You already hold explicit mutex when you do mutations, so memory_order_relaxed
would be enough, you only need atomicity. This will likely just result with something like xchg [RAX], RCX
on x86_64, no memory fence, same as __sync_fetch*
would. If you solve the problem only for C++11, you leave out most clients without a fix.
—
Reply to this email directly or view it on GitHub.
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1