codegen: fix memset unroll for small bytearrays, add 64-bit sets

Fixes #16052

When the offset in `setByteArray#` is statically known, we can provide
better alignment guarantees then just 1 byte.

Also, memset can now do 64-bit wide sets.

The current memset intrinsic is not optimal however and can be
improved for the case when we know that we deal with

(baseAddress at known alignment) + offset

For instance, on 64-bit

`setByteArray# s 1# 23# 0#`

given that bytearray is 8 bytes aligned could be unrolled into
`movb, movw, movl, movq, movq`; but currently it is
`movb x23` since alignment of 1 is all we can embed into MO_Memset op.

1 files changed, 57 insertions, 2 deletions

diff --git a/testsuite/driver/testlib.py b/testsuite/driver/testlib.py
index 3fefb5274d..95274f30e5 100644
--- a/testsuite/driver/testlib.py
+++ b/testsuite/driver/testlib.py
@@ -1131,9 +1131,9 @@ def do_compile(name, way, should_fail, top_mod, extra_mods, extra_hc_opts, **kwa
     # no problems found, this test passed
     return passed()
 
-def compile_cmp_asm( name, way, extra_hc_opts ):
+def compile_cmp_asm( name, way, ext, extra_hc_opts ):
     print('Compile only, extra args = ', extra_hc_opts)
-    result = simple_build(name + '.cmm', way, '-keep-s-files -O ' + extra_hc_opts, 0, '', 0, 0)
+    result = simple_build(name + '.' + ext, way, '-keep-s-files -O ' + extra_hc_opts, 0, '', 0, 0)
 
     if badResult(result):
         return result
@@ -1153,6 +1153,24 @@ def compile_cmp_asm( name, way, extra_hc_opts ):
     # no problems found, this test passed
     return passed()
 
+def compile_grep_asm( name, way, ext, is_substring, extra_hc_opts ):
+    print('Compile only, extra args = ', extra_hc_opts)
+    result = simple_build(name + '.' + ext, way, '-keep-s-files -O ' + extra_hc_opts, 0, '', 0, 0)
+
+    if badResult(result):
+        return result
+
+    expected_pat_file = find_expected_file(name, 'asm')
+    actual_asm_file = add_suffix(name, 's')
+
+    if not grep_output(join_normalisers(normalise_errmsg),
+                       expected_pat_file, actual_asm_file,
+                       is_substring):
+        return failBecause('asm mismatch')
+
+    # no problems found, this test passed
+    return passed()
+
 # -----------------------------------------------------------------------------
 # Compile-and-run tests
 
@@ -1735,6 +1753,43 @@ def compare_outputs(way, kind, normaliser, expected_file, actual_file, diff_file
         else:
             return False
 
+# Checks that each line from pattern_file is present in actual_file as
+# a substring or regex pattern depending on is_substring.
+def grep_output(normaliser, pattern_file, actual_file, is_substring=True):
+    expected_path = in_srcdir(pattern_file)
+    actual_path = in_testdir(actual_file)
+
+    expected_patterns = read_no_crs(expected_path).strip().split('\n')
+    actual_raw = read_no_crs(actual_path)
+    actual_str = normaliser(actual_raw)
+
+    success = True
+    failed_patterns = []
+
+    def regex_match(pat, actual):
+        return re.search(pat, actual) is not None
+
+    def substring_match(pat, actual):
+        return pat in actual
+
+    def is_match(pat, actual):
+        if is_substring:
+            return substring_match(pat, actual)
+        else:
+            return regex_match(pat, actual)
+
+    for pat in expected_patterns:
+        if not is_match(pat, actual_str):
+            success = False
+            failed_patterns.append(pat)
+
+    if not success:
+        print('Actual output does not contain the following patterns:')
+        for pat in failed_patterns:
+            print(pat)
+
+    return success
+
 # Note [Output comparison]
 #
 # We do two types of output comparison: