mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-26 12:43:36 +01:00
5f547ad156
The killed flag is not always set. E.g. when a variable is used in a loop, it is never marked as killed, although it is unused in following basic blocks. Also, we try to deprecate kill flags and not use them. Check if the register is live in the endif block. If not, consider it killed in the then and else blocks. The vgpr-liverange tests have two new tests with loops (pre-committed, so the diff is visible). I also needed to change the subtarget to gfx10.1, otherwise calls are not working. Differential Revision: https://reviews.llvm.org/D106291
3431 lines
154 KiB
LLVM
3431 lines
154 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -global-isel -amdgpu-codegenprepare-disable-idiv-expansion=1 -mtriple=amdgcn-amd-amdpal -denormal-fp-math-f32=preserve-sign -mattr=+mad-mac-f32-insts < %s | FileCheck -check-prefixes=CHECK,GISEL %s
|
|
; RUN: llc -global-isel -amdgpu-codegenprepare-disable-idiv-expansion=0 -mtriple=amdgcn-amd-amdpal -denormal-fp-math-f32=preserve-sign -mattr=+mad-mac-f32-insts < %s | FileCheck -check-prefixes=CHECK,CGP %s
|
|
|
|
; The same 32-bit expansion is implemented in the legalizer and in AMDGPUCodeGenPrepare.
|
|
|
|
define i64 @v_udiv_i64(i64 %num, i64 %den) {
|
|
; CHECK-LABEL: v_udiv_i64:
|
|
; CHECK: ; %bb.0:
|
|
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CHECK-NEXT: v_or_b32_e32 v5, v1, v3
|
|
; CHECK-NEXT: v_mov_b32_e32 v4, 0
|
|
; CHECK-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[4:5]
|
|
; CHECK-NEXT: ; implicit-def: $vgpr4_vgpr5
|
|
; CHECK-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CHECK-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CHECK-NEXT: s_cbranch_execz BB0_2
|
|
; CHECK-NEXT: ; %bb.1:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v4, v2
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v5, v3
|
|
; CHECK-NEXT: v_sub_i32_e32 v6, vcc, 0, v2
|
|
; CHECK-NEXT: v_subb_u32_e32 v7, vcc, 0, v3, vcc
|
|
; CHECK-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; CHECK-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; CHECK-NEXT: v_mul_f32_e32 v5, 0x2f800000, v4
|
|
; CHECK-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_mac_f32_e32 v4, 0xcf800000, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v6, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v6, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v7, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v11, v6, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v10, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v5, v9
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v4, v9
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, v5, v9
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v4, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v13, v5, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v14, v4, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, v5, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v9, v14
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v11, vcc, v13, v12
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CHECK-NEXT: v_addc_u32_e64 v9, s[4:5], v5, v8, vcc
|
|
; CHECK-NEXT: v_add_i32_e64 v5, s[4:5], v5, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v6, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v7, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v6, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v6, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v9, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v4, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v10
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v4, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v9, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v13, v4, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v9, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v11, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v10, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v7, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v13
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v9, v7
|
|
; CHECK-NEXT: v_add_i32_e64 v9, s[4:5], v10, v11
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v8, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v8
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, v5, v6, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v7
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v1, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v0, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v4, v1, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v0, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v1, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v9, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v9, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v2, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v3, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, v2, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v5, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v2, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, 1, v4
|
|
; CHECK-NEXT: v_addc_u32_e32 v11, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, 1, v10
|
|
; CHECK-NEXT: v_addc_u32_e32 v12, vcc, 0, v11, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v7
|
|
; CHECK-NEXT: v_subb_u32_e64 v7, s[4:5], v1, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v6
|
|
; CHECK-NEXT: v_cmp_ge_u32_e64 s[4:5], v0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_cmp_ge_u32_e64 s[4:5], v7, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_subb_u32_e32 v1, vcc, v1, v3, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, v7, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v6, v9, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v2
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v1, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, v1, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v10, v8, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v11, v12, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v4, v4, v0, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v5, v5, v1, vcc
|
|
; CHECK-NEXT: ; implicit-def: $vgpr2
|
|
; CHECK-NEXT: ; implicit-def: $vgpr0
|
|
; CHECK-NEXT: BB0_2: ; %Flow
|
|
; CHECK-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CHECK-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CHECK-NEXT: s_cbranch_execz BB0_4
|
|
; CHECK-NEXT: ; %bb.3:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v1, v2
|
|
; CHECK-NEXT: v_sub_i32_e32 v3, vcc, 0, v2
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v1, v1
|
|
; CHECK-NEXT: v_mul_f32_e32 v1, 0x4f7ffffe, v1
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; CHECK-NEXT: v_mul_lo_u32 v3, v3, v1
|
|
; CHECK-NEXT: v_mul_hi_u32 v3, v1, v3
|
|
; CHECK-NEXT: v_add_i32_e32 v1, vcc, v1, v3
|
|
; CHECK-NEXT: v_mul_hi_u32 v1, v0, v1
|
|
; CHECK-NEXT: v_mul_lo_u32 v3, v1, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, 1, v1
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v3
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v1, v4, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v3, s[4:5], v0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v0, v3, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v3, vcc, 1, v1
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v4, v1, v3, vcc
|
|
; CHECK-NEXT: v_mov_b32_e32 v5, 0
|
|
; CHECK-NEXT: BB0_4:
|
|
; CHECK-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CHECK-NEXT: v_mov_b32_e32 v0, v4
|
|
; CHECK-NEXT: v_mov_b32_e32 v1, v5
|
|
; CHECK-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv i64 %num, %den
|
|
ret i64 %result
|
|
}
|
|
|
|
; FIXME: This is a workaround for not handling uniform VGPR case.
|
|
declare i32 @llvm.amdgcn.readfirstlane(i32)
|
|
|
|
define amdgpu_ps i64 @s_udiv_i64(i64 inreg %num, i64 inreg %den) {
|
|
; CHECK-LABEL: s_udiv_i64:
|
|
; CHECK: ; %bb.0:
|
|
; CHECK-NEXT: s_or_b64 s[6:7], s[0:1], s[2:3]
|
|
; CHECK-NEXT: s_mov_b32 s4, 0
|
|
; CHECK-NEXT: s_mov_b32 s5, -1
|
|
; CHECK-NEXT: s_and_b64 s[6:7], s[6:7], s[4:5]
|
|
; CHECK-NEXT: v_cmp_ne_u64_e64 vcc, s[6:7], 0
|
|
; CHECK-NEXT: s_cbranch_vccz BB1_2
|
|
; CHECK-NEXT: ; %bb.1:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v0, s2
|
|
; CHECK-NEXT: v_mov_b32_e32 v1, s3
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v2, s3
|
|
; CHECK-NEXT: s_sub_u32 s6, 0, s2
|
|
; CHECK-NEXT: s_cselect_b32 s4, 1, 0
|
|
; CHECK-NEXT: v_mov_b32_e32 v3, s1
|
|
; CHECK-NEXT: v_mac_f32_e32 v0, 0x4f800000, v2
|
|
; CHECK-NEXT: s_and_b32 s4, s4, 1
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CHECK-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0
|
|
; CHECK-NEXT: s_cmp_lg_u32 s4, 0
|
|
; CHECK-NEXT: s_subb_u32 s7, 0, s3
|
|
; CHECK-NEXT: v_mul_f32_e32 v2, 0x2f800000, v0
|
|
; CHECK-NEXT: v_trunc_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mac_f32_e32 v0, 0xcf800000, v2
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v4, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v5, s6, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s7, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, s6, v0
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v6, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v2, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, v0, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, v2, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v7
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v0, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v2, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v4, v2, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v9, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v5, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v5, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v0, v5
|
|
; CHECK-NEXT: v_addc_u32_e64 v5, s[4:5], v2, v4, vcc
|
|
; CHECK-NEXT: v_add_i32_e64 v2, s[4:5], v2, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v4, s6, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s7, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, s6, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, s6, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v5, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v4
|
|
; CHECK-NEXT: v_mul_hi_u32 v4, v5, v4
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v7
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v0, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v5, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v11, v0, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, v5, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v9, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v4, s[4:5], v8, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v4, s[4:5], v4, v11
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v8, v9
|
|
; CHECK-NEXT: v_add_i32_e64 v4, s[4:5], v4, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v5, s[4:5], v5, v6
|
|
; CHECK-NEXT: v_addc_u32_e32 v2, vcc, v2, v5, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v0, v4
|
|
; CHECK-NEXT: v_addc_u32_e32 v2, vcc, 0, v2, vcc
|
|
; CHECK-NEXT: v_mul_lo_u32 v4, s1, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, s0, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v0, s1, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s0, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, s1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, s0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v2, s1, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v7, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v4, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v4, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v0, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v5, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v6, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v7, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v4, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, v5, v4
|
|
; CHECK-NEXT: v_mul_lo_u32 v5, s2, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s3, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, s2, v0
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, 1, v0
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v4, vcc, 1, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v2, s2, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v6, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v7
|
|
; CHECK-NEXT: v_sub_i32_e32 v5, vcc, s0, v5
|
|
; CHECK-NEXT: v_subb_u32_e64 v3, s[4:5], v3, v2, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v2, s[4:5], s1, v2
|
|
; CHECK-NEXT: v_cmp_le_u32_e64 s[4:5], s2, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_cmp_le_u32_e64 s[4:5], s3, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_subb_u32_e32 v1, vcc, v2, v1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, s3, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v2, v7, v6, vcc
|
|
; CHECK-NEXT: v_subrev_i32_e32 v3, vcc, s2, v5
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s2, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v3, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s3, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v5, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, s3, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v5, v3, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v8, v4, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v0, v1, vcc
|
|
; CHECK-NEXT: s_mov_b32 s5, 0
|
|
; CHECK-NEXT: s_branch BB1_3
|
|
; CHECK-NEXT: BB1_2:
|
|
; CHECK-NEXT: ; implicit-def: $vgpr0_vgpr1
|
|
; CHECK-NEXT: BB1_3: ; %Flow
|
|
; CHECK-NEXT: s_xor_b32 s1, s5, -1
|
|
; CHECK-NEXT: s_and_b32 s1, s1, 1
|
|
; CHECK-NEXT: s_cmp_lg_u32 s1, 0
|
|
; CHECK-NEXT: s_cbranch_scc1 BB1_5
|
|
; CHECK-NEXT: ; %bb.4:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v0, s2
|
|
; CHECK-NEXT: s_sub_i32 s1, 0, s2
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CHECK-NEXT: v_mul_f32_e32 v0, 0x4f7ffffe, v0
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v1, s1, v0
|
|
; CHECK-NEXT: v_mul_hi_u32 v1, v0, v1
|
|
; CHECK-NEXT: v_add_i32_e32 v0, vcc, v0, v1
|
|
; CHECK-NEXT: v_mul_hi_u32 v0, s0, v0
|
|
; CHECK-NEXT: v_mul_lo_u32 v1, v0, s2
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, 1, v0
|
|
; CHECK-NEXT: v_sub_i32_e32 v1, vcc, s0, v1
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s2, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v0, v2, vcc
|
|
; CHECK-NEXT: v_subrev_i32_e64 v2, s[0:1], s2, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v1, v2, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, 1, v0
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s2, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v0, v2, vcc
|
|
; CHECK-NEXT: BB1_5:
|
|
; CHECK-NEXT: v_readfirstlane_b32 s0, v0
|
|
; CHECK-NEXT: s_mov_b32 s1, s0
|
|
; CHECK-NEXT: ; return to shader part epilog
|
|
%result = udiv i64 %num, %den
|
|
%cast = bitcast i64 %result to <2 x i32>
|
|
%elt.0 = extractelement <2 x i32> %cast, i32 0
|
|
%elt.1 = extractelement <2 x i32> %cast, i32 1
|
|
%res.0 = call i32 @llvm.amdgcn.readfirstlane(i32 %elt.0)
|
|
%res.1 = call i32 @llvm.amdgcn.readfirstlane(i32 %elt.1)
|
|
%ins.0 = insertelement <2 x i32> undef, i32 %res.0, i32 0
|
|
%ins.1 = insertelement <2 x i32> %ins.0, i32 %res.0, i32 1
|
|
%cast.back = bitcast <2 x i32> %ins.1 to i64
|
|
ret i64 %cast.back
|
|
}
|
|
|
|
define <2 x i64> @v_udiv_v2i64(<2 x i64> %num, <2 x i64> %den) {
|
|
; GISEL-LABEL: v_udiv_v2i64:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v8, v4
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v9, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v8, 0x4f800000, v9
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_mul_f32_e32 v8, 0x5f7ffffc, v8
|
|
; GISEL-NEXT: v_mul_f32_e32 v9, 0x2f800000, v8
|
|
; GISEL-NEXT: v_trunc_f32_e32 v9, v9
|
|
; GISEL-NEXT: v_mac_f32_e32 v8, 0xcf800000, v9
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v9, v9
|
|
; GISEL-NEXT: v_sub_i32_e32 v10, vcc, 0, v4
|
|
; GISEL-NEXT: v_subb_u32_e32 v11, vcc, 0, v5, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v10, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v10, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v9, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v8, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v8, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v9, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v9, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v8, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v15, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v15, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v9, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_addc_u32_e64 v12, s[4:5], v9, v13, vcc
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v10, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v14
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v11, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v12, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v8, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v8, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v14, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v12, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v12, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v8, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[4:5], v14, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v13, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v12, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v10, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v9, vcc, v9, v10, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v9, vcc, 0, v9, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v1, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v0, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v1, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v1, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v4, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v5, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v13
|
|
; GISEL-NEXT: v_sub_i32_e32 v0, vcc, v0, v10
|
|
; GISEL-NEXT: v_subb_u32_e64 v10, s[4:5], v1, v11, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v11
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v10, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v0, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], v10, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, v11, v12, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v4
|
|
; GISEL-NEXT: v_subb_u32_e32 v1, vcc, v1, v5, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, 1, v8
|
|
; GISEL-NEXT: v_addc_u32_e32 v12, vcc, 0, v9, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v1, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v0, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, v1, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v13, v0, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v1, vcc, 1, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v4, vcc, 0, v12, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v11, v1, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v12, v4, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v8, v0, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v9, v1, vcc
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v4, v6
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v5, v7
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v5, 0x2f800000, v4
|
|
; GISEL-NEXT: v_trunc_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0xcf800000, v5
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_sub_i32_e32 v8, vcc, 0, v6
|
|
; GISEL-NEXT: v_subb_u32_e32 v9, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v8, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v9, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v8, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v8, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v5, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v4, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v4, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v5, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v5, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v13, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v5, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_addc_u32_e64 v10, s[4:5], v5, v11, vcc
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v5, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v8, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v9, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v8, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v8, v4
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v10, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v12, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v10, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v10, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v12, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[4:5], v12, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v11, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v12, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v4, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v5, vcc, v5, v8, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v5, vcc, 0, v5, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, v3, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v2, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v3, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v9, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v5, v3, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, v6, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v7, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v6, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v6, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v11
|
|
; GISEL-NEXT: v_sub_i32_e32 v2, vcc, v2, v8
|
|
; GISEL-NEXT: v_subb_u32_e64 v8, s[4:5], v3, v9, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v8, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v6
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], v8, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, v9, v10, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v2, s[4:5], v2, v6
|
|
; GISEL-NEXT: v_subb_u32_e32 v3, vcc, v3, v7, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v3, vcc, 0, v3, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, 1, v4
|
|
; GISEL-NEXT: v_addc_u32_e32 v10, vcc, 0, v5, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v3, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v2, v6
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, v3, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v11, v2, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v3, vcc, 1, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, 0, v10, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v9, v3, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v10, v6, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v4, v2, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v5, v3, vcc
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_v2i64:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: v_mov_b32_e32 v8, v0
|
|
; CGP-NEXT: v_mov_b32_e32 v9, v1
|
|
; CGP-NEXT: v_or_b32_e32 v1, v9, v5
|
|
; CGP-NEXT: v_mov_b32_e32 v0, 0
|
|
; CGP-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[0:1]
|
|
; CGP-NEXT: ; implicit-def: $vgpr0_vgpr1
|
|
; CGP-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CGP-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CGP-NEXT: s_cbranch_execz BB2_2
|
|
; CGP-NEXT: ; %bb.1:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v4
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v1, v5
|
|
; CGP-NEXT: v_sub_i32_e32 v10, vcc, 0, v4
|
|
; CGP-NEXT: v_subb_u32_e32 v11, vcc, 0, v5, vcc
|
|
; CGP-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0
|
|
; CGP-NEXT: v_trunc_f32_e32 v1, v1
|
|
; CGP-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v10, v1
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v10, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v11, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v15, v10, v0
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v14, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v1, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v0, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v1, v13
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v0, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v17, v1, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v0, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v1, v12
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v17, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, v17, v16
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v13, s[4:5], v1, v12, vcc
|
|
; CGP-NEXT: v_add_i32_e64 v1, s[4:5], v1, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v10, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v11, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v10, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v10, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v13, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v0, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v13, v12
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v11, v10
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v14
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v0, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v13, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v0, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v13, v10
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v15, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v14, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v12, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v13, v11
|
|
; CGP-NEXT: v_add_i32_e64 v13, s[4:5], v14, v15
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v12, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v13, v12
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v12
|
|
; CGP-NEXT: v_addc_u32_e32 v1, vcc, v1, v10, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v11
|
|
; CGP-NEXT: v_addc_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v9, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v8, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v0, v9, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v8, v1
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v9, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v8, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v1, v9, v1
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v13, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v12, v10
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v13, v11
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v4, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v5, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v4, v0
|
|
; CGP-NEXT: v_add_i32_e32 v1, vcc, v1, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v4, v1
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, 1, v0
|
|
; CGP-NEXT: v_addc_u32_e32 v15, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v12, v10
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, 1, v14
|
|
; CGP-NEXT: v_addc_u32_e32 v16, vcc, 0, v15, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v13
|
|
; CGP-NEXT: v_sub_i32_e32 v8, vcc, v8, v11
|
|
; CGP-NEXT: v_subb_u32_e64 v11, s[4:5], v9, v10, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v9, s[4:5], v9, v10
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v8, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v11, v5
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_subb_u32_e32 v9, vcc, v9, v5, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v11, v5
|
|
; CGP-NEXT: v_cndmask_b32_e32 v10, v13, v10, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v8, vcc, v8, v4
|
|
; CGP-NEXT: v_subbrev_u32_e32 v9, vcc, 0, v9, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v8, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v4, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v9, v5
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v9, v5
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v8, v4, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v4
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v14, v12, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v5, v15, v16, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v10
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v4, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v1, v5, vcc
|
|
; CGP-NEXT: ; implicit-def: $vgpr4
|
|
; CGP-NEXT: ; implicit-def: $vgpr8
|
|
; CGP-NEXT: BB2_2: ; %Flow2
|
|
; CGP-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CGP-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: s_cbranch_execz BB2_4
|
|
; CGP-NEXT: ; %bb.3:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v4
|
|
; CGP-NEXT: v_sub_i32_e32 v1, vcc, 0, v4
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v0, 0x4f7ffffe, v0
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v1, v1, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v1, v0, v1
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v0, v8, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v1, v0, v4
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, 1, v0
|
|
; CGP-NEXT: v_sub_i32_e32 v1, vcc, v8, v1
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v1, v4
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v5, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v5, s[4:5], v1, v4
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v1, v5, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, 1, v0
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v1, v4
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v5, vcc
|
|
; CGP-NEXT: v_mov_b32_e32 v1, 0
|
|
; CGP-NEXT: BB2_4:
|
|
; CGP-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: v_or_b32_e32 v5, v3, v7
|
|
; CGP-NEXT: v_mov_b32_e32 v4, 0
|
|
; CGP-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[4:5]
|
|
; CGP-NEXT: ; implicit-def: $vgpr4_vgpr5
|
|
; CGP-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CGP-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CGP-NEXT: s_cbranch_execz BB2_6
|
|
; CGP-NEXT: ; %bb.5:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v4, v6
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v5, v7
|
|
; CGP-NEXT: v_sub_i32_e32 v8, vcc, 0, v6
|
|
; CGP-NEXT: v_subb_u32_e32 v9, vcc, 0, v7, vcc
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v5, 0x2f800000, v4
|
|
; CGP-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0xcf800000, v5
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v9, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v8, v4
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v12, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v5, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v4, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v5, v11
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v4, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v5, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v5, v10
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v15, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v11
|
|
; CGP-NEXT: v_addc_u32_e64 v11, s[4:5], v5, v10, vcc
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v5, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v9, v9, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v8, v8, v11
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v11, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v11, v10
|
|
; CGP-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CGP-NEXT: v_add_i32_e64 v8, s[4:5], v8, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v9, v4, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v11, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v15, v4, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v8, v11, v8
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v13, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v12, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v9, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v11, v9
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v12, v13
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v10, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v11, v10
|
|
; CGP-NEXT: v_add_i32_e64 v8, s[4:5], v8, v10
|
|
; CGP-NEXT: v_addc_u32_e32 v5, vcc, v5, v8, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CGP-NEXT: v_addc_u32_e32 v5, vcc, 0, v5, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v8, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v9, v2, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v2, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v3, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v2, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v5, v3, v5
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v11, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v10, v8
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v11, v9
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v8
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v9, v6, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v7, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v6, v4
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v8, v6, v5
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, 1, v4
|
|
; CGP-NEXT: v_addc_u32_e32 v13, vcc, 0, v5, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v10, v8
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, 1, v12
|
|
; CGP-NEXT: v_addc_u32_e32 v14, vcc, 0, v13, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v9
|
|
; CGP-NEXT: v_subb_u32_e64 v9, s[4:5], v3, v8, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v8
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v9, v7
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_subb_u32_e32 v3, vcc, v3, v7, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v9, v7
|
|
; CGP-NEXT: v_cndmask_b32_e32 v8, v11, v8, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v6
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v3, v7
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v3, v7
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v6, v2, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v12, v10, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v13, v14, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v4, v2, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v5, v5, v3, vcc
|
|
; CGP-NEXT: ; implicit-def: $vgpr6
|
|
; CGP-NEXT: ; implicit-def: $vgpr2
|
|
; CGP-NEXT: BB2_6: ; %Flow
|
|
; CGP-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CGP-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: s_cbranch_execz BB2_8
|
|
; CGP-NEXT: ; %bb.7:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v3, v6
|
|
; CGP-NEXT: v_sub_i32_e32 v4, vcc, 0, v6
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v3, v3
|
|
; CGP-NEXT: v_mul_f32_e32 v3, 0x4f7ffffe, v3
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v4, v3
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_add_i32_e32 v3, vcc, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v3, v2, v3
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v3, v6
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, 1, v3
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v4
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v6
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v3, v5, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v4, s[4:5], v2, v6
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v2, v4, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, 1, v3
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v6
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v3, v4, vcc
|
|
; CGP-NEXT: v_mov_b32_e32 v5, 0
|
|
; CGP-NEXT: BB2_8:
|
|
; CGP-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: v_mov_b32_e32 v2, v4
|
|
; CGP-NEXT: v_mov_b32_e32 v3, v5
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv <2 x i64> %num, %den
|
|
ret <2 x i64> %result
|
|
}
|
|
|
|
define i64 @v_udiv_i64_pow2k_denom(i64 %num) {
|
|
; CHECK-LABEL: v_udiv_i64_pow2k_denom:
|
|
; CHECK: ; %bb.0:
|
|
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v2, 0x1000
|
|
; CHECK-NEXT: v_cvt_f32_ubyte0_e32 v3, 0
|
|
; CHECK-NEXT: s_movk_i32 s6, 0xf000
|
|
; CHECK-NEXT: s_movk_i32 s7, 0x1000
|
|
; CHECK-NEXT: s_bfe_i32 s4, -1, 0x10000
|
|
; CHECK-NEXT: s_bfe_i32 s5, -1, 0x10000
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0x4f800000, v3
|
|
; CHECK-NEXT: v_mov_b32_e32 v3, s4
|
|
; CHECK-NEXT: v_mov_b32_e32 v4, s5
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v2, 0x5f7ffffc, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v5, 0x2f800000, v2
|
|
; CHECK-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0xcf800000, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s6, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, -1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s6, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v5, v7
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v2, v7
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v5, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v2, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v5, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v5, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v11, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v7, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v7, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v7
|
|
; CHECK-NEXT: v_addc_u32_e64 v7, s[4:5], v5, v6, vcc
|
|
; CHECK-NEXT: v_add_i32_e64 v5, s[4:5], v5, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, -1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, s6, v7
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v7, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v10
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v2, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v7, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v13, v2, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v7, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v11, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v10, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v13
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v9, s[4:5], v10, v11
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v7, v8
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, v5, v7, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v6
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v2, v1, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v0, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v1, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v9, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v9, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, s7, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, 0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s7, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v5, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s7, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, 1, v2
|
|
; CHECK-NEXT: v_addc_u32_e32 v11, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, 1, v10
|
|
; CHECK-NEXT: v_addc_u32_e32 v12, vcc, 0, v11, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v7
|
|
; CHECK-NEXT: v_subb_u32_e64 v7, s[4:5], v1, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v6
|
|
; CHECK-NEXT: v_cmp_le_u32_e64 s[4:5], s7, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_cmp_eq_u32_e64 s[4:5], 0, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v3, v3, v6, s[4:5]
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_subrev_i32_e32 v0, vcc, s7, v0
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s7, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v4, v0, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v10, v8, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v11, v12, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v5, v1, vcc
|
|
; CHECK-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv i64 %num, 4096
|
|
ret i64 %result
|
|
}
|
|
|
|
define <2 x i64> @v_udiv_v2i64_pow2k_denom(<2 x i64> %num) {
|
|
; GISEL-LABEL: v_udiv_v2i64_pow2k_denom:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: s_movk_i32 s12, 0x1000
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v4, s12
|
|
; GISEL-NEXT: s_sub_u32 s8, 0, s12
|
|
; GISEL-NEXT: s_cselect_b32 s4, 1, 0
|
|
; GISEL-NEXT: v_cvt_f32_ubyte0_e32 v5, 0
|
|
; GISEL-NEXT: v_mov_b32_e32 v6, v4
|
|
; GISEL-NEXT: s_and_b32 s4, s4, 1
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v6, 0x4f800000, v5
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v5, v6
|
|
; GISEL-NEXT: s_cmp_lg_u32 s4, 0
|
|
; GISEL-NEXT: s_subb_u32 s9, 0, 0
|
|
; GISEL-NEXT: s_bfe_i32 s10, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s11, -1, 0x10000
|
|
; GISEL-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v5, 0x5f7ffffc, v5
|
|
; GISEL-NEXT: v_mul_f32_e32 v6, 0x2f800000, v4
|
|
; GISEL-NEXT: s_sub_u32 s13, 0, s12
|
|
; GISEL-NEXT: s_cselect_b32 s4, 1, 0
|
|
; GISEL-NEXT: v_mul_f32_e32 v7, 0x2f800000, v5
|
|
; GISEL-NEXT: v_trunc_f32_e32 v6, v6
|
|
; GISEL-NEXT: s_and_b32 s4, s4, 1
|
|
; GISEL-NEXT: v_trunc_f32_e32 v7, v7
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0xcf800000, v6
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v6, v6
|
|
; GISEL-NEXT: v_mac_f32_e32 v5, 0xcf800000, v7
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v7, v7
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; GISEL-NEXT: s_cmp_lg_u32 s4, 0
|
|
; GISEL-NEXT: s_subb_u32 s6, 0, 0
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, s13, v6
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, s8, v7
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s6, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, s9, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, s8, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v6, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v6, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v14, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v7, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v5, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v7, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v6, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v4, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v6, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v5, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v19
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v7, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v5, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v7, v9
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v15, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v10, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v15, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v19, v18
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v14, v15
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_addc_u32_e64 v10, s[4:5], v6, v8, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, s6, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, s13, v4
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v5, v13
|
|
; GISEL-NEXT: v_addc_u32_e64 v13, s[6:7], v7, v9, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, s9, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, s8, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v13, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v16, v18
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v5, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v16, v17
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, v5, v16
|
|
; GISEL-NEXT: v_add_i32_e64 v17, s[6:7], v19, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v17, s[6:7], v17, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, s13, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v10, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v12, v17
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v12, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v4, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[8:9], v18, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[8:9], v14, v17
|
|
; GISEL-NEXT: v_mov_b32_e32 v14, s10
|
|
; GISEL-NEXT: v_mov_b32_e32 v17, s11
|
|
; GISEL-NEXT: s_bfe_i32 s13, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s14, -1, 0x10000
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[10:11], v6, v8
|
|
; GISEL-NEXT: v_mov_b32_e32 v8, s13
|
|
; GISEL-NEXT: v_add_i32_e64 v7, s[10:11], v7, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v10, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[8:9], v18, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v4, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[8:9], v18, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[8:9], v9, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v18, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v13, v16
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v13, v16
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v5, v16
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v19, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v11, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v19, v16
|
|
; GISEL-NEXT: v_mov_b32_e32 v19, s14
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v9, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v11, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], v12, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[6:7], v16, v18
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[6:7], v10, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], v13, v15
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, v6, v10, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v7, vcc, v7, v12, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, 0, v6, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v3, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v7, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v2, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v3, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v2, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v6, v3, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v0, v7
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v7
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v7
|
|
; GISEL-NEXT: v_mul_hi_u32 v7, v1, v7
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v15, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v11, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v4, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v15, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v18, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v10, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, s12, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, 0, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, s12, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, s12, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, 0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, s12, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v7, vcc, v7, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, s12, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s12, v7
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v16, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, 1, v4
|
|
; GISEL-NEXT: v_addc_u32_e32 v16, vcc, 0, v6, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, 1, v5
|
|
; GISEL-NEXT: v_addc_u32_e32 v18, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_sub_i32_e32 v2, vcc, v2, v10
|
|
; GISEL-NEXT: v_subb_u32_e64 v10, s[4:5], v3, v9, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; GISEL-NEXT: v_cmp_le_u32_e64 s[4:5], s12, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v12
|
|
; GISEL-NEXT: v_subb_u32_e64 v12, s[6:7], v1, v11, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v1, s[6:7], v1, v11
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[8:9], 1, v13
|
|
; GISEL-NEXT: v_addc_u32_e64 v11, s[8:9], 0, v16, s[8:9]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, v8, v9, s[6:7]
|
|
; GISEL-NEXT: v_cmp_le_u32_e64 s[6:7], s12, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[6:7]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, v14, v9, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], 1, v15
|
|
; GISEL-NEXT: v_addc_u32_e64 v14, s[6:7], 0, v18, s[6:7]
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; GISEL-NEXT: v_subrev_i32_e32 v2, vcc, s12, v2
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; GISEL-NEXT: v_cmp_le_u32_e32 vcc, s12, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; GISEL-NEXT: v_subrev_i32_e32 v0, vcc, s12, v0
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; GISEL-NEXT: v_cmp_le_u32_e32 vcc, s12, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v3
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v19, v2, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v17, v0, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v13, v10, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, v15, v12, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v16, v11, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v4, v1, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v1, v18, v14, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, v5, v0, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v1, v7, v1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v6, v3, vcc
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_v2i64_pow2k_denom:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v4, 0x1000
|
|
; CGP-NEXT: v_cvt_f32_ubyte0_e32 v5, 0
|
|
; CGP-NEXT: s_movk_i32 s8, 0xf000
|
|
; CGP-NEXT: s_movk_i32 s12, 0x1000
|
|
; CGP-NEXT: s_bfe_i32 s10, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s11, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s13, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s14, -1, 0x10000
|
|
; CGP-NEXT: v_mov_b32_e32 v6, v4
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; CGP-NEXT: v_mac_f32_e32 v6, 0x4f800000, v5
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v5, v6
|
|
; CGP-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v5, 0x5f7ffffc, v5
|
|
; CGP-NEXT: v_mul_f32_e32 v6, 0x2f800000, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v7, 0x2f800000, v5
|
|
; CGP-NEXT: v_trunc_f32_e32 v6, v6
|
|
; CGP-NEXT: v_trunc_f32_e32 v7, v7
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0xcf800000, v6
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v6, v6
|
|
; CGP-NEXT: v_mac_f32_e32 v5, 0xcf800000, v7
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v7, v7
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v8, s8, v6
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v9, s8, v7
|
|
; CGP-NEXT: v_mul_lo_u32 v10, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v11, -1, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v12, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v13, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v14, -1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v15, s8, v5
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v6, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v6, v10
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v14, v9
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v7, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v5, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v7, v13
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v6, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v4, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v8, v6, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v5, v9
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v19
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v7, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v5, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v9, v7, v9
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v11, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v15, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v15, v16
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, v19, v18
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v14, v15
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; CGP-NEXT: v_addc_u32_e64 v10, s[4:5], v6, v8, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v11, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v12, -1, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v14, s8, v4
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v5, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v13, s[6:7], v7, v9, s[4:5]
|
|
; CGP-NEXT: v_mul_lo_u32 v15, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v16, -1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v17, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v18, s8, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v13, v15
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v16, v18
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v5, v15
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v16, v17
|
|
; CGP-NEXT: v_mul_lo_u32 v17, v5, v16
|
|
; CGP-NEXT: v_add_i32_e64 v17, s[6:7], v19, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v17, s[6:7], v17, v18
|
|
; CGP-NEXT: v_mul_lo_u32 v17, s8, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v10, v11
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v12, v17
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v4, v11
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v12, v14
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v4, v12
|
|
; CGP-NEXT: v_add_i32_e64 v14, s[8:9], v18, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v14, s[8:9], v14, v17
|
|
; CGP-NEXT: v_mov_b32_e32 v14, s10
|
|
; CGP-NEXT: v_mov_b32_e32 v17, s11
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[10:11], v6, v8
|
|
; CGP-NEXT: v_mov_b32_e32 v8, s13
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[10:11], v7, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v9, v10, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v13, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v15, s[8:9], v18, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v10, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v10, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v4, v12
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[8:9], v18, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[8:9], v9, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v18, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v13, v16
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v13, v16
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v5, v16
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v19, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v19, v16
|
|
; CGP-NEXT: v_mov_b32_e32 v19, s14
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[6:7], v9, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v11, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], v12, v15
|
|
; CGP-NEXT: v_add_i32_e64 v15, s[6:7], v16, v18
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[6:7], v10, v12
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], v13, v15
|
|
; CGP-NEXT: v_addc_u32_e32 v6, vcc, v6, v10, vcc
|
|
; CGP-NEXT: v_addc_u32_e64 v7, vcc, v7, v12, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CGP-NEXT: v_addc_u32_e32 v6, vcc, 0, v6, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v9, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v11
|
|
; CGP-NEXT: v_addc_u32_e32 v7, vcc, 0, v7, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v0, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v2, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v3, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v2, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v6, v3, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v0, v7
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v1, v7
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v0, v7
|
|
; CGP-NEXT: v_mul_hi_u32 v7, v1, v7
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v15, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v11, v5
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v9, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v4, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v15, v10
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v18, v16
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v10, v9
|
|
; CGP-NEXT: v_mul_lo_u32 v10, s12, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v13, 0, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v15, s12, v4
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v12, s12, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v16, 0, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v18, s12, v5
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CGP-NEXT: v_add_i32_e32 v7, vcc, v7, v11
|
|
; CGP-NEXT: v_mul_lo_u32 v9, s12, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v11, s12, v7
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v16, v11
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, 1, v4
|
|
; CGP-NEXT: v_addc_u32_e32 v16, vcc, 0, v6, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, 1, v5
|
|
; CGP-NEXT: v_addc_u32_e32 v18, vcc, 0, v7, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v10
|
|
; CGP-NEXT: v_subb_u32_e64 v10, s[4:5], v3, v9, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; CGP-NEXT: v_cmp_le_u32_e64 s[4:5], s12, v2
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v12
|
|
; CGP-NEXT: v_subb_u32_e64 v12, s[6:7], v1, v11, s[4:5]
|
|
; CGP-NEXT: v_sub_i32_e64 v1, s[6:7], v1, v11
|
|
; CGP-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v10
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[8:9], 1, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v11, s[8:9], 0, v16, s[8:9]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, v8, v9, s[6:7]
|
|
; CGP-NEXT: v_cmp_le_u32_e64 s[6:7], s12, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[6:7]
|
|
; CGP-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, v14, v9, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], 1, v15
|
|
; CGP-NEXT: v_addc_u32_e64 v14, s[6:7], 0, v18, s[6:7]
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; CGP-NEXT: v_subrev_i32_e32 v2, vcc, s12, v2
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_cmp_le_u32_e32 vcc, s12, v2
|
|
; CGP-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; CGP-NEXT: v_subrev_i32_e32 v0, vcc, s12, v0
|
|
; CGP-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_cmp_le_u32_e32 vcc, s12, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, 0, v3
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v19, v2, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v17, v0, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v13, v10, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, v15, v12, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v16, v11, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v4, v1, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e64 v1, v18, v14, s[4:5]
|
|
; CGP-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, v5, v0, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v1, v7, v1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v6, v3, vcc
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv <2 x i64> %num, <i64 4096, i64 4096>
|
|
ret <2 x i64> %result
|
|
}
|
|
|
|
define i64 @v_udiv_i64_oddk_denom(i64 %num) {
|
|
; CHECK-LABEL: v_udiv_i64_oddk_denom:
|
|
; CHECK: ; %bb.0:
|
|
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v2, 0x12d8fb
|
|
; CHECK-NEXT: v_cvt_f32_ubyte0_e32 v3, 0
|
|
; CHECK-NEXT: s_mov_b32 s6, 0xffed2705
|
|
; CHECK-NEXT: s_mov_b32 s7, 0x12d8fb
|
|
; CHECK-NEXT: s_bfe_i32 s4, -1, 0x10000
|
|
; CHECK-NEXT: s_bfe_i32 s5, -1, 0x10000
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0x4f800000, v3
|
|
; CHECK-NEXT: v_mov_b32_e32 v3, s4
|
|
; CHECK-NEXT: v_mov_b32_e32 v4, s5
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v2, 0x5f7ffffc, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v5, 0x2f800000, v2
|
|
; CHECK-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0xcf800000, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s6, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, -1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s6, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v5, v7
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v2, v7
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v5, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v2, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v5, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v5, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v11, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v7, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v7, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v7
|
|
; CHECK-NEXT: v_addc_u32_e64 v7, s[4:5], v5, v6, vcc
|
|
; CHECK-NEXT: v_add_i32_e64 v5, s[4:5], v5, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, -1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, s6, v7
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v7, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v10
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v2, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v7, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v13, v2, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v7, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v11, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v10, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v13
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v9, s[4:5], v10, v11
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v7, v8
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, v5, v7, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v6
|
|
; CHECK-NEXT: v_addc_u32_e32 v5, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v2, v1, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v0, v5
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v1, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v5
|
|
; CHECK-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v9, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v9, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, s7, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, 0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, s7, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v5, vcc, v5, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, s7, v5
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, 1, v2
|
|
; CHECK-NEXT: v_addc_u32_e32 v11, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, 1, v10
|
|
; CHECK-NEXT: v_addc_u32_e32 v12, vcc, 0, v11, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v7
|
|
; CHECK-NEXT: v_subb_u32_e64 v7, s[4:5], v1, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v6
|
|
; CHECK-NEXT: v_cmp_le_u32_e64 s[4:5], s7, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_cmp_eq_u32_e64 s[4:5], 0, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v3, v3, v6, s[4:5]
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_subrev_i32_e32 v0, vcc, s7, v0
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_cmp_le_u32_e32 vcc, s7, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v4, v0, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v10, v8, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v11, v12, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v3
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v2, v0, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v5, v1, vcc
|
|
; CHECK-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv i64 %num, 1235195
|
|
ret i64 %result
|
|
}
|
|
|
|
define <2 x i64> @v_udiv_v2i64_oddk_denom(<2 x i64> %num) {
|
|
; GISEL-LABEL: v_udiv_v2i64_oddk_denom:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: s_mov_b32 s12, 0x12d8fb
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v4, s12
|
|
; GISEL-NEXT: s_sub_u32 s8, 0, s12
|
|
; GISEL-NEXT: s_cselect_b32 s4, 1, 0
|
|
; GISEL-NEXT: v_cvt_f32_ubyte0_e32 v5, 0
|
|
; GISEL-NEXT: v_mov_b32_e32 v6, v4
|
|
; GISEL-NEXT: s_and_b32 s4, s4, 1
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v6, 0x4f800000, v5
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v5, v6
|
|
; GISEL-NEXT: s_cmp_lg_u32 s4, 0
|
|
; GISEL-NEXT: s_subb_u32 s9, 0, 0
|
|
; GISEL-NEXT: s_bfe_i32 s10, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s11, -1, 0x10000
|
|
; GISEL-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v5, 0x5f7ffffc, v5
|
|
; GISEL-NEXT: v_mul_f32_e32 v6, 0x2f800000, v4
|
|
; GISEL-NEXT: s_sub_u32 s13, 0, s12
|
|
; GISEL-NEXT: s_cselect_b32 s4, 1, 0
|
|
; GISEL-NEXT: v_mul_f32_e32 v7, 0x2f800000, v5
|
|
; GISEL-NEXT: v_trunc_f32_e32 v6, v6
|
|
; GISEL-NEXT: s_and_b32 s4, s4, 1
|
|
; GISEL-NEXT: v_trunc_f32_e32 v7, v7
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0xcf800000, v6
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v6, v6
|
|
; GISEL-NEXT: v_mac_f32_e32 v5, 0xcf800000, v7
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v7, v7
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; GISEL-NEXT: s_cmp_lg_u32 s4, 0
|
|
; GISEL-NEXT: s_subb_u32 s6, 0, 0
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, s13, v6
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, s8, v7
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s6, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, s9, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, s8, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v6, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v6, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v14, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v7, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v5, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v7, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v6, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v4, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v6, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v5, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v19
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v7, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v5, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v7, v9
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v15, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v10, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v15, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v19, v18
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v14, v15
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_addc_u32_e64 v10, s[4:5], v6, v8, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s13, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, s6, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, s13, v4
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v5, v13
|
|
; GISEL-NEXT: v_addc_u32_e64 v13, s[6:7], v7, v9, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, s9, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, s8, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, s8, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v13, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v16, v18
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v5, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v16, v17
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, v5, v16
|
|
; GISEL-NEXT: v_add_i32_e64 v17, s[6:7], v19, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v17, s[6:7], v17, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, s13, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v10, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v12, v17
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v12, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v4, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[8:9], v18, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[8:9], v14, v17
|
|
; GISEL-NEXT: v_mov_b32_e32 v14, s10
|
|
; GISEL-NEXT: v_mov_b32_e32 v17, s11
|
|
; GISEL-NEXT: s_bfe_i32 s13, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s14, -1, 0x10000
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[10:11], v6, v8
|
|
; GISEL-NEXT: v_mov_b32_e32 v8, s13
|
|
; GISEL-NEXT: v_add_i32_e64 v7, s[10:11], v7, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v10, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[8:9], v18, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v4, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[8:9], v18, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[8:9], v9, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[8:9], v18, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v13, v16
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v13, v16
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v5, v16
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v19, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v11, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[6:7], v19, v16
|
|
; GISEL-NEXT: v_mov_b32_e32 v19, s14
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v9, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[6:7], v11, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], v12, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[6:7], v16, v18
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[6:7], v10, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], v13, v15
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, v6, v10, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v7, vcc, v7, v12, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, 0, v6, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v3, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v7, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v2, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v3, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v2, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v6, v3, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v0, v7
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v7
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v7
|
|
; GISEL-NEXT: v_mul_hi_u32 v7, v1, v7
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v15, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v11, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v4, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v15, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v18, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v10, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, s12, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, 0, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, s12, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, s12, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, 0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, s12, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v7, vcc, v7, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, s12, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, s12, v7
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v16, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, 1, v4
|
|
; GISEL-NEXT: v_addc_u32_e32 v16, vcc, 0, v6, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, 1, v5
|
|
; GISEL-NEXT: v_addc_u32_e32 v18, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_sub_i32_e32 v2, vcc, v2, v10
|
|
; GISEL-NEXT: v_subb_u32_e64 v10, s[4:5], v3, v9, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; GISEL-NEXT: v_cmp_le_u32_e64 s[4:5], s12, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v12
|
|
; GISEL-NEXT: v_subb_u32_e64 v12, s[6:7], v1, v11, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v1, s[6:7], v1, v11
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[8:9], 1, v13
|
|
; GISEL-NEXT: v_addc_u32_e64 v11, s[8:9], 0, v16, s[8:9]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, v8, v9, s[6:7]
|
|
; GISEL-NEXT: v_cmp_le_u32_e64 s[6:7], s12, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[6:7]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, v14, v9, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[6:7], 1, v15
|
|
; GISEL-NEXT: v_addc_u32_e64 v14, s[6:7], 0, v18, s[6:7]
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; GISEL-NEXT: v_subrev_i32_e32 v2, vcc, s12, v2
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; GISEL-NEXT: v_cmp_le_u32_e32 vcc, s12, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; GISEL-NEXT: v_subrev_i32_e32 v0, vcc, s12, v0
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; GISEL-NEXT: v_cmp_le_u32_e32 vcc, s12, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v3
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v19, v2, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v17, v0, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v13, v10, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, v15, v12, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v16, v11, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v4, v1, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v1, v18, v14, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, v5, v0, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v1, v7, v1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v6, v3, vcc
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_v2i64_oddk_denom:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v4, 0x12d8fb
|
|
; CGP-NEXT: v_cvt_f32_ubyte0_e32 v5, 0
|
|
; CGP-NEXT: s_mov_b32 s8, 0xffed2705
|
|
; CGP-NEXT: s_mov_b32 s12, 0x12d8fb
|
|
; CGP-NEXT: s_bfe_i32 s10, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s11, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s13, -1, 0x10000
|
|
; CGP-NEXT: s_bfe_i32 s14, -1, 0x10000
|
|
; CGP-NEXT: v_mov_b32_e32 v6, v4
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; CGP-NEXT: v_mac_f32_e32 v6, 0x4f800000, v5
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v5, v6
|
|
; CGP-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v5, 0x5f7ffffc, v5
|
|
; CGP-NEXT: v_mul_f32_e32 v6, 0x2f800000, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v7, 0x2f800000, v5
|
|
; CGP-NEXT: v_trunc_f32_e32 v6, v6
|
|
; CGP-NEXT: v_trunc_f32_e32 v7, v7
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0xcf800000, v6
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v6, v6
|
|
; CGP-NEXT: v_mac_f32_e32 v5, 0xcf800000, v7
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v7, v7
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v8, s8, v6
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v9, s8, v7
|
|
; CGP-NEXT: v_mul_lo_u32 v10, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v11, -1, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v12, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v13, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v14, -1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v15, s8, v5
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v6, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v6, v10
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v14, v9
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v7, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v5, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v7, v13
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v6, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v4, v8
|
|
; CGP-NEXT: v_mul_hi_u32 v8, v6, v8
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v5, v9
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v19
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v7, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v5, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v9, v7, v9
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v11, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v15, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v15, v16
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, v19, v18
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v17
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v12, v11
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v14, v15
|
|
; CGP-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; CGP-NEXT: v_addc_u32_e64 v10, s[4:5], v6, v8, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v11, s8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v12, -1, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v14, s8, v4
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v5, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v13, s[6:7], v7, v9, s[4:5]
|
|
; CGP-NEXT: v_mul_lo_u32 v15, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v16, -1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v17, s8, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v18, s8, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v13, v15
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v16, v18
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v5, v15
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v16, v17
|
|
; CGP-NEXT: v_mul_lo_u32 v17, v5, v16
|
|
; CGP-NEXT: v_add_i32_e64 v17, s[6:7], v19, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v17, s[6:7], v17, v18
|
|
; CGP-NEXT: v_mul_lo_u32 v17, s8, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v10, v11
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v12, v17
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v4, v11
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v12, v14
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v4, v12
|
|
; CGP-NEXT: v_add_i32_e64 v14, s[8:9], v18, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v14, s[8:9], v14, v17
|
|
; CGP-NEXT: v_mov_b32_e32 v14, s10
|
|
; CGP-NEXT: v_mov_b32_e32 v17, s11
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[10:11], v6, v8
|
|
; CGP-NEXT: v_mov_b32_e32 v8, s13
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[10:11], v7, v9
|
|
; CGP-NEXT: v_mul_hi_u32 v9, v10, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v13, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v15, s[8:9], v18, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v10, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v10, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v4, v12
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[8:9], v18, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[8:9], v9, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[8:9]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[8:9], v18, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; CGP-NEXT: v_mul_lo_u32 v19, v13, v16
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v13, v16
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v5, v16
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v19, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v16, s[6:7], v19, v16
|
|
; CGP-NEXT: v_mov_b32_e32 v19, s14
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[6:7], v9, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[6:7], v11, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], v12, v15
|
|
; CGP-NEXT: v_add_i32_e64 v15, s[6:7], v16, v18
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[6:7], v10, v12
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], v13, v15
|
|
; CGP-NEXT: v_addc_u32_e32 v6, vcc, v6, v10, vcc
|
|
; CGP-NEXT: v_addc_u32_e64 v7, vcc, v7, v12, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CGP-NEXT: v_addc_u32_e32 v6, vcc, 0, v6, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v9, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v11
|
|
; CGP-NEXT: v_addc_u32_e32 v7, vcc, 0, v7, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v1, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v0, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v5, v1, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v2, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v3, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v2, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v6, v3, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v18, v0, v7
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v1, v7
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v0, v7
|
|
; CGP-NEXT: v_mul_hi_u32 v7, v1, v7
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v15, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v11, v5
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v9, s[4:5], v9, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v4, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v15, v10
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v18, v16
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v10, v9
|
|
; CGP-NEXT: v_mul_lo_u32 v10, s12, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v13, 0, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v15, s12, v4
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v12, s12, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v16, 0, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v18, s12, v5
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CGP-NEXT: v_add_i32_e32 v7, vcc, v7, v11
|
|
; CGP-NEXT: v_mul_lo_u32 v9, s12, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v11, s12, v7
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v16, v11
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, 1, v4
|
|
; CGP-NEXT: v_addc_u32_e32 v16, vcc, 0, v6, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v9, vcc, v9, v15
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, 1, v5
|
|
; CGP-NEXT: v_addc_u32_e32 v18, vcc, 0, v7, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v10
|
|
; CGP-NEXT: v_subb_u32_e64 v10, s[4:5], v3, v9, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; CGP-NEXT: v_cmp_le_u32_e64 s[4:5], s12, v2
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v12
|
|
; CGP-NEXT: v_subb_u32_e64 v12, s[6:7], v1, v11, s[4:5]
|
|
; CGP-NEXT: v_sub_i32_e64 v1, s[6:7], v1, v11
|
|
; CGP-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v10
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[8:9], 1, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v11, s[8:9], 0, v16, s[8:9]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v8, v8, v9, s[6:7]
|
|
; CGP-NEXT: v_cmp_le_u32_e64 s[6:7], s12, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[6:7]
|
|
; CGP-NEXT: v_cmp_eq_u32_e64 s[6:7], 0, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v9, v14, v9, s[6:7]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[6:7], 1, v15
|
|
; CGP-NEXT: v_addc_u32_e64 v14, s[6:7], 0, v18, s[6:7]
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; CGP-NEXT: v_subrev_i32_e32 v2, vcc, s12, v2
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_cmp_le_u32_e32 vcc, s12, v2
|
|
; CGP-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; CGP-NEXT: v_subrev_i32_e32 v0, vcc, s12, v0
|
|
; CGP-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_cmp_le_u32_e32 vcc, s12, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, 0, v3
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v19, v2, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, 0, v1
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v17, v0, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v13, v10, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, v15, v12, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v16, v11, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v4, v1, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e64 v1, v18, v14, s[4:5]
|
|
; CGP-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, v5, v0, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e64 v1, v7, v1, s[4:5]
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v6, v3, vcc
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%result = udiv <2 x i64> %num, <i64 1235195, i64 1235195>
|
|
ret <2 x i64> %result
|
|
}
|
|
|
|
define i64 @v_udiv_i64_pow2_shl_denom(i64 %x, i64 %y) {
|
|
; CHECK-LABEL: v_udiv_i64_pow2_shl_denom:
|
|
; CHECK: ; %bb.0:
|
|
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CHECK-NEXT: s_mov_b64 s[4:5], 0x1000
|
|
; CHECK-NEXT: v_lshl_b64 v[4:5], s[4:5], v2
|
|
; CHECK-NEXT: v_or_b32_e32 v3, v1, v5
|
|
; CHECK-NEXT: v_mov_b32_e32 v2, 0
|
|
; CHECK-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[2:3]
|
|
; CHECK-NEXT: ; implicit-def: $vgpr2_vgpr3
|
|
; CHECK-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CHECK-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CHECK-NEXT: s_cbranch_execz BB7_2
|
|
; CHECK-NEXT: ; %bb.1:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v2, v4
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v3, v5
|
|
; CHECK-NEXT: v_sub_i32_e32 v6, vcc, 0, v4
|
|
; CHECK-NEXT: v_subb_u32_e32 v7, vcc, 0, v5, vcc
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0x4f800000, v3
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v2, 0x5f7ffffc, v2
|
|
; CHECK-NEXT: v_mul_f32_e32 v3, 0x2f800000, v2
|
|
; CHECK-NEXT: v_trunc_f32_e32 v3, v3
|
|
; CHECK-NEXT: v_mac_f32_e32 v2, 0xcf800000, v3
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v6, v3
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v7, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v11, v6, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v10, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v3, v9
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v9
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, v3, v9
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v11
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v2, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v13, v3, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v14, v2, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, v3, v8
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v13, v9
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v9, v14
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v11, vcc, v13, v12
|
|
; CHECK-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v9
|
|
; CHECK-NEXT: v_addc_u32_e64 v9, s[4:5], v3, v8, vcc
|
|
; CHECK-NEXT: v_add_i32_e64 v3, s[4:5], v3, v8
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v7, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v6, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v6, v9
|
|
; CHECK-NEXT: v_mul_lo_u32 v11, v9, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v12, v2, v8
|
|
; CHECK-NEXT: v_mul_hi_u32 v8, v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v7, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v10
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v2, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v10, v9, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v13, v2, v6
|
|
; CHECK-NEXT: v_mul_hi_u32 v6, v9, v6
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v11, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v10, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v7, v12
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v8, v13
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v9, v7
|
|
; CHECK-NEXT: v_add_i32_e64 v9, s[4:5], v10, v11
|
|
; CHECK-NEXT: v_add_i32_e64 v7, s[4:5], v8, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; CHECK-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; CHECK-NEXT: v_add_i32_e64 v6, s[4:5], v6, v8
|
|
; CHECK-NEXT: v_addc_u32_e32 v3, vcc, v3, v6, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v7
|
|
; CHECK-NEXT: v_addc_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v7, v0, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v2, v1, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v0, v3
|
|
; CHECK-NEXT: v_mul_lo_u32 v9, v1, v3
|
|
; CHECK-NEXT: v_mul_hi_u32 v10, v0, v3
|
|
; CHECK-NEXT: v_mul_hi_u32 v3, v1, v3
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v9, v2
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v10
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v7, vcc, v9, v7
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, v2, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v7, v4, v2
|
|
; CHECK-NEXT: v_mul_lo_u32 v8, v5, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v9, v4, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v3, vcc, v3, v6
|
|
; CHECK-NEXT: v_mul_lo_u32 v6, v4, v3
|
|
; CHECK-NEXT: v_add_i32_e32 v10, vcc, 1, v2
|
|
; CHECK-NEXT: v_addc_u32_e32 v11, vcc, 0, v3, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v8, v6
|
|
; CHECK-NEXT: v_add_i32_e32 v8, vcc, 1, v10
|
|
; CHECK-NEXT: v_addc_u32_e32 v12, vcc, 0, v11, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v6, vcc, v6, v9
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v7
|
|
; CHECK-NEXT: v_subb_u32_e64 v7, s[4:5], v1, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v6
|
|
; CHECK-NEXT: v_cmp_ge_u32_e64 s[4:5], v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_cmp_ge_u32_e64 s[4:5], v7, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; CHECK-NEXT: v_subb_u32_e32 v1, vcc, v1, v5, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, v7, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v6, v9, v6, vcc
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v4
|
|
; CHECK-NEXT: v_subbrev_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v1, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e64 v4, 0, -1, vcc
|
|
; CHECK-NEXT: v_cmp_eq_u32_e32 vcc, v1, v5
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v4, v0, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v10, v8, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v11, v12, vcc
|
|
; CHECK-NEXT: v_cmp_ne_u32_e32 vcc, 0, v6
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v2, v2, v0, vcc
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v3, v3, v1, vcc
|
|
; CHECK-NEXT: ; implicit-def: $vgpr4_vgpr5
|
|
; CHECK-NEXT: ; implicit-def: $vgpr0
|
|
; CHECK-NEXT: BB7_2: ; %Flow
|
|
; CHECK-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CHECK-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CHECK-NEXT: s_cbranch_execz BB7_4
|
|
; CHECK-NEXT: ; %bb.3:
|
|
; CHECK-NEXT: v_cvt_f32_u32_e32 v1, v4
|
|
; CHECK-NEXT: v_sub_i32_e32 v2, vcc, 0, v4
|
|
; CHECK-NEXT: v_rcp_iflag_f32_e32 v1, v1
|
|
; CHECK-NEXT: v_mul_f32_e32 v1, 0x4f7ffffe, v1
|
|
; CHECK-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; CHECK-NEXT: v_mul_lo_u32 v2, v2, v1
|
|
; CHECK-NEXT: v_mul_hi_u32 v2, v1, v2
|
|
; CHECK-NEXT: v_add_i32_e32 v1, vcc, v1, v2
|
|
; CHECK-NEXT: v_mul_hi_u32 v1, v0, v1
|
|
; CHECK-NEXT: v_mul_lo_u32 v2, v1, v4
|
|
; CHECK-NEXT: v_add_i32_e32 v3, vcc, 1, v1
|
|
; CHECK-NEXT: v_sub_i32_e32 v0, vcc, v0, v2
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v1, v1, v3, vcc
|
|
; CHECK-NEXT: v_sub_i32_e64 v2, s[4:5], v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v0, v0, v2, vcc
|
|
; CHECK-NEXT: v_add_i32_e32 v2, vcc, 1, v1
|
|
; CHECK-NEXT: v_cmp_ge_u32_e32 vcc, v0, v4
|
|
; CHECK-NEXT: v_cndmask_b32_e32 v2, v1, v2, vcc
|
|
; CHECK-NEXT: v_mov_b32_e32 v3, 0
|
|
; CHECK-NEXT: BB7_4:
|
|
; CHECK-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CHECK-NEXT: v_mov_b32_e32 v0, v2
|
|
; CHECK-NEXT: v_mov_b32_e32 v1, v3
|
|
; CHECK-NEXT: s_setpc_b64 s[30:31]
|
|
%shl.y = shl i64 4096, %y
|
|
%r = udiv i64 %x, %shl.y
|
|
ret i64 %r
|
|
}
|
|
|
|
define <2 x i64> @v_udiv_v2i64_pow2_shl_denom(<2 x i64> %x, <2 x i64> %y) {
|
|
; GISEL-LABEL: v_udiv_v2i64_pow2_shl_denom:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: s_mov_b64 s[4:5], 0x1000
|
|
; GISEL-NEXT: v_lshl_b64 v[4:5], s[4:5], v4
|
|
; GISEL-NEXT: v_lshl_b64 v[6:7], s[4:5], v6
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v8, v4
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v9, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v8, 0x4f800000, v9
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_mul_f32_e32 v8, 0x5f7ffffc, v8
|
|
; GISEL-NEXT: v_mul_f32_e32 v9, 0x2f800000, v8
|
|
; GISEL-NEXT: v_trunc_f32_e32 v9, v9
|
|
; GISEL-NEXT: v_mac_f32_e32 v8, 0xcf800000, v9
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v9, v9
|
|
; GISEL-NEXT: v_sub_i32_e32 v10, vcc, 0, v4
|
|
; GISEL-NEXT: v_subb_u32_e32 v11, vcc, 0, v5, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v10, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v10, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v15
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v9, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v8, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v8, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v9, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v9, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v8, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v15, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v15, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v9, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_addc_u32_e64 v12, s[4:5], v9, v13, vcc
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v10, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v11, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v10, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v14
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v11, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v12, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v8, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v8, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v14, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v12, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v12, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v8, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v13, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[4:5], v14, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v13, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[4:5], v14, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v12, v10
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[4:5], v10, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v9, vcc, v9, v10, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v9, vcc, 0, v9, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v1, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v0, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v1, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v1, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v0, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v11, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v11, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v1, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v4, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v5, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v13
|
|
; GISEL-NEXT: v_sub_i32_e32 v0, vcc, v0, v10
|
|
; GISEL-NEXT: v_subb_u32_e64 v10, s[4:5], v1, v11, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v1, s[4:5], v1, v11
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v10, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v0, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], v10, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, v11, v12, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v0, s[4:5], v0, v4
|
|
; GISEL-NEXT: v_subb_u32_e32 v1, vcc, v1, v5, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v1, vcc, 0, v1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, 1, v8
|
|
; GISEL-NEXT: v_addc_u32_e32 v12, vcc, 0, v9, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v1, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v0, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, v1, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v13, v0, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v1, vcc, 1, v11
|
|
; GISEL-NEXT: v_addc_u32_e32 v4, vcc, 0, v12, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v11, v1, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v12, v4, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v8, v0, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v9, v1, vcc
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v4, v6
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v5, v7
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; GISEL-NEXT: v_mul_f32_e32 v5, 0x2f800000, v4
|
|
; GISEL-NEXT: v_trunc_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_mac_f32_e32 v4, 0xcf800000, v5
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_sub_i32_e32 v8, vcc, 0, v6
|
|
; GISEL-NEXT: v_subb_u32_e32 v9, vcc, 0, v7, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v8, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v9, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v8, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v8, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v5, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v4, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v4, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v5, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v5, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v13, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v5, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_addc_u32_e64 v10, s[4:5], v5, v11, vcc
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v5, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v8, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v9, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v8, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v8, v4
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v9, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v10, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v4, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v9, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v12, v9
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v10, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v10, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v4, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v12, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v11, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[4:5], v12, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[4:5], v11, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[4:5], v12, v11
|
|
; GISEL-NEXT: v_mul_hi_u32 v8, v10, v8
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v11
|
|
; GISEL-NEXT: v_add_i32_e64 v4, s[4:5], v4, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v5, vcc, v5, v8, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v5, vcc, 0, v5, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, v3, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v2, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v8, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v3, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v2, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v9, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, v4, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v9, v8
|
|
; GISEL-NEXT: v_mul_hi_u32 v5, v3, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, v6, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v7, v4
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v6, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v11, v6, v4
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v10
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v11
|
|
; GISEL-NEXT: v_sub_i32_e32 v2, vcc, v2, v8
|
|
; GISEL-NEXT: v_subb_u32_e64 v8, s[4:5], v3, v9, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v9
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v8, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v9, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v6
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v10, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], v8, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, v9, v10, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v2, s[4:5], v2, v6
|
|
; GISEL-NEXT: v_subb_u32_e32 v3, vcc, v3, v7, vcc
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v3, vcc, 0, v3, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, 1, v4
|
|
; GISEL-NEXT: v_addc_u32_e32 v10, vcc, 0, v5, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v3, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v2, v6
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, v3, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v11, v2, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v3, vcc, 1, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, 0, v10, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v9, v3, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v10, v6, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v4, v2, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v5, v3, vcc
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_v2i64_pow2_shl_denom:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: v_mov_b32_e32 v5, v0
|
|
; CGP-NEXT: v_mov_b32_e32 v7, v1
|
|
; CGP-NEXT: s_mov_b64 s[4:5], 0x1000
|
|
; CGP-NEXT: v_lshl_b64 v[10:11], s[4:5], v4
|
|
; CGP-NEXT: v_lshl_b64 v[8:9], s[4:5], v6
|
|
; CGP-NEXT: v_or_b32_e32 v1, v7, v11
|
|
; CGP-NEXT: v_mov_b32_e32 v0, 0
|
|
; CGP-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[0:1]
|
|
; CGP-NEXT: ; implicit-def: $vgpr0_vgpr1
|
|
; CGP-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CGP-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CGP-NEXT: s_cbranch_execz BB8_2
|
|
; CGP-NEXT: ; %bb.1:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v10
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v1, v11
|
|
; CGP-NEXT: v_sub_i32_e32 v4, vcc, 0, v10
|
|
; CGP-NEXT: v_subb_u32_e32 v6, vcc, 0, v11, vcc
|
|
; CGP-NEXT: v_mac_f32_e32 v0, 0x4f800000, v1
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v0, 0x5f7ffffc, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v1, 0x2f800000, v0
|
|
; CGP-NEXT: v_trunc_f32_e32 v1, v1
|
|
; CGP-NEXT: v_mac_f32_e32 v0, 0xcf800000, v1
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v4, v1
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v4, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v6, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v15, v4, v0
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v14, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v1, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v0, v13
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v1, v13
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v15
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v0, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v17, v1, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v18, v0, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v1, v12
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v17, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v14, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v18
|
|
; CGP-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v15, vcc, v17, v16
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v13, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v13
|
|
; CGP-NEXT: v_addc_u32_e64 v13, s[4:5], v1, v12, vcc
|
|
; CGP-NEXT: v_add_i32_e64 v1, s[4:5], v1, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v4, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v6, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v4, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v4, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v13, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v0, v12
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v13, v12
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v6, v4
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v4, v14
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v0, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v14, v13, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v17, v0, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v13, v4
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v15, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v14, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v6, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v12, v17
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v13, v6
|
|
; CGP-NEXT: v_add_i32_e64 v13, s[4:5], v14, v15
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v12, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v12, s[4:5], v13, v12
|
|
; CGP-NEXT: v_add_i32_e64 v4, s[4:5], v4, v12
|
|
; CGP-NEXT: v_addc_u32_e32 v1, vcc, v1, v4, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v6
|
|
; CGP-NEXT: v_addc_u32_e32 v1, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v7, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v6, v5, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v0, v7, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v5, v1
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v7, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v5, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v1, v7, v1
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v13, v0
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v4, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v12, v4
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v13, v6
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v4, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v6, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v10, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v11, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v10, v0
|
|
; CGP-NEXT: v_add_i32_e32 v1, vcc, v1, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v10, v1
|
|
; CGP-NEXT: v_add_i32_e32 v14, vcc, 1, v0
|
|
; CGP-NEXT: v_addc_u32_e32 v15, vcc, 0, v1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v12, v4
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, 1, v14
|
|
; CGP-NEXT: v_addc_u32_e32 v16, vcc, 0, v15, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v13
|
|
; CGP-NEXT: v_sub_i32_e32 v5, vcc, v5, v6
|
|
; CGP-NEXT: v_subb_u32_e64 v6, s[4:5], v7, v4, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v4, s[4:5], v7, v4
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v5, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v7, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v6, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_subb_u32_e32 v4, vcc, v4, v11, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v6, v11
|
|
; CGP-NEXT: v_cndmask_b32_e32 v6, v13, v7, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v5, vcc, v5, v10
|
|
; CGP-NEXT: v_subbrev_u32_e32 v4, vcc, 0, v4, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v5, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v5, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v4, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v7, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v4, v11
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v7, v5, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v4
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v14, v12, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v5, v15, v16, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v6
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v4, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v1, v5, vcc
|
|
; CGP-NEXT: ; implicit-def: $vgpr10_vgpr11
|
|
; CGP-NEXT: ; implicit-def: $vgpr5
|
|
; CGP-NEXT: BB8_2: ; %Flow2
|
|
; CGP-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CGP-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: s_cbranch_execz BB8_4
|
|
; CGP-NEXT: ; %bb.3:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v10
|
|
; CGP-NEXT: v_sub_i32_e32 v1, vcc, 0, v10
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_f32_e32 v0, 0x4f7ffffe, v0
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v0, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v1, v1, v0
|
|
; CGP-NEXT: v_mul_hi_u32 v1, v0, v1
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v0, v1
|
|
; CGP-NEXT: v_mul_hi_u32 v0, v5, v0
|
|
; CGP-NEXT: v_mul_lo_u32 v1, v0, v10
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, 1, v0
|
|
; CGP-NEXT: v_sub_i32_e32 v1, vcc, v5, v1
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v1, v10
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v4, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v4, s[4:5], v1, v10
|
|
; CGP-NEXT: v_cndmask_b32_e32 v1, v1, v4, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, 1, v0
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v1, v10
|
|
; CGP-NEXT: v_cndmask_b32_e32 v0, v0, v4, vcc
|
|
; CGP-NEXT: v_mov_b32_e32 v1, 0
|
|
; CGP-NEXT: BB8_4:
|
|
; CGP-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: v_or_b32_e32 v5, v3, v9
|
|
; CGP-NEXT: v_mov_b32_e32 v4, 0
|
|
; CGP-NEXT: v_cmp_ne_u64_e32 vcc, 0, v[4:5]
|
|
; CGP-NEXT: ; implicit-def: $vgpr4_vgpr5
|
|
; CGP-NEXT: s_and_saveexec_b64 s[4:5], vcc
|
|
; CGP-NEXT: s_xor_b64 s[6:7], exec, s[4:5]
|
|
; CGP-NEXT: s_cbranch_execz BB8_6
|
|
; CGP-NEXT: ; %bb.5:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v4, v8
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v5, v9
|
|
; CGP-NEXT: v_sub_i32_e32 v6, vcc, 0, v8
|
|
; CGP-NEXT: v_subb_u32_e32 v7, vcc, 0, v9, vcc
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0x4f800000, v5
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v4, 0x5f7ffffc, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v5, 0x2f800000, v4
|
|
; CGP-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CGP-NEXT: v_mac_f32_e32 v4, 0xcf800000, v5
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v6, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v6, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v7, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v13, v6, v4
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v12, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v5, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v4, v11
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v5, v11
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v13
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v4, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v15, v5, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v16, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v5, v10
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v13
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v15, v11
|
|
; CGP-NEXT: v_cndmask_b32_e64 v15, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v12, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v16
|
|
; CGP-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; CGP-NEXT: v_add_i32_e32 v13, vcc, v15, v14
|
|
; CGP-NEXT: v_add_i32_e32 v11, vcc, v11, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, v13, v12
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, v10, v12
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v11
|
|
; CGP-NEXT: v_addc_u32_e64 v11, s[4:5], v5, v10, vcc
|
|
; CGP-NEXT: v_add_i32_e64 v5, s[4:5], v5, v10
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v6, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v7, v7, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v6, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v6, v11
|
|
; CGP-NEXT: v_mul_lo_u32 v13, v11, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v14, v4, v10
|
|
; CGP-NEXT: v_mul_hi_u32 v10, v11, v10
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v7, v6
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v6, v12
|
|
; CGP-NEXT: v_mul_lo_u32 v7, v4, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v12, v11, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v15, v4, v6
|
|
; CGP-NEXT: v_mul_hi_u32 v6, v11, v6
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[4:5], v13, v7
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v12, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v12, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[4:5], v7, v14
|
|
; CGP-NEXT: v_cndmask_b32_e64 v7, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v10, v15
|
|
; CGP-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[4:5], v11, v7
|
|
; CGP-NEXT: v_add_i32_e64 v11, s[4:5], v12, v13
|
|
; CGP-NEXT: v_add_i32_e64 v7, s[4:5], v10, v7
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e64 v10, s[4:5], v11, v10
|
|
; CGP-NEXT: v_add_i32_e64 v6, s[4:5], v6, v10
|
|
; CGP-NEXT: v_addc_u32_e32 v5, vcc, v5, v6, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v7
|
|
; CGP-NEXT: v_addc_u32_e32 v5, vcc, 0, v5, vcc
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v7, v2, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v2, v5
|
|
; CGP-NEXT: v_mul_lo_u32 v11, v3, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v12, v2, v5
|
|
; CGP-NEXT: v_mul_hi_u32 v5, v3, v5
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v6, v10
|
|
; CGP-NEXT: v_cndmask_b32_e64 v10, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v11, v4
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v6, v7
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v12
|
|
; CGP-NEXT: v_cndmask_b32_e64 v7, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v10, v6
|
|
; CGP-NEXT: v_add_i32_e32 v7, vcc, v11, v7
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, v4, v6
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, 1, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v7, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v7, v8, v4
|
|
; CGP-NEXT: v_mul_lo_u32 v10, v9, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v11, v8, v4
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, v5, v6
|
|
; CGP-NEXT: v_mul_lo_u32 v6, v8, v5
|
|
; CGP-NEXT: v_add_i32_e32 v12, vcc, 1, v4
|
|
; CGP-NEXT: v_addc_u32_e32 v13, vcc, 0, v5, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v10, v6
|
|
; CGP-NEXT: v_add_i32_e32 v10, vcc, 1, v12
|
|
; CGP-NEXT: v_addc_u32_e32 v14, vcc, 0, v13, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v6, vcc, v6, v11
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v7
|
|
; CGP-NEXT: v_subb_u32_e64 v7, s[4:5], v3, v6, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v3, s[4:5], v3, v6
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v8
|
|
; CGP-NEXT: v_cndmask_b32_e64 v6, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_cmp_ge_u32_e64 s[4:5], v7, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v11, 0, -1, s[4:5]
|
|
; CGP-NEXT: v_subb_u32_e32 v3, vcc, v3, v9, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v7, v9
|
|
; CGP-NEXT: v_cndmask_b32_e32 v6, v11, v6, vcc
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v8
|
|
; CGP-NEXT: v_subbrev_u32_e32 v3, vcc, 0, v3, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v8
|
|
; CGP-NEXT: v_cndmask_b32_e64 v2, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v3, v9
|
|
; CGP-NEXT: v_cndmask_b32_e64 v7, 0, -1, vcc
|
|
; CGP-NEXT: v_cmp_eq_u32_e32 vcc, v3, v9
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v7, v2, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v2
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v12, v10, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v13, v14, vcc
|
|
; CGP-NEXT: v_cmp_ne_u32_e32 vcc, 0, v6
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v4, v2, vcc
|
|
; CGP-NEXT: v_cndmask_b32_e32 v5, v5, v3, vcc
|
|
; CGP-NEXT: ; implicit-def: $vgpr8_vgpr9
|
|
; CGP-NEXT: ; implicit-def: $vgpr2
|
|
; CGP-NEXT: BB8_6: ; %Flow
|
|
; CGP-NEXT: s_or_saveexec_b64 s[6:7], s[6:7]
|
|
; CGP-NEXT: s_xor_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: s_cbranch_execz BB8_8
|
|
; CGP-NEXT: ; %bb.7:
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v3, v8
|
|
; CGP-NEXT: v_sub_i32_e32 v4, vcc, 0, v8
|
|
; CGP-NEXT: v_rcp_iflag_f32_e32 v3, v3
|
|
; CGP-NEXT: v_mul_f32_e32 v3, 0x4f7ffffe, v3
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v3, v3
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v4, v3
|
|
; CGP-NEXT: v_mul_hi_u32 v4, v3, v4
|
|
; CGP-NEXT: v_add_i32_e32 v3, vcc, v3, v4
|
|
; CGP-NEXT: v_mul_hi_u32 v3, v2, v3
|
|
; CGP-NEXT: v_mul_lo_u32 v4, v3, v8
|
|
; CGP-NEXT: v_add_i32_e32 v5, vcc, 1, v3
|
|
; CGP-NEXT: v_sub_i32_e32 v2, vcc, v2, v4
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v3, v3, v5, vcc
|
|
; CGP-NEXT: v_sub_i32_e64 v4, s[4:5], v2, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v2, v2, v4, vcc
|
|
; CGP-NEXT: v_add_i32_e32 v4, vcc, 1, v3
|
|
; CGP-NEXT: v_cmp_ge_u32_e32 vcc, v2, v8
|
|
; CGP-NEXT: v_cndmask_b32_e32 v4, v3, v4, vcc
|
|
; CGP-NEXT: v_mov_b32_e32 v5, 0
|
|
; CGP-NEXT: BB8_8:
|
|
; CGP-NEXT: s_or_b64 exec, exec, s[6:7]
|
|
; CGP-NEXT: v_mov_b32_e32 v2, v4
|
|
; CGP-NEXT: v_mov_b32_e32 v3, v5
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%shl.y = shl <2 x i64> <i64 4096, i64 4096>, %y
|
|
%r = udiv <2 x i64> %x, %shl.y
|
|
ret <2 x i64> %r
|
|
}
|
|
|
|
define i64 @v_udiv_i64_24bit(i64 %num, i64 %den) {
|
|
; GISEL-LABEL: v_udiv_i64_24bit:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: s_mov_b32 s4, 0xffffff
|
|
; GISEL-NEXT: v_and_b32_e32 v0, s4, v0
|
|
; GISEL-NEXT: v_and_b32_e32 v1, s4, v2
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v2, v1
|
|
; GISEL-NEXT: v_sub_i32_e32 v3, vcc, 0, v1
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v2, v2
|
|
; GISEL-NEXT: v_mul_f32_e32 v2, 0x4f7ffffe, v2
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; GISEL-NEXT: v_mul_lo_u32 v3, v3, v2
|
|
; GISEL-NEXT: v_mul_hi_u32 v3, v2, v3
|
|
; GISEL-NEXT: v_add_i32_e32 v2, vcc, v2, v3
|
|
; GISEL-NEXT: v_mul_hi_u32 v2, v0, v2
|
|
; GISEL-NEXT: v_mul_lo_u32 v3, v2, v1
|
|
; GISEL-NEXT: v_add_i32_e32 v4, vcc, 1, v2
|
|
; GISEL-NEXT: v_sub_i32_e32 v0, vcc, v0, v3
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v0, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v2, v2, v4, vcc
|
|
; GISEL-NEXT: v_sub_i32_e64 v3, s[4:5], v0, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v0, v3, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v3, vcc, 1, v2
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v0, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v2, v3, vcc
|
|
; GISEL-NEXT: v_mov_b32_e32 v1, 0
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_i64_24bit:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: s_mov_b32 s4, 0xffffff
|
|
; CGP-NEXT: v_and_b32_e32 v0, s4, v0
|
|
; CGP-NEXT: v_and_b32_e32 v1, s4, v2
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v0
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v1, v1
|
|
; CGP-NEXT: v_rcp_f32_e32 v2, v1
|
|
; CGP-NEXT: v_mul_f32_e32 v2, v0, v2
|
|
; CGP-NEXT: v_trunc_f32_e32 v2, v2
|
|
; CGP-NEXT: v_mad_f32 v0, -v2, v1, v0
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v2, v2
|
|
; CGP-NEXT: v_cmp_ge_f32_e64 s[4:5], |v0|, v1
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v2, v0
|
|
; CGP-NEXT: v_and_b32_e32 v0, 0xffffff, v0
|
|
; CGP-NEXT: v_mov_b32_e32 v1, 0
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%num.mask = and i64 %num, 16777215
|
|
%den.mask = and i64 %den, 16777215
|
|
%result = udiv i64 %num.mask, %den.mask
|
|
ret i64 %result
|
|
}
|
|
|
|
define <2 x i64> @v_udiv_v2i64_24bit(<2 x i64> %num, <2 x i64> %den) {
|
|
; GISEL-LABEL: v_udiv_v2i64_24bit:
|
|
; GISEL: ; %bb.0:
|
|
; GISEL-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; GISEL-NEXT: s_mov_b32 s6, 0xffffff
|
|
; GISEL-NEXT: v_cvt_f32_ubyte0_e32 v1, 0
|
|
; GISEL-NEXT: v_and_b32_e32 v3, s6, v4
|
|
; GISEL-NEXT: v_and_b32_e32 v4, s6, v6
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v5, v3
|
|
; GISEL-NEXT: v_sub_i32_e32 v6, vcc, 0, v3
|
|
; GISEL-NEXT: v_subb_u32_e64 v7, s[4:5], 0, 0, vcc
|
|
; GISEL-NEXT: v_cvt_f32_u32_e32 v8, v4
|
|
; GISEL-NEXT: v_sub_i32_e32 v9, vcc, 0, v4
|
|
; GISEL-NEXT: v_subb_u32_e64 v10, s[4:5], 0, 0, vcc
|
|
; GISEL-NEXT: v_mac_f32_e32 v5, 0x4f800000, v1
|
|
; GISEL-NEXT: v_mac_f32_e32 v8, 0x4f800000, v1
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v1, v5
|
|
; GISEL-NEXT: v_rcp_iflag_f32_e32 v5, v8
|
|
; GISEL-NEXT: v_mul_f32_e32 v1, 0x5f7ffffc, v1
|
|
; GISEL-NEXT: v_mul_f32_e32 v5, 0x5f7ffffc, v5
|
|
; GISEL-NEXT: v_mul_f32_e32 v8, 0x2f800000, v1
|
|
; GISEL-NEXT: v_mul_f32_e32 v11, 0x2f800000, v5
|
|
; GISEL-NEXT: v_trunc_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_trunc_f32_e32 v11, v11
|
|
; GISEL-NEXT: v_mac_f32_e32 v1, 0xcf800000, v8
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v8, v8
|
|
; GISEL-NEXT: v_mac_f32_e32 v5, 0xcf800000, v11
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v11, v11
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v1, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v12, v6, v8
|
|
; GISEL-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v9, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v6, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v7, v1
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v6, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, v9, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v10, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v19, v9, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v15, v12
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v18, v13
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v11, v17
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v5, v17
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v19
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v5, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v15, v19
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v15, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v8, v14
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v1, v14
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v8, v14
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v11, v17
|
|
; GISEL-NEXT: v_add_i32_e64 v12, s[4:5], v12, v16
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, v1, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[4:5], v15, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[4:5], v15, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v8, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v16, s[4:5], v16, v18
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v1, v12
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[4:5], v15, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[4:5], v14, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[4:5], v15, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v18, vcc, v19, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v11, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v17, vcc, v19, v17
|
|
; GISEL-NEXT: v_mul_hi_u32 v19, v5, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v20, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v17, vcc, v17, v19
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v19, vcc, v20, v19
|
|
; GISEL-NEXT: s_bfe_i32 s10, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s11, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s12, -1, 0x10000
|
|
; GISEL-NEXT: s_bfe_i32 s13, -1, 0x10000
|
|
; GISEL-NEXT: v_and_b32_e32 v0, s6, v0
|
|
; GISEL-NEXT: v_and_b32_e32 v2, s6, v2
|
|
; GISEL-NEXT: v_mul_hi_u32 v12, v8, v12
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v11, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v14, vcc, v14, v16
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v17, vcc, v17, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v15, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v16, vcc, v19, v18
|
|
; GISEL-NEXT: v_add_i32_e32 v12, vcc, v12, v15
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v13, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v1, vcc, v1, v14
|
|
; GISEL-NEXT: v_addc_u32_e64 v14, s[4:5], v8, v12, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v6, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v7, v7, v1
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v5, v17
|
|
; GISEL-NEXT: v_addc_u32_e64 v16, s[6:7], v11, v13, s[4:5]
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, v9, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v10, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v9, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v9, v16
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v16, v17
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v10, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v5, v17
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v9, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v5, v9
|
|
; GISEL-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[6:7], v18, v10
|
|
; GISEL-NEXT: v_mul_hi_u32 v10, v6, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v6, v6, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v14, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[8:9], v7, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v7, v1, v15
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[8:9], v6, v10
|
|
; GISEL-NEXT: v_mul_lo_u32 v10, v1, v6
|
|
; GISEL-NEXT: v_add_i32_e64 v10, s[8:9], v18, v10
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v7, s[8:9], v10, v7
|
|
; GISEL-NEXT: v_mov_b32_e32 v7, s10
|
|
; GISEL-NEXT: v_mov_b32_e32 v10, s11
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[10:11], v8, v12
|
|
; GISEL-NEXT: v_mov_b32_e32 v12, s12
|
|
; GISEL-NEXT: v_add_i32_e64 v11, s[10:11], v11, v13
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v14, v15
|
|
; GISEL-NEXT: v_mul_hi_u32 v15, v16, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v17, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v17, s[8:9], v18, v17
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v14, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v14, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v6, v1, v6
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[8:9], v18, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[8:9], v13, v6
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[8:9]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[8:9], v18, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v18, s[6:7], v19, v18
|
|
; GISEL-NEXT: v_mul_lo_u32 v19, v16, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v16, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, v5, v9
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[6:7], v19, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v19, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v15, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[6:7], v19, v15
|
|
; GISEL-NEXT: v_mov_b32_e32 v19, s13
|
|
; GISEL-NEXT: v_add_i32_e64 v6, s[6:7], v6, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v17, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v9, s[6:7], v9, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, s[6:7]
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[6:7], v13, v17
|
|
; GISEL-NEXT: v_add_i32_e64 v15, s[6:7], v15, v18
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[6:7], v14, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[6:7], v16, v15
|
|
; GISEL-NEXT: v_addc_u32_e32 v8, vcc, v8, v13, vcc
|
|
; GISEL-NEXT: v_addc_u32_e64 v11, vcc, v11, v14, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e32 v1, vcc, v1, v6
|
|
; GISEL-NEXT: v_addc_u32_e32 v6, vcc, 0, v8, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v8, 0, v1
|
|
; GISEL-NEXT: v_mul_hi_u32 v13, v0, v1
|
|
; GISEL-NEXT: v_mul_hi_u32 v1, 0, v1
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v9
|
|
; GISEL-NEXT: v_addc_u32_e32 v9, vcc, 0, v11, vcc
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, 0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v2, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v5, 0, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, v0, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v16, 0, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v17, v0, v6
|
|
; GISEL-NEXT: v_mul_hi_u32 v6, 0, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v18, v2, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v18, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, 0, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v14, v2, v9
|
|
; GISEL-NEXT: v_mul_hi_u32 v9, 0, v9
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v15, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v1, s[4:5], v16, v1
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v5, s[4:5], v11, v5
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v11, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v8, s[4:5], v8, v13
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v1, s[4:5], v1, v17
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v13, 0, 1, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v17, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v15, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v13, vcc, v16, v13
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, v18, v17
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v14
|
|
; GISEL-NEXT: v_add_i32_e32 v1, vcc, v1, v8
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v8, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v5, vcc, v5, v15
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v14, 0, 1, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v13, v8
|
|
; GISEL-NEXT: v_mul_lo_u32 v13, v3, v1
|
|
; GISEL-NEXT: v_mul_lo_u32 v15, 0, v1
|
|
; GISEL-NEXT: v_mul_hi_u32 v16, v3, v1
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v14
|
|
; GISEL-NEXT: v_mul_lo_u32 v14, v4, v5
|
|
; GISEL-NEXT: v_mul_lo_u32 v17, 0, v5
|
|
; GISEL-NEXT: v_mul_hi_u32 v18, v4, v5
|
|
; GISEL-NEXT: v_add_i32_e32 v6, vcc, v6, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v8, vcc, v9, v11
|
|
; GISEL-NEXT: v_mul_lo_u32 v9, v3, v6
|
|
; GISEL-NEXT: v_mul_lo_u32 v11, v4, v8
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v15, v9
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v17, v11
|
|
; GISEL-NEXT: v_add_i32_e32 v15, vcc, 1, v1
|
|
; GISEL-NEXT: v_addc_u32_e32 v17, vcc, 0, v6, vcc
|
|
; GISEL-NEXT: v_add_i32_e32 v9, vcc, v9, v16
|
|
; GISEL-NEXT: v_add_i32_e32 v11, vcc, v11, v18
|
|
; GISEL-NEXT: v_sub_i32_e32 v0, vcc, v0, v13
|
|
; GISEL-NEXT: v_subb_u32_e64 v13, s[4:5], 0, v9, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v0, v3
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], 0, v13
|
|
; GISEL-NEXT: v_add_i32_e64 v13, s[6:7], 1, v5
|
|
; GISEL-NEXT: v_addc_u32_e64 v18, s[6:7], 0, v8, s[6:7]
|
|
; GISEL-NEXT: v_sub_i32_e64 v2, s[6:7], v2, v14
|
|
; GISEL-NEXT: v_subb_u32_e64 v14, s[8:9], 0, v11, s[6:7]
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v7, v7, v16, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v4
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v16, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e64 s[4:5], 0, v14
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v12, v12, v16, s[4:5]
|
|
; GISEL-NEXT: v_add_i32_e64 v14, s[4:5], 1, v15
|
|
; GISEL-NEXT: v_addc_u32_e64 v16, s[4:5], 0, v17, s[4:5]
|
|
; GISEL-NEXT: v_sub_i32_e64 v11, s[4:5], 0, v11
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v11, s[4:5], 0, v11, s[6:7]
|
|
; GISEL-NEXT: v_sub_i32_e64 v2, s[4:5], v2, v4
|
|
; GISEL-NEXT: v_subbrev_u32_e64 v11, s[4:5], 0, v11, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ge_u32_e64 s[4:5], v2, v4
|
|
; GISEL-NEXT: v_add_i32_e64 v2, s[6:7], 1, v13
|
|
; GISEL-NEXT: v_addc_u32_e64 v4, s[6:7], 0, v18, s[6:7]
|
|
; GISEL-NEXT: v_sub_i32_e64 v9, s[6:7], 0, v9
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v9, vcc, 0, v9, vcc
|
|
; GISEL-NEXT: v_sub_i32_e32 v0, vcc, v0, v3
|
|
; GISEL-NEXT: v_subbrev_u32_e32 v9, vcc, 0, v9, vcc
|
|
; GISEL-NEXT: v_cmp_ge_u32_e32 vcc, v0, v3
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v0, 0, -1, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v3, 0, -1, s[4:5]
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v9
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v10, v0, vcc
|
|
; GISEL-NEXT: v_cmp_eq_u32_e32 vcc, 0, v11
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v19, v3, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v0
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v15, v14, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v3
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, v13, v2, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v3, v17, v16, vcc
|
|
; GISEL-NEXT: v_cmp_ne_u32_e32 vcc, 0, v7
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v0, v1, v0, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v4, v18, v4, s[4:5]
|
|
; GISEL-NEXT: v_cmp_ne_u32_e64 s[4:5], 0, v12
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v2, v5, v2, s[4:5]
|
|
; GISEL-NEXT: v_cndmask_b32_e32 v1, v6, v3, vcc
|
|
; GISEL-NEXT: v_cndmask_b32_e64 v3, v8, v4, s[4:5]
|
|
; GISEL-NEXT: s_setpc_b64 s[30:31]
|
|
;
|
|
; CGP-LABEL: v_udiv_v2i64_24bit:
|
|
; CGP: ; %bb.0:
|
|
; CGP-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
|
|
; CGP-NEXT: s_mov_b32 s6, 0xffffff
|
|
; CGP-NEXT: v_and_b32_e32 v0, s6, v0
|
|
; CGP-NEXT: v_and_b32_e32 v1, s6, v2
|
|
; CGP-NEXT: v_and_b32_e32 v2, s6, v4
|
|
; CGP-NEXT: v_and_b32_e32 v3, s6, v6
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v0, v0
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v2, v2
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v1, v1
|
|
; CGP-NEXT: v_cvt_f32_u32_e32 v3, v3
|
|
; CGP-NEXT: v_rcp_f32_e32 v4, v2
|
|
; CGP-NEXT: v_rcp_f32_e32 v5, v3
|
|
; CGP-NEXT: v_mul_f32_e32 v4, v0, v4
|
|
; CGP-NEXT: v_mul_f32_e32 v5, v1, v5
|
|
; CGP-NEXT: v_trunc_f32_e32 v4, v4
|
|
; CGP-NEXT: v_trunc_f32_e32 v5, v5
|
|
; CGP-NEXT: v_mad_f32 v0, -v4, v2, v0
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v4, v4
|
|
; CGP-NEXT: v_mad_f32 v1, -v5, v3, v1
|
|
; CGP-NEXT: v_cvt_u32_f32_e32 v5, v5
|
|
; CGP-NEXT: v_cmp_ge_f32_e64 s[4:5], |v0|, v2
|
|
; CGP-NEXT: v_cndmask_b32_e64 v0, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_cmp_ge_f32_e64 s[4:5], |v1|, v3
|
|
; CGP-NEXT: v_cndmask_b32_e64 v1, 0, 1, s[4:5]
|
|
; CGP-NEXT: v_add_i32_e32 v0, vcc, v4, v0
|
|
; CGP-NEXT: v_add_i32_e32 v1, vcc, v5, v1
|
|
; CGP-NEXT: v_and_b32_e32 v0, s6, v0
|
|
; CGP-NEXT: v_and_b32_e32 v2, s6, v1
|
|
; CGP-NEXT: v_mov_b32_e32 v1, 0
|
|
; CGP-NEXT: v_mov_b32_e32 v3, 0
|
|
; CGP-NEXT: s_setpc_b64 s[30:31]
|
|
%num.mask = and <2 x i64> %num, <i64 16777215, i64 16777215>
|
|
%den.mask = and <2 x i64> %den, <i64 16777215, i64 16777215>
|
|
%result = udiv <2 x i64> %num.mask, %den.mask
|
|
ret <2 x i64> %result
|
|
}
|