[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH] exec: fix ram_list dirty map optimization
From: |
Alexey Kardashevskiy |
Subject: |
[Qemu-devel] [PATCH] exec: fix ram_list dirty map optimization |
Date: |
Thu, 30 Jan 2014 23:03:50 +1100 |
The ae2810c4bb3b383176e8e1b33931b16c01483aab patch introduced
optimization for ram_list.dirty_memory update. However it can only
work correctly if hpratio is 1 as the @bitmap parameter stores 1 bits
per system page size (may vary, 4K or 64K on PPC64) and
ram_list.dirty_memory stores 1 bit per TARGET_PAGE_SIZE
(which is hardcoded to 4K).
This fixes hpratio!=1 case to fall back to the slow path.
Signed-off-by: Alexey Kardashevskiy <address@hidden>
---
include/exec/ram_addr.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/include/exec/ram_addr.h b/include/exec/ram_addr.h
index 33c8acc..6e83772 100644
--- a/include/exec/ram_addr.h
+++ b/include/exec/ram_addr.h
@@ -92,7 +92,8 @@ static inline void
cpu_physical_memory_set_dirty_lebitmap(unsigned long *bitmap,
unsigned long page = BIT_WORD(start >> TARGET_PAGE_BITS);
/* start address is aligned at the start of a word? */
- if (((page * BITS_PER_LONG) << TARGET_PAGE_BITS) == start) {
+ if ((((page * BITS_PER_LONG) << TARGET_PAGE_BITS) == start) &&
+ (hpratio == 1)) {
long k;
long nr = BITS_TO_LONGS(pages);
--
1.8.4.rc4
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Qemu-devel] [PATCH] exec: fix ram_list dirty map optimization,
Alexey Kardashevskiy <=