Re: AHCI driver progress report - hot-swap PM and parallel probe

看板DFBSD_kernel作者時間16年前 (2009/06/14 12:32), 編輯推噓0(000)
留言0則, 0人參與, 最新討論串5/11 (看更多)
On 13/06/2009, Justin C. Sherrill <justin@shiningsilence.com> wrote: > On Sat, June 13, 2009 8:13 pm, Matthew Dillon wrote: > > > How about a kernel environment variable that can be set from the > > bootloader to disable the driver? > > > > Then a person can upgrade and reboot and if they made a mistake > > its a simple 'set' command at the loader prompt and they're up and > > running again. > > > That'll work - perhaps default to the non-AHCI behavior for 2.4, and then > change at 2.6? My pet issue here is having an upgrade scramble - or > appear to scramble - someone's machine. Our upgrades have been remarkably > painless, and I'd hate to lose that pattern. That's a bad idea. The longer you wait, the greater will be the install base that is using AHCI hardware with non-AHCI drivers, and the more people will have problems with upgrades in the future. As an example, the switchover in OpenBSD, back in the day, was done so early that, from my understanding, practically noone has even had to deal with the renaming issue. http://www.openbsd.org/faq/upgrade42.html#ahci br, cnst.su.
文章代碼(AID): #1AD7r600 (DFBSD_kernel)
討論串 (同標題文章)
完整討論串 (本文為第 5 之 11 篇):
文章代碼(AID): #1AD7r600 (DFBSD_kernel)