From: Chen Gang Subject: Re: [PATCH] synclink fix ldisc buffer argument Date: Wed, 19 Dec 2012 10:23:29 +0800 Message-ID: <50D12521.8010808@asianux.com> References: <50B6E751.9000000@asianux.com> <20121129051335.GA4375@kroah.com> <50B6F967.3050000@asianux.com> <20121129183207.GA4688@kroah.com> <50B81F76.8020508@asianux.com> <50B8DDAC.8070901@microgate.com> <50B90D0D.9040401@microgate.com> <20121202151332.3b6a6504@pyramind.ukuu.org.uk> <20121202181057.097012c6@pyramind.ukuu.org.uk> <989CB961-79F8-479B-B16C-41358A60AC94@microgate.com> <50BC0C84.4060802@asianux.com> <50BCCD4C.8050209@microgate.com> <50BEAA0E.9000603@asianux.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from intranet.asianux.com ([58.214.24.6]:23427 "EHLO intranet.asianux.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753636Ab2LSCWj (ORCPT ); Tue, 18 Dec 2012 21:22:39 -0500 In-Reply-To: <50BEAA0E.9000603@asianux.com> Sender: linux-serial-owner@vger.kernel.org List-Id: linux-serial@vger.kernel.org To: Paul Fulghum Cc: Alan Cox , Greg KH , Linux Kernel Mailing List , linux-serial@vger.kernel.org Hello Paul Fulghum: it seems you are very busy, and can not get your reply for "checking length in function rx_get_bu= f". 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_r= bufs). unit test plan: write a simple driver to integrate relative functions (already modi= fied 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 h= as 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 pro= vided. and then add the "checking length in function rx_get_buf". Regards gchen. =E4=BA=8E 2012=E5=B9=B412=E6=9C=8805=E6=97=A5 09:57, Chen Gang =E5=86=99= =E9=81=93: > by the way: > does it also need check the length in function rx_get_buf ?=20 > (it seems not, but I am not quite sure, can you give a confirm ?) >=20 > 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 =3D info->rbuf_current; > 4786 unsigned int count; > 4787=20 > 4788 if (!desc_complete(info->rbufs[i])) > 4789 return false; > 4790 count =3D 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=3D%d\n", count)); > 4802 if (count) > 4803 ldisc_receive_buf(info->port.tty, info->rbufs[i]= =2Ebuf, > 4804 info->flag_buf, count); > 4805 free_rbufs(info, i, i); > 4806 return true; > 4807 } --=20 Chen Gang Asianux Corporation -- To unsubscribe from this list: send the line "unsubscribe linux-serial"= in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html