derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
87 lines
2.0 KiB
ArmAsm
87 lines
2.0 KiB
ArmAsm
//===-- clzdi2.c - Implement __clzdi2 -------------------------------------===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file implements count leading zeros for 64bit arguments.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "../assembly.h"
|
|
|
|
.syntax unified
|
|
.text
|
|
DEFINE_CODE_STATE
|
|
|
|
.p2align 2
|
|
DEFINE_COMPILERRT_FUNCTION(__clzdi2)
|
|
#ifdef __ARM_FEATURE_CLZ
|
|
#ifdef __ARMEB__
|
|
cmp r0, 0
|
|
itee ne
|
|
clzne r0, r0
|
|
clzeq r0, r1
|
|
addeq r0, r0, 32
|
|
#else
|
|
cmp r1, 0
|
|
itee ne
|
|
clzne r0, r1
|
|
clzeq r0, r0
|
|
addeq r0, r0, 32
|
|
#endif
|
|
JMP(lr)
|
|
#else
|
|
// Assumption: n != 0
|
|
|
|
// r0: n
|
|
// r1: upper half of n, overwritten after check
|
|
// r1: count of leading zeros in n + 1
|
|
// r2: scratch register for shifted r0
|
|
#ifdef __ARMEB__
|
|
cmp r0, 0
|
|
moveq r0, r1
|
|
#else
|
|
cmp r1, 0
|
|
movne r0, r1
|
|
#endif
|
|
movne r1, 1
|
|
moveq r1, 33
|
|
|
|
// Basic block:
|
|
// if ((r0 >> SHIFT) == 0)
|
|
// r1 += SHIFT;
|
|
// else
|
|
// r0 >>= SHIFT;
|
|
// for descending powers of two as SHIFT.
|
|
#define BLOCK(shift) \
|
|
lsrs r2, r0, shift; \
|
|
movne r0, r2; \
|
|
addeq r1, shift \
|
|
|
|
BLOCK(16)
|
|
BLOCK(8)
|
|
BLOCK(4)
|
|
BLOCK(2)
|
|
|
|
// The basic block invariants at this point are (r0 >> 2) == 0 and
|
|
// r0 != 0. This means 1 <= r0 <= 3 and 0 <= (r0 >> 1) <= 1.
|
|
//
|
|
// r0 | (r0 >> 1) == 0 | (r0 >> 1) == 1 | -(r0 >> 1) | 1 - (r0 >> 1)
|
|
// ---+----------------+----------------+------------+--------------
|
|
// 1 | 1 | 0 | 0 | 1
|
|
// 2 | 0 | 1 | -1 | 0
|
|
// 3 | 0 | 1 | -1 | 0
|
|
//
|
|
// The r1's initial value of 1 compensates for the 1 here.
|
|
sub r0, r1, r0, lsr #1
|
|
|
|
JMP(lr)
|
|
#endif // __ARM_FEATURE_CLZ
|
|
END_COMPILERRT_FUNCTION(__clzdi2)
|
|
|
|
NO_EXEC_STACK_DIRECTIVE
|
|
|