Re: Improving spin-lock implementation on ARM. - Mailing list pgsql-hackers

From Krunal Bauskar
Subject Re: Improving spin-lock implementation on ARM.
Date
Msg-id CAB10pyaq3M7VUZWYELxVbPTUC_DXksFfEY+MHsH9-Psc0mn6Fw@mail.gmail.com
Whole thread Raw
In response to Re: Improving spin-lock implementation on ARM.  (Alexander Korotkov <aekorotkov@gmail.com>)
Responses Re: Improving spin-lock implementation on ARM.
List pgsql-hackers


On Tue, 1 Dec 2020 at 02:31, Alexander Korotkov <aekorotkov@gmail.com> wrote:
On Mon, Nov 30, 2020 at 7:00 AM Krunal Bauskar <krunalbauskar@gmail.com> wrote:
> 3. Problem with GCC approach is still a lot of distro don't support gcc 9.4 as default.
>     To use this approach:
>     * PGSQL will have to roll out its packages using gcc-9.4+ only so that they are compatible with all aarch64 machines
>     * but this continues to affect all other users who tend to build pgsql using standard distro based compiler. (unless they upgrade compiler).

I think if a user, who runs PostgreSQL on a multicore machine with
high-concurrent load, can take care about installing the appropriate
package/using the appropriate compiler (especially if we publish
explicit instructions for that).  In the same way such advanced users
tune Linux kernel etc.

BTW, how do you get that required gcc version is 9.4?  I've managed to
use LSE with gcc 9.3.

Did they backported it to 9.3?
I am just looking at the gcc guide.

GCC 9.4

Target Specific Changes

AArch64

  • The option -moutline-atomics has been added to aid deployment of the Large System Extensions (LSE)
 

------
Regards,
Alexander Korotkov


--
Regards,
Krunal Bauskar

pgsql-hackers by date:

Previous
From: Craig Ringer
Date:
Subject: Re: Printing backtrace of postgres processes
Next
From: Andres Freund
Date:
Subject: Re: Printing backtrace of postgres processes