When I try to boot the LiveCD on Dell PW 610 it detects my aic7xxx correctly, but then it continue to scan for aic7xxx_old and hangs. Reproducible: Always Steps to Reproduce: 1. Boot the LiveCD Actual Results: It hangs. Expected Results: Not hang.
This refers to bug #12185 which was previously closed, but that problem now exists in the latest LiveCD 1.4rc4. I have a scsi cd on an Adaptec AHA-2940 UW (aic7880) with the latest firmware that will not boot the LiveCD because it hangs while detecting the aic7xxx_old driver and after successfully detecting the aic7xxx driver. Any chance that the aic7xxx_old driver may be removed from the LiveCD?
Preliminary fix: Have the LiveCD boot with noscsi, to get it in "Oye, no drives, here's a terminal, *you* figure it out", modprobe aic7xxx, Ctrl+D. This one has bitten me too. Removing the aic7xxx_old driver wouldn't hurt, I can't think of a configuration that works with aic7xxx_old but fails with aic7xxx.
I've got the same on a home made Dual Athlon system. I agree that the aic7xxx_old module should be removed as it appears to be causing a fair few problems for people using hardware supported by the aic7xxx module.
What's really interesting is that, while the aic7xxx_old driver causes the described problem in the LiveCD, one has to use it when one builds a production kernel, as the "new" aic7xxx driver will randomly hard-lock SMP systems. See http://forums.gentoo.org/viewtopic.php?t=91175. I can confirm both problems: I have a Dell Precision 530 (dual Xeon 1.7Ghz). The aic7xxx_old driver causes the LiveCD to hang. However, after getting the entire system setup and building the "new" aic7xxx driver into the gentoo-sources kernel, the system would randomly hard-lock. Using the aic7xxx_old driver in my SMP kernel resolves that problem.
I got a Dell PowerEdge 4300 with 2 CPUS (500MHz), it also hang up with 1.4 CD.
I can't duplicate the SMP hangup of the aic7xxx driver here, though. Not on an NCR WorldMark with two AIC7880U controllers and not on a dual Athlon MP 1800+ with one 29160 controller either.
This problem also happened when I booted into my freshly generated kernel. During initrd processing, both the aic7xxx and aic7xxx_old modules get loaded, and the latter gets hung retrying on a bus timeout. I had to remove aic7xxx_old from the linuxrc script, now the system boots ok. - My hardware IS old: the 7870 is about 10 years old, but obviously too new for that module...
I was able to resolve this by running gentoo noscsi cd /modules/storage modprobe ./aic7xxx.o It resolved it for me
This bug has been open for a while now, might as well RESO WONT and shove the 2.6 LiveCD's forward? If 2.4 LiveCD remain the standard for another few months, this does seem worthy of getting handled....
has this been tried w/ the new 2004.0 livecd using 2.6?
The old driver will be no longer supported
