Rollup merge of rust-lang#141993 - tgross35:use-in-tree-builtins, r=bjorn3

Use the in-tree `compiler-builtins` for the sysroot

Many of `std`'s dependency have a dependency on the crates.io `compiler-builtins` when used with the feature `rustc-std-workspace-core`. Use a Cargo patch to select the in-tree version instead.

`compiler-builtins` is also added as a dependency of `rustc-std-workspace-core` so these crates can remove their crates.io dependency in the future.

Zulip discussion: [#t-compiler > Using in-tree compiler-builtins](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/Using.20in-tree.20compiler-builtins/with/522445336)

Once this merges, the following PRs will need to make it to a release for the relevant crates:

- rust-lang/getopts#119 (can merge at any time)
- rust-lang/hashbrown#625 (can merge at any time)
- rust-lang/stdarch#1825
- rust-lang/rustc-demangle#80
- rust-lang/cfg-if#84
- unicode-rs/unicode-width#77

The above should cover all tier 1 targets with no `std` features enabled. The remaining cover the rest:

- alexcrichton/dlmalloc-rs#50 (wasm, xous, sgx)
- gimli-rs/gimli#769
- r-efi/r-efi#89 (efi)
- r-efi/r-efi-alloc#9 (efi)
- fortanix/rust-sgx#770 (sgx)
- hermit-os/hermit-rs#718 (hermit)
- bytecodealliance/wasi-rs#108 (wasi)
- gimli-rs/addr2line#345
- oyvindln/adler2#2
- BurntSushi/memchr#180
- Frommi/miniz_oxide#173
- gimli-rs/object#777

try-job: x86_64-gnu
try-job: test-various

Author: tgross35

Cargo.lock

@@ -50,3 +50,4 @@ rustc-demangle.opt-level = "s"
 rustc-std-workspace-core = { path = 'rustc-std-workspace-core' }
 rustc-std-workspace-alloc = { path = 'rustc-std-workspace-alloc' }
 rustc-std-workspace-std = { path = 'rustc-std-workspace-std' }
+compiler_builtins = { path = "compiler-builtins/compiler-builtins" }

Cargo.toml

@@ -16,7 +16,7 @@ bench = false
 
 [dependencies]
 core = { path = "../core", public = true }
-compiler_builtins = { version = "=0.1.160", features = ['rustc-dep-of-std'] }
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins", features = ["rustc-dep-of-std"] }
 
 [features]
 compiler-builtins-mem = ['compiler_builtins/mem']

alloc/Cargo.toml

@@ -17,9 +17,7 @@ doctest = false
 test = false
 
 [dependencies]
-# For more information on this dependency see
-# https://github.com/rust-lang/rust/tree/master/library/rustc-std-workspace-core
-core = { version = "1.0.1", optional = true, package = "rustc-std-workspace-core" }
+core = { path = "../../core", optional = true }
 
 [build-dependencies]
 cc = { optional = true, version = "1.2" }

compiler-builtins/compiler-builtins/Cargo.toml

@@ -22,6 +22,9 @@ fn main() {
 
     println!("cargo:compiler-rt={}", cwd.join("compiler-rt").display());
 
+    println!("cargo::rustc-check-cfg=cfg(kernel_user_helpers)");
+    println!("cargo::rustc-check-cfg=cfg(feature, values(\"mem-unaligned\"))");
+
     // Emscripten's runtime includes all the builtins
     if target.os == "emscripten" {
         return;
@@ -47,7 +50,6 @@ fn main() {
     }
 
     // These targets have hardware unaligned access support.
-    println!("cargo::rustc-check-cfg=cfg(feature, values(\"mem-unaligned\"))");
     if target.arch.contains("x86_64")
         || target.arch.contains("x86")
         || target.arch.contains("aarch64")
@@ -78,7 +80,6 @@ fn main() {
     // Only emit the ARM Linux atomic emulation on pre-ARMv6 architectures. This
     // includes the old androideabi. It is deprecated but it is available as a
     // rustc target (arm-linux-androideabi).
-    println!("cargo::rustc-check-cfg=cfg(kernel_user_helpers)");
     if llvm_target[0] == "armv4t"
         || llvm_target[0] == "armv5te"
         || target.triple == "arm-linux-androideabi"

compiler-builtins/compiler-builtins/build.rs

@@ -4,7 +4,7 @@
 //! To avoid breaking backwards compat, C toolchains introduced a concept of "outlined atomics",
 //! where atomic operations call into the compiler runtime to dispatch between two depending on
 //! which is supported on the current CPU.
-//! See https://community.arm.com/arm-community-blogs/b/tools-software-ides-blog/posts/making-the-most-of-the-arm-architecture-in-gcc-10#:~:text=out%20of%20line%20atomics for more discussion.
+//! See <https://community.arm.com/arm-community-blogs/b/tools-software-ides-blog/posts/making-the-most-of-the-arm-architecture-in-gcc-10#:~:text=out%20of%20line%20atomics> for more discussion.
 //!
 //! Currently we only support LL/SC, because LSE requires `getauxval` from libc in order to do runtime detection.
 //! Use the `compiler-rt` intrinsics if you want LSE support.

compiler-builtins/compiler-builtins/src/aarch64_linux.rs

@@ -4,12 +4,17 @@ use core::{arch, mem};
 // Kernel-provided user-mode helper functions:
 // https://www.kernel.org/doc/Documentation/arm/kernel_user_helpers.txt
 unsafe fn __kuser_cmpxchg(oldval: u32, newval: u32, ptr: *mut u32) -> bool {
-    let f: extern "C" fn(u32, u32, *mut u32) -> u32 = mem::transmute(0xffff0fc0usize as *const ());
+    // FIXME(volatile): the third parameter is a volatile pointer
+    // SAFETY: kernel docs specify a known address with the given signature
+    let f = unsafe {
+        mem::transmute::<_, extern "C" fn(u32, u32, *mut u32) -> u32>(0xffff0fc0usize as *const ())
+    };
     f(oldval, newval, ptr) == 0
 }
 
 unsafe fn __kuser_memory_barrier() {
-    let f: extern "C" fn() = mem::transmute(0xffff0fa0usize as *const ());
+    // SAFETY: kernel docs specify a known address with the given signature
+    let f = unsafe { mem::transmute::<_, extern "C" fn()>(0xffff0fa0usize as *const ()) };
     f();
 }
 
@@ -67,8 +72,10 @@ fn insert_aligned(aligned: u32, val: u32, shift: u32, mask: u32) -> u32 {
 /// - if `size_of::<T>() == 2`, `ptr` or `ptr` offset by 2 bytes must be valid for a relaxed atomic
 ///   read of 2 bytes.
 /// - if `size_of::<T>() == 4`, `ptr` must be valid for a relaxed atomic read of 4 bytes.
+// FIXME: assert some of the preconditions in debug mode
 unsafe fn atomic_load_aligned<T>(ptr: *mut u32) -> u32 {
-    if mem::size_of::<T>() == 4 {
+    const { assert!(size_of::<T>() <= 4) };
+    if size_of::<T>() == 4 {
         // SAFETY: As `T` has a size of 4, the caller garantees this is sound.
         unsafe { AtomicU32::from_ptr(ptr).load(Ordering::Relaxed) }
     } else {
@@ -100,11 +107,13 @@ unsafe fn atomic_rmw<T, F: Fn(u32) -> u32, G: Fn(u32, u32) -> u32>(ptr: *mut T,
     let (shift, mask) = get_shift_mask(ptr);
 
     loop {
-        let curval_aligned = atomic_load_aligned::<T>(aligned_ptr);
+        // FIXME(safety): preconditions review needed
+        let curval_aligned = unsafe { atomic_load_aligned::<T>(aligned_ptr) };
         let curval = extract_aligned(curval_aligned, shift, mask);
         let newval = f(curval);
         let newval_aligned = insert_aligned(curval_aligned, newval, shift, mask);
-        if __kuser_cmpxchg(curval_aligned, newval_aligned, aligned_ptr) {
+        // FIXME(safety): preconditions review needed
+        if unsafe { __kuser_cmpxchg(curval_aligned, newval_aligned, aligned_ptr) } {
             return g(curval, newval);
         }
     }
@@ -116,13 +125,15 @@ unsafe fn atomic_cmpxchg<T>(ptr: *mut T, oldval: u32, newval: u32) -> u32 {
     let (shift, mask) = get_shift_mask(ptr);
 
     loop {
-        let curval_aligned = atomic_load_aligned::<T>(aligned_ptr);
+        // FIXME(safety): preconditions review needed
+        let curval_aligned = unsafe { atomic_load_aligned::<T>(aligned_ptr) };
         let curval = extract_aligned(curval_aligned, shift, mask);
         if curval != oldval {
             return curval;
         }
         let newval_aligned = insert_aligned(curval_aligned, newval, shift, mask);
-        if __kuser_cmpxchg(curval_aligned, newval_aligned, aligned_ptr) {
+        // FIXME(safety): preconditions review needed
+        if unsafe { __kuser_cmpxchg(curval_aligned, newval_aligned, aligned_ptr) } {
             return oldval;
         }
     }
@@ -132,7 +143,14 @@ macro_rules! atomic_rmw {
     ($name:ident, $ty:ty, $op:expr, $fetch:expr) => {
         intrinsics! {
             pub unsafe extern "C" fn $name(ptr: *mut $ty, val: $ty) -> $ty {
-                atomic_rmw(ptr, |x| $op(x as $ty, val) as u32, |old, new| $fetch(old, new)) as $ty
+                // FIXME(safety): preconditions review needed
+                unsafe {
+                    atomic_rmw(
+                        ptr,
+                        |x| $op(x as $ty, val) as u32,
+                        |old, new| $fetch(old, new)
+                    ) as $ty
+                }
             }
         }
     };
@@ -149,7 +167,8 @@ macro_rules! atomic_cmpxchg {
     ($name:ident, $ty:ty) => {
         intrinsics! {
             pub unsafe extern "C" fn $name(ptr: *mut $ty, oldval: $ty, newval: $ty) -> $ty {
-                atomic_cmpxchg(ptr, oldval as u32, newval as u32) as $ty
+                // FIXME(safety): preconditions review needed
+                unsafe { atomic_cmpxchg(ptr, oldval as u32, newval as u32) as $ty }
             }
         }
     };
@@ -285,6 +304,7 @@ atomic_cmpxchg!(__sync_val_compare_and_swap_4, u32);
 
 intrinsics! {
     pub unsafe extern "C" fn __sync_synchronize() {
-        __kuser_memory_barrier();
+       // SAFETY: preconditions are the same as the calling function.
+       unsafe {  __kuser_memory_barrier() };
     }
 }

compiler-builtins/compiler-builtins/src/arm_linux.rs

@@ -69,7 +69,7 @@ pub unsafe fn copy_backward(dest: *mut u8, src: *const u8, count: usize) {
         "rep movsb",
         "sub $7, %rsi",
         "sub $7, %rdi",
-        "mov {qword_count}, %rcx",
+        "mov {qword_count:r}, %rcx",
         "rep movsq",
         "test {pre_byte_count:e}, {pre_byte_count:e}",
         "add $7, %rsi",
@@ -212,7 +212,7 @@ pub unsafe fn c_string_length(mut s: *const core::ffi::c_char) -> usize {
     let x = {
         let r;
         asm!(
-            "movdqa ({addr}), {dest}",
+            "movdqa ({addr:r}), {dest}",
             addr = in(reg) s,
             dest = out(xmm_reg) r,
             options(att_syntax, nostack),
@@ -232,7 +232,7 @@ pub unsafe fn c_string_length(mut s: *const core::ffi::c_char) -> usize {
         let x = {
             let r;
             asm!(
-                "movdqa ({addr}), {dest}",
+                "movdqa ({addr:r}), {dest}",
                 addr = in(reg) s,
                 dest = out(xmm_reg) r,
                 options(att_syntax, nostack),

compiler-builtins/compiler-builtins/src/mem/x86_64.rs

@@ -13,7 +13,7 @@ doc = false
 
 [dependencies]
 core = { path = "../core" }
-compiler_builtins = "0.1.0"
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins" }
 
 [target.'cfg(target_os = "android")'.dependencies]
 libc = { version = "0.2", default-features = false }

panic_abort/Cargo.toml

@@ -15,7 +15,7 @@ doc = false
 alloc = { path = "../alloc" }
 core = { path = "../core" }
 unwind = { path = "../unwind" }
-compiler_builtins = "0.1.0"
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins" }
 cfg-if = { version = "1.0", features = ['rustc-dep-of-std'] }
 
 [target.'cfg(not(all(windows, target_env = "msvc")))'.dependencies]

panic_unwind/Cargo.toml

@@ -12,3 +12,4 @@ path = "lib.rs"
 
 [dependencies]
 core = { path = "../core" }
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins", features = ["compiler-builtins"] }

rustc-std-workspace-core/Cargo.toml

@@ -2,3 +2,7 @@
 #![no_core]
 
 pub use core::*;
+
+// Crate must be brought into scope so it appears in the crate graph for anything that
+// depends on `rustc-std-workspace-core`.
+use compiler_builtins as _;

rustc-std-workspace-core/lib.rs

@@ -18,7 +18,7 @@ cfg-if = { version = "1.0", features = ['rustc-dep-of-std'] }
 panic_unwind = { path = "../panic_unwind", optional = true }
 panic_abort = { path = "../panic_abort" }
 core = { path = "../core", public = true }
-compiler_builtins = { version = "=0.1.160" }
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins" }
 unwind = { path = "../unwind" }
 hashbrown = { version = "0.15", default-features = false, features = [
     'rustc-dep-of-std',

std/Cargo.toml

@@ -15,7 +15,7 @@ doc = false
 
 [dependencies]
 core = { path = "../core" }
-compiler_builtins = "0.1.0"
+compiler_builtins = { path = "../compiler-builtins/compiler-builtins" }
 cfg-if = "1.0"
 
 [target.'cfg(not(all(windows, target_env = "msvc")))'.dependencies]