Merge tag 'metag-for-v3.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag

Pull arch/metag update from James Hogan:

 - Various fixes for the interrupting perf counter handling in metag's
   perf backend.

 - Add OProfile support based on perf.

 - Sets up cache partitions for SMP so bootloader doesn't have to.

 - Patch from Paul Bolle to remove ARCH_POPULATES_NODE_MAP again
   (touches microblaze too).

 - Add TLS pointer regset to metag ptrace api.

 - Add exported metag DSP extended context handling header <asm/ech.h>.

 - Increase defconfig log buffer size to 128KiB.

 - Various fixes, typos, missing exports.

* tag 'metag-for-v3.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag:
  metag: defconfigs: increase log buffer 8KiB => 128KiB
  metag: avoid unnecessary builtin dtb rebuilds
  metag: add exported <asm/ech.h> for extended context handling
  metag: export _metag_da_present and cpu_2_hwthread_id
  metag: ptrace: Implement NT_METAG_TLS
  memblock: Kill ARCH_POPULATES_NODE_MAP once more
  metag: cachepart: fix get_global_dcache_size() typo
  metag: cachepart: take into account small cache bits
  metag: smp: copy cache partition and enable GCOn
  metag: OProfile support
  metag: perf: prepare for use by oprofile
  metag: perf: don't reset TXTACTCYC
  metag: perf: use hard_processor_id() to get thread
  metag: perf: fix frequency sampling (dynamic period)
  metag: perf: add missing prev_count updates
  metag: perf: fixes for interrupting perf counters
  metag: perf: fix wrap handling in delta calculation
  metag: perf: fix core internal / perf channel mux

1 files changed, 115 insertions, 0 deletions

diff --git a/arch/metag/kernel/smp.c b/arch/metag/kernel/smp.c
index 4de8fc8e31a5..f443ec9a7cbe 100644
--- a/arch/metag/kernel/smp.c
+++ b/arch/metag/kernel/smp.c
@@ -28,6 +28,8 @@
 #include <asm/cachepart.h>
 #include <asm/core_reg.h>
 #include <asm/cpu.h>
+#include <asm/global_lock.h>
+#include <asm/metag_mem.h>
 #include <asm/mmu_context.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -37,6 +39,9 @@
 #include <asm/hwthread.h>
 #include <asm/traps.h>
 
+#define SYSC_DCPART(n)	(SYSC_DCPART0 + SYSC_xCPARTn_STRIDE * (n))
+#define SYSC_ICPART(n)	(SYSC_ICPART0 + SYSC_xCPARTn_STRIDE * (n))
+
 DECLARE_PER_CPU(PTBI, pTBI);
 
 void *secondary_data_stack;
@@ -99,6 +104,114 @@ int __cpuinit boot_secondary(unsigned int thread, struct task_struct *idle)
 	return 0;
 }
 
+/**
+ * describe_cachepart_change: describe a change to cache partitions.
+ * @thread:	Hardware thread number.
+ * @label:	Label of cache type, e.g. "dcache" or "icache".
+ * @sz:		Total size of the cache.
+ * @old:	Old cache partition configuration (*CPART* register).
+ * @new:	New cache partition configuration (*CPART* register).
+ *
+ * If the cache partition has changed, prints a message to the log describing
+ * those changes.
+ */
+static __cpuinit void describe_cachepart_change(unsigned int thread,
+						const char *label,
+						unsigned int sz,
+						unsigned int old,
+						unsigned int new)
+{
+	unsigned int lor1, land1, gor1, gand1;
+	unsigned int lor2, land2, gor2, gand2;
+	unsigned int diff = old ^ new;
+
+	if (!diff)
+		return;
+
+	pr_info("Thread %d: %s partition changed:", thread, label);
+	if (diff & (SYSC_xCPARTL_OR_BITS | SYSC_xCPARTL_AND_BITS)) {
+		lor1   = (old & SYSC_xCPARTL_OR_BITS)  >> SYSC_xCPARTL_OR_S;
+		lor2   = (new & SYSC_xCPARTL_OR_BITS)  >> SYSC_xCPARTL_OR_S;
+		land1  = (old & SYSC_xCPARTL_AND_BITS) >> SYSC_xCPARTL_AND_S;
+		land2  = (new & SYSC_xCPARTL_AND_BITS) >> SYSC_xCPARTL_AND_S;
+		pr_cont(" L:%#x+%#x->%#x+%#x",
+			(lor1 * sz) >> 4,
+			((land1 + 1) * sz) >> 4,
+			(lor2 * sz) >> 4,
+			((land2 + 1) * sz) >> 4);
+	}
+	if (diff & (SYSC_xCPARTG_OR_BITS | SYSC_xCPARTG_AND_BITS)) {
+		gor1   = (old & SYSC_xCPARTG_OR_BITS)  >> SYSC_xCPARTG_OR_S;
+		gor2   = (new & SYSC_xCPARTG_OR_BITS)  >> SYSC_xCPARTG_OR_S;
+		gand1  = (old & SYSC_xCPARTG_AND_BITS) >> SYSC_xCPARTG_AND_S;
+		gand2  = (new & SYSC_xCPARTG_AND_BITS) >> SYSC_xCPARTG_AND_S;
+		pr_cont(" G:%#x+%#x->%#x+%#x",
+			(gor1 * sz) >> 4,
+			((gand1 + 1) * sz) >> 4,
+			(gor2 * sz) >> 4,
+			((gand2 + 1) * sz) >> 4);
+	}
+	if (diff & SYSC_CWRMODE_BIT)
+		pr_cont(" %sWR",
+			(new & SYSC_CWRMODE_BIT) ? "+" : "-");
+	if (diff & SYSC_DCPART_GCON_BIT)
+		pr_cont(" %sGCOn",
+			(new & SYSC_DCPART_GCON_BIT) ? "+" : "-");
+	pr_cont("\n");
+}
+
+/**
+ * setup_smp_cache: ensure cache coherency for new SMP thread.
+ * @thread:	New hardware thread number.
+ *
+ * Ensures that coherency is enabled and that the threads share the same cache
+ * partitions.
+ */
+static __cpuinit void setup_smp_cache(unsigned int thread)
+{
+	unsigned int this_thread, lflags;
+	unsigned int dcsz, dcpart_this, dcpart_old, dcpart_new;
+	unsigned int icsz, icpart_old, icpart_new;
+
+	/*
+	 * Copy over the current thread's cache partition configuration to the
+	 * new thread so that they share cache partitions.
+	 */
+	__global_lock2(lflags);
+	this_thread = hard_processor_id();
+	/* Share dcache partition */
+	dcpart_this = metag_in32(SYSC_DCPART(this_thread));
+	dcpart_old = metag_in32(SYSC_DCPART(thread));
+	dcpart_new = dcpart_this;
+#if PAGE_OFFSET < LINGLOBAL_BASE
+	/*
+	 * For the local data cache to be coherent the threads must also have
+	 * GCOn enabled.
+	 */
+	dcpart_new |= SYSC_DCPART_GCON_BIT;
+	metag_out32(dcpart_new, SYSC_DCPART(this_thread));
+#endif
+	metag_out32(dcpart_new, SYSC_DCPART(thread));
+	/* Share icache partition too */
+	icpart_new = metag_in32(SYSC_ICPART(this_thread));
+	icpart_old = metag_in32(SYSC_ICPART(thread));
+	metag_out32(icpart_new, SYSC_ICPART(thread));
+	__global_unlock2(lflags);
+
+	/*
+	 * Log if the cache partitions were altered so the user is aware of any
+	 * potential unintentional cache wastage.
+	 */
+	dcsz = get_dcache_size();
+	icsz = get_dcache_size();
+	describe_cachepart_change(this_thread, "dcache", dcsz,
+				  dcpart_this, dcpart_new);
+	describe_cachepart_change(thread, "dcache", dcsz,
+				  dcpart_old, dcpart_new);
+	describe_cachepart_change(thread, "icache", icsz,
+				  icpart_old, icpart_new);
+}
+
 int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle)
 {
 	unsigned int thread = cpu_2_hwthread_id[cpu];
@@ -108,6 +221,8 @@ int __cpuinit __cpu_up(unsigned int cpu, struct task_struct *idle)
 
 	flush_tlb_all();
 
+	setup_smp_cache(thread);
+
 	/*
 	 * Tell the secondary CPU where to find its idle thread's stack.
 	 */