diff options
| author | Kevin Ryde <user42@zip.com.au> | 2000-03-19 00:10:16 +0100 |
|---|---|---|
| committer | Kevin Ryde <user42@zip.com.au> | 2000-03-19 00:10:16 +0100 |
| commit | be36a4377db4dd03d82157b7eb7584623b162489 (patch) | |
| tree | bec4f449677ebe870535cb8498ecbfd0064e979f /mpn | |
| parent | 08ed4c6cb6a2325cb74736337a96e94ae9728a47 (diff) | |
| download | gmp-be36a4377db4dd03d82157b7eb7584623b162489.tar.gz | |
* mpn/x86/k6/sqr_basecase.asm: Tiny amendments for 3x3 case.
Diffstat (limited to 'mpn')
| -rw-r--r-- | mpn/x86/k6/sqr_basecase.asm | 34 |
1 files changed, 18 insertions, 16 deletions
diff --git a/mpn/x86/k6/sqr_basecase.asm b/mpn/x86/k6/sqr_basecase.asm index 6e02fadd1..646d9bdcc 100644 --- a/mpn/x86/k6/sqr_basecase.asm +++ b/mpn/x86/k6/sqr_basecase.asm @@ -1,8 +1,8 @@ # AMD K6 mpn_sqr_basecase -- square an mpn number. # -# K6: 12-13 cycles per triangular product. +# K6: approx 6.5 cycles per crossproduct, or 12 cycles per triangular product. # -# Future: Unrolling to help sizes 10 to 20 is in progress. +# Future: Some unrolling will be needed to help sizes 10 to 20. # Copyright (C) 1999-2000 Free Software Foundation, Inc. @@ -45,13 +45,14 @@ PROLOGUE(mpn_sqr_basecase) deflit(`FRAME',0) movl PARAM_SIZE, %ecx + movl PARAM_SRC, %eax - movl PARAM_DST, %edx cmpl $2, %ecx + movl PARAM_DST, %edx je L(two_limbs) - ja L(two_or_more) + ja L(three_or_more) #------------------------------------------------------------------------------ @@ -113,7 +114,7 @@ L(two_limbs): #------------------------------------------------------------------------------ -L(two_or_more): +L(three_or_more): cmpl $4, %ecx jae L(four_or_more) @@ -133,35 +134,35 @@ L(two_or_more): mull %eax # src[0] ^ 2 movl %eax, (%ecx) + movl 4(%ebx), %eax + movl %edx, 4(%ecx) - movl 4(%ebx), %eax pushl %esi mull %eax # src[1] ^ 2 movl %eax, 8(%ecx) + movl 8(%ebx), %eax + movl %edx, 12(%ecx) - movl 8(%ebx), %eax pushl %edi mull %eax # src[2] ^ 2 movl %eax, 16(%ecx) - movl %edx, 20(%ecx) - - movl (%ebx), %eax + + movl %edx, 20(%ecx) movl 4(%ebx), %edx mull %edx # src[0] * src[1] movl %eax, %esi - movl %edx, %edi - - movl (%ebx), %eax + + movl %edx, %edi movl 8(%ebx), %edx pushl %ebp @@ -170,16 +171,17 @@ L(two_or_more): mull %edx # src[0] * src[2] addl %eax, %edi - adcl $0, %edx - movl %edx, %ebp + movl 4(%ebx), %eax + adcl $0, %edx - movl 4(%ebx), %eax + movl %edx, %ebp movl 8(%ebx), %edx mull %edx # src[1] * src[2] addl %eax, %ebp + adcl $0, %edx |