| From: | Andreas Karlsson <andreas(at)proxel(dot)se> |
|---|---|
| To: | Anthonin Bonnefoy <anthonin(dot)bonnefoy(at)datadoghq(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Add missing JIT inline pass for llvm>=17 |
| Date: | 2026-01-15 03:20:45 |
| Message-ID: | a60e000e-7ebb-4ab5-9fff-73c0491d1154@proxel.se |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Great find! Sadly shows how little people actually use JIT.
On 1/14/26 9:02 AM, Anthonin Bonnefoy wrote:
> I didn't add the always-inline pass as I don't think this can have any
> effect. From what I understand, the functions need to be imported
> through llvm_inline for LLVM to be able to inline them. I've tested by
> tagging int4mod as always inline, and the generated bc was still
> calling the function despite the always-inline pass.
If inline-always does not do anything it should be removed on older LLVM
versions too. I do not think we should be having pre- and post-LLVM 17
run different passes. But as Thomas pointed out inline-always is likely
used for tuple deforming.
Andreas
| From | Date | Subject | |
|---|---|---|---|
| Next Message | jian he | 2026-01-15 03:30:41 | Re: Enable partitionwise join for partition keys wrapped by RelabelType |
| Previous Message | Thomas Munro | 2026-01-15 02:33:13 | Re: Add missing JIT inline pass for llvm>=17 |