Skip to contentSkip to conversation

buffer_is_zero: use vector optimizations if possible

RFC PATCH In review Peter Lieven 2 participants 4 messages Mar 12, 2013

lore
Share & tools
lore
#0 lore raw

RFC PATCH 4/9 buffer_is_zero: use vector optimizations if possible

performance gain on SSE2 is approx. 20-25%. altivec is not tested. performance for unsigned long arithmetic is unchanged.

---
 util/cutils.c |    5 +++++
 1 file changed, 5 insertions(+)
util/cutils.c +5 -0
index a09d8e8..23f0cd6 100644
@@ -186,6 +186,11 @@ bool buffer_is_zero(const void *buf, size_t len)
      * latency.
      */
 
+    if (((uintptr_t) buf) % sizeof(VECTYPE) == 0
+          && len % 8*sizeof(VECTYPE) == 0) {
+        return buffer_find_nonzero_offset(buf, len)==len;
+    }
+
     size_t i;
     long d0, d1, d2, d3;
     const long * const data = buf;
Signature · 1 line
1.7.9.5
#1 lore raw
Quoted 14 lines · Peter Lieven

performance gain on SSE2 is approx. 20-25%. altivec is not tested. performance for unsigned long arithmetic is unchanged.

---
 util/cutils.c |    5 +++++
 1 file changed, 5 insertions(+)
util/cutils.c +0 -0
index a09d8e8..23f0cd6 100644

util/cutils.c bool buffer_is_zero(const void *buf, size_t len) · line 190

Quoted 5 lines
util/cutils.c +2 -0
@@ -186,6 +186,11 @@ bool buffer_is_zero(const void *buf, size_t len)
      * latency.
      */
 
+    if (((uintptr_t) buf) % sizeof(VECTYPE) == 0
+          && len % 8*sizeof(VECTYPE) == 0) {

Space around binary operators. Use CHAR_BITS instead of a magic number
8. Also, did you mean:

len % (CHAR_BITS * sizeof(VECTYPE))

instead of what you wrote as '(len % 8) * sizeof(VECTYPE)'?

Quoted 6 lines
util/cutils.c +3 -0
+        return buffer_find_nonzero_offset(buf, len)==len;
+    }
+
     size_t i;
     long d0, d1, d2, d3;
     const long * const data = buf;
Signature · 2 lines
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org
#2 lore raw

Am 12.03.2013 um 17:01 schrieb Eric Blake <[email protected]>:

Quoted 24 lines · Peter Lieven
Quoted 14 lines

performance gain on SSE2 is approx. 20-25%. altivec is not tested. performance for unsigned long arithmetic is unchanged.

---
util/cutils.c | 5 +++++

1 file changed, 5 insertions(+)
util/cutils.c +0 -0
index a09d8e8..23f0cd6 100644

util/cutils.c bool buffer_is_zero(const void *buf, size_t len) · line 190

Quoted 5 lines
util/cutils.c +2 -0
@@ -186,6 +186,11 @@ bool buffer_is_zero(const void *buf, size_t len)
     * latency.
     */
 
+    if (((uintptr_t) buf) % sizeof(VECTYPE) == 0
+          && len % 8*sizeof(VECTYPE) == 0) {

Space around binary operators. Use CHAR_BITS instead of a magic number
8. Also, did you mean:

Quoted 3 lines

len % (CHAR_BITS * sizeof(VECTYPE))

instead of what you wrote as '(len % 8) * sizeof(VECTYPE)'?

the 8 is not BITS_PER_BYTE or CHAR_BITS its the number of vectors in one loop in buffer_find_nonzero_offset(). I will add a constant for this to make it clearer.

Peter

Quoted 12 lines
Quoted 7 lines
util/cutils.c +3 -0
+        return buffer_find_nonzero_offset(buf, len)==len;
+    }
+
    size_t i;
    long d0, d1, d2, d3;
    const long * const data = buf;
Signature · 2 lines
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org
#3 lore raw
Quoted 9 lines · Peter Lieven
Quoted 5 lines

util/cutils.c bool buffer_is_zero(const void *buf, size_t len) · line 190

Quoted 2 lines
util/cutils.c +2 -0
+    if (((uintptr_t) buf) % sizeof(VECTYPE) == 0
+          && len % 8*sizeof(VECTYPE) == 0) {

Space around binary operators. Use CHAR_BITS instead of a magic number 8.

the 8 is not BITS_PER_BYTE or CHAR_BITS its the number of vectors in one loop in buffer_find_nonzero_offset(). I will add a constant for this to make it clearer.

Indeed, now I see it - 8 is the unroll factor. Well, all the more evidence that a named constant makes the code easier to read, compared to me mis-interpreting the magic number.

Signature · 2 lines
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org