Re: LLVM 16 (opaque pointers)

Andres Freund <andres@anarazel.de>

From: Andres Freund <andres@anarazel.de>

To: Thomas Munro <thomas.munro@gmail.com>

Cc: Devrim Gündüz <devrim@gunduz.org>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Fabien COELHO <coelho@cri.ensmp.fr>, Ronan Dunklau <ronan.dunklau@aiven.io>, Dmitry Dolgov <9erthalion6@gmail.com>

Date: 2023-10-12T23:31:20Z

Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

jit: Adjust back-patch of f90b4a84 to 12 and 13.
- 9ad98627614b 12.17 landed
- 7c070e224c4a 13.13 landed
Log LLVM library version in configure output.
- 0a6e2c7f8694 11.22 landed
- 527e62a5e5d7 12.17 landed
- a13b63c788a3 13.13 landed
- 03e749c2b89e 14.10 landed
- 0e49e23782c5 15.5 landed
- 3b0f20f6ce14 16.1 landed
- 5e4dacb9878c 17.0 landed
jit: Changes for LLVM 17.
- 7da915e33abf 12.17 landed
- 53c4dabe18d3 13.13 landed
- 0a8b7d5c1126 14.10 landed
- b60e3ac7603d 15.5 landed
- 774185056834 16.1 landed
- 76200e5ee469 17.0 landed
jit: Supply LLVMGlobalGetValueType() for LLVM < 8.
- 60596f148a66 16.1 landed
- d701f0d1f3e7 12.17 landed
- 981292c19f38 13.13 landed
- ee3e4c41f354 14.10 landed
- b2e097788663 15.5 landed
- f90b4a846b32 17.0 landed
jit: Support opaque pointers in LLVM 16.
- 15ddc9725eb7 12.17 landed
- f28956b239f1 13.13 landed
- 82d9a782a296 14.10 landed
- eed1feb3fee1 15.5 landed
- 74d19ec096df 16.1 landed
- 37d5babb5cfa 17.0 landed
jit: Reference function pointer types via llvmjit_types.c.
- df99ddc70b97 14.0 cited

Hi,

On 2023-10-11 21:59:50 +1300, Thomas Munro wrote:
> +#else
> +	LLVMPassBuilderOptionsRef options;
> +	LLVMErrorRef err;
> +	int			compile_optlevel;
> +	char	   *passes;
> +
> +	if (context->base.flags & PGJIT_OPT3)
> +		compile_optlevel = 3;
> +	else
> +		compile_optlevel = 0;
> +
> +	passes = psprintf("default<O%d>,mem2reg,function(no-op-function),no-op-module",
> +					  compile_optlevel);

I don't think the "function(no-op-function),no-op-module" bit does something
particularly useful?

I also don't think we should add the mem2reg pass outside of -O0 - running it
after a real optimization pipeline doesn't seem useful and might even make the
code worse? mem2reg is included in default<O1> (and obviously also in O3).

Thanks for working on this stuff!

I'm working on setting up buildfarm animals for 16, 17, each once with
a normal and an assertion enabled LLVM build.

Greetings,

Andres Freund