From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH 2.6.11 4/8] tg3: Add msi test Date: Tue, 22 Mar 2005 18:50:36 -0500 Message-ID: <4240AF4C.9030109@pobox.com> References: <424098E2.6090007@pobox.com> <20050322141714.4515258f.davem@davemloft.net> <42409BA4.3090302@pobox.com> <20050322154030.310e85c1.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: mchan@broadcom.com, netdev@oss.sgi.com To: "David S. Miller" In-Reply-To: <20050322154030.310e85c1.davem@davemloft.net> Sender: netdev-bounce@oss.sgi.com Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org David S. Miller wrote: > Just because it's what we've been doing doesn't mean it's the > best solution in this case. > > If you hope users will just report this stuff, most of them won't. > They'll say Linux doesn't work on my computer and I have real life > stuff to take care of so I'll install freebsd or windows or a different > networking card and get on with life. > > Eventually, say over a year or so, the database of bad chipsets will > get built with your scheme, but to much hardship and pain to users. What do you think will happen for all the other MSI drivers/hardware out there? Adding this test to tg3 simply means another piece of hardware doesn't work. The problem disappears for -you-, but not the user. You're just pushing the pain around, for the -few- -early- systems with problems. Sorry, I still disagree. We need to think about how to handle this situation for -all- drivers and -all- users, not just tg3. Jeff