public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/110967] New: c++20 coroutine does not support SIMD avx2 and avx512
@ 2023-08-10 7:32 fangzhuhe at 126 dot com
2023-08-10 7:37 ` [Bug c++/110967] " fangzhuhe at 126 dot com
2023-08-10 7:37 ` pinskia at gcc dot gnu.org
0 siblings, 2 replies; 3+ messages in thread
From: fangzhuhe at 126 dot com @ 2023-08-10 7:32 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110967
Bug ID: 110967
Summary: c++20 coroutine does not support SIMD avx2 and avx512
Product: gcc
Version: 10.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: fangzhuhe at 126 dot com
Target Milestone: ---
coroutine does not support avx2 and avx512 well as code listed below, which can
be checked on https://godbolt.org/z/h3hej1ddq
#include <coroutine>
#include <iostream>
#include <set>
#include <thread>
#include <immintrin.h>
#include <chrono>
using namespace std;
struct HelloCoroutine {
struct HelloPromise {
HelloCoroutine get_return_object() {
return std::coroutine_handle<HelloPromise>::from_promise(*this);
}
std::suspend_never initial_suspend() { return {}; }
std::suspend_always final_suspend() noexcept { return {}; }
void unhandled_exception() {
std::cout <<" exception!" << std::endl;
}
};
using promise_type = HelloPromise;
HelloCoroutine(std::coroutine_handle<HelloPromise> h) : handle(h) {}
std::coroutine_handle<HelloPromise> handle;
};
#define AVX512 0
#define AVX2 1
#define SSE 0
HelloCoroutine hello(int& index, int id, int group_size) {
unsigned res=0;
#if AVX512
// g++ simd.cpp -std=gnu++20 -fcoroutines -O2 -march=skylake -mavx512f
// segment fault
for(auto i= index++; i< 20; i=index++)
{
std::cout <<"step 1" <<std::endl;
__m512i v_offset = _mm512_set1_epi64(int64_t (i));
std::cout <<"step 2" <<std::endl;
__m512i v_size = _mm512_set1_epi64(int64_t(group_size));
std::cout <<"step 3" <<std::endl;
res = _mm512_cmpgt_epi64_mask(v_offset, v_size);
cout <<i << " > " << group_size <<" ? " << (int)res<<endl;
co_await std::suspend_always();
}
#elif AVX2
// g++ simd.cpp -std=gnu++20 -fcoroutines -O2 -march=skylake
// only specify `-O2 -march=skylake` and runs ok on local machine, otherwise
segment fault (also on godbolt)
for(auto i= index++; i< 20; i=index++)
{
std::cout <<"step 1" <<std::endl;
__m256i v_offset = _mm256_set1_epi32(int32_t (i));
std::cout <<"step 2" <<std::endl;
__m256i v_size = _mm256_set1_epi32(int32_t(group_size));
std::cout <<"step 3" <<std::endl;
res = _mm256_movemask_epi8(_mm256_cmpgt_epi32(v_offset, v_size));
cout <<i << " > " << group_size <<" ? " << (int)res<<endl;
co_await std::suspend_always();
}
#elif SSE
for(auto i= index++; i< 20; i=index++)
{
__m128i v_offset = _mm_set1_epi32(int32_t (i));
__m128i v_size = _mm_set1_epi32(int32_t(group_size));
res = _mm_movemask_epi8(_mm_cmpgt_epi32(v_offset, v_size));
cout <<i << " > " << group_size <<" ? " << res<<endl;
co_await std::suspend_always();
}
#else
for(auto i= index++; i< 20; i=index++)
{
res = i > group_size;
cout <<i << " > " << group_size <<" ? " << res<<endl;
co_await std::suspend_always();
}
#endif
}
int main() {
set<std::coroutine_handle<HelloCoroutine::HelloPromise>>handles;
int group_size = 10;
int index =0;
for(auto i=0; i< group_size;++i)
{
handles.insert(hello(index, i, group_size).handle);
}
while(!handles.empty()) {
for (auto it = handles.begin(); it != handles.end();) {
if (it->done()) {
it->destroy();
it = handles.erase(it);
} else {
it->resume();
it++;
}
}
}
return 0;
}
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug c++/110967] c++20 coroutine does not support SIMD avx2 and avx512
2023-08-10 7:32 [Bug c++/110967] New: c++20 coroutine does not support SIMD avx2 and avx512 fangzhuhe at 126 dot com
@ 2023-08-10 7:37 ` fangzhuhe at 126 dot com
2023-08-10 7:37 ` pinskia at gcc dot gnu.org
1 sibling, 0 replies; 3+ messages in thread
From: fangzhuhe at 126 dot com @ 2023-08-10 7:37 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110967
--- Comment #1 from Zhuhe Fang <fangzhuhe at 126 dot com> ---
also discussed on
https://stackoverflow.com/questions/76727854/c-coroutine-runs-avx-simd-code-but-causes-sigsegv
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug c++/110967] c++20 coroutine does not support SIMD avx2 and avx512
2023-08-10 7:32 [Bug c++/110967] New: c++20 coroutine does not support SIMD avx2 and avx512 fangzhuhe at 126 dot com
2023-08-10 7:37 ` [Bug c++/110967] " fangzhuhe at 126 dot com
@ 2023-08-10 7:37 ` pinskia at gcc dot gnu.org
1 sibling, 0 replies; 3+ messages in thread
From: pinskia at gcc dot gnu.org @ 2023-08-10 7:37 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110967
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |DUPLICATE
Status|UNCONFIRMED |RESOLVED
--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Dup of bug 104177.
*** This bug has been marked as a duplicate of bug 104177 ***
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-08-10 7:37 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-10 7:32 [Bug c++/110967] New: c++20 coroutine does not support SIMD avx2 and avx512 fangzhuhe at 126 dot com
2023-08-10 7:37 ` [Bug c++/110967] " fangzhuhe at 126 dot com
2023-08-10 7:37 ` pinskia at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).