Manuel, Increasing the delay of the "ahciprb" tsleep() to 500 seems to work. I have a kernel from February which did not need this delay, any idea why it has changed? From dmesg: ahcisata0 port 0: device present, speed: 1.5Gb/s cd0: 32-bit data port cd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 2 (Ultra/33) cd0(piixide0:0:0): using PIO mode 4, Ultra-DMA mode 2 (Ultra/33) (using DMA) ahcisata0 port 2: PHY offline ahcisata0 port 1: PHY offline is it okay that ahcisata0 port 2 appears before 1? This does not always happen. Thanks. Manuel Bouyer wrote:
On Wed, May 14, 2008 at 05:04:16PM +0200, Yorick Hardy wrote:Manuel, I did a fresh checkout of the sources and tried again. This time ahcisata0 port 2 appears before port 1 in the output, and unfortunately the problem persists. Should the ahcisata0 ports appear in numerical order?Could you give a copy of the outpout ? I'm not sure what you mean. Also you could try to increase the delay of the "ahciprb" tsleep() in ahcisata_core.c
-- Kind regards, Yorick Hardy