Kernel 2.6.14-2 hangs on iMac G5 8,2 (ata1 timeout)
I finally found out what happens when the system hangs, using netconsole. I get this error message twice before the system dies: ata1: command 0x25 timeout, stat 0x50 host_stat 0x4 It seems to happen faster when I work under X, then it is reproducible, I open Thunderbird, type my password and it hangs. With the console I've had uptimes of some hours, it's harder to trigger it. The problem is also described here [1], although it's not exactly the same. Find attached the kernel.log for that session. Eduardo. PS: As Ben suggested I added an entry for PowerMac8,2 in arch/ppc64/kernel/pmac_feature.c. [1] http://bugzilla.ubuntu.com/show_bug.cgi?id=12904 kern.log.bz2 Description: application/bzip2
Re: How to compile custom kernel 2.6 for ppc64
On Wed, Dec 14, 2005 at 09:08:26PM +0100, Wolfgang Pfeiffer wrote: On Wed, Dec 14, 2005 at 11:03:21AM +0200, Eduardo Trápani wrote: I've tried everything I could think of, apt-get source --build, make-kpk, debian/rules binary, dpkg-buildpackage -b -uc but either the kernel does not boot or [ ... ] you might want to search for exactly these topics between April 8 through 15 this year, where compile questions were discussed in great detail, partly ad nauseam: ... no, not ad nauseam, but one could think it came *close* to that ... :) And please come back if the links don't help ... Best Regards Wolfgang -- Wolfgang Pfeiffer: /ICQ: 286585973/ + + + /AIM: crashinglinux/ http://profiles.yahoo.com/wolfgangpfeiffer Key ID: E3037113 http://keyserver.mine.nu/pks/lookup?search=0xE3037113fingerprint=on -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: stock kernel 2.6.12 eth0/eth1 swapped
On 2005-12-11 10:18:58 +0900, Charles Plessy wrote: just to say that this swapping can be very difficult to diagnose to the basic user. Coming from the i386 world, I would only expect this to happen with two similar interfaces. I installed a debian (testing) on my iMac G5 yesterday (only one eth physical inteface), and if this thread had not been started, I would still be trying to figure out why the eth0 used by debian-installer was unable to find its IP adress by dhcp after rebooting in the new system... Not only the installer... When I start pump from the command line, it used to work reliably, but no longer seems to work without a configuration file. -- Vincent Lefèvre [EMAIL PROTECTED] - Web: http://www.vinc17.org/ 100% accessible validated (X)HTML - Blog: http://www.vinc17.org/blog/ Work: CR INRIA - computer arithmetic / SPACES project at LORIA -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Ibook G4 freezes
Le Mercredi 14 Décembre 2005 1:mM:sS AM, Benjamin Herrenschmidt a écrit : DRI is disabled, even when the radeon module is not loaded, it freezes. From xorg log : (II) RADEON(0): Direct rendering disabled Complete xorg log at the end of this mail. Ok, two other things you can try: - Disable cpufreq, make sure it's completely disabled (ie, the kernel option too, rebuild your kernel if necessary) and tell us if it makes a difference. I rebuilt my kernel without cpufreq, and still it freezes. - Do you have xmon enabled ? if not, enable it (CONFIG_XMON). First check that it works, that is when pressing command-power you enter xmon (you see the mon prompt, exit with x return). xmon is a low level kernel debugger. Also, make sure X uses UseFBDev so that xmon can reliably access the screen even while running X. Now, try to reproduce the freeze. Does it enter xmon at the point of the freeze ? If not, try pressing command-power when it's frozen. Do you manage to get into xmon that way ? If yes to any of these, then, copy what xmon says. Then use xmon t command to get a backtrace and copy that too and send it to me. I enabled xmon, but it didn't enter xmon. While frozen, command-power has no effect. - If none of these works, can you try reproducing the problem in console mode instead of X ? When it happens, do you see anything in the console ? Oops message, kernel panic, whatever ? Thanks, Ben. I have tried to get my laptop crashing in console mode, but I have not succeeded. (Did a 16 hours stress test, with an average load of 28, heavy reads from hard drive, running out of memory, swapping, net overload, several suspends to ram : it didn't freeze) I will keep on trying to crash it in console mode. Kevin
Re: Re: Problem with Bull Escala
I tried it with all vmlinuz-prep.initrd but it hasn't worked so far. I'm sure it has no MCA Subsystem but a PCI. I have also heard that the original AIX Version 4.3.3 won't work with the Bull even it is a RS/6000 clone. It won't boot from the AIX CD's. Maybe this is a hint that there are some differences between these systems. I have also tried the powerpc Mac port because of the openfirmware of this machine. I also tried the netbooting option from the openfirmware but without a result. I had a problem with the networkinterface of Bull.
suggest for installing powerbook
hi all, I found in internet some info about old powerbook and old kernels; with a new powerbook G4 1,67GHz what suggest me to do? installing debian unstable? compile my own kernel with a particular patch? installing ubuntu? I also install macOSX in other partiton, and I do it before installing linux. thanks for reply -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Kernel 2.6.14-2 hangs on iMac G5 8,2 (ata1 timeout)
On Thu, 2005-12-15 at 10:29 +0200, Eduardo Trápani wrote: I finally found out what happens when the system hangs, using netconsole. I get this error message twice before the system dies: ata1: command 0x25 timeout, stat 0x50 host_stat 0x4 It seems to happen faster when I work under X, then it is reproducible, I open Thunderbird, type my password and it hangs. With the console I've had uptimes of some hours, it's harder to trigger it. The problem is also described here [1], although it's not exactly the same. Find attached the kernel.log for that session. Does it happen if you disable the sound driver ? (prevent it from loading or don't compile it at all if you are building your own kernels). Ben. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Ibook G4 freezes
On Thu, 2005-12-15 at 14:54 +0100, Kevin ALLIX wrote:
I have tried to get my laptop crashing in console mode, but I have not
succeeded.
(Did a 16 hours stress test, with an average load of 28, heavy reads from
hard
drive, running out of memory, swapping, net overload, several suspends to
ram : it didn't freeze)
I will keep on trying to crash it in console mode.
Ok. A couple of other things coming to mind:
- In xorg.conf, section Device, try Option DynamicClocks
false (it will increase power consumption of the video chip though).
Check in X log that it indeed disables the dynamic clocks.
- Try the attached kernel patch:
Index: linux-work/arch/powerpc/platforms/powermac/feature.c
===
--- linux-work.orig/arch/powerpc/platforms/powermac/feature.c 2005-11-19
10:56:07.0 +1100
+++ linux-work/arch/powerpc/platforms/powermac/feature.c2005-12-12
12:17:02.0 +1100
@@ -1650,11 +1650,19 @@
*/
if (macio-type == macio_intrepid) {
- if (enable)
- UN_OUT(UNI_N_CLOCK_SPREADING, 2);
- else
- UN_OUT(UNI_N_CLOCK_SPREADING, 0);
- mdelay(40);
+ struct device_node *clock =
+ of_find_node_by_path(/[EMAIL PROTECTED]/hw-clock);
+ if (clock get_property(clock, platform-do-clockspreading,
+ NULL)) {
+ printk(KERN_INFO %sabling clock spreading on Intrepid
+ ASIC\n, enable ? En : Dis);
+ if (enable)
+ UN_OUT(UNI_N_CLOCK_SPREADING, 2);
+ else
+ UN_OUT(UNI_N_CLOCK_SPREADING, 0);
+ mdelay(40);
+ }
+ of_node_put(clock);
}
while (machine_is_compatible(PowerBook5,2) ||
@@ -1724,6 +1732,9 @@
pmac_low_i2c_close(ui2c);
break;
}
+ printk(KERN_INFO %sabling clock spreading on i2c clock chip\n,
+ enable ? En : Dis);
+
pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub);
rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80,
buffer, 9);
DBG(write result: %d,, rc);
Index: linux-work/arch/ppc/platforms/pmac_feature.c
===
--- linux-work.orig/arch/ppc/platforms/pmac_feature.c 2005-11-14
20:32:16.0 +1100
+++ linux-work/arch/ppc/platforms/pmac_feature.c2005-12-12
12:16:38.0 +1100
@@ -1606,11 +1606,19 @@
*/
if (macio-type == macio_intrepid) {
- if (enable)
- UN_OUT(UNI_N_CLOCK_SPREADING, 2);
- else
- UN_OUT(UNI_N_CLOCK_SPREADING, 0);
- mdelay(40);
+ struct device_node *clock =
+ of_find_node_by_path(/[EMAIL PROTECTED]/hw-clock);
+ if (clock get_property(clock, platform-do-clockspreading,
+ NULL)) {
+ printk(KERN_INFO %sabling clock spreading on Intrepid
+ ASIC\n, enable ? En : Dis);
+ if (enable)
+ UN_OUT(UNI_N_CLOCK_SPREADING, 2);
+ else
+ UN_OUT(UNI_N_CLOCK_SPREADING, 0);
+ mdelay(40);
+ }
+ of_node_put(clock);
}
while (machine_is_compatible(PowerBook5,2) ||
@@ -1680,6 +1688,8 @@
pmac_low_i2c_close(ui2c);
break;
}
+ printk(KERN_INFO %sabling clock spreading on i2c clock chip\n,
+ enable ? En : Dis);
pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub);
rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80,
buffer, 9);
DBG(write result: %d,, rc);
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
