Hello Paul Fulghum:
it seems you are very busy,
and can not get your reply for "checking length in function rx_get_buf".
so I suggest:
design:
to give it additional length checking in function rx_get_buf.
if realy > max_frame_size, will return false (also need call free_rbufs).
unit test plan:
write a simple driver to integrate relative functions (already modified as design).
pass compiling, and can succeed loading.
and then
call the relative function (rx_get_buf) with intended values
check the work flow of our modification whether as expected.
at last, I still suggestion Paul Fulghum to provide the patch when he has free time.
for synclink, he is more expert than me.
for synclink, he can give better test.
if still get no reply within an additonal week:
I should provide the patch, it is my duty.
I will be according to the patch which Paul Fulghum has already provided.
and then add the "checking length in function rx_get_buf".
Regards
gchen.
于 2012年12月05日 09:57, Chen Gang 写道:
> by the way:
> does it also need check the length in function rx_get_buf ?
> (it seems not, but I am not quite sure, can you give a confirm ?)
>
> 4779 /*
> 4780 * pass receive buffer (RAW synchronous mode) to tty layer
> 4781 * return true if buffer available, otherwise false
> 4782 */
> 4783 static bool rx_get_buf(struct slgt_info *info)
> 4784 {
> 4785 unsigned int i = info->rbuf_current;
> 4786 unsigned int count;
> 4787
> 4788 if (!desc_complete(info->rbufs[i]))
> 4789 return false;
> 4790 count = desc_count(info->rbufs[i]);
> 4791 switch(info->params.mode) {
> 4792 case MGSL_MODE_MONOSYNC:
> 4793 case MGSL_MODE_BISYNC:
> 4794 case MGSL_MODE_XSYNC:
> 4795 /* ignore residue in byte synchronous modes */
> 4796 if (desc_residue(info->rbufs[i]))
> 4797 count--;
> 4798 break;
> 4799 }
> 4800 DBGDATA(info, info->rbufs[i].buf, count, "rx");
> 4801 DBGINFO(("rx_get_buf size=%d\n", count));
> 4802 if (count)
> 4803 ldisc_receive_buf(info->port.tty, info->rbufs[i].buf,
> 4804 info->flag_buf, count);
> 4805 free_rbufs(info, i, i);
> 4806 return true;
> 4807 }
--
Chen Gang
Asianux Corporation
--
To unsubscribe from this list: send the line "unsubscribe linux-serial" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
