You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
One thing I have noticed is the massive increase in time taken when a variable is allocated with __thread - compared to when it is not. This can be seen by running the test program versus when that variable is not per thread. This happens with both newlib and clib4. The example does iterate 1 billion times, so there is no doubt a lot of overhead of extra instructions etc. It is quite a contrived example.
The text was updated successfully, but these errors were encountered:
When we build our compilers, we use
--disable-tls
, but the test in: https://github.com/3246251196/adtools_testing/tree/main/tests/22_ThreadLocalStorage shows that - at least for that example - it works.Investigate this.
@afxgroup Pinging.
===
One thing I have noticed is the massive increase in time taken when a variable is allocated with
__thread
- compared to when it is not. This can be seen by running the test program versus when that variable is not per thread. This happens with both newlib and clib4. The example does iterate 1 billion times, so there is no doubt a lot of overhead of extra instructions etc. It is quite a contrived example.The text was updated successfully, but these errors were encountered: