Re: [PATCH 1/1] untracked cache: fix off-by-one

From: Jeff King <peff@peff.net>
Date: Apr 10, 2019, 16:20 UTC
Message-ID: <20190410162029.GA30818@sigill.intra.peff.net>
In-Reply-To: <a7f75cefb682546862be4dd8b48f91c4034c5d5a.1554901006.git.gitgitgadget@gmail.com>

On Wed, Apr 10, 2019 at 05:56:48AM -0700, Johannes Schindelin via GitGitGadget wrote:

> Probably in the endeavor to avoid the `calloc()` implied by
> `FLEX_ALLOC_STR()` (it is hard to know why exactly, the commit message
> of that commit is a bit parsimonious with information), it calls
> `malloc()` manually and then `memcpy()`s the bits and pieces into place.

You have a talent for understatement. :)

Show 12 quoted lines

> It allocates the size of `struct untracked_cache_dir` plus the string
> length of the untracked file name, then copies the information in two
> steps: first the fixed-size metadata, then the name. And here lies the
> rub: it includes the trailing NUL byte in the name.
> 
> If `FLEX_ARRAY` is defined as 0, this results in a buffer overrun.
> 
> To fix this, let's just add 1, for the trailing NUL byte. Technically,
> this overallocates on platforms where `FLEX_ARRAY` is 1, but it should
> not matter much in reality, as `malloc()` usually overallocates anyway,
> unless the size to allocate aligns exactly with some internal chunk size
> (see below for more on that).

Yeah, every struct on a platform where FLEX_ARRAY is 1 ends up over-allocated by 1 byte. We could account for that in all of our flex allocations, but I don't it affects enough platforms to be worth the bother.

> The real strange thing is that neither valgrind nor DrMemory catches
> this bug. In this developer's tests, a `memcpy()` (but not a
> `memset()`!) could write up to 4 bytes after the allocated memory range
> before valgrind would start reporting an issue.

I couldn't get it to trigger with ASan, either. I assume it's due to alignment (i.e., those tools are stuck poisoning at the end of a page, but the start of the struct needs to be page-aligned). But that would imply that we could trigger it with different path lengths, which I wasn't able to do. So I dunno.

Show 15 quoted lines

> However, when running Git built with nedmalloc as allocator, under rare
> conditions (and inconsistently at that), this bug triggered an `abort()`
> because nedmalloc rounds up the size to be `malloc()`ed to a multiple of
> a certain chunk size, then adds a few bytes to be used for storing some
> internal state. If there is no rounding up to do (because the size is
> already a multiple of that chunk size), and if the buffer is overrun as
> in the code patched in this commit, the internal state is corrupted.
>
> The scenario that triggered this here bug fix entailed a git.git
> checkout with an extra copy of the source code in an untracked
> subdirectory, meaning that there was an untracked subdirectory called
> "thunderbird-patch-inline" whose name's length is exactly 24 bytes,
> which, added to the size of above-mentioned `struct untracked_cache_dir`
> that weighs in with 104 bytes on a 64-bit system, amounts to 128,
> aligning perfectly with nedmalloc's chunk size.

Right, that makes sense that the length would impact it there.

> As there is no obvious way to trigger this bug reliably, on all
> platforms supported by Git, and as the bug is obvious enough, this patch
> comes without a regression test.

Makes sense. This code path should be well-covered by the existing tests anyway, so even if we could get those tools to trigger, I don't think there would be much point in adding a new test.

Show 12 quoted lines

> diff --git a/dir.c b/dir.c
> index b2cabadf25..f5293a6536 100644
> --- a/dir.c
> +++ b/dir.c
> @@ -2760,7 +2760,7 @@ static int read_one_dir(struct untracked_cache_dir **untracked_,
>  	next = data + len + 1;
>  	if (next > rd->end)
>  		return -1;
> -	*untracked_ = untracked = xmalloc(st_add(sizeof(*untracked), len));
> +	*untracked_ = untracked = xmalloc(st_add3(sizeof(*untracked), len, 1));
>  	memcpy(untracked, &ud, sizeof(ud));
>  	memcpy(untracked->name, data, len + 1);

This is obviously correct, even just from the context.

IMHO it is worth it in cases like this to just use FLEX_ALLOC for simplicity; calloc() is not too expensive. However, there's an interesting subtlety there. Our length comes from this line just above your hunk:

  len = strlen((const char *)data);

how do we know that data actually contains a NUL? It's ultimately pointing to our mmap of the index file. So I think a malformed index would potentially cause us to go off the end of the array and read arbitrary memory.

The right thing is probably something like:

  eos = memchr(data, '\0', end - data);
  if (!eos)
	return error("malformed untracked cache extension");
  len = eos - data;

I wouldn't be at all surprised if other bits of the index code have the same issue, though. And at any rate, thinking about that should definitely not hold up your fix.

-Peff