[RISC-V] coreclr-vm and other directories in coreclr #82381
Conversation
clamp03
commented
Feb 20, 2023
clamp03
commented
- Successfully cross-build for RISC-V.
- Run A simple application "helloworld"
- Fail a test in clr.paltest
ghost
added
the
community-contribution
clamp03
force-pushed
the
riscv_vm
branch
2 times, most recently
from
ba6f364 to
3db1146
Compare
clamp03
force-pushed
the
riscv_vm
branch
3 times, most recently
from
8ee2012 to
1bcdcc7
Compare
|
@mangod9 Could you review this PR? Actually I don't know why tests failed or cancelled. |
|
@janvorli as well. Also are you asking about why the superpmi tests are failing or something else? |
|
Actually, I am asking reviews from you and owners to merge the PR. To get review, I tried to pass all tests, but some tests failed and I cannot find the reasons. :) |
| @@ -92,7 +92,20 @@ class NativeFieldDescriptor | |||
| return m_category; | |||
| } | |||
|
|
|||
| PTR_MethodTable GetNestedNativeMethodTable() const; | |||
| PTR_MethodTable GetNestedNativeMethodTable() const | |||
There was a problem hiding this comment.
is this definition moved here for consistency?
There was a problem hiding this comment.
This is for resolving a build error.
[ 93%] Building CXX object vm/wks/CMakeFiles/cee_wks_core.dir/__/weakreferencenative.cpp.o
/usr/bin/riscv64-linux-gnu-ld: ../../dlls/mscordac/libmscordaccore.so: undefined reference to `NativeFieldDescriptor::GetNestedNativeMethodTable() const'
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [debug/createdump/CMakeFiles/createdump.dir/build.make:308: debug/createdump/createdump] Error 1
fieldmarshalre.cpp is not included for createdump.
I resolved this error by putting the implemention into header.
clamp03
force-pushed
the
riscv_vm
branch
2 times, most recently
from
72dccdb to
fdaf64f
Compare
- Successfully cross-build for RISC-V. - Run A simple application "helloworld" - Fail a test in clr.paltest
Updated by review in dotnet#82380
This reverts commit cd1ea45.
|
Revert |
In coreclr-jit patch, it makes getRISCV64PassStructInRegisterFlags for RISCV64. So restore getLoongArch64PassStructInRegisterFlags
|
@mangod9 Could you review it? And if it is good, could you please merge it? I fixed all the things requested by the reviewers and waited 50 days. Other two PRs (#82379 , #82382) are already merged. Actually, I cannot fix all bugs and unimplemented code with just this PR. After the PR is merged, I will continue to fix and my colleagues and community members can help to support RISC-V in coreclr better. |
Thank you for the approval. If it is okay, could you merge this PR? If it needs more approvals or something else, please let me know. |
|
@clamp03 there are some conflicts, can you please resolve them? |
| elseif(clr_cmake_target_arch_loongarch64) | ||
| set(vm_sources_dac_and_wks_arch | ||
| ${arch_sources_dir}/stubs.cpp | ||
| exceptionhandling.cpp | ||
| ) | ||
|
|
||
| set(vm_headers_dac_and_wks_arch | ||
| ${arch_sources_dir}/virtualcallstubcpu.hpp | ||
| exceptionhandling.h | ||
| ) | ||
|
|
||
| set(vm_sources_wks_arch | ||
| ${arch_sources_dir}/profiler.cpp | ||
| gcinfodecoder.cpp | ||
| ) |
There was a problem hiding this comment.
@clamp03 why did you destroy the LoongArch64's config?????
You can copy the LoongArch64's code But please don't destroy the LoongArch64's config.
There was a problem hiding this comment.
@shushanhf I am really sorry. It is just my mistake. I don't want to destroy LoongArch64 at all. Sorry again.
There was a problem hiding this comment.
@shushanhf I am really sorry. It is just my mistake. I don't want to destroy LoongArch64 at all. Sorry again.
shake hands.
ghost
locked as resolved and limited conversation to collaborators
