Auto merge of rust-lang#120069 - Mark-Simulacrum:fast-memcpy, r=<try>

Optimize large array creation in const-eval

This changes repeated memcpy's to a memset for the case that we're propagating a single byte into a region of memory. It also optimizes the element-by-element copies to have a tighter loop; I'm pretty sure the old code was actually doing a multiply within each loop iteration.

For an 8GB array (`static SLICE: [u8; SIZE] = [0u8; 1 << 33];`) this takes us from ~23 seconds to ~6 seconds locally, which is spent roughly 50/50 in (a) memset to zero and (b) memcpy of the original place into a new place, when popping stack frame. The latter seems hard to avoid but is a big memcpy (since we're copying the type rather than initializing a region, so it's pretty fast), and the first is as good as it's going to get without special casing constant-valued arrays.

Closes rust-lang#55795. (That issue's references to lint checking don't appear true anymore, but I think this closes that case as something that is slow due to *time* pretty fully. An 8GB array taking only 6 seconds feels reasonable enough to not merit further tracking).

compiler/rustc_const_eval/src/interpret/memory.rs

@@ -1209,21 +1209,25 @@ impl<'mir, 'tcx: 'mir, M: Machine<'mir, 'tcx>> InterpCx<'mir, 'tcx, M> {
                         throw_ub_custom!(fluent::const_eval_copy_nonoverlapping_overlapping);
                     }
                 }
+            }
 
-                for i in 0..num_copies {
-                    ptr::copy(
-                        src_bytes,
-                        dest_bytes.add((size * i).bytes_usize()), // `Size` multiplication
-                        size.bytes_usize(),
-                    );
+            let size_in_bytes = size.bytes_usize();
+            // For particularly large arrays (where this is perf-sensitive) it's common that
+            // we're writing a single byte repeatedly. So, optimize that case to a memset.
+            if size_in_bytes == 1 && num_copies >= 1 {
+                let value = *src_bytes;
+                dest_bytes.write_bytes(value, (size * num_copies).bytes_usize());
+            } else if src_alloc_id == dest_alloc_id {
+                let mut dest_ptr = dest_bytes;
+                for _ in 0..num_copies {
+                    ptr::copy(src_bytes, dest_ptr, size_in_bytes);
+                    dest_ptr = dest_ptr.add(size_in_bytes);
                 }
             } else {
-                for i in 0..num_copies {
-                    ptr::copy_nonoverlapping(
-                        src_bytes,
-                        dest_bytes.add((size * i).bytes_usize()), // `Size` multiplication
-                        size.bytes_usize(),
-                    );
+                let mut dest_ptr = dest_bytes;
+                for _ in 0..num_copies {
+                    ptr::copy_nonoverlapping(src_bytes, dest_ptr, size_in_bytes);
+                    dest_ptr = dest_ptr.add(size_in_bytes);
                 }
             }
         }