Merge branch 'for-linus' of git://one.firstfloor.org/home/andi/git/linux-2.6

* 'for-linus' of git://one.firstfloor.org/home/andi/git/linux-2.6:
  [PATCH] Use early clobber in semaphores
  [PATCH] Define vsyscall cache as blob to make clearer that user space shouldn't use it
  [PATCH] Re-positioning the bss segment
  [PATCH] Use ARRAY_SIZE in setup.c
  [PATCH] i386: replace intermediate array-size definitions with ARRAY_SIZE()
  [PATCH] x86: Clean up x86 NMI sysctls
  [PATCH] Refactor some duplicated code in mpparse.c
  [PATCH] Document iommu=panic
  [PATCH] Fix broken indentation in iommu_setup
  [PATCH] Allow disabling DAC using command line options
  [PATCH] Add proper sparse __user casts to __copy_to_user_inatomic
  [PATCH] i386: Update defconfig
  [PATCH] Update defconfig

Documentation/x86_64/boot-options.txt

@@ -199,6 +199,11 @@ IOMMU
    allowed  overwrite iommu off workarounds for specific chipsets.
    soft	 Use software bounce buffering (default for Intel machines)
    noaperture Don't touch the aperture for AGP.
+   allowdac Allow DMA >4GB
+	    When off all DMA over >4GB is forced through an IOMMU or bounce
+	    buffering.
+   nodac    Forbid DMA >4GB
+   panic    Always panic when IOMMU overflows
 
   swiotlb=pages[,force]
 

arch/i386/defconfig

@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-git5
+# Linux kernel version: 2.6.18-git7
-# Tue Sep 26 09:30:47 2006
+# Wed Sep 27 21:53:10 2006
 #
 CONFIG_X86_32=y
 CONFIG_GENERIC_TIME=y
@@ -210,6 +210,7 @@ CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y
 CONFIG_PM=y
 CONFIG_PM_LEGACY=y
 # CONFIG_PM_DEBUG is not set
+CONFIG_PM_SYSFS_DEPRECATED=y
 
 #
 # ACPI (Advanced Configuration and Power Interface) Support
@@ -292,6 +293,7 @@ CONFIG_PCI_DIRECT=y
 CONFIG_PCI_MMCONFIG=y
 # CONFIG_PCIEPORTBUS is not set
 CONFIG_PCI_MSI=y
+# CONFIG_PCI_MULTITHREAD_PROBE is not set
 # CONFIG_PCI_DEBUG is not set
 CONFIG_ISA_DMA_API=y
 # CONFIG_ISA is not set
@@ -1427,6 +1429,7 @@ CONFIG_KPROBES=y
 #
 CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_UNUSED_SYMBOLS=y
 CONFIG_DEBUG_KERNEL=y

arch/i386/kernel/nmi.c

@@ -31,6 +31,9 @@
 
 #include "mach_traps.h"
 
+int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for

arch/i386/kernel/setup.c

@@ -209,9 +209,6 @@ static struct resource adapter_rom_resources[] = { {
 	.flags	= IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
 } };
 
-#define ADAPTER_ROM_RESOURCES \
-	(sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
-
 static struct resource video_rom_resource = {
 	.name 	= "Video ROM",
 	.start	= 0xc0000,
@@ -273,9 +270,6 @@ static struct resource standard_io_resources[] = { {
 	.flags	= IORESOURCE_BUSY | IORESOURCE_IO
 } };
 
-#define STANDARD_IO_RESOURCES \
-	(sizeof standard_io_resources / sizeof standard_io_resources[0])
-
 #define romsignature(x) (*(unsigned short *)(x) == 0xaa55)
 
 static int __init romchecksum(unsigned char *rom, unsigned long length)
@@ -332,7 +326,7 @@ static void __init probe_roms(void)
 	}
 
 	/* check for adapter roms on 2k boundaries */
-	for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
+	for (i = 0; i < ARRAY_SIZE(adapter_rom_resources) && start < upper; start += 2048) {
 		rom = isa_bus_to_virt(start);
 		if (!romsignature(rom))
 			continue;
@@ -1272,7 +1266,7 @@ static int __init request_standard_resources(void)
 	request_resource(&iomem_resource, &video_ram_resource);
 
 	/* request I/O space for devices used on all i[345]86 PCs */
-	for (i = 0; i < STANDARD_IO_RESOURCES; i++)
+	for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
 		request_resource(&ioport_resource, &standard_io_resources[i]);
 	return 0;
 }

arch/i386/kernel/traps.c

@@ -57,6 +57,8 @@
 
 #include "mach_traps.h"
 
+int panic_on_unrecovered_nmi;
+
 asmlinkage int system_call(void);
 
 struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },

arch/x86_64/defconfig

@@ -1,11 +1,12 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-git5
+# Linux kernel version: 2.6.18-git7
-# Tue Sep 26 09:30:47 2006
+# Wed Sep 27 21:53:10 2006
 #
 CONFIG_X86_64=y
 CONFIG_64BIT=y
 CONFIG_X86=y
+CONFIG_ZONE_DMA32=y
 CONFIG_LOCKDEP_SUPPORT=y
 CONFIG_STACKTRACE_SUPPORT=y
 CONFIG_SEMAPHORE_SLEEPERS=y
@@ -179,6 +180,7 @@ CONFIG_GENERIC_PENDING_IRQ=y
 CONFIG_PM=y
 # CONFIG_PM_LEGACY is not set
 # CONFIG_PM_DEBUG is not set
+# CONFIG_PM_SYSFS_DEPRECATED is not set
 CONFIG_SOFTWARE_SUSPEND=y
 CONFIG_PM_STD_PARTITION=""
 CONFIG_SUSPEND_SMP=y
@@ -251,6 +253,7 @@ CONFIG_PCI_DIRECT=y
 CONFIG_PCI_MMCONFIG=y
 CONFIG_PCIEPORTBUS=y
 CONFIG_PCI_MSI=y
+# CONFIG_PCI_MULTITHREAD_PROBE is not set
 # CONFIG_PCI_DEBUG is not set
 
 #
@@ -1458,6 +1461,7 @@ CONFIG_KPROBES=y
 #
 CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
 CONFIG_MAGIC_SYSRQ=y
 CONFIG_UNUSED_SYMBOLS=y
 CONFIG_DEBUG_KERNEL=y

arch/x86_64/kernel/mpparse.c

@@ -152,6 +152,21 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	}
 }
 
+static int bad_ioapic(unsigned long address)
+{
+	if (nr_ioapics >= MAX_IO_APICS) {
+		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
+			"(found %d)\n", MAX_IO_APICS, nr_ioapics);
+		panic("Recompile kernel with bigger MAX_IO_APICS!\n");
+	}
+	if (!address) {
+		printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
+			" found in table, skipping!\n");
+		return 1;
+	}
+	return 0;
+}
+
 static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
 {
 	if (!(m->mpc_flags & MPC_APIC_USABLE))
@@ -159,16 +174,10 @@ static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
 
 	printk("I/O APIC #%d at 0x%X.\n",
 		m->mpc_apicid, m->mpc_apicaddr);
-	if (nr_ioapics >= MAX_IO_APICS) {
+
-		printk(KERN_ERR "Max # of I/O APICs (%d) exceeded (found %d).\n",
+	if (bad_ioapic(m->mpc_apicaddr))
-			MAX_IO_APICS, nr_ioapics);
-		panic("Recompile kernel with bigger MAX_IO_APICS!.\n");
-	}
-	if (!m->mpc_apicaddr) {
-		printk(KERN_ERR "WARNING: bogus zero I/O APIC address"
-			" found in MP table, skipping!\n");
 		return;
-	}
+
 	mp_ioapics[nr_ioapics] = *m;
 	nr_ioapics++;
 }
@@ -647,16 +656,8 @@ void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 {
 	int idx = 0;
 
-	if (nr_ioapics >= MAX_IO_APICS) {
+	if (bad_ioapic(address))
-		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
-			"(found %d)\n", MAX_IO_APICS, nr_ioapics);
-		panic("Recompile kernel with bigger MAX_IO_APICS!\n");
-	}
-	if (!address) {
-		printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
-			" found in MADT table, skipping!\n");
 		return;
-	}
 
 	idx = nr_ioapics++;
 

arch/x86_64/kernel/nmi.c

@@ -28,6 +28,10 @@
 #include <asm/mce.h>
 #include <asm/intel_arch_perfmon.h>
 
+int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+int panic_on_unrecovered_nmi;
+
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for

arch/x86_64/kernel/pci-dma.c

@@ -170,8 +170,20 @@ void dma_free_coherent(struct device *dev, size_t 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);
 
@@ -231,6 +243,9 @@ EXPORT_SYMBOL(dma_set_mask);
    allowed  overwrite iommu off workarounds for specific chipsets.
    soft	 Use software bounce buffering (default for Intel machines)
    noaperture Don't touch the aperture for AGP.
+   allowdac Allow DMA >4GB
+   nodac    Forbid DMA >4GB
+   panic    Force panic when IOMMU overflows
 */
 __init int iommu_setup(char *p)
 {
@@ -267,6 +282,10 @@ __init int iommu_setup(char *p)
 			iommu_merge = 0;
 		if (!strncmp(p, "forcesac",8))
 			iommu_sac_force = 1;
+		if (!strncmp(p, "allowdac", 8))
+			forbid_dac = 0;
+		if (!strncmp(p, "nodac", 5))
+			forbid_dac = -1;
 
 #ifdef CONFIG_SWIOTLB
 		if (!strncmp(p, "soft",4))

arch/x86_64/kernel/setup.c

@@ -123,9 +123,6 @@ struct resource standard_io_resources[] = {
 		.flags = IORESOURCE_BUSY | IORESOURCE_IO }
 };
 
-#define STANDARD_IO_RESOURCES \
-	(sizeof standard_io_resources / sizeof standard_io_resources[0])
-
 #define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)
 
 struct resource data_resource = {
@@ -172,9 +169,6 @@ static struct resource adapter_rom_resources[] = {
 		.flags = IORESOURCE_ROM }
 };
 
-#define ADAPTER_ROM_RESOURCES \
-	(sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
-
 static struct resource video_rom_resource = {
 	.name = "Video ROM",
 	.start = 0xc0000,
@@ -245,7 +239,8 @@ static void __init probe_roms(void)
 	}
 
 	/* check for adapter roms on 2k boundaries */
-	for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
+	for (i = 0; i < ARRAY_SIZE(adapter_rom_resources) && start < upper;
+	     start += 2048) {
 		rom = isa_bus_to_virt(start);
 		if (!romsignature(rom))
 			continue;
@@ -537,7 +532,7 @@ void __init setup_arch(char **cmdline_p)
 	{
 	unsigned i;
 	/* request I/O space for devices used on all i[345]86 PCs */
-	for (i = 0; i < STANDARD_IO_RESOURCES; i++)
+	for (i = 0; i < ARRAY_SIZE(standard_io_resources); i++)
 		request_resource(&ioport_resource, &standard_io_resources[i]);
 	}
 

arch/x86_64/kernel/vmlinux.lds.S

@@ -67,13 +67,6 @@ SECTIONS
 
   _edata = .;			/* End of data section */
 
-  __bss_start = .;		/* BSS */
-  .bss : AT(ADDR(.bss) - LOAD_OFFSET) {
-	*(.bss.page_aligned)	
-	*(.bss)
-	}
-  __bss_stop = .;
-
   . = ALIGN(PAGE_SIZE);
   . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
   .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) {
@@ -229,6 +222,13 @@ SECTIONS
   . = ALIGN(4096);
   __nosave_end = .;
 
+  __bss_start = .;		/* BSS */
+  .bss : AT(ADDR(.bss) - LOAD_OFFSET) {
+	*(.bss.page_aligned)
+	*(.bss)
+	}
+  __bss_stop = .;
+
   _end = . ;
 
   /* Sections to be discarded */

arch/x86_64/kernel/vsyscall.c

@@ -155,8 +155,8 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 	   We do this here because otherwise user space would do it on
 	   its own in a likely inferior way (no access to jiffies).
 	   If you don't like it pass NULL. */
-	if (tcache && tcache->t0 == (j = __jiffies)) {
+	if (tcache && tcache->blob[0] == (j = __jiffies)) {
-		p = tcache->t1;
+		p = tcache->blob[1];
 	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
 		/* Load per CPU data from RDTSCP */
 		rdtscp(dummy, dummy, p);
@@ -165,8 +165,8 @@ vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
 	}
 	if (tcache) {
-		tcache->t0 = j;
+		tcache->blob[0] = j;
-		tcache->t1 = p;
+		tcache->blob[1] = p;
 	}
 	if (cpu)
 		*cpu = p & 0xfff;

include/asm-i386/nmi.h

@@ -36,4 +36,10 @@ extern unsigned int nmi_watchdog;
 #define NMI_LOCAL_APIC	2
 #define NMI_INVALID	3
 
+struct ctl_table;
+struct file;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
+extern int unknown_nmi_panic;
+
 #endif /* ASM_NMI_H */

include/asm-x86_64/nmi.h

@@ -70,4 +70,11 @@ extern unsigned int nmi_watchdog;
 #define NMI_LOCAL_APIC	2
 #define NMI_INVALID	3
 
+struct ctl_table;
+struct file;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
+
+extern int unknown_nmi_panic;
+
 #endif /* ASM_NMI_H */

include/asm-x86_64/semaphore.h

@@ -132,7 +132,7 @@ static inline int down_interruptible(struct semaphore * sem)
 		"jns 2f\n\t"
 		"call __down_failed_interruptible\n"
 		"2:\n"
-		:"=a" (result), "=m" (sem->count)
+		:"=&a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory");
 	return result;
@@ -153,7 +153,7 @@ static inline int down_trylock(struct semaphore * sem)
 		"jns 2f\n\t"
 		"call __down_failed_trylock\n\t"
 		"2:\n"
-		:"=a" (result), "=m" (sem->count)
+		:"=&a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory","cc");
 	return result;

include/asm-x86_64/uaccess.h

@@ -361,6 +361,11 @@ __must_check unsigned long clear_user(void __user *mem, unsigned long len);
 __must_check unsigned long __clear_user(void __user *mem, unsigned long len);
 
 __must_check long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
-#define __copy_to_user_inatomic copy_user_generic
+
+static __must_check __always_inline int
+__copy_to_user_inatomic(void __user *dst, const void *src, unsigned size)
+{
+	return copy_user_generic((__force void *)dst, src, size);
+}
 
 #endif /* __X86_64_UACCESS_H */

include/linux/getcpu.h

@@ -1,16 +1,18 @@
 #ifndef _LINUX_GETCPU_H
 #define _LINUX_GETCPU_H 1
 
-/* Cache for getcpu() to speed it up. Results might be upto a jiffie
+/* Cache for getcpu() to speed it up. Results might be a short time
    out of date, but will be faster.
+
    User programs should not refer to the contents of this structure.
-   It is only a cache for vgetcpu(). It might change in future kernels.
+   I repeat they should not refer to it. If they do they will break
+   in future kernels.
+
+   It is only a private cache for vgetcpu(). It will change in future kernels.
    The user program must store this information per thread (__thread)
    If you want 100% accurate information pass NULL instead. */
 struct getcpu_cache {
-	unsigned long t0;
+	unsigned long blob[128 / sizeof(long)];
-	unsigned long t1;
-	unsigned long res[4];
 };
 
 #endif

kernel/panic.c

@@ -21,7 +21,6 @@
 #include <linux/debug_locks.h>
 
 int panic_on_oops;
-int panic_on_unrecovered_nmi;
 int tainted;
 static int pause_on_oops;
 static int pause_on_oops_flag;

kernel/sys.c

@@ -2083,12 +2083,12 @@ asmlinkage long sys_getcpu(unsigned __user *cpup, unsigned __user *nodep,
 		 * padding
 		 */
 		unsigned long t0, t1;
-		get_user(t0, &cache->t0);
+		get_user(t0, &cache->blob[0]);
-		get_user(t1, &cache->t1);
+		get_user(t1, &cache->blob[1]);
 		t0++;
 		t1++;
-		put_user(t0, &cache->t0);
+		put_user(t0, &cache->blob[0]);
-		put_user(t1, &cache->t1);
+		put_user(t1, &cache->blob[1]);
 	}
 	return err ? -EFAULT : 0;
 }

kernel/sysctl.c

@@ -52,6 +52,10 @@
 extern int proc_nr_files(ctl_table *table, int write, struct file *filp,
                      void __user *buffer, size_t *lenp, loff_t *ppos);
 
+#ifdef CONFIG_X86
+#include <asm/nmi.h>
+#endif
+
 #if defined(CONFIG_SYSCTL)
 
 /* External variables not in a header file. */
@@ -74,13 +78,6 @@ extern int sysctl_drop_caches;
 extern int percpu_pagelist_fraction;
 extern int compat_log;
 
-#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
-int unknown_nmi_panic;
-int nmi_watchdog_enabled;
-extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
-			void __user *, size_t *, loff_t *);
-#endif
-
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
 static int maxolduid = 65535;
 static int minolduid;