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
-	pushl %eax	# end of the image as third argument
+	leal boot_heap(%ebx), %eax
+	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:
-	.fill 4096, 1, 0
-stack_end:
+boot_heap:
+	.fill BOOT_HEAP_SIZE, 1, 0
+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(%rbx), %r9
-	movq	$_end /* - $startup_32 */, %rcx
+	leaq	_end_before_pgt(%rip), %r8
+	leaq	_end_before_pgt(%rbx), %r9
+	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 */
-	.balign 4
-user_stack:
-	.fill 4096,4,0
-user_stack_end:
+
+.bss
+/* Stack and heap for uncompression */
+.balign 4
+boot_heap:
+	.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
  */