On 18 August 2014 09:00, zhanghailiang <zhang.zhanghaili...@huawei.com> wrote:
> From: Li Liu <john.li...@huawei.com>
>
> fopen() may return NULL which will cause setbuf() segmentfault
>
> Signed-off-by: zhanghailiang <zhang.zhanghaili...@huawei.com>
> Signed-off-by: Li Liu <john.li...@huawei.com>
> ---
> block/vvfat.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/block/vvfat.c b/block/vvfat.c
> index 70176b1..62023e1 100644
> --- a/block/vvfat.c
> +++ b/block/vvfat.c
> @@ -1084,6 +1084,7 @@ static int vvfat_open(BlockDriverState *bs, QDict
> *options, int flags,
>
> DLOG(if (stderr == NULL) {
> stderr = fopen("vvfat.log", "a");
> + assert(stderr);
> setbuf(stderr, NULL);
> })
An assertion is no better than a segfault.
Better I think would be to just remove this whole lump of code
entirely. Lots of other files do debug printing to stderr without
attempting to open a file if stderr happens to be NULL, why
should vvfat.c be special?
-- PMM
