Date: Wed, 31 Jul 2013 16:28:17 +0800 From: XiaoQI Ge <ghw@7axu.com> To: Ian Lepore <ian@freebsd.org> Cc: John-Mark Gurney <jmg@funkthat.com>, "freebsd-wireless@freebsd.org" <freebsd-wireless@freebsd.org>, freebsd-arm <freebsd-arm@freebsd.org>, Hans Petter Selasky <hans.petter.selasky@bitfrost.no> Subject: Re: My WLI-UC-GNM up crash Message-ID: <CAKrd9eVzY1GX71G88jk0icivoYNR5Tgx3QewqFdvC_jPi-sMnA@mail.gmail.com> In-Reply-To: <1375151095.45247.54.camel@revolution.hippie.lan> References: <cakrd9exvuqeqz=3ntmh7k3e4_6fjw9jph_tpfchujhqhwwbzsg@mail.gmail.com> <zarafa.51f6ad45.7961.536131947b8b1205@mail.lockless.no> <20130729211540.GZ26412@funkthat.com> <1375151095.45247.54.camel@revolution.hippie.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
Last night, I use the latest FreeBSD source (r253827) and crochet-freebsd compile img My WLI-UC-GNM will be disconnected after a period of operation run0: device timeout run0: at uhub0, port 1, addr 2 (disconnected) Another log prompted ti_mmchs0: Error: current cmd NULL, already done? My BB-Black broken? -- Regards. By: XiaoQI Ge; PGP:8B09D5F7 WWW: https://www.7axu.com/ 2013/7/30 Ian Lepore <ian@freebsd.org>: > On Mon, 2013-07-29 at 14:15 -0700, John-Mark Gurney wrote: >> Hans Petter Selasky wrote this message on Mon, Jul 29, 2013 at 19:58 +02= 00: >> > The aligned will make sure that the structure gets padded properly to = the size specified. Only on ARM/MIPS etc, structures get automatically alig= ned according to the element in the structure requiring the greatest alignm= ent. I've test-compiled the USB WLAN drivers, and the aligned makes a diffe= rence. The problem is that the radiotap header skews some following element= s, so that they are no longer aligned. The radiotap header itself is packed= , and this is not a problem. >> >> Ouch, has anyone looked at the code that caused this? >> >> in ieee80211_radiotap.c, it looks like the original fault was in either >> set_channel, or set_xchannel, and both do (the equivalent of): >> struct { >> uint16_t freq; >> uint16_t flags; >> } *rc =3D p; >> >> rc->freq =3D htole16(c->ic_freq); >> rc->flags =3D htole16(c->ic_flags); >> > > If there's any chance the pointer isn't aligned in code like this, then > htole16() is the wrong function, it should be using le16enc() such as > > le16enc(&rc->freq, c->ic_freq); > le16enc(&rc->flags, c->ic_flags); > > With any luck, an x86 compiler can optimize the inline code for the > endian enc/dec functions to take advantage of the platform's ability to > do unaligned accesses. But that's a side issue to code correctness -- > portable code has to get these things right even when it's inefficient. > > -- Ian > >> And then there is complicated code that calculates offsets, etc, in >> radiotap_offset.. What we probably really need is to mark the above >> as __packed or equiv so that we don't assume that the passed in pointer >> is aligned... >> >> The whole management of this radiochan and radiotap_offset is pretty >> nasty... If marking the structures __packed works, it's probably >> because two bugs are offsetting, and magicly making things align >> again... >> >> > -----Original message----- >> > > From:Warner Losh <imp@bsdimp.com <mailto:imp@bsdimp.com> > >> > > Sent: Monday 29th July 2013 17:04 >> > > To: Adrian Chadd <adrian@freebsd.org <mailto:adrian@freebsd.org> > >> > > Cc: Hans Petter Selasky <hans.petter.selasky@bitfrost.no <mailto:han= s.petter.selasky@bitfrost.no> >; freebsd-arm <freebsd-arm@freebsd.org <mail= to:freebsd-arm@freebsd.org> >; freebsd-wireless@freebsd.org <mailto:freebsd= -wireless@freebsd.org> >> > > Subject: Re: My WLI-UC-GNM up crash >> > > >> > > Aren't structures already aligned to 4 bytes when placed inside othe= r structures (unless marked __packed)? >> > > >> > > Warner >> > > >> > > On Jul 28, 2013, at 11:50 AM, Adrian Chadd wrote: >> > > >> > > > As long as that results in the radiotap structures being 4 or 8 by= te >> > > > padded when it's embedded in the softc - then yes, indeed. >> > > > >> > > > Xiao, can you try? >> > > > >> > > > >> > > > -adrian >> > > > >> > > > On 28 July 2013 03:35, Hans Petter Selasky <hps@bitfrost.no <mailt= o:hps@bitfrost.no> > wrote: >> > > >> Hi, >> > > >> >> > > >> Can you try the attached patch? >> > > >> >> > > >> --HPS >> > > > _______________________________________________ >> > > > freebsd-arm@freebsd.org <mailto:freebsd-arm@freebsd.org> mailing = list >> > > > http://lists.freebsd.org/mailman/listinfo/freebsd-arm <http://list= s.freebsd.org/mailman/listinfo/freebsd-arm> >> > > > To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.= org <mailto:freebsd-arm-unsubscribe@freebsd.org> " >> > > >> > > >> > >> > _______________________________________________ >> > freebsd-arm@freebsd.org mailing list >> > http://lists.freebsd.org/mailman/listinfo/freebsd-arm >> > To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org" >> > > > _______________________________________________ > freebsd-arm@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-arm > To unsubscribe, send any mail to "freebsd-arm-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAKrd9eVzY1GX71G88jk0icivoYNR5Tgx3QewqFdvC_jPi-sMnA>