Merge branch 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm

* 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm:
  ARM: 7182/1: ARM cpu topology: fix warning
  ARM: 7181/1: Restrict kprobes probing SWP instructions to ARMv5 and below
  ARM: 7180/1: Change kprobes testcase with unpredictable STRD instruction
  ARM: 7177/1: GIC: avoid skipping non-existent PPIs in irq_start calculation
  ARM: 7176/1: cpu_pm: register GIC PM notifier only once
  ARM: 7175/1: add subname parameter to mfp_set_groupg callers
  ARM: 7174/1: Fix build error in kprobes test code on Thumb2 kernels
  ARM: 7172/1: dma: Drop GFP_COMP for DMA memory allocations
  ARM: 7171/1: unwind: add unwind directives to bitops assembly macros
  ARM: 7170/2: fix compilation breakage in entry-armv.S
  ARM: 7168/1: use cache type functions for arch_get_unmapped_area
  ARM: perf: check that we have a platform device when reserving PMU
  ARM: 7166/1: Use PMD_SHIFT instead of PGDIR_SHIFT in dma-consistent.c
  ARM: 7165/2: PL330: Fix typo in _prepare_ccr()
  ARM: 7163/2: PL330: Only register usable channels
  ARM: 7162/1: errata: tidy up Kconfig options for PL310 errata workarounds
  ARM: 7161/1: errata: no automatic store buffer drain
  ARM: perf: initialise used_mask for fake PMU during validation
  ARM: PMU: remove pmu_init declaration
  ARM: PMU: re-export release_pmu symbol to modules

Author: torvalds

arch/arm/Kconfig

@@ -1231,7 +1231,7 @@ config ARM_ERRATA_742231
 	  capabilities of the processor.
 
 config PL310_ERRATA_588369
-	bool "Clean & Invalidate maintenance operations do not invalidate clean lines"
+	bool "PL310 errata: Clean & Invalidate maintenance operations do not invalidate clean lines"
 	depends on CACHE_L2X0
 	help
 	   The PL310 L2 cache controller implements three types of Clean &
@@ -1256,7 +1256,7 @@ config ARM_ERRATA_720789
 	  entries regardless of the ASID.
 
 config PL310_ERRATA_727915
-	bool "Background Clean & Invalidate by Way operation can cause data corruption"
+	bool "PL310 errata: Background Clean & Invalidate by Way operation can cause data corruption"
 	depends on CACHE_L2X0
 	help
 	  PL310 implements the Clean & Invalidate by Way L2 cache maintenance
@@ -1289,8 +1289,8 @@ config ARM_ERRATA_751472
 	  operation is received by a CPU before the ICIALLUIS has completed,
 	  potentially leading to corrupted entries in the cache or TLB.
 
-config ARM_ERRATA_753970
-	bool "ARM errata: cache sync operation may be faulty"
+config PL310_ERRATA_753970
+	bool "PL310 errata: cache sync operation may be faulty"
 	depends on CACHE_PL310
 	help
 	  This option enables the workaround for the 753970 PL310 (r3p0) erratum.
@@ -1352,6 +1352,18 @@ config ARM_ERRATA_764369
 	  relevant cache maintenance functions and sets a specific bit
 	  in the diagnostic control register of the SCU.
 
+config PL310_ERRATA_769419
+	bool "PL310 errata: no automatic Store Buffer drain"
+	depends on CACHE_L2X0
+	help
+	  On revisions of the PL310 prior to r3p2, the Store Buffer does
+	  not automatically drain. This can cause normal, non-cacheable
+	  writes to be retained when the memory system is idle, leading
+	  to suboptimal I/O performance for drivers using coherent DMA.
+	  This option adds a write barrier to the cpu_idle loop so that,
+	  on systems with an outer cache, the store buffer is drained
+	  explicitly.
+
 endmenu
 
 source "arch/arm/common/Kconfig"

arch/arm/common/gic.c

@@ -526,7 +526,8 @@ static void __init gic_pm_init(struct gic_chip_data *gic)
 		sizeof(u32));
 	BUG_ON(!gic->saved_ppi_conf);
 
-	cpu_pm_register_notifier(&gic_notifier_block);
+	if (gic == &gic_data[0])
+		cpu_pm_register_notifier(&gic_notifier_block);
 }
 #else
 static void __init gic_pm_init(struct gic_chip_data *gic)
@@ -581,13 +582,16 @@ void __init gic_init(unsigned int gic_nr, int irq_start,
 	 * For primary GICs, skip over SGIs.
 	 * For secondary GICs, skip over PPIs, too.
 	 */
+	domain->hwirq_base = 32;
 	if (gic_nr == 0) {
 		gic_cpu_base_addr = cpu_base;
-		domain->hwirq_base = 16;
-		if (irq_start > 0)
-			irq_start = (irq_start & ~31) + 16;
-	} else
-		domain->hwirq_base = 32;
+
+		if ((irq_start & 31) > 0) {
+			domain->hwirq_base = 16;
+			if (irq_start != -1)
+				irq_start = (irq_start & ~31) + 16;
+		}
+	}
 
 	/*
 	 * Find out how many interrupts are supported.

arch/arm/common/pl330.c

@@ -1211,8 +1211,8 @@ static inline u32 _prepare_ccr(const struct pl330_reqcfg *rqc)
 	ccr |= (rqc->brst_size << CC_SRCBRSTSIZE_SHFT);
 	ccr |= (rqc->brst_size << CC_DSTBRSTSIZE_SHFT);
 
-	ccr |= (rqc->dcctl << CC_SRCCCTRL_SHFT);
-	ccr |= (rqc->scctl << CC_DSTCCTRL_SHFT);
+	ccr |= (rqc->scctl << CC_SRCCCTRL_SHFT);
+	ccr |= (rqc->dcctl << CC_DSTCCTRL_SHFT);
 
 	ccr |= (rqc->swap << CC_SWAP_SHFT);
 
@@ -1623,6 +1623,11 @@ static inline int _alloc_event(struct pl330_thread *thrd)
 	return -1;
 }
 
+static bool _chan_ns(const struct pl330_info *pi, int i)
+{
+	return pi->pcfg.irq_ns & (1 << i);
+}
+
 /* Upon success, returns IdentityToken for the
  * allocated channel, NULL otherwise.
  */
@@ -1647,7 +1652,8 @@ void *pl330_request_channel(const struct pl330_info *pi)
 
 	for (i = 0; i < chans; i++) {
 		thrd = &pl330->channels[i];
-		if (thrd->free) {
+		if ((thrd->free) && (!_manager_ns(thrd) ||
+					_chan_ns(pi, i))) {
 			thrd->ev = _alloc_event(thrd);
 			if (thrd->ev >= 0) {
 				thrd->free = false;

arch/arm/include/asm/pmu.h

@@ -55,16 +55,6 @@ reserve_pmu(enum arm_pmu_type type);
 extern void
 release_pmu(enum arm_pmu_type type);
 
-/**
- * init_pmu() - Initialise the PMU.
- *
- * Initialise the system ready for PMU enabling. This should typically set the
- * IRQ affinity and nothing else. The users (oprofile/perf events etc) will do
- * the actual hardware initialisation.
- */
-extern int
-init_pmu(enum arm_pmu_type type);
-
 #else /* CONFIG_CPU_HAS_PMU */
 
 #include <linux/err.h>

arch/arm/include/asm/topology.h

@@ -25,7 +25,7 @@ extern struct cputopo_arm cpu_topology[NR_CPUS];
 
 void init_cpu_topology(void);
 void store_cpu_topology(unsigned int cpuid);
-const struct cpumask *cpu_coregroup_mask(unsigned int cpu);
+const struct cpumask *cpu_coregroup_mask(int cpu);
 
 #else
 

arch/arm/kernel/entry-armv.S

@@ -497,7 +497,7 @@ ENDPROC(__und_usr)
 	.popsection
 	.pushsection __ex_table,"a"
 	.long	1b, 4b
-#if __LINUX_ARM_ARCH__ >= 7
+#if CONFIG_ARM_THUMB && __LINUX_ARM_ARCH__ >= 6 && CONFIG_CPU_V7
 	.long	2b, 4b
 	.long	3b, 4b
 #endif

arch/arm/kernel/kprobes-arm.c

@@ -519,10 +519,12 @@ static const union decode_item arm_cccc_0000_____1001_table[] = {
 static const union decode_item arm_cccc_0001_____1001_table[] = {
 	/* Synchronization primitives					*/
 
+#if __LINUX_ARM_ARCH__ < 6
+	/* Deprecated on ARMv6 and may be UNDEFINED on v7		*/
 	/* SMP/SWPB		cccc 0001 0x00 xxxx xxxx xxxx 1001 xxxx */
 	DECODE_EMULATEX	(0x0fb000f0, 0x01000090, emulate_rd12rn16rm0_rwflags_nopc,
 						 REGS(NOPC, NOPC, 0, 0, NOPC)),
-
+#endif
 	/* LDREX/STREX{,D,B,H}	cccc 0001 1xxx xxxx xxxx xxxx 1001 xxxx */
 	/* And unallocated instructions...				*/
 	DECODE_END

arch/arm/kernel/kprobes-test-arm.c

@@ -427,18 +427,25 @@ void kprobe_arm_test_cases(void)
 
 	TEST_GROUP("Synchronization primitives")
 
-	/*
-	 * Use hard coded constants for SWP instructions to avoid warnings
-	 * about deprecated instructions.
-	 */
-	TEST_RP( ".word 0xe108e097 @ swp	lr, r",7,VAL2,", [r",8,0,"]")
-	TEST_R(  ".word 0x610d0091 @ swpvs	r0, r",1,VAL1,", [sp]")
-	TEST_RP( ".word 0xe10cd09e @ swp	sp, r",14,VAL2,", [r",12,13*4,"]")
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swp	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvs	r0, r",1,VAL1,", [sp]")
+	TEST_RP("swp	sp, r",14,VAL2,", [r",12,13*4,"]")
+#else
+	TEST_UNSUPPORTED(".word 0xe108e097 @ swp	lr, r7, [r8]")
+	TEST_UNSUPPORTED(".word 0x610d0091 @ swpvs	r0, r1, [sp]")
+	TEST_UNSUPPORTED(".word 0xe10cd09e @ swp	sp, r14 [r12]")
+#endif
 	TEST_UNSUPPORTED(".word 0xe102f091 @ swp pc, r1, [r2]")
 	TEST_UNSUPPORTED(".word 0xe102009f @ swp r0, pc, [r2]")
 	TEST_UNSUPPORTED(".word 0xe10f0091 @ swp r0, r1, [pc]")
-	TEST_RP( ".word 0xe148e097 @ swpb	lr, r",7,VAL2,", [r",8,0,"]")
-	TEST_R(  ".word 0x614d0091 @ swpvsb	r0, r",1,VAL1,", [sp]")
+#if __LINUX_ARM_ARCH__ < 6
+	TEST_RP("swpb	lr, r",7,VAL2,", [r",8,0,"]")
+	TEST_R( "swpvsb	r0, r",1,VAL1,", [sp]")
+#else
+	TEST_UNSUPPORTED(".word 0xe148e097 @ swpb	lr, r7, [r8]")
+	TEST_UNSUPPORTED(".word 0x614d0091 @ swpvsb	r0, r1, [sp]")
+#endif
 	TEST_UNSUPPORTED(".word 0xe142f091 @ swpb pc, r1, [r2]")
 
 	TEST_UNSUPPORTED(".word	0xe1100090") /* Unallocated space */
@@ -550,7 +557,7 @@ void kprobe_arm_test_cases(void)
 	TEST_RPR(  "strccd	r",8, VAL2,", [r",13,0, ", r",12,48,"]")
 	TEST_RPR(  "strd	r",4, VAL1,", [r",2, 24,", r",3, 48,"]!")
 	TEST_RPR(  "strcsd	r",12,VAL2,", [r",11,48,", -r",10,24,"]!")
-	TEST_RPR(  "strd	r",2, VAL1,", [r",3, 24,"], r",4,48,"")
+	TEST_RPR(  "strd	r",2, VAL1,", [r",5, 24,"], r",4,48,"")
 	TEST_RPR(  "strd	r",10,VAL2,", [r",9, 48,"], -r",7,24,"")
 	TEST_UNSUPPORTED(".word 0xe1afc0fa	@ strd r12, [pc, r10]!")
 

arch/arm/kernel/kprobes-test-thumb.c

@@ -222,8 +222,8 @@ void kprobe_thumb16_test_cases(void)
 DONT_TEST_IN_ITBLOCK(
 	TEST_BF_R(  "cbnz	r",0,0, ", 2f")
 	TEST_BF_R(  "cbz	r",2,-1,", 2f")
-	TEST_BF_RX( "cbnz	r",4,1, ", 2f",0x20)
-	TEST_BF_RX( "cbz	r",7,0, ", 2f",0x40)
+	TEST_BF_RX( "cbnz	r",4,1, ", 2f", SPACE_0x20)
+	TEST_BF_RX( "cbz	r",7,0, ", 2f", SPACE_0x40)
 )
 	TEST_R("sxth	r0, r",7, HH1,"")
 	TEST_R("sxth	r7, r",0, HH2,"")
@@ -246,7 +246,7 @@ DONT_TEST_IN_ITBLOCK(
 	TESTCASE_START(code)		\
 	TEST_ARG_PTR(13, offset)	\
 	TEST_ARG_END("")		\
-	TEST_BRANCH_F(code,0)		\
+	TEST_BRANCH_F(code)		\
 	TESTCASE_END
 
 	TEST("push	{r0}")
@@ -319,8 +319,8 @@ CONDITION_INSTRUCTIONS(8,
 
 	TEST_BF(  "b	2f")
 	TEST_BB(  "b	2b")
-	TEST_BF_X("b	2f", 0x400)
-	TEST_BB_X("b	2b", 0x400)
+	TEST_BF_X("b	2f", SPACE_0x400)
+	TEST_BB_X("b	2b", SPACE_0x400)
 
 	TEST_GROUP("Testing instructions in IT blocks")
 
@@ -746,7 +746,7 @@ CONDITION_INSTRUCTIONS(22,
 	TEST_BB("bne.w	2b")
 	TEST_BF("bgt.w	2f")
 	TEST_BB("blt.w	2b")
-	TEST_BF_X("bpl.w	2f",0x1000)
+	TEST_BF_X("bpl.w	2f", SPACE_0x1000)
 )
 
 	TEST_UNSUPPORTED("msr	cpsr, r0")
@@ -786,11 +786,11 @@ CONDITION_INSTRUCTIONS(22,
 
 	TEST_BF(  "b.w	2f")
 	TEST_BB(  "b.w	2b")
-	TEST_BF_X("b.w	2f", 0x1000)
+	TEST_BF_X("b.w	2f", SPACE_0x1000)
 
 	TEST_BF(  "bl.w	2f")
 	TEST_BB(  "bl.w	2b")
-	TEST_BB_X("bl.w	2b", 0x1000)
+	TEST_BB_X("bl.w	2b", SPACE_0x1000)
 
 	TEST_X(	"blx	__dummy_arm_subroutine",
 		".arm				\n\t"