On Tue, Oct 1, 2024 at 11:23 PM Peter Eisentraut wrote:
> I have committed the two remaining patches. I'll go nudge any affected
> buildfarm members to upgrade as needed.
Thanks!
FWIW next on my list of LLVM maintenance work are: fix ARM crash in
the next week or so (CF #5220), and figure out w
On 21.08.24 16:00, Peter Eisentraut wrote:
On 17.05.24 00:54, Thomas Munro wrote:
On Fri, May 17, 2024 at 3:17 AM Nazir Bilal Yavuz
wrote:
Actually, 32 bit builds are working but the Perl version needs to be
updated to 'perl5.36-i386-linux-gnu' in .cirrus.tasks.yml. I changed
0001 with the wor
On 17.05.24 00:54, Thomas Munro wrote:
On Fri, May 17, 2024 at 3:17 AM Nazir Bilal Yavuz wrote:
Actually, 32 bit builds are working but the Perl version needs to be
updated to 'perl5.36-i386-linux-gnu' in .cirrus.tasks.yml. I changed
0001 with the working version of 32 bit builds [1] and the re
On 19.05.24 00:46, Ole Peder Brandtzæg wrote:
On Wed, May 15, 2024 at 07:20:09AM +0200, Peter Eisentraut wrote:
Yes, let's get that v3-0001 patch into PG17.
Upon seeing this get committed in 4dd29b6833, I noticed that the docs
still advertise the llvm-config-$version search dance. That's still
Thomas Munro writes:
> Oops, right I didn't know we had that documented. Thanks. Will hold
> off doing anything until the thaw.
FWIW, I don't think the release freeze precludes docs-only fixes.
But if you prefer to sit on this, that's fine too.
regards, tom lane
On Sun, May 19, 2024 at 11:05:49AM +1200, Thomas Munro wrote:
> Oops, right I didn't know we had that documented. Thanks. Will hold
> off doing anything until the thaw.
No worries, thanks!
> Hmm, I also didn't know that Meson had its own list like our just-removed one:
>
> https://github.com/m
On Sun, May 19, 2024 at 10:46 AM Ole Peder Brandtzæg
wrote:
> On Wed, May 15, 2024 at 07:20:09AM +0200, Peter Eisentraut wrote:
> > Yes, let's get that v3-0001 patch into PG17.
>
> Upon seeing this get committed in 4dd29b6833, I noticed that the docs
> still advertise the llvm-config-$version sear
On Wed, May 15, 2024 at 07:20:09AM +0200, Peter Eisentraut wrote:
> Yes, let's get that v3-0001 patch into PG17.
Upon seeing this get committed in 4dd29b6833, I noticed that the docs
still advertise the llvm-config-$version search dance. That's still
correct for Meson-based builds since we use the
On Fri, May 17, 2024 at 3:17 AM Nazir Bilal Yavuz wrote:
> Actually, 32 bit builds are working but the Perl version needs to be
> updated to 'perl5.36-i386-linux-gnu' in .cirrus.tasks.yml. I changed
> 0001 with the working version of 32 bit builds [1] and the rest is the
> same. All tests pass now
Hi,
On Thu, 16 May 2024 at 05:34, Thomas Munro wrote:
>
> On Wed, May 15, 2024 at 5:20 PM Peter Eisentraut wrote:
> > Yes, let's get that v3-0001 patch into PG17.
>
> Done.
>
> Bilal recently created the CI images for Debian Bookworm[1]. You can
> try them with s/bullseye/bookworm/ in .cirrus.t
On Wed, May 15, 2024 at 5:20 PM Peter Eisentraut wrote:
> Yes, let's get that v3-0001 patch into PG17.
Done.
Bilal recently created the CI images for Debian Bookworm[1]. You can
try them with s/bullseye/bookworm/ in .cirrus.tasks.yml, but it looks
like he is still wrestling with a perl installa
On 15.05.24 06:21, Thomas Munro wrote:
And as I'm looking up how this was previously handled, I notice that
this list of clang-NN versions was last updated equally sneakily as part
of your patch to trim off LLVM <10 (820b5af73dc). I wonder if the
original intention of that configure code was tha
On Mon, May 13, 2024 at 2:33 AM Peter Eisentraut wrote:
> These patches look fine to me. The new cut-off makes sense, and it does
> save quite a bit of code. We do need to get the Cirrus CI Debian images
> updated first, as you had already written.
Thanks for looking!
> As part of this patch,
On 24.04.24 01:43, Thomas Munro wrote:
Rebased over ca89db5f.
These patches look fine to me. The new cut-off makes sense, and it does
save quite a bit of code. We do need to get the Cirrus CI Debian images
updated first, as you had already written.
As part of this patch, you also sneak in
Rebased over ca89db5f.
I looked into whether we could drop the "old pass manager" code
too[1]. Almost, but nope, even the C++ API lacks a way to set the
inline threshold before LLVM 16, so that would cause a regression.
Although we just hard-code the threshold to 512 with a comment that
sounds li
On Wed, Apr 10, 2024 at 1:38 PM Thomas Munro wrote:
> Therefore, some time after the tree re-opens for hacking, we could rip
> out a bunch of support code for LLVM 10-13, and then rip out support
> for pre-opaque-pointer mode. Please see attached.
... or of course closer to the end of the cycle
Hi
PostgreSQL 18 will ship after these vacuum horizon systems reach EOL[1]:
animal | arch | llvm_version | os | os_release | end_of_support
---+-+--+++
branta | s390x | 10.0.0 | Ubuntu | 20.04 | 2025-04-01
splitfin | aarch64 | 10
17 matches
Mail list logo
