Re: [PATCH v2 07/12] packfile, git-zlib: widen `use_pack()` and zstream avail fields to `size_t`
- From
Junio C Hamano <gitster@pobox.com>
- Date
- Aug 7, 2026, 21:41 UTC
- Message-ID
- <xmqqjyq1d2x7.fsf@gitster.g>
- In-Reply-To
- <ca928b457959ab8bfa643c65f83ca1ec4289fdd3.1785946479.git.gitgitgadget@gmail.com>
"Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com> writes:
Show 15 quoted lines
> diff --git a/git-zlib.h b/git-zlib.h
> index 44380e8ad3..0b24b15bd0 100644
> --- a/git-zlib.h
> +++ b/git-zlib.h
> @@ -5,8 +5,8 @@
>
> typedef struct git_zstream {
> struct z_stream_s z;
> - unsigned long avail_in;
> - unsigned long avail_out;
> + size_t avail_in;
> + size_t avail_out;
> size_t total_in;
> size_t total_out;
> unsigned char *next_in;We have these size_t which means we can use a buffer larger than 4GB where size_t is larger than 32-bit ulong. But these are sizes of a single contiguous buffer, so I think that is why the log message mentioned that this is more of type consistency than being able to handle larger data (I do not think people feed >4GB contiguious buffer in one go in practice).
- zlib_buf_cap() is still "unsigned long", and zlib_pre_call() feeds these potentially wider values to it. Is it possible that we trigger truncation before the avail_in/avail_out is compared with ZLIB_BUF_MAX in the zlib_buf_cap() function?
- unpack_object_header_buffer() still takes "unsigned long" length; builtin/pack-objects.c:oe_get_size_slow() passes size_t avail to unpack_object_header_buffer(). This comes from use_pack(), so it is a relatively small value stored in wider size_t but I am unsure if your static checker would not flag for potential truncation?