6 files changed, 69 insertions, 22 deletions

diff --git a/arch/microblaze/kernel/cpu/cpuinfo.c b/arch/microblaze/kernel/cpu/cpuinfo.c
index 44394d8..54194b2 100644
--- a/arch/microblaze/kernel/cpu/cpuinfo.c
+++ b/arch/microblaze/kernel/cpu/cpuinfo.c
@@ -34,6 +34,7 @@ const struct cpu_ver_key cpu_ver_lookup[] = {
 	{"8.00.a", 0x12},
 	{"8.00.b", 0x13},
 	{"8.10.a", 0x14},
+	{"8.20.a", 0x15},
 	{NULL, 0},
 };
 
diff --git a/arch/microblaze/kernel/dma.c b/arch/microblaze/kernel/dma.c
index 393e6b2..dc6416d 100644
--- a/arch/microblaze/kernel/dma.c
+++ b/arch/microblaze/kernel/dma.c
@@ -11,7 +11,6 @@
 #include <linux/gfp.h>
 #include <linux/dma-debug.h>
 #include <asm/bug.h>
-#include <asm/cacheflush.h>
 
 /*
  * Generic direct DMA implementation
@@ -21,21 +20,6 @@
  * can set archdata.dma_data to an unsigned long holding the offset. By
  * default the offset is PCI_DRAM_OFFSET.
  */
-static inline void __dma_sync_page(unsigned long paddr, unsigned long offset,
-				size_t size, enum dma_data_direction direction)
-{
-	switch (direction) {
-	case DMA_TO_DEVICE:
-	case DMA_BIDIRECTIONAL:
-		flush_dcache_range(paddr + offset, paddr + offset + size);
-		break;
-	case DMA_FROM_DEVICE:
-		invalidate_dcache_range(paddr + offset, paddr + offset + size);
-		break;
-	default:
-		BUG();
-	}
-}
 
 static unsigned long get_dma_direct_offset(struct device *dev)
 {
@@ -91,7 +75,7 @@ static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
 	/* FIXME this part of code is untested */
 	for_each_sg(sgl, sg, nents, i) {
 		sg->dma_address = sg_phys(sg) + get_dma_direct_offset(dev);
-		__dma_sync_page(page_to_phys(sg_page(sg)), sg->offset,
+		__dma_sync(page_to_phys(sg_page(sg)) + sg->offset,
 							sg->length, direction);
 	}
 
@@ -116,7 +100,7 @@ static inline dma_addr_t dma_direct_map_page(struct device *dev,
 					     enum dma_data_direction direction,
 					     struct dma_attrs *attrs)
 {
-	__dma_sync_page(page_to_phys(page), offset, size, direction);
+	__dma_sync(page_to_phys(page) + offset, size, direction);
 	return page_to_phys(page) + offset + get_dma_direct_offset(dev);
 }
 
@@ -131,7 +115,63 @@ static inline void dma_direct_unmap_page(struct device *dev,
  * phys_to_virt is here because in __dma_sync_page is __virt_to_phys and
  * dma_address is physical address
  */
-	__dma_sync_page(dma_address, 0 , size, direction);
+	__dma_sync(dma_address, size, direction);
+}
+
+static inline void
+dma_direct_sync_single_for_cpu(struct device *dev,
+			       dma_addr_t dma_handle, size_t size,
+			       enum dma_data_direction direction)
+{
+	/*
+	 * It's pointless to flush the cache as the memory segment
+	 * is given to the CPU
+	 */
+
+	if (direction == DMA_FROM_DEVICE)
+		__dma_sync(dma_handle, size, direction);
+}
+
+static inline void
+dma_direct_sync_single_for_device(struct device *dev,
+				  dma_addr_t dma_handle, size_t size,
+				  enum dma_data_direction direction)
+{
+	/*
+	 * It's pointless to invalidate the cache if the device isn't
+	 * supposed to write to the relevant region
+	 */
+
+	if (direction == DMA_TO_DEVICE)
+		__dma_sync(dma_handle, size, direction);
+}
+
+static inline void
+dma_direct_sync_sg_for_cpu(struct device *dev,
+			   struct scatterlist *sgl, int nents,
+			   enum dma_data_direction direction)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	if (direction == DMA_FROM_DEVICE)
+		for_each_sg(sgl, sg, nents, i)
+			__dma_sync(sg->dma_address, sg->length, direction);
+}
+
+static inline void
+dma_direct_sync_sg_for_device(struct device *dev,
+			      struct scatterlist *sgl, int nents,
+			      enum dma_data_direction direction)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	if (direction == DMA_TO_DEVICE)
+		for_each_sg(sgl, sg, nents, i)
+			__dma_sync(sg->dma_address, sg->length, direction);
 }
 
 struct dma_map_ops dma_direct_ops = {
@@ -142,6 +182,10 @@ struct dma_map_ops dma_direct_ops = {
 	.dma_supported	= dma_direct_dma_supported,
 	.map_page	= dma_direct_map_page,
 	.unmap_page	= dma_direct_unmap_page,
+	.sync_single_for_cpu		= dma_direct_sync_single_for_cpu,
+	.sync_single_for_device		= dma_direct_sync_single_for_device,
+	.sync_sg_for_cpu		= dma_direct_sync_sg_for_cpu,
+	.sync_sg_for_device		= dma_direct_sync_sg_for_device,
 };
 EXPORT_SYMBOL(dma_direct_ops);
 
diff --git a/arch/microblaze/kernel/exceptions.c b/arch/microblaze/kernel/exceptions.c
index 66fad23..6348dc8 100644
--- a/arch/microblaze/kernel/exceptions.c
+++ b/arch/microblaze/kernel/exceptions.c
@@ -119,7 +119,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 	case MICROBLAZE_DIV_ZERO_EXCEPTION:
 		if (user_mode(regs)) {
 			pr_debug("Divide by zero exception in user mode\n");
-			_exception(SIGILL, regs, FPE_INTDIV, addr);
+			_exception(SIGFPE, regs, FPE_INTDIV, addr);
 			return;
 		}
 		printk(KERN_WARNING "Divide by zero exception " \
diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c
index dbb8124..95cc295 100644
--- a/arch/microblaze/kernel/process.c
+++ b/arch/microblaze/kernel/process.c
@@ -179,6 +179,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp,
 
 	ti->cpu_context.msr = (childregs->msr|MSR_VM);
 	ti->cpu_context.msr &= ~MSR_UMS; /* switch_to to kernel mode */
+	ti->cpu_context.msr &= ~MSR_IE;
 #endif
 	ti->cpu_context.r15 = (unsigned long)ret_from_fork - 8;
 
diff --git a/arch/microblaze/kernel/ptrace.c b/arch/microblaze/kernel/ptrace.c
index 6a8e0cc..043cb58 100644
--- a/arch/microblaze/kernel/ptrace.c
+++ b/arch/microblaze/kernel/ptrace.c
@@ -148,7 +148,7 @@ asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
 		ret = -1L;
 
 	if (unlikely(current->audit_context))
-		audit_syscall_entry(EM_XILINX_MICROBLAZE, regs->r12,
+		audit_syscall_entry(EM_MICROBLAZE, regs->r12,
 				    regs->r5, regs->r6,
 				    regs->r7, regs->r8);
 
diff --git a/arch/microblaze/kernel/timer.c b/arch/microblaze/kernel/timer.c
index e5550ce..af74b11 100644
--- a/arch/microblaze/kernel/timer.c
+++ b/arch/microblaze/kernel/timer.c
@@ -308,7 +308,8 @@ unsigned long long notrace sched_clock(void)
 {
 	if (timer_initialized) {
 		struct clocksource *cs = &clocksource_microblaze;
-		cycle_t cyc = cnt32_to_63(cs->read(NULL));
+
+		cycle_t cyc = cnt32_to_63(cs->read(NULL)) & LLONG_MAX;
 		return clocksource_cyc2ns(cyc, cs->mult, cs->shift);
 	}
 	return 0;