* Jakub Jelinek:
> On Wed, Nov 03, 2021 at 05:27:42PM +0100, Florian Weimer wrote:
>> + /* S3. See __condvar_load_64_relaxed. */
>
> Shouldn't that be See __atomic_wide_counter_load_relaxed ?
I'm going to fix the stale reference, thanks.
>> + if (((l >> 31) > 0) && ((h >> 31) > 0))
>
>
On Wed, Nov 03, 2021 at 05:27:42PM +0100, Florian Weimer wrote:
> + /* S3. See __condvar_load_64_relaxed. */
Shouldn't that be See __atomic_wide_counter_load_relaxed ?
> + if (((l >> 31) > 0) && ((h >> 31) > 0))
Any reason not to write it as (int) l < 0 && (int) h < 0?
Yes, the compiler
On 03/11/2021 13:27, Florian Weimer via Libc-alpha wrote:
> And make it an installed header. This addresses a few aliasing
> violations (which do not seem to result in miscompilation due to
> the use of atomics), and also enables use of wide counters in other
> parts of the library.
>
> The
And make it an installed header. This addresses a few aliasing
violations (which do not seem to result in miscompilation due to
the use of atomics), and also enables use of wide counters in other
parts of the library.
The debug output in nptl/tst-cond22 has been adjusted to print
the 32-bit
