Re: [lttng-dev] [PATCH 02/11] urcu/uatomic: Use atomic builtins if configured
Olivier Dion via lttng-dev <[email protected]>
| Newsgroups | org.lttng.lists.lttng-dev |
|---|---|
| Organization | EfficiOS |
| Message-ID | <87v8ffl0ku.fsf@laura> |
On Thu, 22 Jun 2023, Mathieu Desnoyers <[email protected]> wrote: > On 6/22/23 15:53, Olivier Dion wrote: >> On Thu, 22 Jun 2023, "Paul E. McKenney" <[email protected]> wrote: >> >>> I suggest C11 volatile atomic load/store. Load/store fusing is permitted >>> for non-volatile atomic loads and stores, and such fusing can ruin your >>> code's entire day. ;-) >> >> Good catch. Seems like not a problem on GCC (yet), but Clang is extremely >> aggressive and seems to do store fusing on some corner cases [0]. > > I don't think this is an example of store fusing, but rather just that > the compiler can eliminate stores to static variables which are > otherwise unused, making the entire variable useless. Indeed, that is not store fusing. It is however interesting to see that the dead store elimination is avoid when casting one of the &x with a volatile qualifier. TIL dead store can apply to atomics. -- Olivier Dion EfficiOS Inc. https://www.efficios.com _______________________________________________ lttng-dev mailing list [email protected] https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev