Date: Mon, 27 Oct 2003 11:28:56 +0100 From: Marius Strobl <marius@alchemy.franken.de> To: "Sergey V. Belov" <bsv@csite.ru> Cc: freebsd-hackers@freebsd.org Subject: Re: Intel PILA 8460B LAN adapter Message-ID: <20031027112856.A72984@newtrinity.zeist.de> In-Reply-To: <20031027134338.6dc9bca9.bsv@csite.ru>; from bsv@csite.ru on Mon, Oct 27, 2003 at 01:43:38PM %2B0400 References: <20031027010925.661d3edd.bsv@csite.ru> <20031027075327.A68883@newtrinity.zeist.de> <20031027134338.6dc9bca9.bsv@csite.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Oct 27, 2003 at 01:43:38PM +0400, Sergey V. Belov wrote: > On Mon, 27 Oct 2003 07:53:27 +0100 > Marius Strobl <marius@alchemy.franken.de> wrote: > > > On Mon, Oct 27, 2003 at 01:09:25AM +0400, Sergey V. Belov wrote: > > > Hello all, > > > I had an experience of install this LAN adapter on latest FreeBSD RELENG_4_8 (4.8p13): > > > PILA 8460B \ EtherExpress PRO/100+ Server Adapter (PCI; TPE; 100 / 10 Mbps). > > > After install, this card has not been responsed; but with following patch card is worked > > > perfectly over three weeks. > > > Comments? > > > Finally, I should be glad to see this changes in the main source tree. > > > > > > > Apart from that patch being wrong, support for this card was added to > > RELENG_4 on June 12. If you update to 4-stable or upcoming 4.9-release > > you will get it. > > Thanks for the reply, but why that change is not in RELENG_4_8 yet? Because only critical fixes, e.g. security fixes, go into RELENG_4_8 (and the other supported RELENG_4_* "security branches") and not new features, e.g. support for new hardware. > And than, what's wrong with the patch that give me working card? I'm not a "kernel hacker", > so more detailed comments are welcome. > It enables workarounds for fxp(4) controllers with chipid 0x1039 to 0x104F which are not intended for these chips and could have some negative impact. In any case this change is irrelevant for supporting fxp(4) controllers with chipid 0x0150. According to the change that was commited to if_fxp.c in order to support 0x0150 adding this id to the fxp_ident_table apparently is enough. Your patch also does this so that's why fxp(4) attaches to your card, but your patch breaks the sorting of the chipids/devids in fxp_ident_table.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20031027112856.A72984>