sh: Fix up and optimize the kmap_coherent() interface.

This fixes up the kmap_coherent/kunmap_coherent() interface for recent
changes both in the page fault path and the shared cache flushers, as
well as adding in some optimizations.

One of the key things to note here is that the TLB flush itself is
deferred until the unmap, and the call in to update_mmu_cache() itself
goes away, relying on the regular page fault path to handle the lazy
dcache writeback if necessary.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
diff --git a/arch/sh/mm/cache.c b/arch/sh/mm/cache.c
index db2b1c5..8e4a8d1 100644
--- a/arch/sh/mm/cache.c
+++ b/arch/sh/mm/cache.c
@@ -51,7 +51,7 @@
 	    !test_bit(PG_dcache_dirty, &page->flags)) {
 		void *vto = kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK);
 		memcpy(vto, src, len);
-		kunmap_coherent();
+		kunmap_coherent(vto);
 	} else {
 		memcpy(dst, src, len);
 		if (boot_cpu_data.dcache.n_aliases)
@@ -70,7 +70,7 @@
 	    !test_bit(PG_dcache_dirty, &page->flags)) {
 		void *vfrom = kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK);
 		memcpy(dst, vfrom, len);
-		kunmap_coherent();
+		kunmap_coherent(vfrom);
 	} else {
 		memcpy(dst, src, len);
 		if (boot_cpu_data.dcache.n_aliases)
@@ -89,7 +89,7 @@
 	    !test_bit(PG_dcache_dirty, &from->flags)) {
 		vfrom = kmap_coherent(from, vaddr);
 		copy_page(vto, vfrom);
-		kunmap_coherent();
+		kunmap_coherent(vfrom);
 	} else {
 		vfrom = kmap_atomic(from, KM_USER0);
 		copy_page(vto, vfrom);
@@ -150,7 +150,7 @@
 
 			kaddr = kmap_coherent(page, vmaddr);
 			__flush_wback_region((void *)kaddr, PAGE_SIZE);
-			kunmap_coherent();
+			kunmap_coherent(kaddr);
 		} else
 			__flush_wback_region((void *)addr, PAGE_SIZE);
 	}