Add -Z small-data-threshold

This flag allows specifying the threshold size above which LLVM should
not consider placing small objects in a .sdata or .sbss section.

compiler/rustc_codegen_llvm/src/context.rs

@@ -351,6 +351,19 @@ pub unsafe fn create_module<'ll>(
         );
     }
 
+    if let Some(threshold) = sess.opts.unstable_opts.small_data_threshold {
+        // Set up the small-data optimization limit for architectures that use
+        // an LLVM module flag to control this.
+        if sess.target.arch.starts_with("riscv") {
+            llvm::LLVMRustAddModuleFlag(
+                llmod,
+                llvm::LLVMModFlagBehavior::Error,
+                "SmallDataLimit\0".as_ptr().cast(),
+                threshold as u32,
+            )
+        }
+    }
+
     // Insert `llvm.ident` metadata.
     //
     // On the wasm targets it will get hooked up to the "producer" sections

compiler/rustc_codegen_llvm/src/llvm_util.rs

@@ -118,6 +118,24 @@ unsafe fn configure_llvm(sess: &Session) {
         for arg in sess_args {
             add(&(*arg), true);
         }
+
+        if let Some(threshold) = sess.opts.unstable_opts.small_data_threshold {
+            // Set up the small-data optimization limit for architectures that use
+            // an LLVM argument to control this.
+            match sess.target.arch.as_ref() {
+                "hexagon" => add(&format!("--hexagon-small-data-threshold={threshold}"), false),
+                // m68k accepts the --m68k-ssection-threshold argument but then
+                // ignores it, so this currently has no effect on m68k
+                "m68k" => add(&format!("--m68k-ssection-threshold={threshold}"), false),
+                // There's currently no rustc support for the Lanai architecture, so this is untested
+                "lanai" => add(&format!("--lanai-ssection-threshold={threshold}"), false),
+                arch @ _ => {
+                    if arch.starts_with("mips") {
+                        add(&format!("--mips-ssection-threshold={threshold}"), false);
+                    }
+                }
+            }
+        }
     }
 
     if sess.opts.unstable_opts.llvm_time_trace {

compiler/rustc_interface/src/tests.rs

@@ -609,6 +609,7 @@ fn test_codegen_options_tracking_hash() {
     tracked!(profile_generate, SwitchWithOptPath::Enabled(None));
     tracked!(profile_use, Some(PathBuf::from("abc")));
     tracked!(relocation_model, Some(RelocModel::Pic));
+    tracked!(small_data_threshold, Some(16));
     tracked!(soft_float, true);
     tracked!(split_debuginfo, Some(SplitDebuginfo::Packed));
     tracked!(symbol_mangling_version, Some(SymbolManglingVersion::V0));

compiler/rustc_session/src/options.rs

@@ -1792,6 +1792,8 @@ written to standard error output)"),
     simulate_remapped_rust_src_base: Option<PathBuf> = (None, parse_opt_pathbuf, [TRACKED],
         "simulate the effect of remap-debuginfo = true at bootstrapping by remapping path \
         to rust's source base directory. only meant for testing purposes"),
+    small_data_threshold: Option<usize> = (None, parse_opt_number, [TRACKED],
+        "Set the threshold for objects to be stored in a \"small data\" section"),
     span_debug: bool = (false, parse_bool, [UNTRACKED],
         "forward proc_macro::Span's `Debug` impl to `Span`"),
     /// o/w tests have closure@path

src/doc/unstable-book/src/compiler-flags/small-data-threshold.md

@@ -0,0 +1,8 @@
+# `small-data-threshold`
+
+-----------------------
+
+This flag controls the maximum static variable size that may be included in the
+"small data segments" (.sdata, .sbss) supported by some architectures (RISCV,
+MIPS, M68K, Hexagon).  Can be set to `0` to disable the use of small data
+segments.

tests/assembly/small_data_threshold.rs

@@ -0,0 +1,70 @@
+// Test for -Z small_data_threshold=...
+// revisions: RISCV MIPS HEXAGON M68K
+// assembly-output: emit-asm
+// compile-flags: -Z small_data_threshold=4
+// [RISCV] compile-flags:--target=riscv32im-unknown-none-elf
+// [MIPS] compile-flags:--target=mips-unknown-linux-musl -C llvm-args=-relocation-model=static -C llvm-args=-membedded-data -C llvm-args=-mgpopt -C llvm-args=-target-abi=n64
+// [HEXAGON] compile-flags:--target=hexagon-unknown-linux-musl -C target-feature=+small-data -C llvm-args=--hexagon-statics-in-small-data
+// [M68K] compile-flags:--target=m68k-unknown-linux-gnu
+
+#![feature(no_core, lang_items)]
+#![no_std]
+#![no_core]
+#![crate_type = "lib"]
+
+#[lang = "sized"]
+trait Sized {}
+
+#[lang = "drop_in_place"]
+fn drop_in_place<T>(_: *mut T) {}
+
+#[used]
+#[no_mangle]
+// U is below the threshold, should be in sdata
+pub static mut U: u16 = 123;
+
+#[used]
+#[no_mangle]
+// V is below the threshold, should be in sbss
+pub static mut V: u16 = 0;
+
+#[used]
+#[no_mangle]
+// W is at the threshold, should be in sdata
+pub static mut W: u32 = 123;
+
+#[used]
+#[no_mangle]
+// X is at the threshold, should be in sbss
+pub static mut X: u32 = 0;
+
+#[used]
+#[no_mangle]
+// Y is over the threshold, should be in its own .data section
+pub static mut Y: u64 = 123;
+
+#[used]
+#[no_mangle]
+/// Z is over the threshold, should be in its own .bss section
+pub static mut Z: u64 = 0;
+
+// Currently, only RISCV successfully puts any objects in the small data
+// sections so the U/V/W/X tests are skipped on MIPS/Hexagon/M68K
+// RISCV: .section .sdata,
+// RISCV-NOT: .section
+// RISCV: U:
+// RISCV: .section .sbss
+// RISV-NOT: .section
+// RISCV: V:
+// RISCV .section .sdata
+// RISV-NOT: .section
+// RISCV: W:
+// RISCV: .section .sbss
+// RISV-NOT: .section
+// RISCV: X:
+// CHECK: .section .data.Y,
+// CHECK-NOT: .section
+// CHECK: Y:
+// CHECK: .section .bss.Z,
+// CHECK-NOT: .section
+// CHECK: Z: