blob: d1b16ab30d3da07829223bc1dfa070a40ad922df [file] [log] [blame]
* This file has no copyright assigned and is placed in the Public Domain.
* This file is part of the mingw-w64 runtime package.
* No warranty is given; refer to the file DISCLAIMER.PD within this package.
#include <_mingw_mac.h>
.file "floorf.S"
.p2align 4,,15
.globl __MINGW_USYMBOL(floorf)
.def __MINGW_USYMBOL(floorf); .scl 2; .type 32; .endef
#ifdef _WIN64
.seh_proc __MINGW_USYMBOL(floorf)
#if defined(_AMD64_) || defined(__x86_64__)
subq $40, %rsp
.seh_stackalloc 40
unpcklps %xmm0, %xmm0
cvtps2pd %xmm0, %xmm0
call floor
unpcklpd %xmm0, %xmm0
cvtpd2ps %xmm0, %xmm0
addq $40, %rsp
.def __MINGW_USYMBOL(floor); .scl 2; .type 32; .endef
#elif defined(_ARM_) || defined(__arm__)
vmrs r1, fpscr
bic r0, r1, #0x00c00000
orr r0, r0, #0x00800000 /* Round towards Minus Infinity */
vmsr fpscr, r0
vcvt.s32.f32 s0, s0
vcvt.f32.s32 s0, s0
vmsr fpscr, r1
bx lr
#elif defined(_X86_) || defined(__i386__)
flds 4(%esp)
subl $8,%esp
fstcw 4(%esp) /* store fpu control word */
/* We use here %edx although only the low 1 bits are defined.
But none of the operations should care and they are faster
than the 16 bit operations. */
movl $0x400,%edx /* round towards -oo */
orl 4(%esp),%edx
andl $0xf7ff,%edx
movl %edx,(%esp)
fldcw (%esp) /* load modified control word */
frndint /* round */
fldcw 4(%esp) /* restore original control word */
addl $8,%esp