[PATCH 0/2] fixing a few function pointer mismatches

[PATCH 0/2] fixing a few function pointer mismatches

[Jeff King]

I saw that clang-18 hit Debian unstable, so I did my usual "build and
see if it finds anything new to complain about". Here's the result. :)

Note that that first one is a regression in the upcoming v2.42, though I
suspect the fallout would not be very widespread (see comments in the
commit message).

  [1/2]: fsck: use enum object_type for fsck_walk callback
  [2/2]: hashmap: use expected signatures for comparison functions

 builtin/fsck.c    |  2 +-
 compat/terminal.c | 10 ++++++----
 range-diff.c      | 11 +++++++----
 3 files changed, 14 insertions(+), 9 deletions(-)

-Peff

[PATCH 1/2] fsck: use enum object_type for fsck_walk callback

[Jeff King]

We switched the function interface for fsck callbacks in a1aad71601
(fsck.h: use "enum object_type" instead of "int", 2021-03-28). However,
we accidentally flipped the type back to "int" as part of 0b4e9013f1
(fsck: mark unused parameters in various fsck callbacks, 2023-07-03).
The mistake happened because that commit was written before a1aad71601
and rebased forward, and I screwed up while resolving the conflict.

Curiously, the compiler does not warn about this mismatch, at least not
when using gcc and clang on Linux (nor in any of our CI environments).
Based on 28abf260a5 (builtin/fsck.c: don't conflate "int" and "enum" in
callback, 2021-06-01), I'd guess that this would cause the AIX xlc
compiler to complain. I noticed because clang-18's UBSan now identifies
mis-matched function calls at runtime, and does complain of this case
when running the test suite.

I'm not entirely clear on whether this mismatch is a problem in
practice. Compilers are certainly free to make enums smaller than "int"
if they don't need the bits, but I suspect that they have to promote
back to int for function calls (though I didn't dig in the standard, and
I won't be surprised if I'm simply wrong and the real-world impact would
depend on the ABI).

Regardless, switching it back to enum is obviously the right thing to do
here; the switch to "int" was simply a mistake.

Signed-off-by: Jeff King <peff@peff.net>
---
 builtin/fsck.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/builtin/fsck.c b/builtin/fsck.c
index c1d0290026..611925905e 100644
--- a/builtin/fsck.c
+++ b/builtin/fsck.c
@@ -206,7 +206,7 @@ static int traverse_reachable(void)
 	return !!result;
 }
 
-static int mark_used(struct object *obj, int type UNUSED,
+static int mark_used(struct object *obj, enum object_type type UNUSED,
 		     void *data UNUSED, struct fsck_options *options UNUSED)
 {
 	if (!obj)
-- 
2.42.0.rc2.418.g602d5859fc

[PATCH 2/2] hashmap: use expected signatures for comparison functions

[Jeff King]

We prefer for callback functions to match the signature with which
they'll be called, rather than casting them to the correct type when
assigning function pointers. Even though casting often works in the real
world, it is a violation of the standard.

We did a mass conversion in 939af16eac (hashmap_cmp_fn takes
hashmap_entry params, 2019-10-06), but have grown a few new cases since
then. Because of the cast, the compiler does not complain. However, as
of clang-18, UBSan will catch these at run-time, and the case in
range-diff.c triggers when running t3206.

After seeing that one, I scanned the results of:

  git grep '_fn)[^(]' '*.c' | grep -v typedef

and found a similar case in compat/terminal.c (which presumably isn't
called in the test suite, since it doesn't trigger UBSan). There might
be other cases lurking if the cast is done using a typedef that doesn't
end in "_fn", but loosening it finds too many false positives. I also
looked for:

  git grep ' = ([a-z_]*) *[a-z]' '*.c'

to find assignments that cast, but nothing looked like a function.

The resulting code is unfortunately a little longer, but the bonus of
using container_of() is that we are no longer restricted to the
hashmap_entry being at the start of the struct.

Signed-off-by: Jeff King <peff@peff.net>
---
 compat/terminal.c | 10 ++++++----
 range-diff.c      | 11 +++++++----
 2 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/compat/terminal.c b/compat/terminal.c
index 83d95e8656..0afda730f2 100644
--- a/compat/terminal.c
+++ b/compat/terminal.c
@@ -479,10 +479,13 @@ struct escape_sequence_entry {
 };
 
 static int sequence_entry_cmp(const void *hashmap_cmp_fn_data UNUSED,
-			      const struct escape_sequence_entry *e1,
-			      const struct escape_sequence_entry *e2,
+			      const struct hashmap_entry *he1,
+			      const struct hashmap_entry *he2,
 			      const void *keydata)
 {
+	const struct escape_sequence_entry
+		*e1 = container_of(he1, const struct escape_sequence_entry, entry),
+		*e2 = container_of(he2, const struct escape_sequence_entry, entry);
 	return strcmp(e1->sequence, keydata ? keydata : e2->sequence);
 }
 
@@ -496,8 +499,7 @@ static int is_known_escape_sequence(const char *sequence)
 		struct strbuf buf = STRBUF_INIT;
 		char *p, *eol;
 
-		hashmap_init(&sequences, (hashmap_cmp_fn)sequence_entry_cmp,
-			     NULL, 0);
+		hashmap_init(&sequences, sequence_entry_cmp, NULL, 0);
 
 		strvec_pushl(&cp.args, "infocmp", "-L", "-1", NULL);
 		if (pipe_command(&cp, NULL, 0, &buf, 0, NULL, 0))
diff --git a/range-diff.c b/range-diff.c
index 2e86063491..ca5493984a 100644
--- a/range-diff.c
+++ b/range-diff.c
@@ -230,16 +230,19 @@ static int read_patches(const char *range, struct string_list *list,
 }
 
 static int patch_util_cmp(const void *cmp_data UNUSED,
-			  const struct patch_util *a,
-			  const struct patch_util *b,
-			  const char *keydata)
+			  const struct hashmap_entry *ha,
+			  const struct hashmap_entry *hb,
+			  const void *keydata)
 {
+	const struct patch_util
+		*a = container_of(ha, const struct patch_util, e),
+		*b = container_of(hb, const struct patch_util, e);
 	return strcmp(a->diff, keydata ? keydata : b->diff);
 }
 
 static void find_exact_matches(struct string_list *a, struct string_list *b)
 {
-	struct hashmap map = HASHMAP_INIT((hashmap_cmp_fn)patch_util_cmp, NULL);
+	struct hashmap map = HASHMAP_INIT(patch_util_cmp, NULL);
 	int i;
 
 	/* First, add the patches of a to a hash map */
-- 
2.42.0.rc2.418.g602d5859fc

Re: [PATCH 0/2] fixing a few function pointer mismatches

[Junio C Hamano]

Jeff King <peff@peff.net> writes:

> I saw that clang-18 hit Debian unstable, so I did my usual "build and
> see if it finds anything new to complain about". Here's the result. :)
>
> Note that that first one is a regression in the upcoming v2.42, though I
> suspect the fallout would not be very widespread (see comments in the
> commit message).

Both changes do look innocuous.  Queued.

Thanks.