| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Andres Freund <andres(at)anarazel(dot)de> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: [COMMITTERS] pgsql: Allow Pin/UnpinBuffer to operate in a lockfree manner. |
| Date: | 2016-04-12 14:42:11 |
| Message-ID: | 32012.1460472131@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-committers pgsql-hackers |
Andres Freund <andres(at)anarazel(dot)de> writes:
> On 2016-04-12 00:32:13 -0400, Tom Lane wrote:
>> Apparently, assigning the result of init_spin_delay() is not as portable
>> as you thought.
> Hm. I'm not sure why not though? Is it because delayStatus isn't static
> or global scope?
It looks like that compiler adheres to the C89 restriction that an
initializer for an array or struct must contain only link-time-constant
expressions, even if the target object is of dynamic scope.
The macro works with a link-time-constant pointer argument, but not
with one that must be evaluated at runtime.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2016-04-12 14:53:12 | pgsql: Remove unnecessary definition of _WIN64 in libpq/win32.mak. |
| Previous Message | Pavel Stehule | 2016-04-12 13:15:21 | Re: [COMMITTERS] pgsql: Code cleanup in the wake of recent LWLock refactoring. |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2016-04-12 14:53:58 | Re: Preprocessor condition fix |
| Previous Message | Amit Kapila | 2016-04-12 14:12:11 | Re: Move PinBuffer and UnpinBuffer to atomics |