NA
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
[lldb] Stop testing LLDB on Clang changes in pre-commit CI (#95537) This is a temporary measure to alleviate Linux pre-commit CI waiting times that started snowballing [recently](https://discourse.llvm.org/t/long-wait-for-linux-presubmit-testing/79547/5). My [initial estimate](https://github.com/llvm/llvm-project/pull/94208#issuecomment-2155972973) of 4 additional minutes spent per built seems to be in the right ballpark, but looks like that was the last straw to break camel's back. It seems that CI load got past the tipping point, and now it's not able to burn through the queue over the night on workdays. I don't intend to overthrow the consensus we reached in #94208, but it shouldn't come at the expense of the whole LLVM community. I'll enable this back as soon as we have news that we got more capacity for Linux pre-commit CI. | 2 年前 | |
| 6 个月前 | ||
[Github] Fix LLVM Project Tests Workflow on Linux (#122221) This patch fixes the LLVM project tests workflow on Linux. Two changes were needed. Firstly, some commands need to be performed with sudo now that the container executes as a non-root user. Second, we needed to change from ubuntu-latest to ubuntu-22.04 as ubuntu-latest not defaults to ubuntu-24.04 which causes setup-python to install a python executable linked against a newer version of glibc that is not found on ubuntu 22.04, which causes failures when CMake cannot execute the python interpreter that it finds. (cherry picked from commit a75917679549109fcbf92cb63ef61638517713d6) | 1 年前 | |
[bolt][tests] Skip tests that use perf when perf counters are unavailable (#107892) On the GitHub Action runners, perf always fails with the error below , so we need to skip the perf tests on platforms like this that have limited access to the perf counters. Access to performance monitoring and observability operations is limited. Consider adjusting /proc/sys/kernel/perf_event_paranoid setting to open access to performance monitoring and observability operations for processes without CAP_PERFMON, CAP_SYS_PTRACE or CAP_SYS_ADMIN Linux capability. More information can be found at 'Perf events and tool security' document: https://www.kernel.org/doc/html/latest/admin-guide/perf-security.html perf_event_paranoid setting is 4: -1: Allow use of (almost) all events by all users Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK >= 0: Disallow raw and ftrace function tracepoint access >= 1: Disallow CPU event access >= 2: Disallow kernel profiling To make the adjusted perf_event_paranoid setting permanent preserve it in /etc/sysctl.conf (e.g. kernel.perf_event_paranoid = <setting>) (cherry picked from commit 773353b20a49bfa0dab608d415c1b4734d037fce) | 1 年前 | |
[clang-tidy] Avoid capturing a local variable in a static lambda in UseRangesCheck (#111282) Fixes https://github.com/llvm/llvm-project/issues/109367 (cherry picked from commit acf92a47c0ece8562fd745215c478fe2d4ab5896) | 1 年前 | |
added all modifications made for msdebug | 7 个月前 | |
[bugfix] 修复msdebug-mi无法使用 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !74 merge fix_msdebug_mi into master [bugfix] 修复msdebug-mi无法使用 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 蓝区msdebug构建工程里,msdebug-mi从原先的shell脚本变成二进制,二进制是无法直接使用的 - **修改方案:** msdebug-mi目录下,增加msdebug_mi_script.sh,构建时拷贝成msdebug_mi - **修改内容:** 修改msdebug-mi/src/CMakeLists.txt - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [x] **功能自验截图**(请确保不体现个人信息)  - [x] **冒烟是否通过** NA ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** ---- See merge request: Ascend/msdebug!74 | 4 个月前 | |
[test][compiler-rt] Mark dlsym_alloc.c as unsupported on macos (#108439) With #106912, the test now fails on macos, e.g. https://green.lab.llvm.org/job/llvm.org/job/clang-stage1-RA/2058/. (cherry picked from commit d9ed8b018df725faec4076a3efdfcbd7a24c99f0) | 1 年前 | |
| 2 年前 | ||
【doc】资料易用性更新 Co-authored-by: luyq11<luyiqian4@h-partners.com> # message auto-generated for no-merge-commit merge: !125 merge master into master 【doc】资料易用性更新 Created-by: luyq11 Commit-by: luyq11 Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 资料易用性更新 - **修改方案:** - **修改内容:** - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** https://gitcode.com/Ascend/msot/issues/38 ---- See merge request: Ascend/msdebug!125 | 3 个月前 | |
added all modifications made for msdebug | 7 个月前 | |
[flang][debug] Set scope of internal functions correctly. (#99531) Summary: The functions internal to subroutine should have the scope set to the parent function. This allows a user to evaluate local variables of parent function when control is stopped in the child. Fixes #96314 Test Plan: Reviewers: Subscribers: Tasks: Tags: Differential Revision: https://phabricator.intern.facebook.com/D60250527 (cherry picked from commit 626022bfd18f335ef62a461992a05dfed4e6d715) | 2 年前 | |
Update libc/docs/configure.rst | 2 年前 | |
libclc: add half version of 'sign' (#99841) | 2 年前 | |
Bump version to 19.1.7 | 1 年前 | |
[libc++] Fix broken configuration system-libcxxabi on Apple (#110920) On Apple platforms, using system-libcxxabi as an ABI library wouldn't work because we'd try to re-export symbols from libc++abi that the system libc++abi.dylib might not have. Instead, only re-export those symbols when we're using the in-tree libc++abi. This does mean that libc++.dylib won't re-export any libc++abi symbols when building against the system libc++abi, which could be fixed in various ways. However, the best solution really depends on the intended use case, so this patch doesn't try to solve that problem. As a drive-by, also improve the diagnostic message when the user forgets to set the LIBCXX_CXX_ABI_INCLUDE_PATHS variable, which would previously lead to a confusing error. Closes #104672 (cherry picked from commit 21da4e7f51c7adfd0b1c5defc8bd0d16ea1ce759) | 1 年前 | |
[libunwind] Stop installing the mach-o module map (#105616) libunwind shouldn't know that compact_unwind_encoding.h is part of a MachO module that it doesn't own. Delete the mach-o module map, and let whatever is in charge of the mach-o directory be the one to say how its module is organized and where compact_unwind_encoding.h fits in. (cherry picked from commit 172c4a4a147833f1c08df1555f3170aa9ccb6cbe) | 1 年前 | |
[lld][WebAssembly] Fix use of uninitialized stack data with --wasm64 (#107780) In the case of --wasm64 we were setting the type of the init expression to be 64-bit but were only setting the low 32-bits of the value (by assigning to Int32). Fixes: https://github.com/emscripten-core/emscripten/issues/22538 (cherry picked from commit 5c8fd1eece8fff69871cef57a2363dc0f734a7d1) | 1 年前 | |
remove useless code | 3 个月前 | |
[runtimes] Correctly apply libdir subdir for multilib (#93354) We weren't applying the libdir subdir to header directories but this is necessary for correctness when building e.g. ASan variant. This change also updates path construction logic accross all runtimes and ensures they're consistent. | 2 年前 | |
[feature]根据断点位置自动下发软硬断点 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !130 merge fix_breapoint into master [feature]根据断点位置自动下发软硬断点 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 1. 当前simd_vf, simt_vf上只支持硬断点设置,用户需要自己识别并修改断点设置参数,本mr用于自动识别断点位置是否处于simt_vf, simd_vf来 2. 当前断点ID会由于多个相同的kernel object递增,这里做了删除操作。 - **修改方案:** 编译器提供每个函数的类型信息,simd/simt等,msdebug解析出pc所在的函数类型信息。具体地:在Function类里增加function_class字段,表示函数类型信息;在设置断点的时候,获取pc的SymbolContext里的function类型,识别出是simt/simd函数类型后,修改断点类型。 - **修改内容:** 1. lldb/include/lldb/Symbol/Function.h 里的Function类里增加function_class字段,每个bit有自己的语义,和编译器对齐 2. 增加SymbolContext的判断在Breakpoint::AddLocation函数里,这个位置属于比较底层了,函数名/文件名行号/地址,断点解析后都会走到这里 3. 增加函数将NotifyModuleUpdated能力释放出来,使用新的module替换老的module, 这样当用户没跑算子时,工具识别到的断点和运行后工具产生的断点能够合并。因为后者才是真正的断点 - [ ] **涉及代码双合**(贴上另一个PR链接):NA ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息)  - [ ] **冒烟是否通过**  ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** NA - [ ] **是否需要在sig会议中进行代码检视** NA - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** NA ---- See merge request: Ascend/msdebug!130 | 3 个月前 | |
Fix sign of largest known divisor of div. (#100081) There's a missing abs, so it returns a negative value if the divisor is negative. Later this is then cast to uint. | 2 年前 | |
[feature] mi提供断点是否simt位置信息 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !136 merge fix_msdebug_mi into master [feature] mi提供断点是否simt位置信息 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 可视化调试时, ascend info threads、ascend thread <id>命令需要判断当前是否是simt函数内,所以mi需要提供这样的信息给插件使用 - **修改方案:** 仿照之前的is-device,增加is-simt的字段;若是切核,会输出Switching字段,可判断有无Thread字段来区分是否simt - **修改内容:** 提供SBProcess::IsSBStopInSimt()给MI使用 - [ ] **涉及代码双合**(贴上另一个PR链接):NA ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) mi命中断点后输出:  - [ ] **冒烟是否通过** 不影响 ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** NA - [ ] **是否需要在sig会议中进行代码检视** NA - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** NA ---- See merge request: Ascend/msdebug!136 | 3 个月前 | |
[cmake] Use DetectTestCompiler from openmp directory Fix the DetectTestCompiler project use to reference the openmp source tree, since the respective files were not copied to offload, and there is no point in duplicating them. Fixes #90333 | 1 年前 | |
[OpenMP][AArch64] Fix branch protection in microtasks (#102317) Start __kmp_invoke_microtask with PACBTI in order to identify the function as a valid branch target. Before returning, SP is authenticated. Also add the BTI and PAC markers to z_Linux_asm.S. With this patch, libomp.so can now be generated with DT_AARCH64_BTI_PLT when built with -mbranch-protection=standard. The implementation is based on the code available in compiler-rt. (cherry picked from commit 0aa22dcd2f6ec5f46b8ef18fee88066463734935) | 1 年前 | |
[feature] 更新版本号 Co-authored-by: l00617896<liusa2@huawei.com> # message auto-generated for no-merge-commit merge: !49 merge asc_version into master [feature] 更新版本号 Created-by: liusa002 Commit-by: l00617896 Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 版本号更新 - **修改方案:** 从1.0.0更新为26.0.0 - **修改内容:** 出包version - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) 执行 python build.py,回显的版本号正确: Self-extractable archive "mindstudio-debugger_26.0.0_aarch64.run" successfully created. output下的run包名称也正确: mindstudio-debugger_26.0.0_aarch64.run 执行msdebug -v,回显正确: msdebug version 26.0.0-9d8def7 - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** ---- See merge request: Ascend/msdebug!49 | 4 个月前 | |
[IRBuilder] Don't include Module.h (NFC) (#97159) This used to be necessary to fetch the DataLayout, but isn't anymore. | 2 年前 | |
[feature] 新增precommit配置文件检查代码 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !105 merge add_pre_commit into master [feature] 新增precommit配置文件检查代码 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 新增pre-commit配置文件用于之后的流水线上的代码检查,本地也支持代码检查 - **修改方案:** 直接新增配置文件,代码检查规则复用仓库里的.clang-format,并且这里只对增量代码做检查,和llvm-project的代码检查规则保持一致。 - **修改内容:** 新增.pre-commit-config.yaml文件和pre-commit/typos.toml文件,只保留c++代码和拼写等检查,移除python语言,不涉及 https://gitcode.com/Ascend/msot/issues/41 - [ ] **涉及代码双合**(贴上另一个PR链接):NA ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) NA - [ ] **冒烟是否通过** NA ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** NA - [ ] **是否需要在sig会议中进行代码检视** NA - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** NA ---- See merge request: Ascend/msdebug!105 | 3 个月前 | |
Revise IDE folder structure (#89755) Update the folder titles for targets in the monorepository that have not seen taken care of for some time. These are the folders that targets are organized in Visual Studio and XCode ( set_property(TARGET <target> PROPERTY FOLDER "<title>")) when using the respective CMake's IDE generator. * Ensure that every target is in a folder * Use a folder hierarchy with each LLVM subproject as a top-level folder * Use consistent folder names between subprojects * When using target-creating functions from AddLLVM.cmake, automatically deduce the folder. This reduces the number of set_property/set_target_property, but are still necessary when add_custom_target, add_executable, add_library, etc. are used. A LLVM_SUBPROJECT_TITLE definition is used for that in each subproject's root CMakeLists.txt. | 2 年前 | |
[CMake] Use Clang to infer the target triple (#89425) When using Clang as a compiler, use Clang to normalize the triple that's used to construct path for runtime library build and install paths. This ensures that paths are consistent and avoids the issue where the build uses a different triple spelling. Differential Revision: https://reviews.llvm.org/D140925 | 2 年前 | |
[feature]: makeself仓切换到v2.5.0.x分支 Co-authored-by: h-glue<huangguang9@h-partners.com> # message auto-generated for no-merge-commit merge: !78 merge submodule_makeself into master [feature]: makeself仓切换到v2.5.0.x分支 Created-by: h-glue Commit-by: h-glue Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 原makeself仓会有grep stray \ before -告警 https://gitcode.com/Ascend/msot/issues/34 - **修改方案:** 切换到新分支v2.5.0.x - **修改内容:** - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** ---- See merge request: Ascend/msdebug!78 | 4 个月前 | |
Revert "demangle function names in trace files (#87626)" This reverts commit 0fa20c55b58deb94090985a5c5ffda4d5ceb3cd1. Storing raw symbol names is generally preferred in profile files. Demangling might lose information. Language frontends might use demangling schemes not supported by LLVMDemangle (https://github.com/llvm/llvm-project/issues/45901#issuecomment-2008686663). In addition, calling demangle for each function has a significant performance overhead (#102222). I believe that even if we decide to provide a producer-side demangling, it would not be on by default. Pull Request: https://github.com/llvm/llvm-project/pull/102274 (cherry picked from commit 72b73e23b6c36537db730ebea00f92798108a6e5) | 1 年前 | |
Revert "Title: [RISCV] Add missing part of instruction vmsge {u}. VX Review By: craig.topper Differential Revision : https://reviews.llvm.org/D100115" This reverts commit 4d9ccb18f50803a1aa9c0332dc72472bdfec8bc4. | 5 年前 | |
Disable clang-tidy misc-include-cleaner (#83945) This does not apply well to LLVM which intentionally rely on forward declarations. Also depending on the config flags passed to CMake the result can be different. | 2 年前 | |
[lldb][nfc] add an nfc entry to the .git-blame-ignore-revs. | 2 年前 | |
[libc++] Format the code base (#74334) This patch runs clang-format on all of libcxx/include and libcxx/src, in accordance with the RFC discussed at [1]. Follow-up patches will format the benchmarks, the test suite and remaining parts of the code. I'm splitting this one into its own patch so the diff is a bit easier to review. This patch was generated with: find libcxx/include libcxx/src -type f \ | grep -v 'module.modulemap.in' \ | grep -v 'CMakeLists.txt' \ | grep -v 'README.txt' \ | grep -v 'libcxx.imp' \ | grep -v '__config_site.in' \ | xargs clang-format -i A Git merge driver is available in libcxx/utils/clang-format-merge-driver.sh to help resolve merge and rebase issues across these formatting changes. [1]: https://discourse.llvm.org/t/rfc-clang-formatting-all-of-libc-once-and-for-all | 2 年前 | |
added all modifications made for msdebug | 7 个月前 | |
[feature]: makeself仓切换到v2.5.0.x分支 Co-authored-by: h-glue<huangguang9@h-partners.com> # message auto-generated for no-merge-commit merge: !78 merge submodule_makeself into master [feature]: makeself仓切换到v2.5.0.x分支 Created-by: h-glue Commit-by: h-glue Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 原makeself仓会有grep stray \ before -告警 https://gitcode.com/Ascend/msot/issues/34 - **修改方案:** 切换到新分支v2.5.0.x - **修改内容:** - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** ---- See merge request: Ascend/msdebug!78 | 4 个月前 | |
.mailmap: add second entry for self | 2 年前 | |
[feature] 支持core文件展示&切换线程信息 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !106 merge fix_coredump_register into master [feature] 支持core文件展示&切换线程信息 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 如果coredump位置是在simt_vf里,工具需要展示、切换线程信息 - **修改方案:** 给ProcessElfCoreDevice类实现正确的GetCoresInfo, GetWarpsInfo接口。通过判断VEC_ERR_INFO寄存器,识别当前pc所在的是否是simt - **修改内容:** 1. ProcessElfCoreDevice类新增ProcessElfCoreDevice::GetThreadDim接口,用于从寄存器里拿thread_dim信息,从而获取warp_num的值。 新增ProcessElfCoreDevice::FocusToActiveThreadInWarp,用于识别到当前error发生在simt_vf时,聚焦到对应的warp_id 实现ProcessElfCoreDevice::GetWarpsInfo接口,获取当前的所有warp信息 2. 修正pc替换时所用的mask数值类型为uint64_t 2. 使用clang-format,自动修改了很多枚举定义的格式。 3. Coredump的寄存器展示的顺序和上板保持一致 4. 移除DeviceContext系列的类的无用函数GetRegisterAddr和GetRegisterList - [ ] **涉及代码双合**(贴上另一个PR链接):NA ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** NA - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** NA ---- See merge request: Ascend/msdebug!106 | 3 个月前 | |
[feature] 支持Unix Makefiles构建 Co-authored-by: wangyixian<wangyixian3@huawei.com> # message auto-generated for no-merge-commit merge: !51 merge fix_build into master [feature] 支持Unix Makefiles构建 Created-by: wiyr0 Commit-by: wangyixian Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 当前编译要求msdebug一定要用ninja版本 - **修改方案:** msdebug实际上可以用make编译的,这里优先使用ninja,如果不存在ninja就使用make - **修改内容:** 1. CMakelists.txt里如果检测到当前是用ninja编译的,先判断ninja是否存在以及版本正确性,否则报错 2. build.py里判断当前ninja是否存在,否则就用Unix Makefiles - [ ] **涉及代码双合**(贴上另一个PR链接):NA ---- ### 2. 功能验证 - [x] **功能自验截图**(请确保不体现个人信息) ninja不存在:   - [x] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [x] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** NA - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** NA ---- See merge request: Ascend/msdebug!51 | 4 个月前 | |
[llvm] Add CODE_OF_CONDUCT.md (#65816) Add a CODE_OF_CONDUCT.md file to the root of the repository. The file itself references the LLVM Community Code of Conduct. GitHub will recognize this file and put a link to it to the right of the repository, similar to the license and security policy, making the CoC easier to discover. | 2 年前 | |
Update CONTRIBUTING.md to remove the not about not accepting PR | 2 年前 | |
【doc】资料易用性优化更新 Co-authored-by: luyq11<luyiqian4@h-partners.com> # message auto-generated for no-merge-commit merge: !82 merge master into master 【doc】资料易用性优化更新 Created-by: luyq11 Commit-by: luyq11 Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 大模型检测低错修改 - **修改方案:** 详见资料修改 - **修改内容:** 见资料修改 - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** 见资料修改 ---- See merge request: Ascend/msdebug!82 | 3 个月前 | |
Initial commit Signed-off-by: gong-siwei <gongsiwei@huawei.com> | 7 个月前 | |
【doc】资料易用性更新 Co-authored-by: luyq11<luyiqian4@h-partners.com> # message auto-generated for no-merge-commit merge: !125 merge master into master 【doc】资料易用性更新 Created-by: luyq11 Commit-by: luyq11 Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 资料易用性更新 - **修改方案:** - **修改内容:** - [ ] **涉及代码双合**(贴上另一个PR链接): ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** - [ ] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** https://gitcode.com/Ascend/msot/issues/38 ---- See merge request: Ascend/msdebug!125 | 3 个月前 | |
[docs] Describe reporting security issues on the chromium tracker. To track security issues, we're starting with the chromium bug tracker (using the llvm project there). We considered using Github Security Advisories. However, they are currently intended as a way for project owners to publicize their security advisories, and aren't well-suited to reporting issues. This also moves the issue-reporting paragraph to the beginning of the document, in part to make it more discoverable, in part to allow the anchor-linking to actually display the paragraph at the top of the page. Note that this doesn't update the concrete list of security-sensitive areas, which is still an open item. When we do, we may want to move the list of security-sensitive areas next to the issue-reporting paragraph as well, as it seems like relevant information needed in the reporting process. Finally, when describing the discission medium, this splits the topics discussed into two: the concrete security issues, discussed in the issue tracker, and the logistics of the group, in our mailing list, as patches on public lists, and in the monthly sync-up call. While there, add a SECURITY.md page linking to the relevant paragraph. Differential Revision: https://reviews.llvm.org/D100873 | 5 年前 | |
[Installation]version修复 Co-authored-by: RainbowWhp<wanghaopeng3@huawei.com> # message auto-generated for no-merge-commit merge: !94 merge br_mas into master [Installation]version修复 Created-by: RainbowWhp Commit-by: RainbowWhp Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** CI/CD 基于 Tag 的版本构建流程 https://gitcode.com/Ascend/msot/issues/37 - **修改方案:** version默认用26.0.0 - **修改内容:** version默认用26.0.0 - [x] **涉及代码双合**(贴上另一个PR链接): https://gitcode.com/Ascend/msdebug/pull/95 ---- ### 2. 功能验证 - [x] **功能自验截图**(请确保不体现个人信息)  - [x] **冒烟是否通过** ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [x] **是否经过代码检视** - [x] **是否具备UT测试用例看护** - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** ---- See merge request: Ascend/msdebug!94 | 3 个月前 | |
【feature】统一优化下载机制,解决UT编译、挂死及失败问题 Co-authored-by: mengguangxin<mgx0018@163.com> # message auto-generated for no-merge-commit merge: !32 merge dev_0301 into master 【feature】统一优化下载机制,解决UT编译、挂死及失败问题 Created-by: mengguangxin Commit-by: mengguangxin Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 原有的单元测试代码存在多处问题导致测试无法通过: (1) AscendProcessLinuxTest 中的 AscendProcessLinux 构造函数签名已变更(从 Factory+MainLoop+socket 方式改为 Manager 方式),但测试未同步适配,导致编译失败; (2) SetBreakpoint 测试中 soc_version 使用了 "test" 这一未注册的值,无法通过工厂方法创建 DeviceContext,导致运行时报 "device context is null!" 错误; (3) GDB Remote 相关测试使用 GDBRemoteCommunication::ConnectLocally 进行 TCP 监听连接,在某些环境下不稳定或不可用; (4) TestClient 使用 --reverse-connect + TCP Listen/Accept 方式启动 lldb-server,存在端口竞争和环境兼容性问题; (5) DWARFExpressionTest 的 ReadMemory 方法签名缺少新增的 MemoryReaderParamClient 参数,导致编译失败; (6) NativeProcessTestUtils.h 中的 mock delegate 缺少新增的接口方法声明; (7) UT 运行时缺少 LD_LIBRARY_PATH 环境变量,导致动态库找不到。 (8) 新的依赖下载机制需要各仓统一同步 - **修改方案:** (1) 适配 AscendProcessLinux 新的构造函数接口,将 Factory+MainLoop+socket 替换为 Manager,去除 socket 通信相关的逻辑,改为直接调用 m_parser.ParseMessage 进行消息注入; (2) 新增 FakeDeviceContext 类继承 DeviceContext,重写关键虚方法以绕过真实设备初始化和地址校验,在 SetBreakpoint 测试中直接注入到 process->m_device_context; (3) 在 GDBRemoteTestUtils.h 中新增 ConnectLocallyViaSocketPair 工具函数,使用 socketpair 替代 TCP 监听方式建立本地连接,并将 GDB Remote 测试中的 ConnectLocally 调用统一替换; (4) 重构 TestClient 启动逻辑,使用 socketpair + --fd 参数替代 --reverse-connect + TCP Listen/Accept; (5) 补全 DWARFExpressionTest 中 ReadMemory 方法的新参数; (6) 在 NativeProcessTestUtils.h 的 mock delegate 中补充 SetSingleCoreRunFlag、SetClientDeviceId、ReadDeviceRegisterValueByName、ReadDeviceRegisterList 等新增接口的 mock 声明; (7) 在 lit.cfg.py 中追加 LD_LIBRARY_PATH 环境变量指向构建产物的 lib 目录。 (8) 同步统一依赖下载机制 - **修改内容:** (1) **lldb/unittests/Process/Linux/AscendProcessLinuxTest.cpp**:将 #include "AscendProcessLinux.h" 移至 #define private public 之后以暴露私有成员;删除 NativeProcessFactory 类型别名,新增 FakeDeviceContext 类(实现 Init、ReadGlobalMemory、WriteGlobalMemory、InvalidInstrCache 等方法的桩/mock 逻辑);HandleStubMessage 和 SetBreakpoint 两个测试用例中,将构造函数从 (pid, fd, gdb_server, arch, mainloop, tids, socket) 改为 (pid, fd, gdb_server, arch, manager, tids),去除 socket 读写方式的消息传递,改为直接调用 m_parser.ParseMessage;SetBreakpoint 测试新增 process->m_device_context = std::make_shared<FakeDeviceContext>() 注入,移除不再需要的 CMD_SQ_SEND/CMD_CQ_RECV mock 期望。 (2) **lldb/unittests/tools/lldb-server/tests/TestClient.cpp**:新增 #include <sys/socket.h>;移除 --reverse-connect 参数和 TCP Listen/Accept 逻辑,改用 socketpair 创建 socket 对,通过 --fd= 参数将 server 端 fd 传递给 lldb-server 子进程;使用 AppendDuplicateFileAction 保留 fd 继承;连接建立后关闭 server 端 fd,用 client 端 fd 构造 TCPSocket 作为通信连接;inferior 参数改为通过 SetInferior 方法设置。 (3) **lldb/unittests/Process/gdb-remote/GDBRemoteTestUtils.h**:新增 #include <sys/socket.h>、TCPSocket.h、ConnectionFileDescriptorPosix.h;新增 ConnectLocallyViaSocketPair 内联函数,使用 socketpair 创建 Unix 域 socket 对并分别设置为 client/server 的连接。 (4) **lldb/unittests/Process/gdb-remote/GDBRemoteClientBaseTest.cpp、GDBRemoteCommunicationClientTest.cpp、GDBRemoteCommunicationTest.cpp**:将 SetUp 中的 GDBRemoteCommunication::ConnectLocally 替换为 ConnectLocallyViaSocketPair。 (5) **lldb/unittests/Expression/DWARFExpressionTest.cpp**:ReadMemory 方法签名新增 MemoryReaderParamClient param = {} 默认参数,适配上游接口变更。 (6) **lldb/unittests/TestingSupport/Host/NativeProcessTestUtils.h**:在 MS_DEBUGGER 宏分支内新增 SetSingleCoreRunFlag、SetClientDeviceId、ReadDeviceRegisterValueByName、ReadDeviceRegisterList 的 MOCK 方法声明,并添加 ReadDeviceRegisterValue(StringRef, uint64_t&) 的转发实现。 (7) **lldb/test/Unit/lit.cfg.py**:新增 LD_LIBRARY_PATH 环境变量配置,将 config.llvm_obj_root/lib 追加到库搜索路径中,确保 UT 运行时能找到动态链接库。 - [ ] **涉及代码双合**(贴上另一个PR链接):无 ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息)  - [ ] **冒烟是否通过** 是 ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** 是 - [ ] **是否具备UT测试用例看护** 是 - [ ] **是否需要在sig会议中进行代码检视** 否 - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** 不涉及 ---- See merge request: Ascend/msdebug!32 | 5 个月前 | |
[feature] 文档优化重构,提升易用性 Co-authored-by: mengguangxin<mgx0018@163.com> # message auto-generated for no-merge-commit merge: !61 merge dev_0331 into master [feature] 文档优化重构,提升易用性 Created-by: mengguangxin Commit-by: mengguangxin Merged-by: ascend-robot Description: ### 1. 修改描述 - **修改原因:** 按资料模板刷新,同步统一的download_dependencies.py修改 - **修改方案:** 按最新模板刷新资料,同步统一的修改 - **修改内容:** 按最新模板刷新资料,同步统一的修改 - [ ] **涉及代码双合**(贴上另一个PR链接):无 ---- ### 2. 功能验证 - [ ] **功能自验截图**(请确保不体现个人信息) - [ ] **冒烟是否通过** 不涉及 ---- ### 3. 代码检视 - **要求:** - 合入功能代码大于 200 行,需要sig会议申报代码检视议题,并在PR中标注会议。 - committer评估是否需要在sig会议进行代码检视。 - 参与检视的committer人员名单与检视时间。 - 大于 1000 行代码原则上不允许合入,需进行备案。 - [ ] **是否经过代码检视** 是 - [ ] **是否具备UT测试用例看护** 不涉及 - [ ] **是否需要在sig会议中进行代码检视** - **检视committer人员名单与检视时间:** ---- ### 4. 资料修改自检 - **资料修改:** 需要资料人员luyq11评审 ---- See merge request: Ascend/msdebug!61 | 4 个月前 | |
[Py Reformat] Exclude third-party from reformat (#83491) | 2 年前 |
MindStudio Debugger
✨ 最新消息
🔹 [2025.12.31]:MindStudio Debugger 项目全面开源
️ ℹ️ 简介
MindStudio Debugger(算子调试工具,msDebug)是基于 LLVM 编译器基础设施构建、面向昇腾设备的算子调试工具,用于调试在 NPU 侧运行的算子程序,为开发者提供关键调试能力,包括读取昇腾设备内存与寄存器、暂停与恢复程序执行状态等。
⚙️ 功能介绍
msDebug工具支持调试所有的昇腾算子,包含Ascend C算子(Vector、Cube以及Mix融合算子)程序,用户可根据实际情况进行选择,支持的功能如下:
| 功能名称 | 功能描述 |
|---|---|
| 断点设置 | 可在算子的运行程序上设置行断点,即在算子代码文件的特定行号上设置断点。 |
| 打印变量和内存 | 根据变量类型和用法,变量可以存储在寄存器中或存储在Local Memory、Global Memory内存中,用户可以打印变量的地址以找出它的存储位置并进一步打印关联的内存。 |
| 单步调试 | 需要了解代码执行具体情况时,可进行单步调试。 |
| 中断运行 | 当算子运行程序卡顿时,手动中断算子运行程序并回显中断位置信息。 |
| 核切换 | 可将当前聚焦的核切换至指定的核,切核后会自动展示指定核代码中断处的位置。 |
| 检查程序状态 | 当调起算子后,可读取当前断点所在设备的寄存器值,检查程序状态。 |
| 调试信息展示 | 查询算子运行的设备信息。 |
| 解析Core dump文件 | 通过对异常算子dump文件的解析,即使在没有主动压测的情况下也能收集到足够的数据用于问题分析。 |
🚀 快速入门
详细操作步骤请参见《msDebug 快速入门》。
📦 安装指南
介绍工具的环境依赖与安装方法,请参见《msDebug 安装指南》。
📘 使用指南
工具的详细使用方法,请参见《msDebug 使用指南》。
💡 典型案例
通过典型问题场景帮助用户理解并掌握工具使用,请参见《msDebug 典型案例》。
❓ FAQ
常见问题及解决方案,请参见《msDebug FAQ》。
🛠️ 贡献指南
欢迎参与项目贡献,请参见《贡献指南》。
⚖️ 相关说明
🔹《版本说明》
🔹《许可证声明》
🔹《安全声明》
🔹《免责声明》
🤝 建议与交流
欢迎大家为社区做贡献。如果有任何疑问或建议,请提交Issues,我们会尽快回复。感谢您的支持。
| 📱 关注 MindStudio 公众号 | 💬 更多交流与支持 |
|---|---|
![]() 扫码关注获取最新动态 |
💡 加入微信交流群: 关注公众号,回复“交流群”即可获取入群二维码。 🛠️ 其他渠道: 👉 昇腾助手: 👉 昇腾论坛: |
🙏 致谢
本工具由华为公司的下列部门联合贡献:
🔹 昇腾计算MindStudio开发部
🔹 昇腾计算生态使能部
🔹 华为云昇腾云服务
🔹 2012编译器实验室
🔹 2012马尔科夫实验室
感谢来自社区的每一个 PR,欢迎贡献!
