This is a NFC patch. This patch run a bulk update on CodeGen tests that are impacted by the true16 features. This patch applies: 1. duplicate GFX11plus runlines and apply them with "+mattr=+real-true16" and "+mattr=-real-true16" 2. update the test with the update script For some GISEL runlines, the current CodeGen do not fully support the true16 version. Still update the runlines, but comment out the failing one, and added a "FIXME-TRUE16" comment to that test for easier tracking. These test will be fixed in the following patches. This is in a transition state that we support both "+real-true16/-real-true16" in our code base. We plan to move to "+real-true16" as default, and finally remove "-real-true16" mode and test lines.
1414 lines
72 KiB
LLVM
1414 lines
72 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --extra_scrub --version 5
|
|
; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GFX9-SDAG %s
|
|
; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck -check-prefixes=GFX9-GISEL %s
|
|
; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GFX10-SDAG %s
|
|
; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GFX10-GISEL %s
|
|
; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX11-SDAG,GFX11-SDAG-TRUE16 %s
|
|
; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1100 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX11-SDAG,GFX11-SDAG-FAKE16 %s
|
|
; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1100 -mattr=+real-true16 < %s | FileCheck -check-prefixes=GFX11-GISEL,GFX11-GISEL-TRUE16 %s
|
|
; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1100 -mattr=-real-true16 < %s | FileCheck -check-prefixes=GFX11-GISEL,GFX11-GISEL-FAKE16 %s
|
|
|
|
define i32 @intrinsic_lround_i32_f32(float %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s4, v2, v0
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_bfrev_b32_e32 v3, 1
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, v3, v2
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_i32_f32:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i32 @llvm.lround.i32.f32(float %arg)
|
|
ret i32 %res
|
|
}
|
|
|
|
define i32 @intrinsic_lround_i32_f64(double %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v0, 0x3ff00000
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e32 v0, 0, v0, vcc
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v1, s4, v0, v1
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v6, 0x3ff00000
|
|
; GFX9-GISEL-NEXT: s_brev_b32 s4, 1
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e32 v4, 0, v6, vcc
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v1, v1, s4, v4
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v0, 0, 0x3ff00000, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v0, v1
|
|
; GFX10-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v0, 0, 0x3ff00000, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v0, v1
|
|
; GFX11-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_i32_f64:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f64_e32 v0, v[0:1]
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i32 @llvm.lround.i32.f64(double %arg)
|
|
ret i32 %res
|
|
}
|
|
|
|
define i64 @intrinsic_lround_i64_f32(float %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s4, v2, v0
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0x2f800000
|
|
; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v0|, s4
|
|
; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0xcf800000
|
|
; GFX9-SDAG-NEXT: v_fma_f32 v2, v1, s4, |v0|
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v2, v3
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
|
|
; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_bfrev_b32_e32 v3, 1
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, v3, v2
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0x2f800000
|
|
; GFX9-GISEL-NEXT: v_mul_f32_e64 v2, |v1|, v2
|
|
; GFX9-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0xcf800000
|
|
; GFX9-GISEL-NEXT: v_fma_f32 v1, v2, v3, |v1|
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v1, v3
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v1, v2, v3
|
|
; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
|
|
; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
|
|
; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX10-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX10-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
|
|
; GFX10-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX10-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
|
|
; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX11-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e64 v1, null, v1, v3, vcc_lo
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_i64_f32:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
|
|
; GFX11-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e64 v1, null, v1, v3, vcc_lo
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i64 @llvm.lround.i64.f32(float %arg)
|
|
ret i64 %res
|
|
}
|
|
|
|
define i64 @intrinsic_lround_i64_f64(double %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v6, 0x3ff00000
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e32 v4, 0, v6, vcc
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v1, s4, v4, v1
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: s_movk_i32 s4, 0xffe0
|
|
; GFX9-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], s4
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s5, 0xc1f00000
|
|
; GFX9-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX9-SDAG-NEXT: v_fma_f64 v[0:1], v[2:3], s[4:5], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v6, 0x3ff00000
|
|
; GFX9-GISEL-NEXT: s_brev_b32 s4, 1
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e32 v4, 0, v6, vcc
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v1, v1, s4, v4
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3df00000
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v4, 0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v5, 0xc1f00000
|
|
; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mul_f64 v[2:3], v[0:1], v[2:3]
|
|
; GFX9-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX9-GISEL-NEXT: v_fma_f64 v[0:1], v[2:3], v[4:5], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v4, v1
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
|
|
; GFX10-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX10-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
|
|
; GFX10-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX10-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v4, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
|
|
; GFX11-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_i64_f64:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
|
|
; GFX11-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i64 @llvm.lround.i64.f64(double %arg)
|
|
ret i64 %res
|
|
}
|
|
|
|
define i64 @intrinsic_llround_i64_f32(float %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s4, v2, v0
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0x2f800000
|
|
; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v0|, s4
|
|
; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0xcf800000
|
|
; GFX9-SDAG-NEXT: v_fma_f32 v2, v1, s4, |v0|
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v2, v3
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
|
|
; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v2|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_bfrev_b32_e32 v3, 1
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, v3, v2
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0x2f800000
|
|
; GFX9-GISEL-NEXT: v_mul_f32_e64 v2, |v1|, v2
|
|
; GFX9-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0xcf800000
|
|
; GFX9-GISEL-NEXT: v_fma_f32 v1, v2, v3, |v1|
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v1, v3
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v1, v2, v3
|
|
; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v3
|
|
; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v1, v3, vcc
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
|
|
; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX10-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v2|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX10-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
|
|
; GFX10-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX10-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v1, v3, vcc_lo
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v2, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_mul_f32_e64 v1, 0x2f800000, |v0|
|
|
; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX11-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_fma_f32 v2, 0xcf800000, v1, |v0|
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v2
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e64 v1, null, v1, v3, vcc_lo
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_llround_i64_f32:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_f32_e32 v2, v0, v1
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v2|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f32_e32 v0, v1, v0
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v1, v0
|
|
; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v1|
|
|
; GFX11-GISEL-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_fma_f32 v1, 0xcf800000, v2, |v1|
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v1
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v0, v0, v3
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v3
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v0, v3
|
|
; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e64 v1, null, v1, v3, vcc_lo
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i64 @llvm.llround.i64.f32(float %arg)
|
|
ret i64 %res
|
|
}
|
|
|
|
define i64 @intrinsic_llround_i64_f64(double %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v6, 0x3ff00000
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s4, -2
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e32 v4, 0, v6, vcc
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v1, s4, v4, v1
|
|
; GFX9-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-SDAG-NEXT: s_movk_i32 s4, 0xffe0
|
|
; GFX9-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], s4
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s4, 0
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s5, 0xc1f00000
|
|
; GFX9-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX9-SDAG-NEXT: v_fma_f64 v[0:1], v[2:3], s[4:5], v[0:1]
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v6, 0x3ff00000
|
|
; GFX9-GISEL-NEXT: s_brev_b32 s4, 1
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f64_e64 vcc, |v[4:5]|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e32 v4, 0, v6, vcc
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v1, v1, s4, v4
|
|
; GFX9-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v2, 0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3df00000
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v4, 0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v5, 0xc1f00000
|
|
; GFX9-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_mul_f64 v[2:3], v[0:1], v[2:3]
|
|
; GFX9-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX9-GISEL-NEXT: v_fma_f64 v[0:1], v[2:3], v[4:5], v[0:1]
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v4, v1
|
|
; GFX10-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
|
|
; GFX10-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX10-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f64_e64 s4, |v[4:5]|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s4
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX10-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
|
|
; GFX10-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX10-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-SDAG-NEXT: v_mov_b32_e32 v0, 0
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s0
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v4, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_ldexp_f64 v[2:3], v[0:1], 0xffffffe0
|
|
; GFX11-SDAG-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_llround_i64_f64:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[4:5], v[0:1], -v[2:3]
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, v0, 0, 0
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f64_e64 s0, |v[4:5]|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 0x3ff00000, s0
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v4
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_add_f64 v[0:1], v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: v_trunc_f64_e32 v[0:1], v[0:1]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_mul_f64 v[2:3], 0x3df00000, v[0:1]
|
|
; GFX11-GISEL-NEXT: v_floor_f64_e32 v[2:3], v[2:3]
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_fma_f64 v[0:1], 0xc1f00000, v[2:3], v[0:1]
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f64_e32 v0, v[0:1]
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f64_e32 v1, v[2:3]
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i64 @llvm.llround.i64.f64(double %arg)
|
|
ret i64 %res
|
|
}
|
|
|
|
define half @intrinsic_fround_half(half %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_fround_half:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v3, 0x3c00
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f16_e64 vcc, |v2|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e32 v2, 0, v3, vcc
|
|
; GFX9-SDAG-NEXT: s_movk_i32 s4, 0x7fff
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s4, v2, v0
|
|
; GFX9-SDAG-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_fround_half:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3c00
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f16_e64 vcc, |v2|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e32 v2, 0, v3, vcc
|
|
; GFX9-GISEL-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX9-GISEL-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX9-GISEL-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_fround_half:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX10-SDAG-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f16_e64 s4, |v2|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX10-SDAG-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_fround_half:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX10-GISEL-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f16_e64 s4, |v2|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s4
|
|
; GFX10-GISEL-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX10-GISEL-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-TRUE16-LABEL: intrinsic_fround_half:
|
|
; GFX11-SDAG-TRUE16: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_mov_b16_e32 v1.l, v0.l
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_trunc_f16_e32 v1.h, v1.l
|
|
; GFX11-SDAG-TRUE16-NEXT: v_sub_f16_e32 v1.l, v1.l, v1.h
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cmp_ge_f16_e64 s0, |v1.l|, 0.5
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cndmask_b16 v1.l, 0, 0x3c00, s0
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_mov_b16_e32 v2.l, v1.l
|
|
; GFX11-SDAG-TRUE16-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_add_f16_e32 v0.l, v1.h, v0.l
|
|
; GFX11-SDAG-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-FAKE16-LABEL: intrinsic_fround_half:
|
|
; GFX11-SDAG-FAKE16: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cmp_ge_f16_e64 s0, |v2|, 0.5
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s0
|
|
; GFX11-SDAG-FAKE16-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-TRUE16-LABEL: intrinsic_fround_half:
|
|
; GFX11-GISEL-TRUE16: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_trunc_f16_e32 v0.h, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_sub_f16_e32 v1.l, v0.l, v0.h
|
|
; GFX11-GISEL-TRUE16-NEXT: v_and_b16 v0.l, 0x8000, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cmp_ge_f16_e64 s0, |v1.l|, 0.5
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cndmask_b16 v1.l, 0, 0x3c00, s0
|
|
; GFX11-GISEL-TRUE16-NEXT: v_or_b16 v0.l, v1.l, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_add_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-FAKE16-LABEL: intrinsic_fround_half:
|
|
; GFX11-GISEL-FAKE16: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX11-GISEL-FAKE16-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cmp_ge_f16_e64 s0, |v2|, 0.5
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s0
|
|
; GFX11-GISEL-FAKE16-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call half @llvm.round.f16(half %arg)
|
|
ret half %res
|
|
}
|
|
|
|
define i32 @intrinsic_lround_i32_f16(half %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX9-SDAG-NEXT: v_mov_b32_e32 v3, 0x3c00
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f16_e64 vcc, |v2|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e32 v2, 0, v3, vcc
|
|
; GFX9-SDAG-NEXT: s_movk_i32 s4, 0x7fff
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s4, v2, v0
|
|
; GFX9-SDAG-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX9-SDAG-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x3c00
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f16_e64 vcc, |v2|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e32 v2, 0, v3, vcc
|
|
; GFX9-GISEL-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX9-GISEL-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX9-GISEL-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX9-GISEL-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX10-SDAG-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f16_e64 s4, |v2|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s4
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX10-SDAG-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX10-SDAG-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX10-GISEL-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX10-GISEL-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f16_e64 s4, |v2|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s4
|
|
; GFX10-GISEL-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX10-GISEL-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX10-GISEL-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-TRUE16-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX11-SDAG-TRUE16: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_mov_b16_e32 v1.l, v0.l
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_trunc_f16_e32 v1.h, v1.l
|
|
; GFX11-SDAG-TRUE16-NEXT: v_sub_f16_e32 v1.l, v1.l, v1.h
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cmp_ge_f16_e64 s0, |v1.l|, 0.5
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cndmask_b16 v1.l, 0, 0x3c00, s0
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_mov_b16_e32 v2.l, v1.l
|
|
; GFX11-SDAG-TRUE16-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_add_f16_e32 v0.l, v1.h, v0.l
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cvt_f32_f16_e32 v0, v0.l
|
|
; GFX11-SDAG-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-SDAG-TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-SDAG-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-FAKE16-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX11-SDAG-FAKE16: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cmp_ge_f16_e64 s0, |v2|, 0.5
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s0
|
|
; GFX11-SDAG-FAKE16-NEXT: v_bfi_b32 v0, 0x7fff, v2, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-SDAG-FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-SDAG-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-TRUE16-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX11-GISEL-TRUE16: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-TRUE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_trunc_f16_e32 v0.h, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_sub_f16_e32 v1.l, v0.l, v0.h
|
|
; GFX11-GISEL-TRUE16-NEXT: v_and_b16 v0.l, 0x8000, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cmp_ge_f16_e64 s0, |v1.l|, 0.5
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cndmask_b16 v1.l, 0, 0x3c00, s0
|
|
; GFX11-GISEL-TRUE16-NEXT: v_or_b16 v0.l, v1.l, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_add_f16_e32 v0.l, v0.h, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cvt_f32_f16_e32 v0, v0.l
|
|
; GFX11-GISEL-TRUE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-GISEL-TRUE16-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-GISEL-TRUE16-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-FAKE16-LABEL: intrinsic_lround_i32_f16:
|
|
; GFX11-GISEL-FAKE16: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-FAKE16-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_trunc_f16_e32 v1, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_sub_f16_e32 v2, v0, v1
|
|
; GFX11-GISEL-FAKE16-NEXT: v_and_b32_e32 v0, 0xffff8000, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cmp_ge_f16_e64 s0, |v2|, 0.5
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cndmask_b32_e64 v2, 0, 0x3c00, s0
|
|
; GFX11-GISEL-FAKE16-NEXT: v_or_b32_e32 v0, v2, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_add_f16_e32 v0, v1, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cvt_f32_f16_e32 v0, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-GISEL-FAKE16-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-GISEL-FAKE16-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call i32 @llvm.lround.i32.f16(half %arg)
|
|
ret i32 %res
|
|
}
|
|
|
|
define <2 x i32> @intrinsic_lround_v2i32_v2f32(<2 x float> %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v3, v0, v2
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s6, -2
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s6, v3, v0
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v2, v1
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v3, v1, v2
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v1, s6, v3, v1
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v1, v2, v1
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v3, v0, v2
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_bfrev_b32_e32 v4, 1
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, v4, v3
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v2, v1
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v3, v1, v2
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v1, v1, v4, v3
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v1, v2, v1
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX9-GISEL-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v4, v0, v2
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v5, v1, v3
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v4|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v5|, 0.5
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v4, v0
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v5, v1
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v1, v3, v1
|
|
; GFX10-SDAG-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v4, v0, v2
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v5, v1, v3
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v4|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v5|, 0.5
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v4
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v5
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v1, v3, v1
|
|
; GFX10-GISEL-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_dual_sub_f32 v4, v0, v2 :: v_dual_sub_f32 v5, v1, v3
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v4|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v5|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v4, v0
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v5, v1
|
|
; GFX11-SDAG-NEXT: v_dual_add_f32 v0, v2, v0 :: v_dual_add_f32 v1, v3, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_v2i32_v2f32:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_dual_sub_f32 v4, v0, v2 :: v_dual_sub_f32 v5, v1, v3
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v4|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v5|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v4
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v5
|
|
; GFX11-GISEL-NEXT: v_dual_add_f32 v0, v2, v0 :: v_dual_add_f32 v1, v3, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f32_e32 v0, v0
|
|
; GFX11-GISEL-NEXT: v_cvt_i32_f32_e32 v1, v1
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call <2 x i32> @llvm.lround.v2i32.v2f32(<2 x float> %arg)
|
|
ret <2 x i32> %res
|
|
}
|
|
|
|
define <2 x i64> @intrinsic_lround_v2i64_v2f32(<2 x float> %arg) {
|
|
; GFX9-SDAG-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX9-SDAG: ; %bb.0: ; %entry
|
|
; GFX9-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v3, v0, v2
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: s_brev_b32 s6, -2
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v0, s6, v3, v0
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s7, 0x2f800000
|
|
; GFX9-SDAG-NEXT: v_mul_f32_e64 v2, |v0|, s7
|
|
; GFX9-SDAG-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX9-SDAG-NEXT: s_mov_b32 s8, 0xcf800000
|
|
; GFX9-SDAG-NEXT: v_fma_f32 v3, v2, s8, |v0|
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v4, 31, v0
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v0, v3, v4
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX9-SDAG-NEXT: v_sub_f32_e32 v5, v1, v3
|
|
; GFX9-SDAG-NEXT: v_cmp_ge_f32_e64 s[4:5], |v5|, 0.5
|
|
; GFX9-SDAG-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s[4:5]
|
|
; GFX9-SDAG-NEXT: v_bfi_b32 v1, s6, v5, v1
|
|
; GFX9-SDAG-NEXT: v_add_f32_e32 v1, v3, v1
|
|
; GFX9-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX9-SDAG-NEXT: v_mul_f32_e64 v1, |v3|, s7
|
|
; GFX9-SDAG-NEXT: v_floor_f32_e32 v1, v1
|
|
; GFX9-SDAG-NEXT: v_fma_f32 v5, v1, s8, |v3|
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GFX9-SDAG-NEXT: v_cvt_u32_f32_e32 v6, v1
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v2, v2, v4
|
|
; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v4
|
|
; GFX9-SDAG-NEXT: v_ashrrev_i32_e32 v3, 31, v3
|
|
; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v1, vcc, v2, v4, vcc
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v2, v5, v3
|
|
; GFX9-SDAG-NEXT: v_xor_b32_e32 v4, v6, v3
|
|
; GFX9-SDAG-NEXT: v_sub_co_u32_e32 v2, vcc, v2, v3
|
|
; GFX9-SDAG-NEXT: v_subb_co_u32_e32 v3, vcc, v4, v3, vcc
|
|
; GFX9-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX9-GISEL-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX9-GISEL: ; %bb.0: ; %entry
|
|
; GFX9-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v3, v0, v2
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v3|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v3, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_bfrev_b32_e32 v4, 1
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v0, v0, v4, v3
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v3, 0x2f800000
|
|
; GFX9-GISEL-NEXT: v_mul_f32_e64 v5, |v2|, v3
|
|
; GFX9-GISEL-NEXT: v_floor_f32_e32 v5, v5
|
|
; GFX9-GISEL-NEXT: v_mov_b32_e32 v6, 0xcf800000
|
|
; GFX9-GISEL-NEXT: v_fma_f32 v2, v5, v6, |v2|
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v7, 31, v0
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v0, v2, v7
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v2, v5, v7
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v5, v1
|
|
; GFX9-GISEL-NEXT: v_sub_f32_e32 v8, v1, v5
|
|
; GFX9-GISEL-NEXT: v_cmp_ge_f32_e64 s[4:5], |v8|, 0.5
|
|
; GFX9-GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1.0, s[4:5]
|
|
; GFX9-GISEL-NEXT: v_and_or_b32 v1, v1, v4, v8
|
|
; GFX9-GISEL-NEXT: v_add_f32_e32 v4, v5, v1
|
|
; GFX9-GISEL-NEXT: v_trunc_f32_e32 v1, v4
|
|
; GFX9-GISEL-NEXT: v_mul_f32_e64 v3, |v1|, v3
|
|
; GFX9-GISEL-NEXT: v_floor_f32_e32 v3, v3
|
|
; GFX9-GISEL-NEXT: v_fma_f32 v1, v3, v6, |v1|
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v5, v1
|
|
; GFX9-GISEL-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v0, vcc, v0, v7
|
|
; GFX9-GISEL-NEXT: v_ashrrev_i32_e32 v4, 31, v4
|
|
; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v1, vcc, v2, v7, vcc
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v2, v5, v4
|
|
; GFX9-GISEL-NEXT: v_xor_b32_e32 v3, v3, v4
|
|
; GFX9-GISEL-NEXT: v_sub_co_u32_e32 v2, vcc, v2, v4
|
|
; GFX9-GISEL-NEXT: v_subb_co_u32_e32 v3, vcc, v3, v4, vcc
|
|
; GFX9-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-SDAG-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX10-SDAG: ; %bb.0: ; %entry
|
|
; GFX10-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v4, v0, v2
|
|
; GFX10-SDAG-NEXT: v_sub_f32_e32 v5, v1, v3
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v4|, 0.5
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_cmp_ge_f32_e64 s4, |v5|, 0.5
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v4, v0
|
|
; GFX10-SDAG-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s4
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX10-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v5, v1
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_add_f32_e32 v1, v3, v1
|
|
; GFX10-SDAG-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v0|
|
|
; GFX10-SDAG-NEXT: v_trunc_f32_e32 v1, v1
|
|
; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v5, 31, v0
|
|
; GFX10-SDAG-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX10-SDAG-NEXT: v_mul_f32_e64 v3, 0x2f800000, |v1|
|
|
; GFX10-SDAG-NEXT: v_ashrrev_i32_e32 v6, 31, v1
|
|
; GFX10-SDAG-NEXT: v_fma_f32 v4, 0xcf800000, v2, |v0|
|
|
; GFX10-SDAG-NEXT: v_floor_f32_e32 v3, v3
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX10-SDAG-NEXT: v_fma_f32 v0, 0xcf800000, v3, |v1|
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v4
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v2, v2, v5
|
|
; GFX10-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v1, v1, v5
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v3, v3, v6
|
|
; GFX10-SDAG-NEXT: v_xor_b32_e32 v4, v0, v6
|
|
; GFX10-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v5
|
|
; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v5, vcc_lo
|
|
; GFX10-SDAG-NEXT: v_sub_co_u32 v2, vcc_lo, v4, v6
|
|
; GFX10-SDAG-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v3, v6, vcc_lo
|
|
; GFX10-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX10-GISEL-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX10-GISEL: ; %bb.0: ; %entry
|
|
; GFX10-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v4, v0, v2
|
|
; GFX10-GISEL-NEXT: v_sub_f32_e32 v5, v1, v3
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v4|, 0.5
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_cmp_ge_f32_e64 s4, |v5|, 0.5
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v4
|
|
; GFX10-GISEL-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s4
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v0, v2, v0
|
|
; GFX10-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v5
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX10-GISEL-NEXT: v_add_f32_e32 v1, v3, v1
|
|
; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v6, 31, v0
|
|
; GFX10-GISEL-NEXT: v_mul_f32_e64 v4, 0x2f800000, |v2|
|
|
; GFX10-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX10-GISEL-NEXT: v_floor_f32_e32 v4, v4
|
|
; GFX10-GISEL-NEXT: v_mul_f32_e64 v5, 0x2f800000, |v3|
|
|
; GFX10-GISEL-NEXT: v_fma_f32 v2, 0xcf800000, v4, |v2|
|
|
; GFX10-GISEL-NEXT: v_floor_f32_e32 v5, v5
|
|
; GFX10-GISEL-NEXT: v_fma_f32 v0, 0xcf800000, v5, |v3|
|
|
; GFX10-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v1
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v4
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v4, v5
|
|
; GFX10-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v1, v1, v6
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v2, v2, v6
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v4, v4, v3
|
|
; GFX10-GISEL-NEXT: v_xor_b32_e32 v5, v0, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v6
|
|
; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v1, vcc_lo, v2, v6, vcc_lo
|
|
; GFX10-GISEL-NEXT: v_sub_co_u32 v2, vcc_lo, v5, v3
|
|
; GFX10-GISEL-NEXT: v_sub_co_ci_u32_e32 v3, vcc_lo, v4, v3, vcc_lo
|
|
; GFX10-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-SDAG-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX11-SDAG: ; %bb.0: ; %entry
|
|
; GFX11-SDAG-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_dual_sub_f32 v4, v0, v2 :: v_dual_sub_f32 v5, v1, v3
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v4|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: v_cmp_ge_f32_e64 s0, |v5|, 0.5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v0, 0x7fffffff, v4, v0
|
|
; GFX11-SDAG-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s0
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_bfi_b32 v1, 0x7fffffff, v5, v1
|
|
; GFX11-SDAG-NEXT: v_dual_add_f32 v0, v2, v0 :: v_dual_add_f32 v1, v3, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: v_trunc_f32_e32 v1, v1
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_mul_f32_e64 v2, 0x2f800000, |v0|
|
|
; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v5, 31, v0
|
|
; GFX11-SDAG-NEXT: v_mul_f32_e64 v3, 0x2f800000, |v1|
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_floor_f32_e32 v2, v2
|
|
; GFX11-SDAG-NEXT: v_ashrrev_i32_e32 v6, 31, v1
|
|
; GFX11-SDAG-NEXT: v_floor_f32_e32 v3, v3
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_fma_f32 v4, 0xcf800000, v2, |v0|
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GFX11-SDAG-NEXT: v_fma_f32 v0, 0xcf800000, v3, |v1|
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_1) | instid1(VALU_DEP_4)
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v1, v4
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v2, v2, v5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
|
|
; GFX11-SDAG-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v1, v1, v5
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v3, v3, v6
|
|
; GFX11-SDAG-NEXT: v_xor_b32_e32 v4, v0, v6
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v5
|
|
; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e64 v1, null, v2, v5, vcc_lo
|
|
; GFX11-SDAG-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-SDAG-NEXT: v_sub_co_u32 v2, vcc_lo, v4, v6
|
|
; GFX11-SDAG-NEXT: v_sub_co_ci_u32_e64 v3, null, v3, v6, vcc_lo
|
|
; GFX11-SDAG-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; GFX11-GISEL-LABEL: intrinsic_lround_v2i64_v2f32:
|
|
; GFX11-GISEL: ; %bb.0: ; %entry
|
|
; GFX11-GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_dual_sub_f32 v4, v0, v2 :: v_dual_sub_f32 v5, v1, v3
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v4|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v4, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: v_cmp_ge_f32_e64 s0, |v5|, 0.5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v0, 0x80000000, v0, v4
|
|
; GFX11-GISEL-NEXT: v_cndmask_b32_e64 v5, 0, 1.0, s0
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_and_or_b32 v1, 0x80000000, v1, v5
|
|
; GFX11-GISEL-NEXT: v_dual_add_f32 v0, v2, v0 :: v_dual_add_f32 v1, v3, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_3)
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v2, v0
|
|
; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v6, 31, v0
|
|
; GFX11-GISEL-NEXT: v_trunc_f32_e32 v3, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_mul_f32_e64 v4, 0x2f800000, |v2|
|
|
; GFX11-GISEL-NEXT: v_mul_f32_e64 v5, 0x2f800000, |v3|
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_floor_f32_e32 v4, v4
|
|
; GFX11-GISEL-NEXT: v_floor_f32_e32 v5, v5
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
|
|
; GFX11-GISEL-NEXT: v_fma_f32 v2, 0xcf800000, v4, |v2|
|
|
; GFX11-GISEL-NEXT: v_fma_f32 v0, 0xcf800000, v5, |v3|
|
|
; GFX11-GISEL-NEXT: v_ashrrev_i32_e32 v3, 31, v1
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_3) | instskip(SKIP_3) | instid1(VALU_DEP_4)
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v1, v2
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v2, v4
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v4, v5
|
|
; GFX11-GISEL-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v1, v1, v6
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v2, v2, v6
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v4, v4, v3
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_4) | instskip(NEXT) | instid1(VALU_DEP_4)
|
|
; GFX11-GISEL-NEXT: v_xor_b32_e32 v5, v0, v3
|
|
; GFX11-GISEL-NEXT: v_sub_co_u32 v0, vcc_lo, v1, v6
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_3)
|
|
; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e64 v1, null, v2, v6, vcc_lo
|
|
; GFX11-GISEL-NEXT: v_sub_co_u32 v2, vcc_lo, v5, v3
|
|
; GFX11-GISEL-NEXT: s_delay_alu instid0(VALU_DEP_1)
|
|
; GFX11-GISEL-NEXT: v_sub_co_ci_u32_e64 v3, null, v4, v3, vcc_lo
|
|
; GFX11-GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
entry:
|
|
%res = tail call <2 x i64> @llvm.lround.v2i64.v2f32(<2 x float> %arg)
|
|
ret <2 x i64> %res
|
|
}
|