Merge /spare/repo/linux-2.6/

Documentation/block/ioprio.txt

+Block io priorities
+===================
+
+
+Intro
+-----
+
+With the introduction of cfq v3 (aka cfq-ts or time sliced cfq), basic io
+priorities is supported for reads on files. This enables users to io nice
+processes or process groups, similar to what has been possible to cpu
+scheduling for ages. This document mainly details the current possibilites
+with cfq, other io schedulers do not support io priorities so far.
+
+Scheduling classes
+------------------
+
+CFQ implements three generic scheduling classes that determine how io is
+served for a process.
+
+IOPRIO_CLASS_RT: This is the realtime io class. This scheduling class is given
+higher priority than any other in the system, processes from this class are
+given first access to the disk every time. Thus it needs to be used with some
+care, one io RT process can starve the entire system. Within the RT class,
+there are 8 levels of class data that determine exactly how much time this
+process needs the disk for on each service. In the future this might change
+to be more directly mappable to performance, by passing in a wanted data
+rate instead.
+
+IOPRIO_CLASS_BE: This is the best-effort scheduling class, which is the default
+for any process that hasn't set a specific io priority. The class data
+determines how much io bandwidth the process will get, it's directly mappable
+to the cpu nice levels just more coarsely implemented. 0 is the highest
+BE prio level, 7 is the lowest. The mapping between cpu nice level and io
+nice level is determined as: io_nice = (cpu_nice + 20) / 5.
+
+IOPRIO_CLASS_IDLE: This is the idle scheduling class, processes running at this
+level only get io time when no one else needs the disk. The idle class has no
+class data, since it doesn't really apply here.
+
+Tools
+-----
+
+See below for a sample ionice tool. Usage:
+
+# ionice -c<class> -n<level> -p<pid>
+
+If pid isn't given, the current process is assumed. IO priority settings
+are inherited on fork, so you can use ionice to start the process at a given
+level:
+
+# ionice -c2 -n0 /bin/ls
+
+will run ls at the best-effort scheduling class at the highest priority.
+For a running process, you can give the pid instead:
+
+# ionice -c1 -n2 -p100
+
+will change pid 100 to run at the realtime scheduling class, at priority 2.
+
+---> snip ionice.c tool <---
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <getopt.h>
+#include <unistd.h>
+#include <sys/ptrace.h>
+#include <asm/unistd.h>
+
+extern int sys_ioprio_set(int, int, int);
+extern int sys_ioprio_get(int, int);
+
+#if defined(__i386__)
+#define __NR_ioprio_set		289
+#define __NR_ioprio_get		290
+#elif defined(__ppc__)
+#define __NR_ioprio_set		273
+#define __NR_ioprio_get		274
+#elif defined(__x86_64__)
+#define __NR_ioprio_set		251
+#define __NR_ioprio_get		252
+#elif defined(__ia64__)
+#define __NR_ioprio_set		1274
+#define __NR_ioprio_get		1275
+#else
+#error "Unsupported arch"
+#endif
+
+_syscall3(int, ioprio_set, int, which, int, who, int, ioprio);
+_syscall2(int, ioprio_get, int, which, int, who);
+
+enum {
+	IOPRIO_CLASS_NONE,
+	IOPRIO_CLASS_RT,
+	IOPRIO_CLASS_BE,
+	IOPRIO_CLASS_IDLE,
+};
+
+enum {
+	IOPRIO_WHO_PROCESS = 1,
+	IOPRIO_WHO_PGRP,
+	IOPRIO_WHO_USER,
+};
+
+#define IOPRIO_CLASS_SHIFT	13
+
+const char *to_prio[] = { "none", "realtime", "best-effort", "idle", };
+
+int main(int argc, char *argv[])
+{
+	int ioprio = 4, set = 0, ioprio_class = IOPRIO_CLASS_BE;
+	int c, pid = 0;
+
+	while ((c = getopt(argc, argv, "+n:c:p:")) != EOF) {
+		switch (c) {
+		case 'n':
+			ioprio = strtol(optarg, NULL, 10);
+			set = 1;
+			break;
+		case 'c':
+			ioprio_class = strtol(optarg, NULL, 10);
+			set = 1;
+			break;
+		case 'p':
+			pid = strtol(optarg, NULL, 10);
+			break;
+		}
+	}
+
+	switch (ioprio_class) {
+		case IOPRIO_CLASS_NONE:
+			ioprio_class = IOPRIO_CLASS_BE;
+			break;
+		case IOPRIO_CLASS_RT:
+		case IOPRIO_CLASS_BE:
+			break;
+		case IOPRIO_CLASS_IDLE:
+			ioprio = 7;
+			break;
+		default:
+			printf("bad prio class %d\n", ioprio_class);
+			return 1;
+	}
+
+	if (!set) {
+		if (!pid && argv[optind])
+			pid = strtol(argv[optind], NULL, 10);
+
+		ioprio = ioprio_get(IOPRIO_WHO_PROCESS, pid);
+
+		printf("pid=%d, %d\n", pid, ioprio);
+
+		if (ioprio == -1)
+			perror("ioprio_get");
+		else {
+			ioprio_class = ioprio >> IOPRIO_CLASS_SHIFT;
+			ioprio = ioprio & 0xff;
+			printf("%s: prio %d\n", to_prio[ioprio_class], ioprio);
+		}
+	} else {
+		if (ioprio_set(IOPRIO_WHO_PROCESS, pid, ioprio | ioprio_class << IOPRIO_CLASS_SHIFT) == -1) {
+			perror("ioprio_set");
+			return 1;
+		}
+
+		if (argv[optind])
+			execvp(argv[optind], &argv[optind]);
+	}
+
+	return 0;
+}
+
+---> snip ionice.c tool <---
+
+
+March 11 2005, Jens Axboe <axboe@suse.de>

Documentation/cciss.txt

@@ -17,6 +17,7 @@ This driver is known to work with the following cards:
 	* SA P600
 	* SA P800
 	* SA E400
+	* SA E300
 
 If nodes are not already created in the /dev/cciss directory, run as root:
 

Documentation/kernel-parameters.txt

@@ -1115,7 +1115,7 @@ running once the system is up.
 			See Documentation/ramdisk.txt.
 
 	psmouse.proto=  [HW,MOUSE] Highest PS2 mouse protocol extension to
-			probe for (bare|imps|exps).
+			probe for (bare|imps|exps|lifebook|any).
 	psmouse.rate=	[HW,MOUSE] Set desired mouse report rate, in reports
 			per second.
 	psmouse.resetafter=

MAINTAINERS

@@ -1149,7 +1149,7 @@ S:	Maintained
 
 INFINIBAND SUBSYSTEM
 P:	Roland Dreier
-M:	roland@topspin.com
+M:	rolandd@cisco.com
 P:	Sean Hefty
 M:	mshefty@ichips.intel.com
 P:	Hal Rosenstock

arch/arm/kernel/process.c

@@ -32,6 +32,7 @@
 #include <asm/leds.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/mach/time.h>
 
 extern const char *processor_modes[];
 extern void setup_mm_for_reboot(char mode);
@@ -85,8 +86,10 @@ EXPORT_SYMBOL(pm_power_off);
 void default_idle(void)
 {
 	local_irq_disable();
-	if (!need_resched() && !hlt_counter)
+	if (!need_resched() && !hlt_counter) {
+		timer_dyn_reprogram();
 		arch_idle();
+	}
 	local_irq_enable();
 }
 

arch/arm/kernel/time.c

@@ -424,15 +424,19 @@ static int timer_dyn_tick_disable(void)
 	return ret;
 }
 
+/*
+ * Reprogram the system timer for at least the calculated time interval.
+ * This function should be called from the idle thread with IRQs disabled,
+ * immediately before sleeping.
+ */
 void timer_dyn_reprogram(void)
 {
 	struct dyn_tick_timer *dyn_tick = system_timer->dyn_tick;
-	unsigned long flags;
 
-	write_seqlock_irqsave(&xtime_lock, flags);
+	write_seqlock(&xtime_lock);
 	if (dyn_tick->state & DYN_TICK_ENABLED)
 		dyn_tick->reprogram(next_timer_interrupt() - jiffies);
-	write_sequnlock_irqrestore(&xtime_lock, flags);
+	write_sequnlock(&xtime_lock);
 }
 
 static ssize_t timer_show_dyn_tick(struct sys_device *dev, char *buf)

arch/arm/mach-aaec2000/Makefile.boot

+	zreladdr-y := 0xf0008000

arch/arm/mach-omap/usb.c

@@ -288,8 +288,8 @@ static void usb_release(struct device *dev)
 static struct resource udc_resources[] = {
 	/* order is significant! */
 	{		/* registers */
-		.start		= IO_ADDRESS(UDC_BASE),
-		.end		= IO_ADDRESS(UDC_BASE + 0xff),
+		.start		= UDC_BASE,
+		.end		= UDC_BASE + 0xff,
 		.flags		= IORESOURCE_MEM,
 	}, {		/* general IRQ */
 		.start		= IH2_BASE + 20,
@@ -355,8 +355,8 @@ static struct platform_device ohci_device = {
 static struct resource otg_resources[] = {
 	/* order is significant! */
 	{
-		.start		= IO_ADDRESS(OTG_BASE),
-		.end		= IO_ADDRESS(OTG_BASE + 0xff),
+		.start		= OTG_BASE,
+		.end		= OTG_BASE + 0xff,
 		.flags		= IORESOURCE_MEM,
 	}, {
 		.start		= IH2_BASE + 8,

arch/arm/mm/init.c

@@ -522,6 +522,69 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s)
 		printk(KERN_INFO "Freeing %s memory: %dK\n", s, size);
 }
 
+static inline void
+free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
+{
+	struct page *start_pg, *end_pg;
+	unsigned long pg, pgend;
+
+	/*
+	 * Convert start_pfn/end_pfn to a struct page pointer.
+	 */
+	start_pg = pfn_to_page(start_pfn);
+	end_pg = pfn_to_page(end_pfn);
+
+	/*
+	 * Convert to physical addresses, and
+	 * round start upwards and end downwards.
+	 */
+	pg = PAGE_ALIGN(__pa(start_pg));
+	pgend = __pa(end_pg) & PAGE_MASK;
+
+	/*
+	 * If there are free pages between these,
+	 * free the section of the memmap array.
+	 */
+	if (pg < pgend)
+		free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
+}
+
+/*
+ * The mem_map array can get very big.  Free the unused area of the memory map.
+ */
+static void __init free_unused_memmap_node(int node, struct meminfo *mi)
+{
+	unsigned long bank_start, prev_bank_end = 0;
+	unsigned int i;
+
+	/*
+	 * [FIXME] This relies on each bank being in address order.  This
+	 * may not be the case, especially if the user has provided the
+	 * information on the command line.
+	 */
+	for (i = 0; i < mi->nr_banks; i++) {
+		if (mi->bank[i].size == 0 || mi->bank[i].node != node)
+			continue;
+
+		bank_start = mi->bank[i].start >> PAGE_SHIFT;
+		if (bank_start < prev_bank_end) {
+			printk(KERN_ERR "MEM: unordered memory banks.  "
+				"Not freeing memmap.\n");
+			break;
+		}
+
+		/*
+		 * If we had a previous bank, and there is a space
+		 * between the current bank and the previous, free it.
+		 */
+		if (prev_bank_end && prev_bank_end != bank_start)
+			free_memmap(node, prev_bank_end, bank_start);
+
+		prev_bank_end = (mi->bank[i].start +
+				 mi->bank[i].size) >> PAGE_SHIFT;
+	}
+}
+
 /*
  * mem_init() marks the free areas in the mem_map and tells us how much
  * memory is free.  This is done after various parts of the system have
@@ -540,16 +603,12 @@ void __init mem_init(void)
 	max_mapnr   = virt_to_page(high_memory) - mem_map;
 #endif
 
-	/*
-	 * We may have non-contiguous memory.
-	 */
-	if (meminfo.nr_banks != 1)
-		create_memmap_holes(&meminfo);
-
 	/* this will put all unused low memory onto the freelists */
 	for_each_online_node(node) {
 		pg_data_t *pgdat = NODE_DATA(node);
 
+		free_unused_memmap_node(node, &meminfo);
+
 		if (pgdat->node_spanned_pages != 0)
 			totalram_pages += free_all_bootmem_node(pgdat);
 	}

arch/arm/mm/mm-armv.c

@@ -169,7 +169,14 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
 
 	memzero(new_pgd, FIRST_KERNEL_PGD_NR * sizeof(pgd_t));
 
+	/*
+	 * Copy over the kernel and IO PGD entries
+	 */
 	init_pgd = pgd_offset_k(0);
+	memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
+		       (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
+
+	clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
 
 	if (!vectors_high()) {
 		/*
@@ -198,14 +205,6 @@ pgd_t *get_pgd_slow(struct mm_struct *mm)
 		spin_unlock(&mm->page_table_lock);
 	}
 
-	/*
-	 * Copy over the kernel and IO PGD entries
-	 */
-	memcpy(new_pgd + FIRST_KERNEL_PGD_NR, init_pgd + FIRST_KERNEL_PGD_NR,
-		       (PTRS_PER_PGD - FIRST_KERNEL_PGD_NR) * sizeof(pgd_t));
-
-	clean_dcache_area(new_pgd, PTRS_PER_PGD * sizeof(pgd_t));
-
 	return new_pgd;
 
 no_pte:
@@ -698,75 +697,3 @@ void __init iotable_init(struct map_desc *io_desc, int nr)
 	for (i = 0; i < nr; i++)
 		create_mapping(io_desc + i);
 }
-
-static inline void
-free_memmap(int node, unsigned long start_pfn, unsigned long end_pfn)
-{
-	struct page *start_pg, *end_pg;
-	unsigned long pg, pgend;
-
-	/*
-	 * Convert start_pfn/end_pfn to a struct page pointer.
-	 */
-	start_pg = pfn_to_page(start_pfn);
-	end_pg = pfn_to_page(end_pfn);
-
-	/*
-	 * Convert to physical addresses, and
-	 * round start upwards and end downwards.
-	 */
-	pg = PAGE_ALIGN(__pa(start_pg));
-	pgend = __pa(end_pg) & PAGE_MASK;
-
-	/*
-	 * If there are free pages between these,
-	 * free the section of the memmap array.
-	 */
-	if (pg < pgend)
-		free_bootmem_node(NODE_DATA(node), pg, pgend - pg);
-}
-
-static inline void free_unused_memmap_node(int node, struct meminfo *mi)
-{
-	unsigned long bank_start, prev_bank_end = 0;
-	unsigned int i;
-
-	/*
-	 * [FIXME] This relies on each bank being in address order.  This
-	 * may not be the case, especially if the user has provided the
-	 * information on the command line.
-	 */
-	for (i = 0; i < mi->nr_banks; i++) {
-		if (mi->bank[i].size == 0 || mi->bank[i].node != node)
-			continue;
-
-		bank_start = mi->bank[i].start >> PAGE_SHIFT;
-		if (bank_start < prev_bank_end) {
-			printk(KERN_ERR "MEM: unordered memory banks.  "
-				"Not freeing memmap.\n");
-			break;
-		}
-
-		/*
-		 * If we had a previous bank, and there is a space
-		 * between the current bank and the previous, free it.
-		 */
-		if (prev_bank_end && prev_bank_end != bank_start)
-			free_memmap(node, prev_bank_end, bank_start);
-
-		prev_bank_end = PAGE_ALIGN(mi->bank[i].start +
-					   mi->bank[i].size) >> PAGE_SHIFT;
-	}
-}
-
-/*
- * The mem_map array can get very big.  Free
- * the unused area of the memory map.
- */
-void __init create_memmap_holes(struct meminfo *mi)
-{
-	int node;
-
-	for_each_online_node(node)
-		free_unused_memmap_node(node, mi);
-}

arch/arm/tools/mach-types

@@ -6,7 +6,7 @@
 # To add an entry into this database, please see Documentation/arm/README,
 # or contact rmk@arm.linux.org.uk
 #
-# Last update: Thu Mar 24 14:34:50 2005
+# Last update: Thu Jun 23 20:19:33 2005
 #
 # machine_is_xxx	CONFIG_xxxx		MACH_TYPE_xxx		number
 #
@@ -243,7 +243,7 @@ yoho			ARCH_YOHO		YOHO			231
 jasper			ARCH_JASPER		JASPER			232
 dsc25			ARCH_DSC25		DSC25			233
 omap_innovator		MACH_OMAP_INNOVATOR	OMAP_INNOVATOR		234
-ramses			ARCH_RAMSES		RAMSES			235
+mnci			ARCH_RAMSES		RAMSES			235
 s28x			ARCH_S28X		S28X			236
 mport3			ARCH_MPORT3		MPORT3			237
 pxa_eagle250		ARCH_PXA_EAGLE250	PXA_EAGLE250		238
@@ -323,7 +323,7 @@ nimbra29x		ARCH_NIMBRA29X		NIMBRA29X		311
 nimbra210		ARCH_NIMBRA210		NIMBRA210		312
 hhp_d95xx		ARCH_HHP_D95XX		HHP_D95XX		313
 labarm			ARCH_LABARM		LABARM			314
-m825xx			ARCH_M825XX		M825XX			315
+comcerto		ARCH_M825XX		M825XX			315
 m7100			SA1100_M7100		M7100			316
 nipc2			ARCH_NIPC2		NIPC2			317
 fu7202			ARCH_FU7202		FU7202			318
@@ -724,3 +724,66 @@ lpc22xx			MACH_LPC22XX		LPC22XX			715
 omap_comet3		MACH_COMET3		COMET3			716
 omap_comet4		MACH_COMET4		COMET4			717
 csb625			MACH_CSB625		CSB625			718
+fortunet2		MACH_FORTUNET2		FORTUNET2		719
+s5h2200			MACH_S5H2200		S5H2200			720
+optorm920		MACH_OPTORM920		OPTORM920		721
+adsbitsyxb		MACH_ADSBITSYXB		ADSBITSYXB		722
+adssphere		MACH_ADSSPHERE		ADSSPHERE		723
+adsportal		MACH_ADSPORTAL		ADSPORTAL		724
+ln2410sbc		MACH_LN2410SBC		LN2410SBC		725
+cb3rufc			MACH_CB3RUFC		CB3RUFC			726
+mp2usb			MACH_MP2USB		MP2USB			727
+ntnp425c		MACH_NTNP425C		NTNP425C		728
+colibri			MACH_COLIBRI		COLIBRI			729
+pcm7220			MACH_PCM7220		PCM7220			730
+gateway7001		MACH_GATEWAY7001	GATEWAY7001		731
+pcm027			MACH_PCM027		PCM027			732
+cmpxa			MACH_CMPXA		CMPXA			733
+anubis			MACH_ANUBIS		ANUBIS			734
+ite8152			MACH_ITE8152		ITE8152			735
+lpc3xxx			MACH_LPC3XXX		LPC3XXX			736
+puppeteer		MACH_PUPPETEER		PUPPETEER		737
+vt001			MACH_MACH_VADATECH	MACH_VADATECH		738
+e570			MACH_E570		E570			739
+x50			MACH_X50		X50			740
+recon			MACH_RECON		RECON			741
+xboardgp8		MACH_XBOARDGP8		XBOARDGP8		742
+fpic2			MACH_FPIC2		FPIC2			743
+akita			MACH_AKITA		AKITA			744
+a81			MACH_A81		A81			745
+svm_sc25x		MACH_SVM_SC25X		SVM_SC25X		746
+vt020			MACH_VADATECH020	VADATECH020		747
+tli			MACH_TLI		TLI			748
+edb9315lc		MACH_EDB9315LC		EDB9315LC		749
+passec			MACH_PASSEC		PASSEC			750
+ds_tiger		MACH_DS_TIGER		DS_TIGER		751
+e310			MACH_E310		E310			752
+e330			MACH_E330		E330			753
+rt3000			MACH_RT3000		RT3000			754
+nokia770		MACH_NOKIA770		NOKIA770		755
+pnx0106			MACH_PNX0106		PNX0106			756
+hx21xx			MACH_HX21XX		HX21XX			757
+faraday			MACH_FARADAY		FARADAY			758
+sbc9312			MACH_SBC9312		SBC9312			759
+batman			MACH_BATMAN		BATMAN			760
+jpd201			MACH_JPD201		JPD201			761
+mipsa			MACH_MIPSA		MIPSA			762
+kacom			MACH_KACOM		KACOM			763
+swarcocpu		MACH_SWARCOCPU		SWARCOCPU		764
+swarcodsl		MACH_SWARCODSL		SWARCODSL		765
+blueangel		MACH_BLUEANGEL		BLUEANGEL		766
+hairygrama		MACH_HAIRYGRAMA		HAIRYGRAMA		767
+banff			MACH_BANFF		BANFF			768
+carmeva			MACH_CARMEVA		CARMEVA			769
+sam255			MACH_SAM255		SAM255			770
+ppm10			MACH_PPM10		PPM10			771
+edb9315a		MACH_EDB9315A		EDB9315A		772
+sunset			MACH_SUNSET		SUNSET			773
+stargate2		MACH_STARGATE2		STARGATE2		774
+intelmote2		MACH_INTELMOTE2		INTELMOTE2		775
+trizeps4		MACH_TRIZEPS4		TRIZEPS4		776
+mainstone2		MACH_MAINSTONE2		MAINSTONE2		777
+ez_ixp42x		MACH_EZ_IXP42X		EZ_IXP42X		778
+tapwave_zodiac		MACH_TAPWAVE_ZODIAC	TAPWAVE_ZODIAC		779
+universalmeter		MACH_UNIVERSALMETER	UNIVERSALMETER		780
+hicoarm9		MACH_HICOARM9		HICOARM9		781

arch/i386/kernel/kprobes.c

@@ -127,48 +127,23 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->eip = (unsigned long)&p->ainsn.insn;
 }
 
-struct task_struct  *arch_get_kprobe_task(void *ptr)
-{
-	return ((struct thread_info *) (((unsigned long) ptr) &
-					(~(THREAD_SIZE -1))))->task;
-}
-
 void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
 {
 	unsigned long *sara = (unsigned long *)&regs->esp;
-	struct kretprobe_instance *ri;
-	static void *orig_ret_addr;
+        struct kretprobe_instance *ri;
+
+        if ((ri = get_free_rp_inst(rp)) != NULL) {
+                ri->rp = rp;
+                ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *) *sara;
 
-	/*
-	 * Save the return address when the return probe hits
-	 * the first time, and use it to populate the (krprobe
-	 * instance)->ret_addr for subsequent return probes at
-	 * the same addrress since stack address would have
-	 * the kretprobe_trampoline by then.
-	 */
-	if (((void*) *sara) != kretprobe_trampoline)
-		orig_ret_addr = (void*) *sara;
-
-	if ((ri = get_free_rp_inst(rp)) != NULL) {
-		ri->rp = rp;
-		ri->stack_addr = sara;
-		ri->ret_addr = orig_ret_addr;
-		add_rp_inst(ri);
 		/* Replace the return addr with trampoline addr */
 		*sara = (unsigned long) &kretprobe_trampoline;
-	} else {
-		rp->nmissed++;
-	}
-}
 
-void arch_kprobe_flush_task(struct task_struct *tk)
-{
-	struct kretprobe_instance *ri;
-	while ((ri = get_rp_inst_tsk(tk)) != NULL) {
-		*((unsigned long *)(ri->stack_addr)) =
-					(unsigned long) ri->ret_addr;
-		recycle_rp_inst(ri);
-	}
+                add_rp_inst(ri);
+        } else {
+                rp->nmissed++;
+        }
 }
 
 /*
@@ -286,36 +261,59 @@ static int kprobe_handler(struct pt_regs *regs)
  */
 int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 {
-	struct task_struct *tsk;
-	struct kretprobe_instance *ri;
-	struct hlist_head *head;
-	struct hlist_node *node;
-	unsigned long *sara = ((unsigned long *) &regs->esp) - 1;
-
-	tsk = arch_get_kprobe_task(sara);
-	head = kretprobe_inst_table_head(tsk);
-
-	hlist_for_each_entry(ri, node, head, hlist) {
-		if (ri->stack_addr == sara && ri->rp) {
-			if (ri->rp->handler)
-				ri->rp->handler(ri, regs);
-		}
-	}
-	return 0;
-}
+        struct kretprobe_instance *ri = NULL;
+        struct hlist_head *head;
+        struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
 
-void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
-						unsigned long flags)
-{
-	struct kretprobe_instance *ri;
-	/* RA already popped */
-	unsigned long *sara = ((unsigned long *)&regs->esp) - 1;
+        head = kretprobe_inst_table_head(current);
 
-	while ((ri = get_rp_inst(sara))) {
-		regs->eip = (unsigned long)ri->ret_addr;
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+         *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
 		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
 	}
-	regs->eflags &= ~TF_MASK;
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->eip = orig_ret_address;
+
+	unlock_kprobes();
+	preempt_enable_no_resched();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
 }
 
 /*
@@ -403,8 +401,7 @@ static inline int post_kprobe_handler(struct pt_regs *regs)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 	}
 
-	if (current_kprobe->post_handler != trampoline_post_handler)
-		resume_execution(current_kprobe, regs);
+	resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_eflags;
 
 	/*Restore back the original saved kprobes variables and continue. */
@@ -534,3 +531,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	}
 	return 0;
 }
+
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+	return register_kprobe(&trampoline_p);
+}

arch/i386/kernel/process.c

@@ -616,6 +616,33 @@ handle_io_bitmap(struct thread_struct *next, struct tss_struct *tss)
 	tss->io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY;
 }
 
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+			       struct task_struct *next_p)
+{
+	struct thread_info *prev, *next;
+
+	/*
+	 * gcc should eliminate the ->thread_info dereference if
+	 * has_secure_computing returns 0 at compile time (SECCOMP=n).
+	 */
+	prev = prev_p->thread_info;
+	next = next_p->thread_info;
+
+	if (has_secure_computing(prev) || has_secure_computing(next)) {
+		/* slow path here */
+		if (has_secure_computing(prev) &&
+		    !has_secure_computing(next)) {
+			write_cr4(read_cr4() & ~X86_CR4_TSD);
+		} else if (!has_secure_computing(prev) &&
+			   has_secure_computing(next))
+			write_cr4(read_cr4() | X86_CR4_TSD);
+	}
+}
+
 /*
  *	switch_to(x,yn) should switch tasks from x to y.
  *
@@ -695,6 +722,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 	if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr))
 		handle_io_bitmap(next, tss);
 
+	disable_tsc(prev_p, next_p);
+
 	return prev_p;
 }
 

arch/i386/kernel/syscall_table.S

@@ -289,3 +289,5 @@ ENTRY(sys_call_table)
 	.long sys_add_key
 	.long sys_request_key
 	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get		/* 290 */

arch/ia64/kernel/entry.S

@@ -1577,8 +1577,8 @@ sys_call_table:
 	data8 sys_add_key
 	data8 sys_request_key
 	data8 sys_keyctl
-	data8 sys_ni_syscall
-	data8 sys_ni_syscall			// 1275
+	data8 sys_ioprio_set
+	data8 sys_ioprio_get			// 1275
 	data8 sys_set_zone_reclaim
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall

arch/ia64/kernel/kprobes.c

@@ -34,6 +34,7 @@
 
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
+#include <asm/sections.h>
 
 extern void jprobe_inst_return(void);
 
@@ -263,13 +264,33 @@ static inline void get_kprobe_inst(bundle_t *bundle, uint slot,
 	}
 }
 
+/* Returns non-zero if the addr is in the Interrupt Vector Table */
+static inline int in_ivt_functions(unsigned long addr)
+{
+	return (addr >= (unsigned long)__start_ivt_text
+		&& addr < (unsigned long)__end_ivt_text);
+}
+
 static int valid_kprobe_addr(int template, int slot, unsigned long addr)
 {
 	if ((slot > 2) || ((bundle_encoding[template][1] == L) && slot > 1)) {
-		printk(KERN_WARNING "Attempting to insert unaligned kprobe at 0x%lx\n",
-				addr);
+		printk(KERN_WARNING "Attempting to insert unaligned kprobe "
+				"at 0x%lx\n", addr);
 		return -EINVAL;
 	}
+
+ 	if (in_ivt_functions(addr)) {
+ 		printk(KERN_WARNING "Kprobes can't be inserted inside "
+				"IVT functions at 0x%lx\n", addr);
+ 		return -EINVAL;
+ 	}
+
+	if (slot == 1 && bundle_encoding[template][1] != L) {
+		printk(KERN_WARNING "Inserting kprobes on slot #1 "
+		       "is not supported\n");
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
@@ -290,6 +311,94 @@ static inline void set_current_kprobe(struct kprobe *p)
 	current_kprobe = p;
 }
 
+static void kretprobe_trampoline(void)
+{
+}
+
+/*
+ * At this point the target function has been tricked into
+ * returning into our trampoline.  Lookup the associated instance
+ * and then:
+ *    - call the handler function
+ *    - cleanup by marking the instance as unused
+ *    - long jump back to the original return address
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri = NULL;
+	struct hlist_head *head;
+	struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =
+		((struct fnptr *)kretprobe_trampoline)->ip;
+
+        head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+	 *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->cr_iip = orig_ret_address;
+
+	unlock_kprobes();
+	preempt_enable_no_resched();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
+}
+
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *)regs->b0;
+
+		/* Replace the return addr with trampoline addr */
+		regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
+
+		add_rp_inst(ri);
+	} else {
+		rp->nmissed++;
+	}
+}
+
 int arch_prepare_kprobe(struct kprobe *p)
 {
 	unsigned long addr = (unsigned long) p->addr;
@@ -492,8 +601,8 @@ static int pre_kprobes_handler(struct die_args *args)
 	if (p->pre_handler && p->pre_handler(p, regs))
 		/*
 		 * Our pre-handler is specifically requesting that we just
-		 * do a return.  This is handling the case where the
-		 * pre-handler is really our special jprobe pre-handler.
+		 * do a return.  This is used for both the jprobe pre-handler
+		 * and the kretprobe trampoline
 		 */
 		return 1;
 
@@ -599,3 +708,14 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	*regs = jprobe_saved_regs;
 	return 1;
 }
+
+static struct kprobe trampoline_p = {
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+	trampoline_p.addr =
+		(kprobe_opcode_t *)((struct fnptr *)kretprobe_trampoline)->ip;
+	return register_kprobe(&trampoline_p);
+}

arch/ia64/kernel/process.c

@@ -27,6 +27,7 @@
 #include <linux/efi.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
+#include <linux/kprobes.h>
 
 #include <asm/cpu.h>
 #include <asm/delay.h>
@@ -707,6 +708,13 @@ kernel_thread_helper (int (*fn)(void *), void *arg)
 void
 flush_thread (void)
 {
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(current);
+
 	/* drop floating-point and debug-register state if it exists: */
 	current->thread.flags &= ~(IA64_THREAD_FPH_VALID | IA64_THREAD_DBG_VALID);
 	ia64_drop_fpu(current);
@@ -721,6 +729,14 @@ flush_thread (void)
 void
 exit_thread (void)
 {
+
+	/*
+	 * Remove function-return probe instances associated with this task
+	 * and put them back on the free list. Do not insert an exit probe for
+	 * this function, it will be disabled by kprobe_flush_task if you do.
+	 */
+	kprobe_flush_task(current);
+
 	ia64_drop_fpu(current);
 #ifdef CONFIG_PERFMON
        /* if needed, stop monitoring and flush state to perfmon context */

arch/ia64/kernel/vmlinux.lds.S

@@ -8,6 +8,11 @@
 #define LOAD_OFFSET	(KERNEL_START - KERNEL_TR_PAGE_SIZE)
 #include <asm-generic/vmlinux.lds.h>
 
+#define IVT_TEXT							\
+		VMLINUX_SYMBOL(__start_ivt_text) = .;			\
+		*(.text.ivt)						\
+		VMLINUX_SYMBOL(__end_ivt_text) = .;
+
 OUTPUT_FORMAT("elf64-ia64-little")
 OUTPUT_ARCH(ia64)
 ENTRY(phys_start)
@@ -39,7 +44,7 @@ SECTIONS
 
   .text : AT(ADDR(.text) - LOAD_OFFSET)
     {
-	*(.text.ivt)
+	IVT_TEXT
 	*(.text)
 	SCHED_TEXT
 	LOCK_TEXT

arch/mips/kernel/signal.c

@@ -457,7 +457,7 @@ static int do_signal(sigset_t *oldset, struct pt_regs *regs)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze(0))
+	if (try_to_freeze())
 		goto no_signal;
 
 	if (!oldset)

arch/ppc/kernel/misc.S

@@ -1449,3 +1449,5 @@ _GLOBAL(sys_call_table)
 	.long sys_request_key		/* 270 */
 	.long sys_keyctl
 	.long sys_waitid
+	.long sys_ioprio_set
+	.long sys_ioprio_get

arch/ppc/mm/init.c

@@ -606,9 +606,19 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address,
 		struct page *page = pfn_to_page(pfn);
 		if (!PageReserved(page)
 		    && !test_bit(PG_arch_1, &page->flags)) {
-			if (vma->vm_mm == current->active_mm)
+			if (vma->vm_mm == current->active_mm) {
+#ifdef CONFIG_8xx
+			/* On 8xx, cache control instructions (particularly 
+		 	 * "dcbst" from flush_dcache_icache) fault as write 
+			 * operation if there is an unpopulated TLB entry 
+			 * for the address in question. To workaround that, 
+			 * we invalidate the TLB here, thus avoiding dcbst 
+			 * misbehaviour.
+			 */
+				_tlbie(address);
+#endif
 				__flush_dcache_icache((void *) address);
-			else
+			} else
 				flush_dcache_icache_page(page);
 			set_bit(PG_arch_1, &page->flags);
 		}

arch/ppc/platforms/pmac_sleep.S

@@ -46,7 +46,7 @@
 	.section .text
 	.align	5
 
-#if defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ_PMAC)
+#if defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ_PMAC)
 
 /* This gets called by via-pmu.c late during the sleep process.
  * The PMU was already send the sleep command and will shut us down
@@ -382,7 +382,7 @@ turn_on_mmu:
 	isync
 	rfi
 
-#endif /* defined(CONFIG_PMAC_PBOOK) || defined(CONFIG_CPU_FREQ) */
+#endif /* defined(CONFIG_PM) || defined(CONFIG_CPU_FREQ) */
 
 	.section .data
 	.balign	L1_CACHE_LINE_SIZE

arch/ppc/platforms/pmac_time.c

@@ -206,7 +206,7 @@ via_calibrate_decr(void)
 	return 1;
 }
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 /*
  * Reset the time after a sleep.
  */
@@ -238,7 +238,7 @@ time_sleep_notify(struct pmu_sleep_notifier *self, int when)
 static struct pmu_sleep_notifier time_sleep_notifier __pmacdata = {
 	time_sleep_notify, SLEEP_LEVEL_MISC,
 };
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 /*
  * Query the OF and get the decr frequency.
@@ -251,9 +251,9 @@ pmac_calibrate_decr(void)
 	struct device_node *cpu;
 	unsigned int freq, *fp;
 
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PM
 	pmu_register_sleep_notifier(&time_sleep_notifier);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PM */
 
 	/* We assume MacRISC2 machines have correct device-tree
 	 * calibration. That's better since the VIA itself seems

arch/ppc/platforms/sandpoint.c

@@ -324,6 +324,7 @@ sandpoint_setup_arch(void)
 			pdata[1].irq = 0;
 			pdata[1].mapbase = 0;
 		}
+	}
 
 	printk(KERN_INFO "Motorola SPS Sandpoint Test Platform\n");
 	printk(KERN_INFO "Port by MontaVista Software, Inc. (source@mvista.com)\n");

arch/ppc/syslib/open_pic.c

@@ -370,8 +370,9 @@ void __init openpic_init(int offset)
 	/* Initialize IPI interrupts */
 	if ( ppc_md.progress ) ppc_md.progress("openpic: ipi",0x3bb);
 	for (i = 0; i < OPENPIC_NUM_IPI; i++) {
-		/* Disabled, Priority 10..13 */
-		openpic_initipi(i, 10+i, OPENPIC_VEC_IPI+i+offset);
+		/* Disabled, increased priorities 10..13 */
+		openpic_initipi(i, OPENPIC_PRIORITY_IPI_BASE+i,
+				OPENPIC_VEC_IPI+i+offset);
 		/* IPIs are per-CPU */
 		irq_desc[OPENPIC_VEC_IPI+i+offset].status |= IRQ_PER_CPU;
 		irq_desc[OPENPIC_VEC_IPI+i+offset].handler = &open_pic_ipi;
@@ -399,8 +400,9 @@ void __init openpic_init(int offset)
 		if (sense & IRQ_SENSE_MASK)
 			irq_desc[i+offset].status = IRQ_LEVEL;
 
-		/* Enabled, Priority 8 */
-		openpic_initirq(i, 8, i+offset, (sense & IRQ_POLARITY_MASK),
+		/* Enabled, Default priority */
+		openpic_initirq(i, OPENPIC_PRIORITY_DEFAULT, i+offset,
+				(sense & IRQ_POLARITY_MASK),
 				(sense & IRQ_SENSE_MASK));
 		/* Processor 0 */
 		openpic_mapirq(i, CPU_MASK_CPU0, CPU_MASK_NONE);
@@ -655,6 +657,18 @@ static void __init openpic_maptimer(u_int timer, cpumask_t cpumask)
 		      cpus_addr(phys)[0]);
 }
 
+/*
+ * Change the priority of an interrupt
+ */
+void __init
+openpic_set_irq_priority(u_int irq, u_int pri)
+{
+	check_arg_irq(irq);
+	openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
+				OPENPIC_PRIORITY_MASK,
+				pri << OPENPIC_PRIORITY_SHIFT);
+}
+
 /*
  * Initalize the interrupt source which will generate an NMI.
  * This raises the interrupt's priority from 8 to 9.
@@ -665,9 +679,7 @@ void __init
 openpic_init_nmi_irq(u_int irq)
 {
 	check_arg_irq(irq);
-	openpic_safe_writefield(&ISR[irq - open_pic_irq_offset]->Vector_Priority,
-				OPENPIC_PRIORITY_MASK,
-				9 << OPENPIC_PRIORITY_SHIFT);
+	openpic_set_irq_priority(irq, OPENPIC_PRIORITY_NMI);
 }
 
 /*

arch/ppc64/kernel/kprobes.c

@@ -36,6 +36,8 @@
 #include <asm/kdebug.h>
 #include <asm/sstep.h>
 
+static DECLARE_MUTEX(kprobe_mutex);
+
 static struct kprobe *current_kprobe;
 static unsigned long kprobe_status, kprobe_saved_msr;
 static struct kprobe *kprobe_prev;
@@ -54,6 +56,15 @@ int arch_prepare_kprobe(struct kprobe *p)
 		printk("Cannot register a kprobe on rfid or mtmsrd\n");
 		ret = -EINVAL;
 	}
+
+	/* insn must be on a special executable page on ppc64 */
+	if (!ret) {
+		up(&kprobe_mutex);
+		p->ainsn.insn = get_insn_slot();
+		down(&kprobe_mutex);
+		if (!p->ainsn.insn)
+			ret = -ENOMEM;
+	}
 	return ret;
 }
 
@@ -79,16 +90,22 @@ void arch_disarm_kprobe(struct kprobe *p)
 
 void arch_remove_kprobe(struct kprobe *p)
 {
+	up(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	down(&kprobe_mutex);
 }
 
 static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 {
+	kprobe_opcode_t insn = *p->ainsn.insn;
+
 	regs->msr |= MSR_SE;
-	/*single step inline if it a breakpoint instruction*/
-	if (p->opcode == BREAKPOINT_INSTRUCTION)
+
+	/* single step inline if it is a trap variant */
+	if (IS_TW(insn) || IS_TD(insn) || IS_TWI(insn) || IS_TDI(insn))
 		regs->nip = (unsigned long)p->addr;
 	else
-		regs->nip = (unsigned long)&p->ainsn.insn;
+		regs->nip = (unsigned long)p->ainsn.insn;
 }
 
 static inline void save_previous_kprobe(void)
@@ -105,6 +122,23 @@ static inline void restore_previous_kprobe(void)
 	kprobe_saved_msr = kprobe_saved_msr_prev;
 }
 
+void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
+{
+	struct kretprobe_instance *ri;
+
+	if ((ri = get_free_rp_inst(rp)) != NULL) {
+		ri->rp = rp;
+		ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *)regs->link;
+
+		/* Replace the return addr with trampoline addr */
+		regs->link = (unsigned long)kretprobe_trampoline;
+		add_rp_inst(ri);
+	} else {
+		rp->nmissed++;
+	}
+}
+
 static inline int kprobe_handler(struct pt_regs *regs)
 {
 	struct kprobe *p;
@@ -194,6 +228,78 @@ static inline int kprobe_handler(struct pt_regs *regs)
 	return ret;
 }
 
+/*
+ * Function return probe trampoline:
+ * 	- init_kprobes() establishes a probepoint here
+ * 	- When the probed function returns, this probe
+ * 		causes the handlers to fire
+ */
+void kretprobe_trampoline_holder(void)
+{
+	asm volatile(".global kretprobe_trampoline\n"
+			"kretprobe_trampoline:\n"
+			"nop\n");
+}
+
+/*
+ * Called when the probe at kretprobe trampoline is hit
+ */
+int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
+{
+        struct kretprobe_instance *ri = NULL;
+        struct hlist_head *head;
+        struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
+
+        head = kretprobe_inst_table_head(current);
+
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+         *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
+		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->nip = orig_ret_address;
+
+	unlock_kprobes();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
+}
+
 /*
  * Called after single-stepping.  p->addr is the address of the
  * instruction whose first byte has been replaced by the "breakpoint"
@@ -205,9 +311,10 @@ static inline int kprobe_handler(struct pt_regs *regs)
 static void resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
 	int ret;
+	unsigned int insn = *p->ainsn.insn;
 
 	regs->nip = (unsigned long)p->addr;
-	ret = emulate_step(regs, p->ainsn.insn[0]);
+	ret = emulate_step(regs, insn);
 	if (ret == 0)
 		regs->nip = (unsigned long)p->addr + 4;
 }
@@ -331,3 +438,13 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
 	return 1;
 }
+
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
+};
+
+int __init arch_init(void)
+{
+	return register_kprobe(&trampoline_p);
+}

arch/ppc64/kernel/ppc_ksyms.c

@@ -75,6 +75,7 @@ EXPORT_SYMBOL(giveup_fpu);
 EXPORT_SYMBOL(giveup_altivec);
 #endif
 EXPORT_SYMBOL(__flush_icache_range);
+EXPORT_SYMBOL(flush_dcache_range);
 
 #ifdef CONFIG_SMP
 #ifdef CONFIG_PPC_ISERIES

arch/ppc64/kernel/process.c

@@ -36,6 +36,7 @@
 #include <linux/kallsyms.h>
 #include <linux/interrupt.h>
 #include <linux/utsname.h>
+#include <linux/kprobes.h>
 
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
@@ -307,6 +308,8 @@ void show_regs(struct pt_regs * regs)
 
 void exit_thread(void)
 {
+	kprobe_flush_task(current);
+
 #ifndef CONFIG_SMP
 	if (last_task_used_math == current)
 		last_task_used_math = NULL;
@@ -321,6 +324,7 @@ void flush_thread(void)
 {
 	struct thread_info *t = current_thread_info();
 
+	kprobe_flush_task(current);
 	if (t->flags & _TIF_ABI_PENDING)
 		t->flags ^= (_TIF_ABI_PENDING | _TIF_32BIT);
 

arch/ppc64/kernel/time.c

@@ -91,6 +91,7 @@ unsigned long tb_to_xs;
 unsigned      tb_to_us;
 unsigned long processor_freq;
 DEFINE_SPINLOCK(rtc_lock);
+EXPORT_SYMBOL_GPL(rtc_lock);
 
 unsigned long tb_to_ns_scale;
 unsigned long tb_to_ns_shift;

arch/x86_64/kernel/kprobes.c

@@ -38,7 +38,7 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/preempt.h>
-#include <linux/moduleloader.h>
+
 #include <asm/cacheflush.h>
 #include <asm/pgtable.h>
 #include <asm/kdebug.h>
@@ -51,8 +51,6 @@ static struct kprobe *kprobe_prev;
 static unsigned long kprobe_status_prev, kprobe_old_rflags_prev, kprobe_saved_rflags_prev;
 static struct pt_regs jprobe_saved_regs;
 static long *jprobe_saved_rsp;
-static kprobe_opcode_t *get_insn_slot(void);
-static void free_insn_slot(kprobe_opcode_t *slot);
 void jprobe_return_end(void);
 
 /* copy of the kernel stack at the probe fire time */
@@ -274,48 +272,23 @@ static void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->rip = (unsigned long)p->ainsn.insn;
 }
 
-struct task_struct  *arch_get_kprobe_task(void *ptr)
-{
-	return ((struct thread_info *) (((unsigned long) ptr) &
-					(~(THREAD_SIZE -1))))->task;
-}
-
 void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs)
 {
 	unsigned long *sara = (unsigned long *)regs->rsp;
-	struct kretprobe_instance *ri;
-	static void *orig_ret_addr;
+        struct kretprobe_instance *ri;
+
+        if ((ri = get_free_rp_inst(rp)) != NULL) {
+                ri->rp = rp;
+                ri->task = current;
+		ri->ret_addr = (kprobe_opcode_t *) *sara;
 
-	/*
-	 * Save the return address when the return probe hits
-	 * the first time, and use it to populate the (krprobe
-	 * instance)->ret_addr for subsequent return probes at
-	 * the same addrress since stack address would have
-	 * the kretprobe_trampoline by then.
-	 */
-	if (((void*) *sara) != kretprobe_trampoline)
-		orig_ret_addr = (void*) *sara;
-
-	if ((ri = get_free_rp_inst(rp)) != NULL) {
-		ri->rp = rp;
-		ri->stack_addr = sara;
-		ri->ret_addr = orig_ret_addr;
-		add_rp_inst(ri);
 		/* Replace the return addr with trampoline addr */
 		*sara = (unsigned long) &kretprobe_trampoline;
-	} else {
-		rp->nmissed++;
-	}
-}
 
-void arch_kprobe_flush_task(struct task_struct *tk)
-{
-	struct kretprobe_instance *ri;
-	while ((ri = get_rp_inst_tsk(tk)) != NULL) {
-		*((unsigned long *)(ri->stack_addr)) =
-					(unsigned long) ri->ret_addr;
-		recycle_rp_inst(ri);
-	}
+                add_rp_inst(ri);
+        } else {
+                rp->nmissed++;
+        }
 }
 
 /*
@@ -428,36 +401,59 @@ int kprobe_handler(struct pt_regs *regs)
  */
 int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 {
-	struct task_struct *tsk;
-	struct kretprobe_instance *ri;
-	struct hlist_head *head;
-	struct hlist_node *node;
-	unsigned long *sara = (unsigned long *)regs->rsp - 1;
-
-	tsk = arch_get_kprobe_task(sara);
-	head = kretprobe_inst_table_head(tsk);
-
-	hlist_for_each_entry(ri, node, head, hlist) {
-		if (ri->stack_addr == sara && ri->rp) {
-			if (ri->rp->handler)
-				ri->rp->handler(ri, regs);
-		}
-	}
-	return 0;
-}
+        struct kretprobe_instance *ri = NULL;
+        struct hlist_head *head;
+        struct hlist_node *node, *tmp;
+	unsigned long orig_ret_address = 0;
+	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
 
-void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs,
-						unsigned long flags)
-{
-	struct kretprobe_instance *ri;
-	/* RA already popped */
-	unsigned long *sara = ((unsigned long *)regs->rsp) - 1;
+        head = kretprobe_inst_table_head(current);
 
-	while ((ri = get_rp_inst(sara))) {
-		regs->rip = (unsigned long)ri->ret_addr;
+	/*
+	 * It is possible to have multiple instances associated with a given
+	 * task either because an multiple functions in the call path
+	 * have a return probe installed on them, and/or more then one return
+	 * return probe was registered for a target function.
+	 *
+	 * We can handle this because:
+	 *     - instances are always inserted at the head of the list
+	 *     - when multiple return probes are registered for the same
+         *       function, the first instance's ret_addr will point to the
+	 *       real return address, and all the rest will point to
+	 *       kretprobe_trampoline
+	 */
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+                if (ri->task != current)
+			/* another task is sharing our hash bucket */
+                        continue;
+
+		if (ri->rp && ri->rp->handler)
+			ri->rp->handler(ri, regs);
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
 		recycle_rp_inst(ri);
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
 	}
-	regs->eflags &= ~TF_MASK;
+
+	BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
+	regs->rip = orig_ret_address;
+
+	unlock_kprobes();
+	preempt_enable_no_resched();
+
+        /*
+         * By returning a non-zero value, we are telling
+         * kprobe_handler() that we have handled unlocking
+         * and re-enabling preemption.
+         */
+        return 1;
 }
 
 /*
@@ -550,8 +546,7 @@ int post_kprobe_handler(struct pt_regs *regs)
 		current_kprobe->post_handler(current_kprobe, regs, 0);
 	}
 
-	if (current_kprobe->post_handler != trampoline_post_handler)
-		resume_execution(current_kprobe, regs);
+	resume_execution(current_kprobe, regs);
 	regs->eflags |= kprobe_saved_rflags;
 
 	/* Restore the original saved kprobes variables and continue. */
@@ -682,111 +677,12 @@ int longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	return 0;
 }
 
-/*
- * kprobe->ainsn.insn points to the copy of the instruction to be single-stepped.
- * By default on x86_64, pages we get from kmalloc or vmalloc are not
- * executable.  Single-stepping an instruction on such a page yields an
- * oops.  So instead of storing the instruction copies in their respective
- * kprobe objects, we allocate a page, map it executable, and store all the
- * instruction copies there.  (We can allocate additional pages if somebody
- * inserts a huge number of probes.)  Each page can hold up to INSNS_PER_PAGE
- * instruction slots, each of which is MAX_INSN_SIZE*sizeof(kprobe_opcode_t)
- * bytes.
- */
-#define INSNS_PER_PAGE (PAGE_SIZE/(MAX_INSN_SIZE*sizeof(kprobe_opcode_t)))
-struct kprobe_insn_page {
-	struct hlist_node hlist;
-	kprobe_opcode_t *insns;		/* page of instruction slots */
-	char slot_used[INSNS_PER_PAGE];
-	int nused;
+static struct kprobe trampoline_p = {
+	.addr = (kprobe_opcode_t *) &kretprobe_trampoline,
+	.pre_handler = trampoline_probe_handler
 };
 
-static struct hlist_head kprobe_insn_pages;
-
-/**
- * get_insn_slot() - Find a slot on an executable page for an instruction.
- * We allocate an executable page if there's no room on existing ones.
- */
-static kprobe_opcode_t *get_insn_slot(void)
-{
-	struct kprobe_insn_page *kip;
-	struct hlist_node *pos;
-
-	hlist_for_each(pos, &kprobe_insn_pages) {
-		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
-		if (kip->nused < INSNS_PER_PAGE) {
-			int i;
-			for (i = 0; i < INSNS_PER_PAGE; i++) {
-				if (!kip->slot_used[i]) {
-					kip->slot_used[i] = 1;
-					kip->nused++;
-					return kip->insns + (i*MAX_INSN_SIZE);
-				}
-			}
-			/* Surprise!  No unused slots.  Fix kip->nused. */
-			kip->nused = INSNS_PER_PAGE;
-		}
-	}
-
-	/* All out of space.  Need to allocate a new page. Use slot 0.*/
-	kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL);
-	if (!kip) {
-		return NULL;
-	}
-
-	/*
-	 * For the %rip-relative displacement fixups to be doable, we
-	 * need our instruction copy to be within +/- 2GB of any data it
-	 * might access via %rip.  That is, within 2GB of where the
-	 * kernel image and loaded module images reside.  So we allocate
-	 * a page in the module loading area.
-	 */
-	kip->insns = module_alloc(PAGE_SIZE);
-	if (!kip->insns) {
-		kfree(kip);
-		return NULL;
-	}
-	INIT_HLIST_NODE(&kip->hlist);
-	hlist_add_head(&kip->hlist, &kprobe_insn_pages);
-	memset(kip->slot_used, 0, INSNS_PER_PAGE);
-	kip->slot_used[0] = 1;
-	kip->nused = 1;
-	return kip->insns;
-}
-
-/**
- * free_insn_slot() - Free instruction slot obtained from get_insn_slot().
- */
-static void free_insn_slot(kprobe_opcode_t *slot)
+int __init arch_init(void)
 {
-	struct kprobe_insn_page *kip;
-	struct hlist_node *pos;
-
-	hlist_for_each(pos, &kprobe_insn_pages) {
-		kip = hlist_entry(pos, struct kprobe_insn_page, hlist);
-		if (kip->insns <= slot
-		    && slot < kip->insns+(INSNS_PER_PAGE*MAX_INSN_SIZE)) {
-			int i = (slot - kip->insns) / MAX_INSN_SIZE;
-			kip->slot_used[i] = 0;
-			kip->nused--;
-			if (kip->nused == 0) {
-				/*
-				 * Page is no longer in use.  Free it unless
-				 * it's the last one.  We keep the last one
-				 * so as not to have to set it up again the
-				 * next time somebody inserts a probe.
-				 */
-				hlist_del(&kip->hlist);
-				if (hlist_empty(&kprobe_insn_pages)) {
-					INIT_HLIST_NODE(&kip->hlist);
-					hlist_add_head(&kip->hlist,
-						&kprobe_insn_pages);
-				} else {
-					module_free(NULL, kip->insns);
-					kfree(kip);
-				}
-			}
-			return;
-		}
-	}
+	return register_kprobe(&trampoline_p);
 }

arch/x86_64/kernel/process.c

@@ -481,6 +481,33 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 	return err;
 }
 
+/*
+ * This function selects if the context switch from prev to next
+ * has to tweak the TSC disable bit in the cr4.
+ */
+static inline void disable_tsc(struct task_struct *prev_p,
+			       struct task_struct *next_p)
+{
+	struct thread_info *prev, *next;
+
+	/*
+	 * gcc should eliminate the ->thread_info dereference if
+	 * has_secure_computing returns 0 at compile time (SECCOMP=n).
+	 */
+	prev = prev_p->thread_info;
+	next = next_p->thread_info;
+
+	if (has_secure_computing(prev) || has_secure_computing(next)) {
+		/* slow path here */
+		if (has_secure_computing(prev) &&
+		    !has_secure_computing(next)) {
+			write_cr4(read_cr4() & ~X86_CR4_TSD);
+		} else if (!has_secure_computing(prev) &&
+			   has_secure_computing(next))
+			write_cr4(read_cr4() | X86_CR4_TSD);
+	}
+}
+
 /*
  * This special macro can be used to load a debugging register
  */
@@ -599,6 +626,8 @@ struct task_struct *__switch_to(struct task_struct *prev_p, struct task_struct *
 		}
 	}
 
+	disable_tsc(prev_p, next_p);
+
 	return prev_p;
 }
 

drivers/block/as-iosched.c

@@ -1806,7 +1806,8 @@ static void as_put_request(request_queue_t *q, struct request *rq)
 	rq->elevator_private = NULL;
 }
 
-static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+static int as_set_request(request_queue_t *q, struct request *rq,
+			  struct bio *bio, int gfp_mask)
 {
 	struct as_data *ad = q->elevator->elevator_data;
 	struct as_rq *arq = mempool_alloc(ad->arq_pool, gfp_mask);
@@ -1827,7 +1828,7 @@ static int as_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
 	return 1;
 }
 
-static int as_may_queue(request_queue_t *q, int rw)
+static int as_may_queue(request_queue_t *q, int rw, struct bio *bio)
 {
 	int ret = ELV_MQUEUE_MAY;
 	struct as_data *ad = q->elevator->elevator_data;

drivers/block/cciss.c

 /*
  *    Disk Array driver for HP SA 5xxx and 6xxx Controllers
- *    Copyright 2000, 2002 Hewlett-Packard Development Company, L.P.
+ *    Copyright 2000, 2005 Hewlett-Packard Development Company, L.P.
  *
  *    This program is free software; you can redistribute it and/or modify
  *    it under the terms of the GNU General Public License as published by
@@ -54,7 +54,7 @@
 MODULE_AUTHOR("Hewlett-Packard Company");
 MODULE_DESCRIPTION("Driver for HP Controller SA5xxx SA6xxx version 2.6.6");
 MODULE_SUPPORTED_DEVICE("HP SA5i SA5i+ SA532 SA5300 SA5312 SA641 SA642 SA6400"
-			" SA6i P600 P800 E400");
+			" SA6i P600 P800 E400 E300");
 MODULE_LICENSE("GPL");
 
 #include "cciss_cmd.h"
@@ -85,8 +85,10 @@ static const struct pci_device_id cciss_pci_device_id[] = {
 		0x103C, 0x3225, 0, 0, 0},
 	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
 		0x103c, 0x3223, 0, 0, 0},
-	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSB,
+	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
 		0x103c, 0x3231, 0, 0, 0},
+	{ PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC,
+		0x103c, 0x3233, 0, 0, 0},
 	{0,}
 };
 MODULE_DEVICE_TABLE(pci, cciss_pci_device_id);
@@ -110,6 +112,7 @@ static struct board_type products[] = {
 	{ 0x3225103C, "Smart Array P600", &SA5_access},
 	{ 0x3223103C, "Smart Array P800", &SA5_access},
 	{ 0x3231103C, "Smart Array E400", &SA5_access},
+	{ 0x3233103C, "Smart Array E300", &SA5_access},
 };
 
 /* How long to wait (in millesconds) for board to go into simple mode */
@@ -635,6 +638,7 @@ static int cciss_ioctl(struct inode *inode, struct file *filep,
 		cciss_pci_info_struct pciinfo;
 
 		if (!arg) return -EINVAL;
+		pciinfo.domain = pci_domain_nr(host->pdev->bus);
 		pciinfo.bus = host->pdev->bus->number;
 		pciinfo.dev_fn = host->pdev->devfn;
 		pciinfo.board_id = host->board_id;
@@ -787,13 +791,6 @@ static int cciss_ioctl(struct inode *inode, struct file *filep,
  		luninfo.LunID = drv->LunID;
  		luninfo.num_opens = drv->usage_count;
  		luninfo.num_parts = 0;
- 		/* count partitions 1 to 15 with sizes > 0 */
- 		for (i = 0; i < MAX_PART - 1; i++) {
-			if (!disk->part[i])
-				continue;
-			if (disk->part[i]->nr_sects != 0)
-				luninfo.num_parts++;
-		}
  		if (copy_to_user(argp, &luninfo,
  				sizeof(LogvolInfo_struct)))
  			return -EFAULT;

drivers/block/deadline-iosched.c

@@ -760,7 +760,8 @@ static void deadline_put_request(request_queue_t *q, struct request *rq)
 }
 
 static int
-deadline_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+deadline_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+		     int gfp_mask)
 {
 	struct deadline_data *dd = q->elevator->elevator_data;
 	struct deadline_rq *drq;

drivers/block/elevator.c

@@ -486,12 +486,13 @@ struct request *elv_former_request(request_queue_t *q, struct request *rq)
 	return NULL;
 }
 
-int elv_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+int elv_set_request(request_queue_t *q, struct request *rq, struct bio *bio,
+		    int gfp_mask)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_set_req_fn)
-		return e->ops->elevator_set_req_fn(q, rq, gfp_mask);
+		return e->ops->elevator_set_req_fn(q, rq, bio, gfp_mask);
 
 	rq->elevator_private = NULL;
 	return 0;
@@ -505,12 +506,12 @@ void elv_put_request(request_queue_t *q, struct request *rq)
 		e->ops->elevator_put_req_fn(q, rq);
 }
 
-int elv_may_queue(request_queue_t *q, int rw)
+int elv_may_queue(request_queue_t *q, int rw, struct bio *bio)
 {
 	elevator_t *e = q->elevator;
 
 	if (e->ops->elevator_may_queue_fn)
-		return e->ops->elevator_may_queue_fn(q, rw);
+		return e->ops->elevator_may_queue_fn(q, rw, bio);
 
 	return ELV_MQUEUE_MAY;
 }

drivers/block/ll_rw_blk.c

@@ -276,6 +276,7 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 	rq->errors = 0;
 	rq->rq_status = RQ_ACTIVE;
 	rq->bio = rq->biotail = NULL;
+	rq->ioprio = 0;
 	rq->buffer = NULL;
 	rq->ref_count = 1;
 	rq->q = q;
@@ -1442,11 +1443,7 @@ void __generic_unplug_device(request_queue_t *q)
 	if (!blk_remove_plug(q))
 		return;
 
-	/*
-	 * was plugged, fire request_fn if queue has stuff to do
-	 */
-	if (elv_next_request(q))
-		q->request_fn(q);
+	q->request_fn(q);
 }
 EXPORT_SYMBOL(__generic_unplug_device);
 
@@ -1776,8 +1773,8 @@ static inline void blk_free_request(request_queue_t *q, struct request *rq)
 	mempool_free(rq, q->rq.rq_pool);
 }
 
-static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
-						int gfp_mask)
+static inline struct request *
+blk_alloc_request(request_queue_t *q, int rw, struct bio *bio, int gfp_mask)
 {
 	struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);
 
@@ -1790,7 +1787,7 @@ static inline struct request *blk_alloc_request(request_queue_t *q, int rw,
 	 */
 	rq->flags = rw;
 
-	if (!elv_set_request(q, rq, gfp_mask))
+	if (!elv_set_request(q, rq, bio, gfp_mask))
 		return rq;
 
 	mempool_free(rq, q->rq.rq_pool);
@@ -1872,7 +1869,8 @@ static void freed_request(request_queue_t *q, int rw)
 /*
  * Get a free request, queue_lock must not be held
  */
-static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
+static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
+				   int gfp_mask)
 {
 	struct request *rq = NULL;
 	struct request_list *rl = &q->rq;
@@ -1895,7 +1893,7 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
 		}
 	}
 
-	switch (elv_may_queue(q, rw)) {
+	switch (elv_may_queue(q, rw, bio)) {
 		case ELV_MQUEUE_NO:
 			goto rq_starved;
 		case ELV_MQUEUE_MAY:
@@ -1920,7 +1918,7 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
 		set_queue_congested(q, rw);
 	spin_unlock_irq(q->queue_lock);
 
-	rq = blk_alloc_request(q, rw, gfp_mask);
+	rq = blk_alloc_request(q, rw, bio, gfp_mask);
 	if (!rq) {
 		/*
 		 * Allocation failed presumably due to memory. Undo anything
@@ -1961,7 +1959,8 @@ static struct request *get_request(request_queue_t *q, int rw, int gfp_mask)
  * No available requests for this queue, unplug the device and wait for some
  * requests to become available.
  */
-static struct request *get_request_wait(request_queue_t *q, int rw)
+static struct request *get_request_wait(request_queue_t *q, int rw,
+					struct bio *bio)
 {
 	DEFINE_WAIT(wait);
 	struct request *rq;
@@ -1972,7 +1971,7 @@ static struct request *get_request_wait(request_queue_t *q, int rw)
 		prepare_to_wait_exclusive(&rl->wait[rw], &wait,
 				TASK_UNINTERRUPTIBLE);
 
-		rq = get_request(q, rw, GFP_NOIO);
+		rq = get_request(q, rw, bio, GFP_NOIO);
 
 		if (!rq) {
 			struct io_context *ioc;
@@ -2003,9 +2002,9 @@ struct request *blk_get_request(request_queue_t *q, int rw, int gfp_mask)
 	BUG_ON(rw != READ && rw != WRITE);
 
 	if (gfp_mask & __GFP_WAIT)
-		rq = get_request_wait(q, rw);
+		rq = get_request_wait(q, rw, NULL);
 	else
-		rq = get_request(q, rw, gfp_mask);
+		rq = get_request(q, rw, NULL, gfp_mask);
 
 	return rq;
 }
@@ -2333,7 +2332,6 @@ static void __blk_put_request(request_queue_t *q, struct request *req)
 		return;
 
 	req->rq_status = RQ_INACTIVE;
-	req->q = NULL;
 	req->rl = NULL;
 
 	/*
@@ -2462,6 +2460,8 @@ static int attempt_merge(request_queue_t *q, struct request *req,
 		req->rq_disk->in_flight--;
 	}
 
+	req->ioprio = ioprio_best(req->ioprio, next->ioprio);
+
 	__blk_put_request(q, next);
 	return 1;
 }
@@ -2514,11 +2514,13 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 {
 	struct request *req, *freereq = NULL;
 	int el_ret, rw, nr_sectors, cur_nr_sectors, barrier, err, sync;
+	unsigned short prio;
 	sector_t sector;
 
 	sector = bio->bi_sector;
 	nr_sectors = bio_sectors(bio);
 	cur_nr_sectors = bio_cur_sectors(bio);
+	prio = bio_prio(bio);
 
 	rw = bio_data_dir(bio);
 	sync = bio_sync(bio);
@@ -2559,6 +2561,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 			req->biotail->bi_next = bio;
 			req->biotail = bio;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_back_merge(q, req))
 				elv_merged_request(q, req);
@@ -2583,6 +2586,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 			req->hard_cur_sectors = cur_nr_sectors;
 			req->sector = req->hard_sector = sector;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
+			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_front_merge(q, req))
 				elv_merged_request(q, req);
@@ -2610,7 +2614,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 		freereq = NULL;
 	} else {
 		spin_unlock_irq(q->queue_lock);
-		if ((freereq = get_request(q, rw, GFP_ATOMIC)) == NULL) {
+		if ((freereq = get_request(q, rw, bio, GFP_ATOMIC)) == NULL) {
 			/*
 			 * READA bit set
 			 */
@@ -2618,7 +2622,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 			if (bio_rw_ahead(bio))
 				goto end_io;
 	
-			freereq = get_request_wait(q, rw);
+			freereq = get_request_wait(q, rw, bio);
 		}
 		goto again;
 	}
@@ -2646,6 +2650,7 @@ static int __make_request(request_queue_t *q, struct bio *bio)
 	req->buffer = bio_data(bio);	/* see ->buffer comment above */
 	req->waiting = NULL;
 	req->bio = req->biotail = bio;
+	req->ioprio = prio;
 	req->rq_disk = bio->bi_bdev->bd_disk;
 	req->start_time = jiffies;
 
@@ -2674,7 +2679,7 @@ static inline void blk_partition_remap(struct bio *bio)
 	if (bdev != bdev->bd_contains) {
 		struct hd_struct *p = bdev->bd_part;
 
-		switch (bio->bi_rw) {
+		switch (bio_data_dir(bio)) {
 		case READ:
 			p->read_sectors += bio_sectors(bio);
 			p->reads++;
@@ -2693,6 +2698,7 @@ void blk_finish_queue_drain(request_queue_t *q)
 {
 	struct request_list *rl = &q->rq;
 	struct request *rq;
+	int requeued = 0;
 
 	spin_lock_irq(q->queue_lock);
 	clear_bit(QUEUE_FLAG_DRAIN, &q->queue_flags);
@@ -2701,9 +2707,13 @@ void blk_finish_queue_drain(request_queue_t *q)
 		rq = list_entry_rq(q->drain_list.next);
 
 		list_del_init(&rq->queuelist);
-		__elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
+		elv_requeue_request(q, rq);
+		requeued++;
 	}
 
+	if (requeued)
+		q->request_fn(q);
+
 	spin_unlock_irq(q->queue_lock);
 
 	wake_up(&rl->wait[0]);
@@ -2900,7 +2910,7 @@ void submit_bio(int rw, struct bio *bio)
 
 	BIO_BUG_ON(!bio->bi_size);
 	BIO_BUG_ON(!bio->bi_io_vec);
-	bio->bi_rw = rw;
+	bio->bi_rw |= rw;
 	if (rw & WRITE)
 		mod_page_state(pgpgout, count);
 	else
@@ -3257,8 +3267,11 @@ void exit_io_context(void)
 	struct io_context *ioc;
 
 	local_irq_save(flags);
+	task_lock(current);
 	ioc = current->io_context;
 	current->io_context = NULL;
+	ioc->task = NULL;
+	task_unlock(current);
 	local_irq_restore(flags);
 
 	if (ioc->aic && ioc->aic->exit)
@@ -3293,12 +3306,12 @@ struct io_context *get_io_context(int gfp_flags)
 	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
-		ret->pid = tsk->pid;
+		ret->task = current;
+		ret->set_ioprio = NULL;
 		ret->last_waited = jiffies; /* doesn't matter... */
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
 		ret->cic = NULL;
-		spin_lock_init(&ret->lock);
 
 		local_irq_save(flags);
 

drivers/block/swim3.c

@@ -253,7 +253,7 @@ static int floppy_revalidate(struct gendisk *disk);
 static int swim3_add_device(struct device_node *swims);
 int swim3_init(void);
 
-#ifndef CONFIG_PMAC_PBOOK
+#ifndef CONFIG_PMAC_MEDIABAY
 #define check_media_bay(which, what)	1
 #endif
 
@@ -297,9 +297,11 @@ static void do_fd_request(request_queue_t * q)
 	int i;
 	for(i=0;i<floppy_count;i++)
 	{
+#ifdef CONFIG_PMAC_MEDIABAY
 		if (floppy_states[i].media_bay &&
 			check_media_bay(floppy_states[i].media_bay, MB_FD))
 			continue;
+#endif /* CONFIG_PMAC_MEDIABAY */
 		start_request(&floppy_states[i]);
 	}
 	sti();
@@ -856,8 +858,10 @@ static int floppy_ioctl(struct inode *inode, struct file *filp,
 	if ((cmd & 0x80) && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 		return -ENXIO;
+#endif
 
 	switch (cmd) {
 	case FDEJECT:
@@ -881,8 +885,10 @@ static int floppy_open(struct inode *inode, struct file *filp)
 	int n, err = 0;
 
 	if (fs->ref_count == 0) {
+#ifdef CONFIG_PMAC_MEDIABAY
 		if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 			return -ENXIO;
+#endif
 		out_8(&sw->setup, S_IBM_DRIVE | S_FCLK_DIV2);
 		out_8(&sw->control_bic, 0xff);
 		out_8(&sw->mode, 0x95);
@@ -967,8 +973,10 @@ static int floppy_revalidate(struct gendisk *disk)
 	struct swim3 __iomem *sw;
 	int ret, n;
 
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (fs->media_bay && check_media_bay(fs->media_bay, MB_FD))
 		return -ENXIO;
+#endif
 
 	sw = fs->swim3;
 	grab_drive(fs, revalidating, 0);

drivers/block/sx8.c

@@ -26,6 +26,7 @@
 #include <linux/delay.h>
 #include <linux/time.h>
 #include <linux/hdreg.h>
+#include <linux/dma-mapping.h>
 #include <asm/io.h>
 #include <asm/semaphore.h>
 #include <asm/uaccess.h>
@@ -1582,9 +1583,9 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_out;
 
 #if IF_64BIT_DMA_IS_POSSIBLE /* grrrr... */
-	rc = pci_set_dma_mask(pdev, 0xffffffffffffffffULL);
+	rc = pci_set_dma_mask(pdev, DMA_64BIT_MASK);
 	if (!rc) {
-		rc = pci_set_consistent_dma_mask(pdev, 0xffffffffffffffffULL);
+		rc = pci_set_consistent_dma_mask(pdev, DMA_64BIT_MASK);
 		if (rc) {
 			printk(KERN_ERR DRV_NAME "(%s): consistent DMA mask failure\n",
 				pci_name(pdev));
@@ -1593,7 +1594,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		pci_dac = 1;
 	} else {
 #endif
-		rc = pci_set_dma_mask(pdev, 0xffffffffULL);
+		rc = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
 		if (rc) {
 			printk(KERN_ERR DRV_NAME "(%s): DMA mask failure\n",
 				pci_name(pdev));

drivers/char/misc.c

@@ -308,9 +308,6 @@ static int __init misc_init(void)
 #endif
 #ifdef CONFIG_BVME6000
 	rtc_DP8570A_init();
-#endif
-#ifdef CONFIG_PMAC_PBOOK
-	pmu_device_init();
 #endif
 	if (register_chrdev(MISC_MAJOR,"misc",&misc_fops)) {
 		printk("unable to get major %d for misc devices\n",

drivers/ide/ppc/pmac.c

@@ -1324,9 +1324,9 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 	/* XXX FIXME: Media bay stuff need re-organizing */
 	if (np->parent && np->parent->name
 	    && strcasecmp(np->parent->name, "media-bay") == 0) {
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
 		media_bay_set_ide_infos(np->parent, pmif->regbase, pmif->irq, hwif->index);
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
 		pmif->mediabay = 1;
 		if (!bidp)
 			pmif->aapl_bus_id = 1;
@@ -1382,10 +1382,10 @@ pmac_ide_setup_device(pmac_ide_hwif_t *pmif, ide_hwif_t *hwif)
 	       hwif->index, model_name[pmif->kind], pmif->aapl_bus_id,
 	       pmif->mediabay ? " (mediabay)" : "", hwif->irq);
 			
-#ifdef CONFIG_PMAC_PBOOK
+#ifdef CONFIG_PMAC_MEDIABAY
 	if (pmif->mediabay && check_media_bay_by_base(pmif->regbase, MB_CD) == 0)
 		hwif->noprobe = 0;
-#endif /* CONFIG_PMAC_PBOOK */
+#endif /* CONFIG_PMAC_MEDIABAY */
 
 	hwif->sg_max_nents = MAX_DCMDS;
 

drivers/ieee1394/ohci1394.c

@@ -3538,8 +3538,8 @@ static void ohci1394_pci_remove(struct pci_dev *pdev)
 
 static int ohci1394_pci_resume (struct pci_dev *pdev)
 {
-#ifdef CONFIG_PMAC_PBOOK
-	{
+#ifdef CONFIG_PPC_PMAC
+	if (_machine == _MACH_Pmac) {
 		struct device_node *of_node;
 
 		/* Re-enable 1394 */
@@ -3547,7 +3547,7 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
 		if (of_node)
 			pmac_call_feature (PMAC_FTR_1394_ENABLE, of_node, 0, 1);
 	}
-#endif
+#endif /* CONFIG_PPC_PMAC */
 
 	pci_enable_device(pdev);
 
@@ -3557,8 +3557,8 @@ static int ohci1394_pci_resume (struct pci_dev *pdev)
 
 static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state)
 {
-#ifdef CONFIG_PMAC_PBOOK
-	{
+#ifdef CONFIG_PPC_PMAC
+	if (_machine == _MACH_Pmac) {
 		struct device_node *of_node;
 
 		/* Disable 1394 */

drivers/infiniband/core/packer.c

@@ -96,7 +96,7 @@ void ib_pack(const struct ib_field        *desc,
 			else
 				val = 0;
 
-			mask = cpu_to_be64(((1ull << desc[i].size_bits) - 1) << shift);
+			mask = cpu_to_be64((~0ull >> (64 - desc[i].size_bits)) << shift);
 			addr = (__be64 *) ((__be32 *) buf + desc[i].offset_words);
 			*addr = (*addr & ~mask) | (cpu_to_be64(val) & mask);
 		} else {
@@ -176,7 +176,7 @@ void ib_unpack(const struct ib_field        *desc,
 			__be64 *addr;
 
 			shift = 64 - desc[i].offset_bits - desc[i].size_bits;
-			mask = ((1ull << desc[i].size_bits) - 1) << shift;
+			mask = (~0ull >> (64 - desc[i].size_bits)) << shift;
 			addr = (__be64 *) buf + desc[i].offset_words;
 			val = (be64_to_cpup(addr) & mask) >> shift;
 			value_write(desc[i].struct_offset_bytes,

drivers/infiniband/core/sa_query.c

@@ -507,7 +507,13 @@ static int send_mad(struct ib_sa_query *query, int timeout_ms)
 		spin_unlock_irqrestore(&idr_lock, flags);
 	}
 
-	return ret;
+	/*
+	 * It's not safe to dereference query any more, because the
+	 * send may already have completed and freed the query in
+	 * another context.  So use wr.wr_id, which has a copy of the
+	 * query's id.
+	 */
+	return ret ? ret : wr.wr_id;
 }
 
 static void ib_sa_path_rec_callback(struct ib_sa_query *sa_query,
@@ -598,14 +604,15 @@ int ib_sa_path_rec_get(struct ib_device *device, u8 port_num,
 		rec, query->sa_query.mad->data);
 
 	*sa_query = &query->sa_query;
+
 	ret = send_mad(&query->sa_query, timeout_ms);
-	if (ret) {
+	if (ret < 0) {
 		*sa_query = NULL;
 		kfree(query->sa_query.mad);
 		kfree(query);
 	}
 
-	return ret ? ret : query->sa_query.id;
+	return ret;
 }
 EXPORT_SYMBOL(ib_sa_path_rec_get);
 
@@ -674,14 +681,15 @@ int ib_sa_mcmember_rec_query(struct ib_device *device, u8 port_num,
 		rec, query->sa_query.mad->data);
 
 	*sa_query = &query->sa_query;
+
 	ret = send_mad(&query->sa_query, timeout_ms);
-	if (ret) {
+	if (ret < 0) {
 		*sa_query = NULL;
 		kfree(query->sa_query.mad);
 		kfree(query);
 	}
 
-	return ret ? ret : query->sa_query.id;
+	return ret;
 }
 EXPORT_SYMBOL(ib_sa_mcmember_rec_query);
 

drivers/infiniband/hw/mthca/mthca_av.c

 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU

drivers/infiniband/hw/mthca/mthca_doorbell.h

 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
+ * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU

drivers/input/gameport/Makefile

@@ -5,9 +5,7 @@
 # Each configuration option enables a list of files.
 
 obj-$(CONFIG_GAMEPORT)		+= gameport.o
-obj-$(CONFIG_GAMEPORT_CS461X)	+= cs461x.o
 obj-$(CONFIG_GAMEPORT_EMU10K1)	+= emu10k1-gp.o
 obj-$(CONFIG_GAMEPORT_FM801)	+= fm801-gp.o
 obj-$(CONFIG_GAMEPORT_L4)	+= lightning.o
 obj-$(CONFIG_GAMEPORT_NS558)	+= ns558.o
-obj-$(CONFIG_GAMEPORT_VORTEX)	+= vortex.o