qemu-trivial
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-trivial] [Qemu-block] [PATCH] dmg: fix binary search


From: Julio Faracco
Subject: Re: [Qemu-trivial] [Qemu-block] [PATCH] dmg: fix binary search
Date: Wed, 2 Jan 2019 12:19:12 -0200

Hi Stefan,

I created a simple DMG file from MacOS to reproduce the problem:
https://bugs.launchpad.net/qemu/+bug/1809304

Em qua, 2 de jan de 2019 às 08:47, Stefan Hajnoczi <address@hidden> escreveu:
On Fri, Dec 21, 2018 at 09:58:03PM +0800, yuchenlin wrote:
> There is a possible hang in original binary searsh implemtation. That is
> if chunk1 = 4, chunk2 = 5, chunk3 = 4, and we go else case.
>
> The chunk1 will be still 4, and so on.
>
> Signed-off-by: yuchenlin <address@hidden>
> ---
>  block/dmg.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)

Do you have an example dmg file that triggers this bug?

> diff --git a/block/dmg.c b/block/dmg.c
> index 50e91aef6d..0e05702f5d 100644
> --- a/block/dmg.c
> +++ b/block/dmg.c
> @@ -572,14 +572,14 @@ static inline uint32_t search_chunk(BDRVDMGState *s, uint64_t sector_num)
>  {
>      /* binary search */
>      uint32_t chunk1 = 0, chunk2 = s->n_chunks, chunk3;
> -    while (chunk1 != chunk2) {
> +    while (chunk1 <= chunk2) {
>          chunk3 = (chunk1 + chunk2) / 2;
>          if (s->sectors[chunk3] > sector_num) {
> -            chunk2 = chunk3;
> +            chunk2 = chunk3 - 1;

What if chunk1 = 0 and chunk2 = 1?  chunk3 = 0 and the new chunk2 value
would be 0xffffffff so the loop continues with out-of-bounds sectors[]
accesses!

reply via email to

[Prev in Thread] Current Thread [Next in Thread]