slub: reduce overhead of slub_debug
slub checks for poison one byte by one, which is highly inefficient
and shows up frequently as a highest cpu-eater in perf top.
Joining reads gives nice speedup:
(Compiling some project with different options)
make -j12 make clean
slub_debug disabled: 1m 27s 1.2 s
slub_debug enabled: 1m 46s 7.6 s
slub_debug enabled + this patch: 1m 33s 3.2 s
check_bytes still shows up high, but not always at the top.
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: linux-mm@kvack.org
Signed-off-by: Pekka Enberg <penberg@kernel.org>
diff --git a/mm/slub.c b/mm/slub.c
index 0e4f4f8..e3403b3 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -589,10 +589,10 @@
memset(p + s->objsize, val, s->inuse - s->objsize);
}
-static u8 *check_bytes(u8 *start, unsigned int value, unsigned int bytes)
+static u8 *check_bytes8(u8 *start, u8 value, unsigned int bytes)
{
while (bytes) {
- if (*start != (u8)value)
+ if (*start != value)
return start;
start++;
bytes--;
@@ -600,6 +600,38 @@
return NULL;
}
+static u8 *check_bytes(u8 *start, u8 value, unsigned int bytes)
+{
+ u64 value64;
+ unsigned int words, prefix;
+
+ if (bytes <= 16)
+ return check_bytes8(start, value, bytes);
+
+ value64 = value | value << 8 | value << 16 | value << 24;
+ value64 = value64 | value64 << 32;
+ prefix = 8 - ((unsigned long)start) % 8;
+
+ if (prefix) {
+ u8 *r = check_bytes8(start, value, prefix);
+ if (r)
+ return r;
+ start += prefix;
+ bytes -= prefix;
+ }
+
+ words = bytes / 8;
+
+ while (words) {
+ if (*(u64 *)start != value64)
+ return check_bytes8(start, value, 8);
+ start += 8;
+ words--;
+ }
+
+ return check_bytes8(start, value, bytes % 8);
+}
+
static void restore_bytes(struct kmem_cache *s, char *message, u8 data,
void *from, void *to)
{