csky: Enable defer flush_dcache_page for abiv2 cpus (807/810/860)

Instead of flushing cache per update_mmu_cache() called, we use
flush_dcache_page to reduce the frequency of flashing the cache.

As abiv2 cpus are all PIPT for icache & dcache, we needn't handle
dcache aliasing problem. But their icache can't snoop dcache, so
we still need sync_icache_dcache in update_mmu_cache().
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>

arch/csky/abiv2/cacheflush.c

@@ -9,20 +9,22 @@
 void update_mmu_cache(struct vm_area_struct *vma, unsigned long address,
 		      pte_t *pte)
 {
-	unsigned long addr, pfn;
+	unsigned long addr;
 	struct page *page;
 
-	pfn = pte_pfn(*pte);
-	if (unlikely(!pfn_valid(pfn)))
+	page = pfn_to_page(pte_pfn(*pte));
+	if (page == ZERO_PAGE(0))
 		return;
 
-	page = pfn_to_page(pfn);
-	if (page == ZERO_PAGE(0))
+	if (test_and_set_bit(PG_dcache_clean, &page->flags))
 		return;
 
 	addr = (unsigned long) kmap_atomic(page);
 
-	cache_wbinv_range(addr, addr + PAGE_SIZE);
+	dcache_wb_range(addr, addr + PAGE_SIZE);
+
+	if (vma->vm_flags & VM_EXEC)
+		icache_inv_range(addr, addr + PAGE_SIZE);
 
 	kunmap_atomic((void *) addr);
 }

arch/csky/abiv2/inc/abi/cacheflush.h

@@ -15,8 +15,16 @@
 #define flush_cache_dup_mm(mm)			do { } while (0)
 #define flush_cache_range(vma, start, end)	do { } while (0)
 #define flush_cache_page(vma, vmaddr, pfn)	do { } while (0)
-#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 0
-#define flush_dcache_page(page)			do { } while (0)
+
+#define PG_dcache_clean		PG_arch_1
+
+#define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE 1
+static inline void flush_dcache_page(struct page *page)
+{
+	if (test_bit(PG_dcache_clean, &page->flags))
+		clear_bit(PG_dcache_clean, &page->flags);
+}
+
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 #define flush_icache_page(vma, page)		do { } while (0)