OpenE2K/qemu-e2k
13
4
Fork 0
Code Issues 4 Pull Requests Projects 2 Releases Activity

cutils: Remove aarch64 buffer zero checking

Browse Source 
The revised integer version is 4 times faster than the neon version
on an AppliedMicro Mustang.  Even with hand scheduling and additional
unrolling I cannot make any neon version run as fast as the integer.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This commit is contained in:
Richard Henderson 2016-08-29 11:46:19 -07:00 committed by Paolo Bonzini
parent 5e33a87222
commit 2250d3a293
1 changed files with 0 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
util/bufferiszero.c
View File

@ -200,21 +200,6 @@ static bool select_accel_fn(const void *buf, size_t len)
return buffer_zero_int(buf, len); return buffer_zero_int(buf, len);
} }
#elif defined(__aarch64__)
#include "arm_neon.h"
#define DO_NONZERO(X) (vgetq_lane_u64((X), 0) | vgetq_lane_u64((X), 1))
ACCEL_BUFFER_ZERO(buffer_zero_neon, 128, uint64x2_t, DO_NONZERO)
static bool select_accel_fn(const void *buf, size_t len)
{
uintptr_t ibuf = (uintptr_t)buf;
if (len % 128 == 0 && ibuf % sizeof(uint64x2_t) == 0) {
return buffer_zero_neon(buf, len);
}
return buffer_zero_int(buf, len);
}
#else #else
#define select_accel_fn buffer_zero_int #define select_accel_fn buffer_zero_int
#endif #endif