[3/6] hash.h: introduce `unsafe_hash_algo()`

Commit Message

Taylor Blau Nov. 20, 2024, 7:13 p.m. UTC

In 253ed9ecff (hash.h: scaffolding for _unsafe hashing variants,
2024-09-26), we introduced "unsafe" variants of the SHA-1 hashing
functions by introducing new functions like "unsafe_init_fn()" and so
on.

This approach has a major shortcoming that callers must remember to
consistently use one variant or the other. Failing to consistently use
(or not use) the unsafe variants can lead to crashes at best, or subtle
memory corruption issues at worst.

In the hashfile API, this isn't difficult to achieve, but verifying that
all callers consistently use the unsafe variants is somewhat of a chore
given how spread out all of the callers are. In the sha1 and sha1-unsafe
test helpers, all of the calls to various hash functions are guarded by
an "if (unsafe)" conditional, which is repetitive and cumbersome.

Address these issues by introducing a new pattern whereby one
'git_hash_algo' can return a pointer to another 'git_hash_algo' that
represents the unsafe version of itself. So instead of having something
like:

    if (unsafe)
      the_hash_algo->init_fn(...);
      the_hash_algo->update_fn(...);
      the_hash_algo->final_fn(...);
    else
      the_hash_algo->unsafe_init_fn(...);
      the_hash_algo->unsafe_update_fn(...);
      the_hash_algo->unsafe_final_fn(...);

we can instead write:

    struct git_hash_algo *algop = the_hash_algo;
    if (unsafe)
      algop = unsafe_hash_algo(algop);

    the_hash_algo->init_fn(...);
    the_hash_algo->update_fn(...);
    the_hash_algo->final_fn(...);

This removes the existing shortcoming by no longer forcing the caller to
"remember" which variant of the hash functions it wants to call, only to
hold onto a 'struct git_hash_algo' pointer that is initialized once.

Similarly, while there currently is still a way to "mix" safe and unsafe
functions, this too will go away after subsequent commits remove all
direct calls to the unsafe_ variants.

Suggested-by: Jeff King <[e-mail]>
Signed-off-by: Taylor Blau <[e-mail]>
---
 hash.h        |  5 +++++
 object-file.c | 26 ++++++++++++++++++++++++++
 2 files changed, 31 insertions(+)

Patch

diff --git a/hash.h b/hash.h
index 756166ce5e8..23cf6876e50 100644
--- a/hash.h
+++ b/hash.h
@@ -305,6 +305,9 @@  struct git_hash_algo {
 
 	/* The all-zeros OID. */
 	const struct object_id *null_oid;
+
+	/* The unsafe variant of this hash function, if one exists. */
+	const struct git_hash_algo *unsafe;
 };
 extern const struct git_hash_algo hash_algos[GIT_HASH_NALGOS];
 
@@ -323,6 +326,8 @@  static inline int hash_algo_by_ptr(const struct git_hash_algo *p)
 	return p - hash_algos;
 }
 
+const struct git_hash_algo *unsafe_hash_algo(const struct git_hash_algo *algop);
+
 const struct object_id *null_oid(void);
 
 static inline int hashcmp(const unsigned char *sha1, const unsigned char *sha2, const struct git_hash_algo *algop)
diff --git a/object-file.c b/object-file.c
index b1a3463852c..fddcdbe9ba6 100644
--- a/object-file.c
+++ b/object-file.c
@@ -204,6 +204,22 @@  static void git_hash_unknown_final_oid(struct object_id *oid UNUSED,
 	BUG("trying to finalize unknown hash");
 }
 
+static const struct git_hash_algo sha1_unsafe_algo = {
+	.name = "sha1",
+	.format_id = GIT_SHA1_FORMAT_ID,
+	.rawsz = GIT_SHA1_RAWSZ,
+	.hexsz = GIT_SHA1_HEXSZ,
+	.blksz = GIT_SHA1_BLKSZ,
+	.init_fn = git_hash_sha1_init_unsafe,
+	.clone_fn = git_hash_sha1_clone_unsafe,
+	.update_fn = git_hash_sha1_update_unsafe,
+	.final_fn = git_hash_sha1_final_unsafe,
+	.final_oid_fn = git_hash_sha1_final_oid_unsafe,
+	.empty_tree = &empty_tree_oid,
+	.empty_blob = &empty_blob_oid,
+	.null_oid = &null_oid_sha1,
+};
+
 const struct git_hash_algo hash_algos[GIT_HASH_NALGOS] = {
 	{
 		.name = NULL,
@@ -241,6 +257,7 @@  const struct git_hash_algo hash_algos[GIT_HASH_NALGOS] = {
 		.unsafe_update_fn = git_hash_sha1_update_unsafe,
 		.unsafe_final_fn = git_hash_sha1_final_unsafe,
 		.unsafe_final_oid_fn = git_hash_sha1_final_oid_unsafe,
+		.unsafe = &sha1_unsafe_algo,
 		.empty_tree = &empty_tree_oid,
 		.empty_blob = &empty_blob_oid,
 		.null_oid = &null_oid_sha1,
@@ -307,6 +324,15 @@  int hash_algo_by_length(int len)
 	return GIT_HASH_UNKNOWN;
 }
 
+const struct git_hash_algo *unsafe_hash_algo(const struct git_hash_algo *algop)
+{
+	/* If we have a faster "unsafe" implementation, use that. */
+	if (algop->unsafe)
+		return algop->unsafe;
+	/* Otherwise use the default one. */
+	return algop;
+}
+
 /*
  * This is meant to hold a *small* number of objects that you would
  * want repo_read_object_file() to be able to return, but yet you do not want