On Monday 21 November 2016 23:10:25 胡聪 wrote:
> Hi,
>
> On Mon, Nov 21, 2016 at 10:26 PM +0300, Maxim Dounin wrote:
>
> >Ah, ok, I see the problem now. Please clarify things in the
> >commit log as well.
> >
> >It also make sense to use sizeof() explicitly, instead of
> >introducing a magic number, e.g.:
>
> Thanks！There is the new patch.
>
> # HG changeset patch
> # User hucongcong <hucon...@foxmail.com>
> # Date 1479740605 -28800
> # Mon Nov 21 23:03:25 2016 +0800
> # Node ID 7a1a6f7b989d4c44523333293265d58663ac7ff4
> # Parent 2c7a2d75938a31044552b0a6cd6edaebdaf0bd69
> Mp4: fixed setting wrong mdat atom size in very rare cases.
>
> Atom size is the sum of atom header size and atom data size. The
> specification says that the first 4 bytes are set to one when
> the atom size is greater than the maximum unsigned 32-bit value.
> Which means atom header size should be considered when the
> comparison takes place between atom data size and 0xffffffff.
>
> diff -r 2c7a2d75938a -r 7a1a6f7b989d src/http/modules/ngx_http_mp4_module.c
> --- a/src/http/modules/ngx_http_mp4_module.c Mon Nov 21 16:49:19 2016 +0300
> +++ b/src/http/modules/ngx_http_mp4_module.c Mon Nov 21 23:03:25 2016 +0800
> @@ -1229,7 +1229,9 @@ ngx_http_mp4_update_mdat_atom(ngx_http_m
>
> atom_header = mp4->mdat_atom_header;
>
> - if ((uint64_t) atom_data_size > (uint64_t) 0xffffffff) {
> + if ((uint64_t) atom_data_size + sizeof(ngx_mp4_atom_header_t)
> + > (uint64_t) 0xffffffff)
> + {
> atom_size = 1;
> atom_header_size = sizeof(ngx_mp4_atom_header64_t);
> ngx_mp4_set_64value(atom_header + sizeof(ngx_mp4_atom_header_t),
Maxim wrote the expression the way that almost all compilers should optimize
it to comparison with one static constant. I'm not sure about your case.
wbr, Valentin V. Bartenev
_______________________________________________
nginx-devel mailing list
nginx-devel@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx-devel
