Auto merge of rust-lang#93009 - matthiaskrgr:rollup-3fkxg6i, r=matthiaskrgr

Rollup of 10 pull requests

Successful merges:

 - rust-lang#90498 (Clarifications in the target tier policy)
 - rust-lang#92164 (Implement `#[rustc_must_implement_one_of]` attribute)
 - rust-lang#92729 (rustc_codegen_llvm: Remove (almost) unused span parameter from many functions in metadata.rs)
 - rust-lang#92752 (Correct minor typos in some long error code explanations)
 - rust-lang#92801 (Enable wrapping words by default)
 - rust-lang#92825 (Rename environment variable for overriding rustc version)
 - rust-lang#92877 (Remove LLVMRustMarkAllFunctionsNounwind)
 - rust-lang#92936 (rustdoc: Remove `collect` in `html::markdown::parse`)
 - rust-lang#92956 (Add `log2` and `log10` to `NonZeroU*`)
 - rust-lang#92960 (Use `carrying_{mul|add}` in `num::bignum`)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

Author: bors

compiler/rustc_codegen_llvm/src/back/lto.rs

@@ -349,13 +349,6 @@ fn fat_lto(
             );
             save_temp_bitcode(cgcx, &module, "lto.after-restriction");
         }
-
-        if cgcx.no_landing_pads {
-            unsafe {
-                llvm::LLVMRustMarkAllFunctionsNounwind(llmod);
-            }
-            save_temp_bitcode(cgcx, &module, "lto.after-nounwind");
-        }
     }
 
     Ok(LtoModuleCodegen::Fat { module: Some(module), _serialized_bitcode: serialized_bitcode })
@@ -770,16 +763,6 @@ pub unsafe fn optimize_thin_module(
             return Err(write::llvm_err(&diag_handler, msg));
         }
 
-        // Like with "fat" LTO, get some better optimizations if landing pads
-        // are disabled by removing all landing pads.
-        if cgcx.no_landing_pads {
-            let _timer = cgcx
-                .prof
-                .generic_activity_with_arg("LLVM_thin_lto_remove_landing_pads", thin_module.name());
-            llvm::LLVMRustMarkAllFunctionsNounwind(llmod);
-            save_temp_bitcode(cgcx, &module, "thin-lto-after-nounwind");
-        }
-
         // Up next comes the per-module local analyses that we do for Thin LTO.
         // Each of these functions is basically copied from the LLVM
         // implementation and then tailored to suit this implementation. Ideally