_reserve_bootmem(void);
 extern void pci_iommu_alloc(void);
 
 /* The PCI address space does equal the physical memory

include/asm-x86/processor.h

@@ -354,7 +354,7 @@ struct i387_soft_struct {
 	u32			entry_eip;
 };
 
-union i387_union {
+union thread_xstate {
 	struct i387_fsave_struct	fsave;
 	struct i387_fxsave_struct	fxsave;
 	struct i387_soft_struct		soft;
@@ -365,6 +365,9 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 #endif
 
 extern void print_cpu_info(struct cpuinfo_x86 *);
+extern unsigned int xstate_size;
+extern void free_thread_xstate(struct task_struct *);
+extern struct kmem_cache *task_xstate_cachep;
 extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
@@ -397,8 +400,8 @@ struct thread_struct {
 	unsigned long		cr2;
 	unsigned long		trap_no;
 	unsigned long		error_code;
-	/* Floating point info: */
+	/* floating point and extended processor state */
-	union i387_union	i387 __attribute__((aligned(16)));;
+	union thread_xstate	*xstate;
 #ifdef CONFIG_X86_32
 	/* Virtual 86 mode info */
 	struct vm86_struct __user *vm86_info;
@@ -918,4 +921,11 @@ extern void start_thread(struct pt_regs *regs, unsigned long new_ip,
 
 #define KSTK_EIP(task)		(task_pt_regs(task)->ip)
 
+/* Get/set a process' ability to use the timestamp counter instruction */
+#define GET_TSC_CTL(adr)	get_tsc_mode((adr))
+#define SET_TSC_CTL(val)	set_tsc_mode((val))
+
+extern int get_tsc_mode(unsigned long adr);
+extern int set_tsc_mode(unsigned int val);
+
 #endif

include/asm-x86/scatterlist.h

@@ -11,9 +11,7 @@ struct scatterlist {
 	unsigned int	offset;
 	unsigned int	length;
 	dma_addr_t	dma_address;
-#ifdef CONFIG_X86_64
 	unsigned int	dma_length;
-#endif
 };
 
 #define ARCH_HAS_SG_CHAIN

include/asm-x86/thread_info.h

@@ -1,5 +1,14 @@
+#ifndef _ASM_X86_THREAD_INFO_H
 #ifdef CONFIG_X86_32
 # include "thread_info_32.h"
 #else
 # include "thread_info_64.h"
 #endif
+
+#ifndef __ASSEMBLY__
+extern void arch_task_cache_init(void);
+extern void free_thread_info(struct thread_info *ti);
+extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src);
+#define arch_task_cache_init arch_task_cache_init
+#endif
+#endif /* _ASM_X86_THREAD_INFO_H */

include/asm-x86/thread_info_32.h

@@ -102,8 +102,6 @@ static inline struct thread_info *current_thread_info(void)
 	__get_free_pages(GFP_KERNEL, get_order(THREAD_SIZE)))
 #endif
 
-#define free_thread_info(info)	free_pages((unsigned long)(info), get_order(THREAD_SIZE))
-
 #else /* !__ASSEMBLY__ */
 
 /* how to get the thread information struct from ASM */

include/asm-x86/thread_info_64.h

@@ -85,8 +85,6 @@ static inline struct thread_info *stack_thread_info(void)
 #define alloc_thread_info(tsk)						\
 	((struct thread_info *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER))
 
-#define free_thread_info(ti) free_pages((unsigned long) (ti), THREAD_ORDER)
-
 #else /* !__ASSEMBLY__ */
 
 /* how to get the thread information struct from ASM */
@@ -126,6 +124,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_DEBUGCTLMSR		25	/* uses thread_struct.debugctlmsr */
 #define TIF_DS_AREA_MSR		26      /* uses thread_struct.ds_area_msr */
 #define TIF_BTS_TRACE_TS	27      /* record scheduling event timestamps */
+#define TIF_NOTSC		28	/* TSC is not accessible in userland */
 
 #define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
 #define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
@@ -147,6 +146,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_DEBUGCTLMSR	(1 << TIF_DEBUGCTLMSR)
 #define _TIF_DS_AREA_MSR	(1 << TIF_DS_AREA_MSR)
 #define _TIF_BTS_TRACE_TS	(1 << TIF_BTS_TRACE_TS)
+#define _TIF_NOTSC		(1 << TIF_NOTSC)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK							\
@@ -160,7 +160,7 @@ static inline struct thread_info *stack_thread_info(void)
 
 /* flags to check in __switch_to() */
 #define _TIF_WORK_CTXSW							\
-	(_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_BTS_TRACE_TS)
+	(_TIF_IO_BITMAP|_TIF_DEBUGCTLMSR|_TIF_DS_AREA_MSR|_TIF_BTS_TRACE_TS|_TIF_NOTSC)
 #define _TIF_WORK_CTXSW_PREV _TIF_WORK_CTXSW
 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW|_TIF_DEBUG)