| From: | Alexander Korotkov <aekorotkov(at)gmail(dot)com> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Krunal Bauskar <krunalbauskar(at)gmail(dot)com>, Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com>, Michael Paquier <michael(at)paquier(dot)xyz>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Improving spin-lock implementation on ARM. |
| Date: | 2020-12-03 19:03:03 |
| Message-ID: | CAPpHfdsrpt-on2+ddCeuR01P6DUVS8Ze8Qxh+bbRN-aJOfYCZA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Dec 3, 2020 at 7:02 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> From a system structural standpoint, I seriously dislike that lwlock.c
> patch: putting machine-specific variant implementations into that file
> seems like a disaster for maintainability. So it would need to show a
> very significant gain across a range of hardware before I'd want to
> consider adopting it ... and it has not shown that.
The current shape of the lwlock patch is experimental. I had quite a
beautiful (in my opinion) idea to wrap platform-dependent parts of
CAS-loops into macros. Then we could provide different low-level
implementations of CAS-loops for Power, ARM and rest platforms with
single code for LWLockAttempLock() and others. However, I see that
modern ARM tends to efficiently implement LSE. Power doesn't seem to
be very popular. So, I'm going to give up with this for now.
------
Regards,
Alexander Korotkov
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Eisentraut | 2020-12-03 19:26:59 | Re: SELECT INTO deprecation |
| Previous Message | Bruce Momjian | 2020-12-03 18:14:56 | Re: Minor documentation error regarding streaming replication protocol |