On 2024-01-26 02:37, Honnappa Nagarahalli wrote:
<snip>
On Thu, Jan 25, 2024 at 11:10:47PM +0100, Morten Br�rup wrote:
From: Mattias R�nnblom [mailto:hof...@lysator.liu.se]
Sent: Thursday, 25 January 2024 19.54
Why do rte_stdatomic.h functions have the suffix "_explicit"?
Especially
since there aren't any wrappers for the implicit variants.
More to type, more to read.
They have the "_explicit" suffix to make their names similar to those in
stdatomic.h.
You might consider their existence somewhat temporary until C11 stdatomics
can be fully phased in, so there's another argument for similar names. (This
probably does not happen as long as compilers generate slower code for C11
stdatomics than with their atomic built-ins.)
yes, there was feedback at the time it was.
* we should *not* have non-explicit versions of the macros
* the atomic generic functions should be named to match C11 standard
with a rte_ prefix.
This was mainly done to ensure that users think through the memory ordering
they want to use. This also matches with the compiler atomic built-ins. Without
explicit, it is sequentially consistent memory order.
"This" is referring to the first bullet only, correct?
You don't have to distinguish between implicit and explicit if you only
have explicit.
When was this API introduced? Shouldn't it say "experimental"
somewhere?
They were introduced as part of the migration to C11.
I suppose they were not marked experimental because they replaced
something we didn't want anymore (the compiler built-ins for atomics, e.g.
__atomic_load_n()). I don't recall if we discussed experimental marking or not.
i don't think we discussed it since they're wrapper macros.
Reverse paper trail:
https://git.dpdk.org/dpdk/log/lib/eal/include/rte_stdatomic.h
https://patchwork.dpdk.org/project/dpdk/patch/1692738045-32363-2-git-
send-email-roret...@linux.microsoft.com/
https://patchwork.dpdk.org/project/dpdk/patch/1692738045-32363-2-git-
send-email-roret...@linux.microsoft.com/
