Support clobber_abi and vector registers (clobber-only) in PowerPC in…

…line assembly
rust-lang · Oct 7, 2024 · fc3dc84 · fc3dc84
1 parent 7caad69
commit fc3dc84
Show file tree

Hide file tree

Showing 6 changed files with 116 additions and 12 deletions.
diff --git a/compiler/rustc_codegen_gcc/src/asm.rs b/compiler/rustc_codegen_gcc/src/asm.rs
@@ -654,7 +654,8 @@ fn reg_to_gcc(reg: InlineAsmRegOrRegClass) -> ConstraintOrRegister {
             InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::reg_nonzero) => "b",
             InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::freg) => "f",
             InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::cr)
-            | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::xer) => {
+            | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::xer)
+            | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::vreg) => {
                 unreachable!("clobber-only")
             }
             InlineAsmRegClass::RiscV(RiscVInlineAsmRegClass::reg) => "r",
@@ -729,7 +730,8 @@ fn dummy_output_type<'gcc, 'tcx>(cx: &CodegenCx<'gcc, 'tcx>, reg: InlineAsmRegCl
         InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::reg_nonzero) => cx.type_i32(),
         InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::freg) => cx.type_f64(),
         InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::cr)
-        | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::xer) => {
+        | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::xer)
+        | InlineAsmRegClass::PowerPC(PowerPCInlineAsmRegClass::vreg) => {
             unreachable!("clobber-only")
         }
         InlineAsmRegClass::RiscV(RiscVInlineAsmRegClass::reg) => cx.type_i32(),

diff --git a/compiler/rustc_codegen_llvm/src/asm.rs b/compiler/rustc_codegen_llvm/src/asm.rs
@@ -679,7 +679,9 @@ fn reg_to_llvm(reg: InlineAsmRegOrRegClass, layout: Option<&TyAndLayout<'_>>) ->
             PowerPC(PowerPCInlineAsmRegClass::reg) => "r",
             PowerPC(PowerPCInlineAsmRegClass::reg_nonzero) => "b",
             PowerPC(PowerPCInlineAsmRegClass::freg) => "f",
-            PowerPC(PowerPCInlineAsmRegClass::cr) | PowerPC(PowerPCInlineAsmRegClass::xer) => {
+            PowerPC(PowerPCInlineAsmRegClass::cr)
+            | PowerPC(PowerPCInlineAsmRegClass::xer)
+            | PowerPC(PowerPCInlineAsmRegClass::vreg) => {
                 unreachable!("clobber-only")
             }
             RiscV(RiscVInlineAsmRegClass::reg) => "r",
@@ -841,7 +843,9 @@ fn dummy_output_type<'ll>(cx: &CodegenCx<'ll, '_>, reg: InlineAsmRegClass) -> &'
         PowerPC(PowerPCInlineAsmRegClass::reg) => cx.type_i32(),
         PowerPC(PowerPCInlineAsmRegClass::reg_nonzero) => cx.type_i32(),
         PowerPC(PowerPCInlineAsmRegClass::freg) => cx.type_f64(),
-        PowerPC(PowerPCInlineAsmRegClass::cr) | PowerPC(PowerPCInlineAsmRegClass::xer) => {
+        PowerPC(PowerPCInlineAsmRegClass::cr)
+        | PowerPC(PowerPCInlineAsmRegClass::xer)
+        | PowerPC(PowerPCInlineAsmRegClass::vreg) => {
             unreachable!("clobber-only")
         }
         RiscV(RiscVInlineAsmRegClass::reg) => cx.type_i32(),

diff --git a/compiler/rustc_target/src/asm/mod.rs b/compiler/rustc_target/src/asm/mod.rs
@@ -892,6 +892,7 @@ pub enum InlineAsmClobberAbi {
     AArch64NoX18,
     RiscV,
     LoongArch,
+    PowerPC,
     S390x,
 }
 
@@ -942,6 +943,10 @@ impl InlineAsmClobberAbi {
                 "C" | "system" => Ok(InlineAsmClobberAbi::LoongArch),
                 _ => Err(&["C", "system"]),
             },
+            InlineAsmArch::PowerPC | InlineAsmArch::PowerPC64 => match name {
+                "C" | "system" => Ok(InlineAsmClobberAbi::PowerPC),
+                _ => Err(&["C", "system"]),
+            },
             InlineAsmArch::S390x => match name {
                 "C" | "system" => Ok(InlineAsmClobberAbi::S390x),
                 _ => Err(&["C", "system"]),
@@ -1103,6 +1108,31 @@ impl InlineAsmClobberAbi {
                     f16, f17, f18, f19, f20, f21, f22, f23,
                 }
             },
+            InlineAsmClobberAbi::PowerPC => clobbered_regs! {
+                PowerPC PowerPCInlineAsmReg {
+                    // r0, r3-r12
+                    r0,
+                    r3, r4, r5, r6, r7,
+                    r8, r9, r10, r11, r12,
+
+                    // f0-f13
+                    f0, f1, f2, f3, f4, f5, f6, f7,
+                    f8, f9, f10, f11, f12, f13,
+
+                    // v0-v19
+                    // FIXME: PPC32 SysV ABI does not mention vector registers processing.
+                    // https://refspecs.linuxfoundation.org/elf/elfspec_ppc.pdf
+                    v0, v1, v2, v3, v4, v5, v6, v7,
+                    v8, v9, v10, v11, v12, v13, v14,
+                    v15, v16, v17, v18, v19,
+
+                    // cr0-cr1, cr5-cr7, xer
+                    cr0, cr1,
+                    cr5, cr6, cr7,
+                    xer,
+                    // lr and ctr are reserved
+                }
+            },
             InlineAsmClobberAbi::S390x => clobbered_regs! {
                 S390x S390xInlineAsmReg {
                     r0, r1, r2, r3, r4, r5,

diff --git a/compiler/rustc_target/src/asm/powerpc.rs b/compiler/rustc_target/src/asm/powerpc.rs
@@ -9,6 +9,7 @@ def_reg_class! {
         reg,
         reg_nonzero,
         freg,
+        vreg,
         cr,
         xer,
     }
@@ -48,6 +49,7 @@ impl PowerPCInlineAsmRegClass {
                 }
             }
             Self::freg => types! { _: F32, F64; },
+            Self::vreg => &[],
             Self::cr | Self::xer => &[],
         }
     }
@@ -113,6 +115,40 @@ def_regs! {
         f29: freg = ["f29", "fr29"],
         f30: freg = ["f30", "fr30"],
         f31: freg = ["f31", "fr31"],
+        v0: vreg = ["v0"],
+        v1: vreg = ["v1"],
+        v2: vreg = ["v2"],
+        v3: vreg = ["v3"],
+        v4: vreg = ["v4"],
+        v5: vreg = ["v5"],
+        v6: vreg = ["v6"],
+        v7: vreg = ["v7"],
+        v8: vreg = ["v8"],
+        v9: vreg = ["v9"],
+        v10: vreg = ["v10"],
+        v11: vreg = ["v11"],
+        v12: vreg = ["v12"],
+        v13: vreg = ["v13"],
+        v14: vreg = ["v14"],
+        v15: vreg = ["v15"],
+        v16: vreg = ["v16"],
+        v17: vreg = ["v17"],
+        v18: vreg = ["v18"],
+        v19: vreg = ["v19"],
+        // FIXME: In AIX, v20-v31 are reserved or nonvolatile depending on the mode.
+        // https://www.ibm.com/docs/en/aix/7.3?topic=concepts-aix-vector-programming
+        v20: vreg = ["v20"],
+        v21: vreg = ["v21"],
+        v22: vreg = ["v22"],
+        v23: vreg = ["v23"],
+        v24: vreg = ["v24"],
+        v25: vreg = ["v25"],
+        v26: vreg = ["v26"],
+        v27: vreg = ["v27"],
+        v28: vreg = ["v28"],
+        v29: vreg = ["v29"],
+        v30: vreg = ["v30"],
+        v31: vreg = ["v31"],
         cr: cr = ["cr"],
         cr0: cr = ["cr0"],
         cr1: cr = ["cr1"],
@@ -170,6 +206,10 @@ impl PowerPCInlineAsmReg {
             (f8, "8"), (f9, "9"), (f10, "10"), (f11, "11"), (f12, "12"), (f13, "13"), (f14, "14"), (f15, "15");
             (f16, "16"), (f17, "17"), (f18, "18"), (f19, "19"), (f20, "20"), (f21, "21"), (f22, "22"), (f23, "23");
             (f24, "24"), (f25, "25"), (f26, "26"), (f27, "27"), (f28, "28"), (f29, "29"), (f30, "30"), (f31, "31");
+            (v0, "0"), (v1, "1"), (v2, "2"), (v3, "3"), (v4, "4"), (v5, "5"), (v6, "6"), (v7, "7");
+            (v8, "8"), (v9, "9"), (v10, "10"), (v11, "11"), (v12, "12"), (v13, "13"), (v14, "14"), (v15, "15");
+            (v16, "16"), (v17, "17"), (v18, "18"), (v19, "19"), (v20, "20"), (v21, "21"), (v22, "22"), (v23, "23");
+            (v24, "24"), (v25, "25"), (v26, "26"), (v27, "27"), (v28, "28"), (v29, "29"), (v30, "30"), (v31, "31");
             (cr, "cr");
             (cr0, "0"), (cr1, "1"), (cr2, "2"), (cr3, "3"), (cr4, "4"), (cr5, "5"), (cr6, "6"), (cr7, "7");
             (xer, "xer");
@@ -201,5 +241,6 @@ impl PowerPCInlineAsmReg {
         reg_conflicts! {
             cr : cr0 cr1 cr2 cr3 cr4 cr5 cr6 cr7;
         }
+        // f0-f31 (vsr0-vsr31) and v0-v31 (vsr32-vsr63) do not conflict.
     }
 }
diff --git a/src/doc/unstable-book/src/language-features/asm-experimental-arch.md b/src/doc/unstable-book/src/language-features/asm-experimental-arch.md
@@ -31,9 +31,10 @@ This feature tracks `asm!` and `global_asm!` support for the following architect
 | NVPTX        | `reg32`        | None\*                             | `r`                  |
 | NVPTX        | `reg64`        | None\*                             | `l`                  |
 | Hexagon      | `reg`          | `r[0-28]`                          | `r`                  |
-| PowerPC      | `reg`          | `r[0-31]`                          | `r`                  |
-| PowerPC      | `reg_nonzero`  | `r[1-31]`                          | `b`                  |
+| PowerPC      | `reg`          | `r0`, `r[3-12]`, `r[14-28]`        | `r`                  |
+| PowerPC      | `reg_nonzero`  | `r[3-12]`, `r[14-28]`              | `b`                  |
 | PowerPC      | `freg`         | `f[0-31]`                          | `f`                  |
+| PowerPC      | `vreg`         | `v[0-31]`                          | Only clobbers        |
 | PowerPC      | `cr`           | `cr[0-7]`, `cr`                    | Only clobbers        |
 | PowerPC      | `xer`          | `xer`                              | Only clobbers        |
 | wasm32       | `local`        | None\*                             | `r`                  |
@@ -79,6 +80,7 @@ This feature tracks `asm!` and `global_asm!` support for the following architect
 | PowerPC      | `reg`                           | None           | `i8`, `i16`, `i32`                      |
 | PowerPC      | `reg_nonzero`                   | None           | `i8`, `i16`, `i32`                      |
 | PowerPC      | `freg`                          | None           | `f32`, `f64`                            |
+| PowerPC      | `vreg`                          | N/A            | Only clobbers                           |
 | PowerPC      | `cr`                            | N/A            | Only clobbers                           |
 | PowerPC      | `xer`                           | N/A            | Only clobbers                           |
 | wasm32       | `local`                         | None           | `i8` `i16` `i32` `i64` `f32` `f64`      |
@@ -105,6 +107,10 @@ This feature tracks `asm!` and `global_asm!` support for the following architect
 | Hexagon      | `r29`         | `sp`      |
 | Hexagon      | `r30`         | `fr`      |
 | Hexagon      | `r31`         | `lr`      |
+| PowerPC      | `r1`          | `sp`      |
+| PowerPC      | `r31`         | `fp`      |
+| PowerPC      | `r[0-31]`     | `[0-31]`  |
+| PowerPC      | `f[0-31]`     | `fr[0-31]`|
 | BPF          | `r[0-10]`     | `w[0-10]` |
 | AVR          | `XH`          | `r27`     |
 | AVR          | `XL`          | `r26`     |
@@ -145,14 +151,19 @@ This feature tracks `asm!` and `global_asm!` support for the following architect
 | Architecture | Unsupported register                    | Reason                                                                                                                                                                              |
 | ------------ | --------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | All          | `sp`, `r15` (s390x)                     | The stack pointer must be restored to its original value at the end of an asm code block.                                                                                           |
-| All          | `fr` (Hexagon), `$fp` (MIPS), `Y` (AVR), `r4` (MSP430), `a6` (M68k), `r11` (s390x), `x29` (Arm64EC) | The frame pointer cannot be used as an input or output.                                                                                                                             |
+| All          | `fr` (Hexagon), `fp` (PowerPC), `$fp` (MIPS), `Y` (AVR), `r4` (MSP430), `a6` (M68k), `r11` (s390x), `x29` (Arm64EC) | The frame pointer cannot be used as an input or output.                                                                                                                             |
 | All          | `r19` (Hexagon), `x19` (Arm64EC)        | This is used internally by LLVM as a "base pointer" for functions with complex stack frames.                                                                                        |
 | MIPS         | `$0` or `$zero`                         | This is a constant zero register which can't be modified.                                                                                                                           |
 | MIPS         | `$1` or `$at`                           | Reserved for assembler.                                                                                                                                                             |
 | MIPS         | `$26`/`$k0`, `$27`/`$k1`                | OS-reserved registers.                                                                                                                                                              |
 | MIPS         | `$28`/`$gp`                             | Global pointer cannot be used as inputs or outputs.                                                                                                                                 |
 | MIPS         | `$ra`                                   | Return address cannot be used as inputs or outputs.                                                                                                                                 |
 | Hexagon      | `lr`                                    | This is the link register which cannot be used as an input or output.                                                                                                               |
+| PowerPC      | `$r2`, `$r13`                           | These are system reserved registers.                                                                                                                                                |
+| PowerPC      | `$r29`, `$r30`                          | These are used internally by LLVM.                                                                                                                                                  |
+| PowerPC      | `lr`                                    | The link register cannot be used as an input or output.                                                                                                                             |
+| PowerPC      | `ctr`                                   | The counter register cannot be used as an input or output.                                                                                                                          |
+| PowerPC      | `vrsave`                                | The vrsave register cannot be used as an input or output.                                                                                                                           |
 | AVR          | `r0`, `r1`, `r1r0`                      | Due to an issue in LLVM, the `r0` and `r1` registers cannot be used as inputs or outputs.  If modified, they must be restored to their original values before the end of the block. |
 |MSP430        | `r0`, `r2`, `r3`                        | These are the program counter, status register, and constant generator respectively. Neither the status register nor constant generator can be written to.                          |
 | M68k         | `a4`, `a5`                              | Used internally by LLVM for the base pointer and global base pointer. |

diff --git a/tests/codegen/asm-powerpc-clobbers.rs b/tests/codegen/asm-powerpc-clobbers.rs
@@ -1,10 +1,12 @@
-//@ revisions: powerpc powerpc64 powerpc64le
+//@ revisions: powerpc powerpc64 powerpc64le aix
 //@[powerpc] compile-flags: --target powerpc-unknown-linux-gnu
 //@[powerpc] needs-llvm-components: powerpc
 //@[powerpc64] compile-flags: --target powerpc64-unknown-linux-gnu
 //@[powerpc64] needs-llvm-components: powerpc
 //@[powerpc64le] compile-flags: --target powerpc64le-unknown-linux-gnu
 //@[powerpc64le] needs-llvm-components: powerpc
+//@[aix] compile-flags: --target powerpc64-ibm-aix
+//@[aix] needs-llvm-components: powerpc
 
 #![crate_type = "rlib"]
 #![feature(no_core, rustc_attrs, lang_items, asm_experimental_arch)]
@@ -22,26 +24,40 @@ macro_rules! asm {
 // CHECK: call void asm sideeffect "", "~{cr}"()
 #[no_mangle]
 pub unsafe fn cr_clobber() {
-    asm!("", out("cr") _, options(nostack, nomem));
+    asm!("", out("cr") _, options(nostack, nomem, preserves_flags));
 }
 
 // CHECK-LABEL: @cr0_clobber
 // CHECK: call void asm sideeffect "", "~{cr0}"()
 #[no_mangle]
 pub unsafe fn cr0_clobber() {
-    asm!("", out("cr0") _, options(nostack, nomem));
+    asm!("", out("cr0") _, options(nostack, nomem, preserves_flags));
 }
 
 // CHECK-LABEL: @cr5_clobber
 // CHECK: call void asm sideeffect "", "~{cr5}"()
 #[no_mangle]
 pub unsafe fn cr5_clobber() {
-    asm!("", out("cr5") _, options(nostack, nomem));
+    asm!("", out("cr5") _, options(nostack, nomem, preserves_flags));
 }
 
 // CHECK-LABEL: @xer_clobber
 // CHECK: call void asm sideeffect "", "~{xer}"()
 #[no_mangle]
 pub unsafe fn xer_clobber() {
-    asm!("", out("xer") _, options(nostack, nomem));
+    asm!("", out("xer") _, options(nostack, nomem, preserves_flags));
+}
+
+// CHECK-LABEL: @v0_clobber
+// CHECK: call void asm sideeffect "", "~{v0}"()
+#[no_mangle]
+pub unsafe fn v0_clobber() {
+    asm!("", out("v0") _, options(nostack, nomem, preserves_flags));
+}
+
+// CHECK-LABEL: @clobber_abi
+// CHECK: asm sideeffect "", "={r0},={r3},={r4},={r5},={r6},={r7},={r8},={r9},={r10},={r11},={r12},={f0},={f1},={f2},={f3},={f4},={f5},={f6},={f7},={f8},={f9},={f10},={f11},={f12},={f13},~{v0},~{v1},~{v2},~{v3},~{v4},~{v5},~{v6},~{v7},~{v8},~{v9},~{v10},~{v11},~{v12},~{v13},~{v14},~{v15},~{v16},~{v17},~{v18},~{v19},~{cr0},~{cr1},~{cr5},~{cr6},~{cr7},~{xer}"()
+#[no_mangle]
+pub unsafe fn clobber_abi() {
+    asm!("", clobber_abi("C"), options(nostack, nomem, preserves_flags));
 }