Merge branch 'master' into gfs2

8 files changed, 103 insertions, 93 deletions

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 20e88f4b564..b8d53dfa993 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/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",
-			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");
+
+	if (bad_ioapic(m->mpc_apicaddr))
 		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) {
-		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");
+	if (bad_ioapic(address))
 		return;
-	}
 
 	idx = nr_ioapics++;
 
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 4d6fb047952..7af9cb3e2d9 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/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
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 4dcb671bd19..f8d857453f8 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/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);
 
@@ -231,57 +243,64 @@ 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)
 {
-    iommu_merge = 1;
+	iommu_merge = 1;
 
 	if (!p)
 		return -EINVAL;
 
-    while (*p) {
-	    if (!strncmp(p,"off",3))
-		    no_iommu = 1;
-	    /* gart_parse_options has more force support */
-	    if (!strncmp(p,"force",5))
-		    force_iommu = 1;
-	    if (!strncmp(p,"noforce",7)) {
-		    iommu_merge = 0;
-		    force_iommu = 0;
-	    }
-
-	    if (!strncmp(p, "biomerge",8)) {
-		    iommu_bio_merge = 4096;
-		    iommu_merge = 1;
-		    force_iommu = 1;
-	    }
-	    if (!strncmp(p, "panic",5))
-		    panic_on_overflow = 1;
-	    if (!strncmp(p, "nopanic",7))
-		    panic_on_overflow = 0;
-	    if (!strncmp(p, "merge",5)) {
-		    iommu_merge = 1;
-		    force_iommu = 1;
-	    }
-	    if (!strncmp(p, "nomerge",7))
-		    iommu_merge = 0;
-	    if (!strncmp(p, "forcesac",8))
-		    iommu_sac_force = 1;
+	while (*p) {
+		if (!strncmp(p,"off",3))
+			no_iommu = 1;
+		/* gart_parse_options has more force support */
+		if (!strncmp(p,"force",5))
+			force_iommu = 1;
+		if (!strncmp(p,"noforce",7)) {
+			iommu_merge = 0;
+			force_iommu = 0;
+		}
+
+		if (!strncmp(p, "biomerge",8)) {
+			iommu_bio_merge = 4096;
+			iommu_merge = 1;
+			force_iommu = 1;
+		}
+		if (!strncmp(p, "panic",5))
+			panic_on_overflow = 1;
+		if (!strncmp(p, "nopanic",7))
+			panic_on_overflow = 0;
+		if (!strncmp(p, "merge",5)) {
+			iommu_merge = 1;
+			force_iommu = 1;
+		}
+		if (!strncmp(p, "nomerge",7))
+			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))
-		    swiotlb = 1;
+		if (!strncmp(p, "soft",4))
+			swiotlb = 1;
 #endif
 
 #ifdef CONFIG_IOMMU
-	    gart_parse_options(p);
+		gart_parse_options(p);
 #endif
 
-	    p += strcspn(p, ",");
-	    if (*p == ',')
-		    ++p;
-    }
-    return 0;
+		p += strcspn(p, ",");
+		if (*p == ',')
+			++p;
+	}
+	return 0;
 }
 early_param("iommu", iommu_setup);
 
diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c
index 6a55f87ba97..697f0aa794b 100644
--- a/arch/x86_64/kernel/pci-swiotlb.c
+++ b/arch/x86_64/kernel/pci-swiotlb.c
@@ -3,7 +3,8 @@
 #include <linux/pci.h>
 #include <linux/cache.h>
 #include <linux/module.h>
-#include <asm/dma-mapping.h>
+#include <linux/dma-mapping.h>
+
 #include <asm/proto.h>
 #include <asm/swiotlb.h>
 #include <asm/dma.h>
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0b00bb2ea57..fc944b5e8f4 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/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]);
 	}
 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 1c255ee76e7..557e92af7be 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -77,7 +77,6 @@ unsigned long long monotonic_base;
 struct vxtime_data __vxtime __section_vxtime;	/* for vsyscalls */
 
 volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
-unsigned long __wall_jiffies __section_wall_jiffies = INITIAL_JIFFIES;
 struct timespec __xtime __section_xtime;
 struct timezone __sys_tz __section_sys_tz;
 
@@ -119,7 +118,7 @@ unsigned int (*do_gettimeoffset)(void) = do_gettimeoffset_tsc;
 
 void do_gettimeofday(struct timeval *tv)
 {
-	unsigned long seq, t;
+	unsigned long seq;
  	unsigned int sec, usec;
 
 	do {
@@ -136,10 +135,7 @@ void do_gettimeofday(struct timeval *tv)
 		   be found. Note when you fix it here you need to do the same
 		   in arch/x86_64/kernel/vsyscall.c and export all needed
 		   variables in vmlinux.lds. -AK */ 
-
-		t = (jiffies - wall_jiffies) * USEC_PER_TICK +
-			do_gettimeoffset();
-		usec += t;
+		usec += do_gettimeoffset();
 
 	} while (read_seqretry(&xtime_lock, seq));
 
@@ -165,8 +161,7 @@ int do_settimeofday(struct timespec *tv)
 
 	write_seqlock_irq(&xtime_lock);
 
-	nsec -= do_gettimeoffset() * NSEC_PER_USEC +
-		(jiffies - wall_jiffies) * NSEC_PER_TICK;
+	nsec -= do_gettimeoffset() * NSEC_PER_USEC;
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
@@ -415,16 +410,16 @@ void main_timer_handler(struct pt_regs *regs)
 				(((long) offset << US_SCALE) / vxtime.tsc_quot) - 1;
 	}
 
-	if (lost > 0) {
+	if (lost > 0)
 		handle_lost_ticks(lost, regs);
-		jiffies += lost;
-	}
+	else
+		lost = 0;
 
 /*
  * Do the timer stuff.
  */
 
-	do_timer(regs);
+	do_timer(lost + 1);
 #ifndef CONFIG_SMP
 	update_process_times(user_mode(regs));
 #endif
@@ -1071,7 +1066,6 @@ static int timer_resume(struct sys_device *dev)
 		vxtime.last_tsc = get_cycles_sync();
 	write_sequnlock_irqrestore(&xtime_lock,flags);
 	jiffies += sleep_length;
-	wall_jiffies += sleep_length;
 	monotonic_base += sleep_length * (NSEC_PER_SEC/HZ);
 	touch_softlockup_watchdog();
 	return 0;
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index d0564f1bcb0..b9df2ab6529 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/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) {
@@ -108,9 +101,6 @@ SECTIONS
   .vgetcpu_mode : AT(VLOAD(.vgetcpu_mode)) { *(.vgetcpu_mode) }
   vgetcpu_mode = VVIRT(.vgetcpu_mode);
 
-  .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) }
-  wall_jiffies = VVIRT(.wall_jiffies);
-
   .sys_tz : AT(VLOAD(.sys_tz)) { *(.sys_tz) }
   sys_tz = VVIRT(.sys_tz);
 
@@ -229,6 +219,13 @@ 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 */
diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index ac48c3857dd..a98b460af6a 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -66,8 +66,7 @@ static __always_inline void do_vgettimeofday(struct timeval * tv)
 		sequence = read_seqbegin(&__xtime_lock);
 		
 		sec = __xtime.tv_sec;
-		usec = (__xtime.tv_nsec / 1000) +
-			(__jiffies - __wall_jiffies) * (1000000 / HZ);
+		usec = __xtime.tv_nsec / 1000;
 
 		if (__vxtime.mode != VXTIME_HPET) {
 			t = get_cycles_sync();
@@ -155,8 +154,8 @@ 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)) {
-		p = tcache->t1;
+	if (tcache && tcache->blob[0] == (j = __jiffies)) {
+		p = tcache->blob[1];
 	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
 		/* Load per CPU data from RDTSCP */
 		rdtscp(dummy, dummy, p);
@@ -165,8 +164,8 @@ 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->t1 = p;
+		tcache->blob[0] = j;
+		tcache->blob[1] = p;
 	}
 	if (cpu)
 		*cpu = p & 0xfff;