Support using const pointers in asm `const` operand #138618

nbdd0121 · 2025-03-17T18:07:31Z

Implements #128464

This adds support of const pointers for asm const in addition to plain integers.

The inline asm! support is implemented using i constraint, and the global_asm! and naked_asm! support is implemented by inserting symbol + offset and make symbol compiler-used. For unnamed consts, it will create additional internal & hidden symbols so that they can be referenced by global_asm.

The feature is also implemented for GCC backend but it's untested.

rustbot · 2025-03-17T18:07:35Z

r? @fmease

rustbot has assigned @fmease.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

fmease · 2025-03-17T18:09:34Z

r? codegen

compiler-errors · 2025-03-18T04:25:43Z

r? compiler-errors

rustbot · 2025-04-10T14:59:14Z

Some changes occurred in compiler/rustc_codegen_ssa

cc @WaffleLapkin

Some changes occurred in compiler/rustc_codegen_gcc

cc @antoyo, @GuillaumeGomez

compiler/rustc_monomorphize/src/collector.rs

compiler-errors · 2025-04-13T16:46:50Z

compiler/rustc_codegen_ssa/src/traits/asm.rs

@@ -25,9 +26,12 @@ pub enum InlineAsmOperandRef<'tcx, B: BackendTypes + ?Sized> {
        in_value: OperandRef<'tcx, B::Value>,
        out_place: Option<PlaceRef<'tcx, B::Value>>,
    },
-    Const {
+    Interpolate {


Please add more doc comment explaining the significance of the difference between Interpolate and Const for backends. It begs the question for why we ever turn const operands into strings, for example.

Added some explaination. Technically we should be able to use Const for integers, although I am a bit conservative here and don't want to change how existing things work.

This might also be useful to have in the future if we decided to add interpolate "some CTFE string to be interpolated" new type of operand.

compiler/rustc_codegen_ssa/src/traits/asm.rs

bors · 2025-04-29T06:25:48Z

☔ The latest upstream changes (presumably #140415) made this pull request unmergeable. Please resolve the merge conflicts.

traviscross · 2025-05-04T09:26:42Z

@rustbot author

rustbot · 2025-05-04T09:26:46Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

rustbot · 2025-05-13T10:00:16Z

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

bors · 2025-05-28T05:09:25Z

☔ The latest upstream changes (presumably #141668) made this pull request unmergeable. Please resolve the merge conflicts.

tests/ui/asm/const-refs-to-static.rs

workingjubilee · 2025-03-17T20:12:55Z

tests/assembly/asm/x86-types.rs

+// and on x86 PIC symbol can't be constant.
+// x86_64-LABEL: const_ptr:
+// x86_64: #APP
+// x86_64: mov al, byte ptr [{{.*}}anon{{.*}}]


// x86_64: mov al, byte ptr [{{.*}}anon{{.*}}]

What was the expansion you observed?

mov al, byte ptr [.Lanon.2e8c0013b01edcd3779e8174c2338a00.0]

oli-obk · 2025-07-09T13:27:58Z

compiler/rustc_codegen_llvm/src/asm.rs

    ) {
        let asm_arch = self.tcx.sess.asm_arch.unwrap();

        // Default to Intel syntax on x86
        let intel_syntax = matches!(asm_arch, InlineAsmArch::X86 | InlineAsmArch::X86_64)
            && !options.contains(InlineAsmOptions::ATT_SYNTAX);

+        // Convert all operands to string interpolations


I think it would be good to take this opportunity and move this into cg_ssa and and add some extra trait methods for the parts that are different. Similar to what I did in #142960

GCC 15+ has ability for file-level assembly to reference symbols and constants, so I think codegen_llvm and codegen_gcc may eventually diverge and use different approaches.

This is used for string interpolation currently, so rename it as so. The name `Const` will be used to denote a general CTFE constant that can either be integer or pointer.

rustbot · 2025-07-09T19:12:04Z

Some changes occurred in src/tools/cargo

cc @ehuss

nbdd0121 · 2025-07-09T19:14:10Z

Oops, committed submodule changes by mistake

compiler-errors · 2025-07-09T20:08:11Z

(remember to mark this as ready when it's ready 😃)

This is intended for supporting passing arbitrary CTFE const into inline assembly.

`global_asm!` themselves don't need symbol names; they currently only have a symbol name during mono collecion to identify them to partitioning algorithm. However it will have shims generated under it which will need unique symbol names. The name themselves ultimately doesn't matter, so they're generated like other shim instances.

This is currently a no-op, but will be useful when const in `global_asm!` can be pointers.

nbdd0121 · 2025-07-09T23:11:58Z

I think this is ready from my side.

One thing that I am not completely certain about is on the aspect of symbol names. To be able to allow global asm to reference a promoted (unnamed) statics, I need give them names and the most sensible way to me is to use the symbol name of containing body as prefix. For naked_asm this is obvious, but for global_asm! this require me to give them unique names. I currently encode them like shims, I wonder that's a good-enough solution?

BTW, this also makes me wonder if we should just desugar global_asm as just naked functions with generated names. Given global_asm! have (fake) bodys now anyway, it feels like desugaring global_asm to naked_asm makes more sense then the other way around...

@rustbot ready

compiler-errors · 2025-07-09T23:19:13Z

compiler/rustc_symbol_mangling/src/v0.rs

@@ -867,6 +867,16 @@ impl<'tcx> Printer<'tcx> for SymbolMangler<'tcx> {
            // are effectively living in their parent modules.
            DefPathData::ForeignMod => return print_prefix(self),

+            // Global asm are handled similar to shims.
+            DefPathData::GlobalAsm => {


Why not just give this a prefix? Then we'll encode it with that prefix + disambiguator + empty name. That is to say, it feels a bit obtuse to encode it like this, not that it's necessarily wrong.

The idea is that existing demangler should be able to demangle this without any changes. The symbol generated for global_asms should not be very meaningful, so similar to shims.

rustbot assigned fmease Mar 17, 2025

rustbot added S-waiting-on-review T-compiler labels Mar 17, 2025

rustbot assigned workingjubilee and unassigned fmease Mar 17, 2025

This comment has been minimized.

Sign in to view

rustbot assigned compiler-errors and unassigned workingjubilee Mar 18, 2025

nikomatsakis mentioned this pull request Mar 26, 2025

Stabilize tooling needed by Rust for Linux rust-lang/rust-project-goals#116

Open

26 tasks

ojeda mentioned this pull request Apr 9, 2025

Rust unstable features needed for the kernel Rust-for-Linux/linux#2

Open

99 tasks

nbdd0121 force-pushed the asm_const_ptr branch from ffba8e8 to 92c0cc4 Compare April 10, 2025 10:05

nbdd0121 marked this pull request as ready for review April 10, 2025 14:59

tmiasko reviewed Apr 10, 2025

View reviewed changes

compiler/rustc_monomorphize/src/collector.rs Show resolved Hide resolved

compiler-errors reviewed Apr 13, 2025

View reviewed changes

rustbot added S-waiting-on-author and removed S-waiting-on-review labels May 4, 2025

nbdd0121 force-pushed the asm_const_ptr branch from 92c0cc4 to c7a2301 Compare May 13, 2025 08:27

This comment has been minimized.

Sign in to view

nbdd0121 force-pushed the asm_const_ptr branch from c7a2301 to 4240004 Compare May 13, 2025 10:00

workingjubilee reviewed Jun 4, 2025

View reviewed changes

oli-obk reviewed Jul 9, 2025

View reviewed changes

nbdd0121 force-pushed the asm_const_ptr branch from 4240004 to 8b5dcdf Compare July 9, 2025 15:56

rustbot added the A-LLVM label Jul 9, 2025

pietroalbini force-pushed the master branch from 048b879 to 6b3ae3f Compare July 9, 2025 16:17

Rename *AsmOperandRef::Const to Interpolate

e47eb32

This is used for string interpolation currently, so rename it as so. The name `Const` will be used to denote a general CTFE constant that can either be integer or pointer.

nbdd0121 force-pushed the asm_const_ptr branch from 8b5dcdf to ce4603c Compare July 9, 2025 19:12

nbdd0121 force-pushed the asm_const_ptr branch from ce4603c to 318fd43 Compare July 9, 2025 19:14

This comment has been minimized.

Sign in to view

nbdd0121 force-pushed the asm_const_ptr branch from 318fd43 to 74ba387 Compare July 9, 2025 19:42

nbdd0121 added 5 commits July 9, 2025 23:58

Add InlineAsmOperandRef::Const

2028157

This is intended for supporting passing arbitrary CTFE const into inline assembly.

Implement asm_const_ptr for inline asm

b38ec0d

Collect constants within global_asm! in collector

363fd49

This is currently a no-op, but will be useful when const in `global_asm!` can be pointers.

Implement asm_const_ptr for global_asm and naked_asm

81a4b3e

nbdd0121 force-pushed the asm_const_ptr branch from 74ba387 to 81a4b3e Compare July 9, 2025 22:58

rustbot added S-waiting-on-review and removed S-waiting-on-author labels Jul 9, 2025

compiler-errors reviewed Jul 9, 2025

View reviewed changes

Support using const pointers in asm const operand #138618

Are you sure you want to change the base?

Support using const pointers in asm const operand #138618

Conversation

nbdd0121 commented Mar 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Mar 17, 2025

Uh oh!

fmease commented Mar 17, 2025

Uh oh!

This comment has been minimized.

compiler-errors commented Mar 18, 2025

Uh oh!

rustbot commented Apr 10, 2025

Uh oh!

Uh oh!

compiler-errors Apr 13, 2025

Choose a reason for hiding this comment

Uh oh!

nbdd0121 Jul 9, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

bors commented Apr 29, 2025

Uh oh!

traviscross commented May 4, 2025

Uh oh!

rustbot commented May 4, 2025

Uh oh!

This comment has been minimized.

rustbot commented May 13, 2025

Uh oh!

bors commented May 28, 2025

Uh oh!

Uh oh!

workingjubilee Mar 17, 2025

Choose a reason for hiding this comment

Uh oh!

nbdd0121 Jul 9, 2025

Choose a reason for hiding this comment

Uh oh!

oli-obk Jul 9, 2025

Choose a reason for hiding this comment

Uh oh!

nbdd0121 Jul 9, 2025

Choose a reason for hiding this comment

Uh oh!

rustbot commented Jul 9, 2025

Uh oh!

nbdd0121 commented Jul 9, 2025

Uh oh!

This comment has been minimized.

compiler-errors commented Jul 9, 2025

Uh oh!

nbdd0121 commented Jul 9, 2025

Uh oh!

compiler-errors Jul 9, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nbdd0121 Jul 9, 2025

Choose a reason for hiding this comment

Uh oh!

Support using const pointers in asm `const` operand #138618

Support using const pointers in asm `const` operand #138618

nbdd0121 commented Mar 17, 2025 •

edited

Loading

compiler-errors Jul 9, 2025 •

edited

Loading