Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc

Pull powerpc updates from Ben Herrenschmidt:
 "This is the powerpc new goodies for 3.17.  The short story:

  The biggest bit is Michael removing all of pre-POWER4 processor
  support from the 64-bit kernel.  POWER3 and rs64.  This gets rid of a
  ton of old cruft that has been bitrotting in a long while.  It was
  broken for quite a few versions already and nobody noticed.  Nobody
  uses those machines anymore.  While at it, he cleaned up a bunch of
  old dusty cabinets, getting rid of a skeletton or two.

  Then, we have some base VFIO support for KVM, which allows assigning
  of PCI devices to KVM guests, support for large 64-bit BARs on
  "powernv" platforms, support for HMI (Hardware Management Interrupts)
  on those same platforms, some sparse-vmemmap improvements (for memory
  hotplug),

  There is the usual batch of Freescale embedded updates (summary in the
  merge commit) and fixes here or there, I think that's it for the
  highlights"

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (102 commits)
  powerpc/eeh: Export eeh_iommu_group_to_pe()
  powerpc/eeh: Add missing #ifdef CONFIG_IOMMU_API
  powerpc: Reduce scariness of interrupt frames in stack traces
  powerpc: start loop at section start of start in vmemmap_populated()
  powerpc: implement vmemmap_free()
  powerpc: implement vmemmap_remove_mapping() for BOOK3S
  powerpc: implement vmemmap_list_free()
  powerpc: Fail remap_4k_pfn() if PFN doesn't fit inside PTE
  powerpc/book3s: Fix endianess issue for HMI handling on napping cpus.
  powerpc/book3s: handle HMIs for cpus in nap mode.
  powerpc/powernv: Invoke opal call to handle hmi.
  powerpc/book3s: Add basic infrastructure to handle HMI in Linux.
  powerpc/iommu: Fix comments with it_page_shift
  powerpc/powernv: Handle compound PE in config accessors
  powerpc/powernv: Handle compound PE for EEH
  powerpc/powernv: Handle compound PE
  powerpc/powernv: Split ioda_eeh_get_state()
  powerpc/powernv: Allow to freeze PE
  powerpc/powernv: Enable M64 aperatus for PHB3
  powerpc/eeh: Aux PE data for error log
  ...

9 files changed, 82 insertions, 35 deletions

diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c
index fe52db2eea6a..b7cd00b0171e 100644
--- a/arch/powerpc/perf/core-book3s.c
+++ b/arch/powerpc/perf/core-book3s.c
@@ -36,7 +36,12 @@ struct cpu_hw_events {
 	struct perf_event *event[MAX_HWEVENTS];
 	u64 events[MAX_HWEVENTS];
 	unsigned int flags[MAX_HWEVENTS];
-	unsigned long mmcr[3];
+	/*
+	 * The order of the MMCR array is:
+	 *  - 64-bit, MMCR0, MMCR1, MMCRA, MMCR2
+	 *  - 32-bit, MMCR0, MMCR1, MMCR2
+	 */
+	unsigned long mmcr[4];
 	struct perf_event *limited_counter[MAX_LIMITED_HWCOUNTERS];
 	u8  limited_hwidx[MAX_LIMITED_HWCOUNTERS];
 	u64 alternatives[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
@@ -112,9 +117,9 @@ static bool is_ebb_event(struct perf_event *event) { return false; }
 static int ebb_event_check(struct perf_event *event) { return 0; }
 static void ebb_event_add(struct perf_event *event) { }
 static void ebb_switch_out(unsigned long mmcr0) { }
-static unsigned long ebb_switch_in(bool ebb, unsigned long mmcr0)
+static unsigned long ebb_switch_in(bool ebb, struct cpu_hw_events *cpuhw)
 {
-	return mmcr0;
+	return cpuhw->mmcr[0];
 }
 
 static inline void power_pmu_bhrb_enable(struct perf_event *event) {}
@@ -542,8 +547,10 @@ static void ebb_switch_out(unsigned long mmcr0)
 	current->thread.mmcr2 = mfspr(SPRN_MMCR2) & MMCR2_USER_MASK;
 }
 
-static unsigned long ebb_switch_in(bool ebb, unsigned long mmcr0)
+static unsigned long ebb_switch_in(bool ebb, struct cpu_hw_events *cpuhw)
 {
+	unsigned long mmcr0 = cpuhw->mmcr[0];
+
 	if (!ebb)
 		goto out;
 
@@ -568,7 +575,15 @@ static unsigned long ebb_switch_in(bool ebb, unsigned long mmcr0)
 	mtspr(SPRN_SIAR, current->thread.siar);
 	mtspr(SPRN_SIER, current->thread.sier);
 	mtspr(SPRN_SDAR, current->thread.sdar);
-	mtspr(SPRN_MMCR2, current->thread.mmcr2);
+
+	/*
+	 * Merge the kernel & user values of MMCR2. The semantics we implement
+	 * are that the user MMCR2 can set bits, ie. cause counters to freeze,
+	 * but not clear bits. If a task wants to be able to clear bits, ie.
+	 * unfreeze counters, it should not set exclude_xxx in its events and
+	 * instead manage the MMCR2 entirely by itself.
+	 */
+	mtspr(SPRN_MMCR2, cpuhw->mmcr[3] | current->thread.mmcr2);
 out:
 	return mmcr0;
 }
@@ -915,6 +930,14 @@ static int check_excludes(struct perf_event **ctrs, unsigned int cflags[],
 	int i, n, first;
 	struct perf_event *event;
 
+	/*
+	 * If the PMU we're on supports per event exclude settings then we
+	 * don't need to do any of this logic. NB. This assumes no PMU has both
+	 * per event exclude and limited PMCs.
+	 */
+	if (ppmu->flags & PPMU_ARCH_207S)
+		return 0;
+
 	n = n_prev + n_new;
 	if (n <= 1)
 		return 0;
@@ -1219,28 +1242,31 @@ static void power_pmu_enable(struct pmu *pmu)
 	}
 
 	/*
-	 * Compute MMCR* values for the new set of events
+	 * Clear all MMCR settings and recompute them for the new set of events.
 	 */
+	memset(cpuhw->mmcr, 0, sizeof(cpuhw->mmcr));
+
 	if (ppmu->compute_mmcr(cpuhw->events, cpuhw->n_events, hwc_index,
-			       cpuhw->mmcr)) {
+			       cpuhw->mmcr, cpuhw->event)) {
 		/* shouldn't ever get here */
 		printk(KERN_ERR "oops compute_mmcr failed\n");
 		goto out;
 	}
 
-	/*
-	 * Add in MMCR0 freeze bits corresponding to the
-	 * attr.exclude_* bits for the first event.
-	 * We have already checked that all events have the
-	 * same values for these bits as the first event.
-	 */
-	event = cpuhw->event[0];
-	if (event->attr.exclude_user)
-		cpuhw->mmcr[0] |= MMCR0_FCP;
-	if (event->attr.exclude_kernel)
-		cpuhw->mmcr[0] |= freeze_events_kernel;
-	if (event->attr.exclude_hv)
-		cpuhw->mmcr[0] |= MMCR0_FCHV;
+	if (!(ppmu->flags & PPMU_ARCH_207S)) {
+		/*
+		 * Add in MMCR0 freeze bits corresponding to the attr.exclude_*
+		 * bits for the first event. We have already checked that all
+		 * events have the same value for these bits as the first event.
+		 */
+		event = cpuhw->event[0];
+		if (event->attr.exclude_user)
+			cpuhw->mmcr[0] |= MMCR0_FCP;
+		if (event->attr.exclude_kernel)
+			cpuhw->mmcr[0] |= freeze_events_kernel;
+		if (event->attr.exclude_hv)
+			cpuhw->mmcr[0] |= MMCR0_FCHV;
+	}
 
 	/*
 	 * Write the new configuration to MMCR* with the freeze
@@ -1252,6 +1278,8 @@ static void power_pmu_enable(struct pmu *pmu)
 	mtspr(SPRN_MMCR1, cpuhw->mmcr[1]);
 	mtspr(SPRN_MMCR0, (cpuhw->mmcr[0] & ~(MMCR0_PMC1CE | MMCR0_PMCjCE))
 				| MMCR0_FC);
+	if (ppmu->flags & PPMU_ARCH_207S)
+		mtspr(SPRN_MMCR2, cpuhw->mmcr[3]);
 
 	/*
 	 * Read off any pre-existing events that need to move
@@ -1307,10 +1335,7 @@ static void power_pmu_enable(struct pmu *pmu)
  out_enable:
 	pmao_restore_workaround(ebb);
 
-	if (ppmu->flags & PPMU_ARCH_207S)
-		mtspr(SPRN_MMCR2, 0);
-
-	mmcr0 = ebb_switch_in(ebb, cpuhw->mmcr[0]);
+	mmcr0 = ebb_switch_in(ebb, cpuhw);
 
 	mb();
 	if (cpuhw->bhrb_users)
diff --git a/arch/powerpc/perf/mpc7450-pmu.c b/arch/powerpc/perf/mpc7450-pmu.c
index fe21b515ca44..d115c5635bf3 100644
--- a/arch/powerpc/perf/mpc7450-pmu.c
+++ b/arch/powerpc/perf/mpc7450-pmu.c
@@ -260,8 +260,9 @@ static const u32 pmcsel_mask[N_COUNTER] = {
 /*
  * Compute MMCR0/1/2 values for a set of events.
  */
-static int mpc7450_compute_mmcr(u64 event[], int n_ev,
-				unsigned int hwc[], unsigned long mmcr[])
+static int mpc7450_compute_mmcr(u64 event[], int n_ev, unsigned int hwc[],
+				unsigned long mmcr[],
+				struct perf_event *pevents[])
 {
 	u8 event_index[N_CLASSES][N_COUNTER];
 	int n_classevent[N_CLASSES];
diff --git a/arch/powerpc/perf/power4-pmu.c b/arch/powerpc/perf/power4-pmu.c
index 9103a1de864d..ce6072fa481b 100644
--- a/arch/powerpc/perf/power4-pmu.c
+++ b/arch/powerpc/perf/power4-pmu.c
@@ -356,7 +356,7 @@ static int p4_get_alternatives(u64 event, unsigned int flags, u64 alt[])
 }
 
 static int p4_compute_mmcr(u64 event[], int n_ev,
-			   unsigned int hwc[], unsigned long mmcr[])
+			   unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr0 = 0, mmcr1 = 0, mmcra = 0;
 	unsigned int pmc, unit, byte, psel, lower;
diff --git a/arch/powerpc/perf/power5+-pmu.c b/arch/powerpc/perf/power5+-pmu.c
index b03b6dc0172d..0526dac66007 100644
--- a/arch/powerpc/perf/power5+-pmu.c
+++ b/arch/powerpc/perf/power5+-pmu.c
@@ -452,7 +452,7 @@ static int power5p_marked_instr_event(u64 event)
 }
 
 static int power5p_compute_mmcr(u64 event[], int n_ev,
-				unsigned int hwc[], unsigned long mmcr[])
+				unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr1 = 0;
 	unsigned long mmcra = 0;
diff --git a/arch/powerpc/perf/power5-pmu.c b/arch/powerpc/perf/power5-pmu.c
index 1e8ce423c3af..4dc99f9f7962 100644
--- a/arch/powerpc/perf/power5-pmu.c
+++ b/arch/powerpc/perf/power5-pmu.c
@@ -383,7 +383,7 @@ static int power5_marked_instr_event(u64 event)
 }
 
 static int power5_compute_mmcr(u64 event[], int n_ev,
-			       unsigned int hwc[], unsigned long mmcr[])
+			       unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr1 = 0;
 	unsigned long mmcra = MMCRA_SDAR_DCACHE_MISS | MMCRA_SDAR_ERAT_MISS;
diff --git a/arch/powerpc/perf/power6-pmu.c b/arch/powerpc/perf/power6-pmu.c
index 31128e086fed..9c9d646b68a1 100644
--- a/arch/powerpc/perf/power6-pmu.c
+++ b/arch/powerpc/perf/power6-pmu.c
@@ -175,7 +175,7 @@ static int power6_marked_instr_event(u64 event)
  * Assign PMC numbers and compute MMCR1 value for a set of events
  */
 static int p6_compute_mmcr(u64 event[], int n_ev,
-			   unsigned int hwc[], unsigned long mmcr[])
+			   unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr1 = 0;
 	unsigned long mmcra = MMCRA_SDAR_DCACHE_MISS | MMCRA_SDAR_ERAT_MISS;
diff --git a/arch/powerpc/perf/power7-pmu.c b/arch/powerpc/perf/power7-pmu.c
index 56c67bca2f75..5b62f2389290 100644
--- a/arch/powerpc/perf/power7-pmu.c
+++ b/arch/powerpc/perf/power7-pmu.c
@@ -245,7 +245,7 @@ static int power7_marked_instr_event(u64 event)
 }
 
 static int power7_compute_mmcr(u64 event[], int n_ev,
-			       unsigned int hwc[], unsigned long mmcr[])
+			       unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr1 = 0;
 	unsigned long mmcra = MMCRA_SDAR_DCACHE_MISS | MMCRA_SDAR_ERAT_MISS;
diff --git a/arch/powerpc/perf/power8-pmu.c b/arch/powerpc/perf/power8-pmu.c
index 639cd9156585..396351db601b 100644
--- a/arch/powerpc/perf/power8-pmu.c
+++ b/arch/powerpc/perf/power8-pmu.c
@@ -15,6 +15,7 @@
 #include <linux/kernel.h>
 #include <linux/perf_event.h>
 #include <asm/firmware.h>
+#include <asm/cputable.h>
 
 
 /*
@@ -266,6 +267,11 @@
 #define MMCRA_SDAR_MODE_TLB		(1ull << 42)
 #define MMCRA_IFM_SHIFT			30
 
+/* Bits in MMCR2 for POWER8 */
+#define MMCR2_FCS(pmc)			(1ull << (63 - (((pmc) - 1) * 9)))
+#define MMCR2_FCP(pmc)			(1ull << (62 - (((pmc) - 1) * 9)))
+#define MMCR2_FCH(pmc)			(1ull << (57 - (((pmc) - 1) * 9)))
+
 
 static inline bool event_is_fab_match(u64 event)
 {
@@ -393,9 +399,10 @@ static int power8_get_constraint(u64 event, unsigned long *maskp, unsigned long
 }
 
 static int power8_compute_mmcr(u64 event[], int n_ev,
-			       unsigned int hwc[], unsigned long mmcr[])
+			       unsigned int hwc[], unsigned long mmcr[],
+			       struct perf_event *pevents[])
 {
-	unsigned long mmcra, mmcr1, unit, combine, psel, cache, val;
+	unsigned long mmcra, mmcr1, mmcr2, unit, combine, psel, cache, val;
 	unsigned int pmc, pmc_inuse;
 	int i;
 
@@ -410,7 +417,7 @@ static int power8_compute_mmcr(u64 event[], int n_ev,
 
 	/* In continous sampling mode, update SDAR on TLB miss */
 	mmcra = MMCRA_SDAR_MODE_TLB;
-	mmcr1 = 0;
+	mmcr1 = mmcr2 = 0;
 
 	/* Second pass: assign PMCs, set all MMCR1 fields */
 	for (i = 0; i < n_ev; ++i) {
@@ -472,6 +479,19 @@ static int power8_compute_mmcr(u64 event[], int n_ev,
 			mmcra |= val << MMCRA_IFM_SHIFT;
 		}
 
+		if (pevents[i]->attr.exclude_user)
+			mmcr2 |= MMCR2_FCP(pmc);
+
+		if (pevents[i]->attr.exclude_hv)
+			mmcr2 |= MMCR2_FCH(pmc);
+
+		if (pevents[i]->attr.exclude_kernel) {
+			if (cpu_has_feature(CPU_FTR_HVMODE))
+				mmcr2 |= MMCR2_FCH(pmc);
+			else
+				mmcr2 |= MMCR2_FCS(pmc);
+		}
+
 		hwc[i] = pmc - 1;
 	}
 
@@ -491,6 +511,7 @@ static int power8_compute_mmcr(u64 event[], int n_ev,
 
 	mmcr[1] = mmcr1;
 	mmcr[2] = mmcra;
+	mmcr[3] = mmcr2;
 
 	return 0;
 }
diff --git a/arch/powerpc/perf/ppc970-pmu.c b/arch/powerpc/perf/ppc970-pmu.c
index 20139ceeacf6..8b6a8a36fa38 100644
--- a/arch/powerpc/perf/ppc970-pmu.c
+++ b/arch/powerpc/perf/ppc970-pmu.c
@@ -257,7 +257,7 @@ static int p970_get_alternatives(u64 event, unsigned int flags, u64 alt[])
 }
 
 static int p970_compute_mmcr(u64 event[], int n_ev,
-			     unsigned int hwc[], unsigned long mmcr[])
+			     unsigned int hwc[], unsigned long mmcr[], struct perf_event *pevents[])
 {
 	unsigned long mmcr0 = 0, mmcr1 = 0, mmcra = 0;
 	unsigned int pmc, unit, byte, psel;