efficient
What's "interesting for Sven" isn't really a consideration, what's reasonable and efficient is. Once this has been tried for some time, I'll be happy to reconsider whether it's effective; if it's going to be dismissed out of hand, though, that indicates a lack of good faith. I am sorry I had to interject here. The first definition of efficient is "Acting directly to produce an effect" <-- http://www.thefreedictionary.com/efficient Making Sven jump through hoops to update code that he had or has to work in is anything but direct or efficient. Another definition under the Thesaurus part of the above page is "being effective without wasting time or effort or expense Now I am not an expert on what happened, and don't claim to be. If you don't like him and don't want him, say that, if he screwed up, then it may be a competancy problem, and say that, but a group should not hide behind an argument of reasonability and efficiency when the whole basis of the argument is inefficient --Mike -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
woody, PB3400
g'day, i've got woody running on my pb3400, and so far the only thing that's given me any serious amount of hassle is XFree 4.0.3. I would be very grateful for a copy of another pb3400 user's working XF86Config-4 file, or any advice on the screen modes i need to use. X dies after finding no workable screens, my 800x600 entry is rejected with "bad mode clock/interlace/doublescan". Searching the archives, I've found references to taking the output of fbset -x, which I've done with no success... cheers mike
Re: woody, PB3400
On Thu, 30 Aug 2001, Michel [iso-8859-1] Dänzer wrote: > > You should no longer need to define modelines at all. When I remove them, all default modes between 640x350 and 800x600 are deleted (bad mode clock/interlace/doublescan), 1024x768 and up are killed due to insufficient memory for mode (not hard to figure that one out :)). I have no idea what to put for HorizSync and VertRefresh, I've tried some based on other powerbook tft display configs and also just didn't put these entries in the config file. in the end, it always fails with one of the two previous errors. The final error is "Screen(s) found, but none have a usable configuration". damn :) I've managed to answer all my other installation/config based questions just from searching around, this is the one problem that I find bugger all about on google (or anywhere else). if it's important to mention, i boot debian via quik and do not have a macos partition (or cd) to boot from. I can send my config file if that would help, aside from my addition of a DefaultDepth 16 line it's a vanilla configuration straight from XFree86 -configure. thanks again, mike
Re: woody, PB3400
hiya, i've found my problem to be the chips driver, switching to fbdev (using 15 bit colour instead of 16) has fixed it. thanks for everyone who replied, aside from the odd crash here and there (particularly with apt-get) everything works well! cheers mike On Fri, 31 Aug 2001, Michael Schmitz wrote: > > I have no idea what to put for HorizSync and VertRefresh, I've tried some > > VertRefresh 35-75 is needed for my Lombard. Anything higher than 35 and I > get an error (might even be the same error as you see). It's been a while > since I installed XFree 4.1 so your mileage may vary. > > Michael > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > >
Beige G3 and USB Mouse
Greetings, (For the time-strapped please skip down to the ***) I have a revision a Beige G3/233 that I have tried repeatedly to install Debian on. Ironically, I have had much more luck with my Power Mac 6500, granted it has no mouse, keyboard, or monitor. I have been able to successfully install LinuxPPC 2000 Q4, Mandrake 8.0, and Yellow Dog 2.0 on this machine, but found these distros to be sloppy, bloated, and quite overwhelming. After installing Debian on the 6500 (I use unstable and find it to be quite stable), I fell in love with apt and dpkg. Those tools, along with Mac OS X and Fink have really helped me learn my way around UNIX. Unfortunately, the lack of multiple kernel versions, as well as USB and video card configuration issues has prevented me from learning the finer points of UNIX. *** I am able to get Debian to install, but haven't had much luck getting it to a useable desktop state. Before I start on the questions, here is my basic system configuration: Rev. A Gossamer G3/233 (over-clocked to 300, but that shouldn't matter ;) 384 MB RAM 60 GB IDE HD OrangeMicro USB/FireWire PCI card ATI Rage Orion 128 16 MB Question 1: I understand I have 2 options, BootX and quik. I prefer BootX, since it doesn't muck with my firmware (Forth is scary stuff), provides me with a graphical prompt, allows me to pick a kernel quickly, and allows me to easily do a worry-free triple boot into OS 9, OS X, and Debian (granted I really only _need_ OS 9). Is there any major advantage to using quik over BootX, and will it help at all with question 2?... Question 2: Short version: How do I get a USB IntelliMouse Optical working with Debian on a Beige G3? (see long version for what I've tried) Long version: Believe it or not, I have been able to "correctly" configure XFree86, although I didn't get it hardware accelerated, only through fbdev. I am quite willing to hack around with this myself. But _not_ with a one-button mouse! Booting Debian under the provided kernel 2.2.19 (from the Debian-iMac package) the boot time messages showed clearly that it recognized my USB card, and even my IntelliMouse (it called it by name!), but I found it nigh impossible to get gpm to recognize it (I really didn't want to try XFree86 again with mouse-button-emulation). I tried this first under straight potato, then upgraded to testing, tried again, and finally under unstable, with the same results each time (nothing). I tried with both mice plugged in (the other is a single-button Apple ADB Mouse), and with only the USB one plugged in (with a shutdown and cold start in between each). With this multitude of configurations, I tried every possible configuration combination of gpm. I also checked for /dev/usbmouse, which to my dismay, never existed under potato, woody, or unstable. I tried making /dev/usbmouse a symlink to /input/mice (which did exist). I tried making /dev/mouse a symlink to both of those. I also (somehow) received instructions on how to properly mknod a mouse dev file, and tried this in numerous spots, running through the above-mentioned procedures several more times. At one point I thought I had it, screamed out with joy, and then realized I was holding the ADB mouse (I was in about my 20th hour of this). I also messed around with loading and unloading kernel modules, but having little experience in that arena, I didn't go too far. One thing I did not try was a 2.4.x kernel, since 2.2.19 was clearly seeing the USB card and mouse. So, obviously, my question is, how to I get this working? -Mike
Re: Beige G3 and USB Mouse [solved]
Thanks for all the replies, it didn't help my immediate problem, but I did learn a bit. I've solved the mouse problem by installing the 2.4.17-rc1 kernel from www.ppckernel.org. My mouse now works great, as well as a bunch of other things I'm discovering I like very much about the 2.4.x kernel series (which I've never used). For anyone else that has the same issues as me, this kernel works fine with no modules enabled, and even dmasound (only module I've enabled thus far) properly routes sound through my attached speakers, instead of the internal speaker (this didn't even work right with Yellow Dog 2.0/kernel 2.4.x). Now I'm faced with two more problems: this new kernel has some pretty ugly redraw problems on the console (which I can live with), and, of course, XFree86 won't start. At least it doesn't crash the machine...well xdm does, but I've removed that. However, I'm perfectly willing to try to figure these out on my own. I will report back if I learn anything not covered in the FAQs, docs, or archives. I will also report back if I'm still stuck after a few days of pounding at it :) Thanks again, -Mike
CamediaPlay and Olympus D-460 Zoom
Greetings, * please skip to next paragraph for actual question * I wrote the list a while back about getting my USB mouse to work with GPM. I just wanted to let those who helped me know that basically the stock Debian-iMac 2.2.19 kernel is about as good at seeing an ADB keyboard/USB mouse combo as stock Mac OS 9 is (ie neither see the mouse). The solution was building a custom 2.4.16 kernel, which I found _delightfully_ simple! I have now got (almost) all my hardware working, including r128 3D acceleration, scanner, FireWire CD-R, Palm V, KeySpan USA-19, and even sound! Honestly, I couldn't be happier, and can now do everything I could do in OS X... _quickly_ ... with a customizable UI, _and_ all my hardware is supported natively. Plus I can't go back to Entourage/Explorer since I've used Evolution/Galeon! Thanks to all the volunteers who make this extraordinary operating system/distribution. I am also finishing up a little website documenting my trials and errors so that it can help other wary Mac users/Linux newbies take the plunge if they wish (can't code a lick yet, but I gotta try to give something back). Now, the one piece of hardware I can't get to work is my old Olympus Camedia D-460 Zoom. I've installed CamediaPlay (20010211-2) and gPhoto (0.4.3-6), tried both, and neither seem to see the camera at all. First I tried gPhoto, selecting D-450Z (it was the closest to my camera, web page says it works with the D-460Z) When I choose "Configure Camera", after a brief pause, it tells me the camera couldn't be found. In CamediaPlay, i tried: camediaplay /dev/ttyS0 camediaplay /dev/ttyS1 These both resulted in a brief pause followed by "camera not ready." The first time I did this, I swear the camera zoom servo made a quick noise (which never happened when using the Mac software), but I couldn't reproduce it. camediaplay /dev/tts/0 camediaplay /dev/tts/1 (Obviously I'm using devfs) These immediately reported "device is busy", however fuser /dev/tts/0 (and 1) reported no one using them! Yes, the camera was on, yes I have the PowerMac serial port kernel modules compiled and enabled, and yes my normal user is a member of the "dialout" group (although I get all the same messages as root). The camera is connected to the serial port via the provided cable, which is a PC RS-232 serial cable with a Mac adapter on the end. The camera has a fresh set of batteries, but no AC adapter (don't have one). The machine is a beige G3/300. I am running the stable BitKeeper 2.4.17 kernel. The only additional modules I have installed are the Mac-on-Linux ones. Does anyone have any pointers or experience with this? I am sure I am missing something relatively simple, but I just can't seem to get this one on my own. Thanks in advance for any answers... Mike
Re: MOL and kernel 2.4
On Fri, 2002-01-04 at 12:10, Eric Deveaud wrote: > I can start mol as root (sudo startmol) but not as a lambda user. I could be totally wrong, but I believe that this is the only way to start MOL. I think it has to do with the fact that a) MOL wants to load its modules by itself and b) it wants to create a new "fake" networking interface. -Mike
Status
-- Virus Warning Message (on the network) Found virus WORM_MIMAIL.R in file readme.pif The file readme.pif is moved to /var/log/virus/virMEW3p1G.O. This is a machine-generated message, please do not reply via email. If you have questions, please contact the Lucent Help Desk at +1 888 300 0770. - test -- Virus Warning Message (on the network) readme.pif is removed from here because it contains a virus. -
Re: Kernel build fails
I sent a mail fixing this a few days ago, but I didn't finalize the patch
according to suggestions, my finger broke.
The thread :
Le 11/01/2022 à 10:32, Mike a écrit :
> I managed to fix it in the end, patch attached, though i should have
> done a $(call cc-option-, -maltivec, -mabi=altivec) in the
> arch/powerpc/mm/Makefile
> I wrongly assumed that the manual i had downloaded at 4.44am was for
> 32bit ppc only and found ptesync to be ppc64 only.
>
> binutils-2.37.50 - GNU assembler version 2.37.50 (powerpc-linux-gnu)
> using BFD version (GNU Binutils for Debian) 2.37.50.20220106
> gcc version 11.2.0 (Debian 11.2.0-13)
> ld.lld is missing but with LLVM/CLANG and LD=ld.bfd
> arch/powerpc/kernel/vdso32/gettimeofday.S:72:8:
> error: unsupported directive '.stabs'.stabs
> "_restgpr_31_x:F-1",36,0,0,_restgpr_31_x; .globl _restgpr_31_x;
> _restgpr_31_x:
>
> Attached the config i'm using, and the debian config 5.15.0-2. It's
> still building.
Ok, I tried with your config on my Fedora Core 35 where I have:
powerpc64-linux-gnu-gcc (GCC) 11.2.1 20210728 (Red Hat Cross 11.2.1-1)
GNU ld version 2.37-3.fc35
From packages:
- binutils-powerpc64-linux-gnu-2.37-3.fc35.x86_64
- gcc-powerpc64-linux-gnu-11.2.1-1.fc35.x86_64
And I don't have the problems you mention, so it must be something
special with Debian GCC.
Your change regarding ptesync is probably OK but is fragile I think,
because for instance there is also a 'ptesync' in
arch/powerpc/mm/pageattr.c and probably many other places.
Also please prefer CONFIG_PPC64 to __powerpc64__
Regarding the DSSALL issue, the following commit will probably help:
d51f86cfd8e3 ("powerpc/mm: Switch obsolete dssall to .long")
Regarding the .stabs with LLVM there is a patch at
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/68932ec2ba6b868d35006b96e90f0890f3da3c05.1638273868.git.christophe.le...@csgroup.eu/
Thanks
Christophe
-Michael
On Wed, Jan 19, 2022, 21:54 John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:
> Hello!
>
> On 1/19/22 14:35, Elimar Riesebieter wrote:
> > kernel build fails on powerpc 32bit no matter what version is used
> > (5.14-5.16):
> >
> > {standard input}: Assembler messages:
> > {standard input}:29: Error: unrecognized opcode: `dssall'
> > make[5]: *** [scripts/Makefile.build:271:
> arch/powerpc/mm/mmu_context.o] Error 1
> > make[4]: *** [scripts/Makefile.build:514: arch/powerpc/mm] Error 2
> > make[3]: *** [Makefile:1861: arch/powerpc] Error 2
> > make[2]: *** [debian/rules:7: build-arch] Error 2
> >
> > I assume the latest binutils (2.37.50.20220106-2) don't work
> > properly. Anyone made the same experience?
>
> This issue shows on the buildds now as well:
>
> >
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=powerpc&ver=5.15.15-1&stamp=1642579068&raw=0
> >
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=ppc64&ver=5.15.15-1&stamp=1642578946&raw=0
>
> I will report it upstream now.
>
> Adrian
>
> --
> .''`. John Paul Adrian Glaubitz
> : :' : Debian Developer - glaub...@debian.org
> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>
>
Author: threader
Date: Tue Jan 11 06:28:27 2022 +0100
arch: powerpc: fix building, 'dssall' in mmu_context.c is an altivec
instruction, so build that acoordingly if such is the case. 'ptesync' is a
PPC64 instruction, so dont go there for ppc32
Signed-off-by: Micahel B Heltne
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index 86f49e3e7cf5..ac8c2d591636 100644
--- a/arch/powerpc/lib/sstep.c
+++ b/arch/powerpc/lib/sstep.c
@@ -3267,9 +3267,11 @@ void emulate_update_regs(struct pt_regs *regs, struct
instruction_op *op)
case BARRIER_LWSYNC:
asm volatile("lwsync" : : : "memory");
break;
+#ifdef __powerpc64__
case BARRIER_PTESYNC:
asm volatile("ptesync" : : : "memory");
break;
+#endif
}
break;
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index df8172da2301..2775b56a77cb 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -4,6 +4,9 @@
#
ccflags-$(CONFIG_PPC64):= $(NO_MINIMAL_TOC)
+ifeq ($(CONFIG_ALTIVEC),y)
+CFLAGS_mmu_context.o += -maltivec
+endif
obj-y := fault.o mem.o pgtable.o mmap.o maccess.o
pageattr.o \
init_$(BITS).o pgtable_$(BITS).o \
Re: Kernel build fails
It did boot, that much I tested before rejoining my couch adventure. On Thu, Jan 20, 2022, 09:28 John Paul Adrian Glaubitz < glaub...@physik.fu-berlin.de> wrote: > Hi Mike! > > On 1/20/22 08:48, Mike wrote: > > I sent a mail fixing this a few days ago, but I didn't finalize > > the patch according to suggestions, my finger broke. > > This is actually a more complex issue, in particular the assembler was > changed on PowerPC when handling the .machine directive, see [1]. > > So, in order to fix this issue, someone needs to update the kernel to > work with the new binutils version. > > Adrian > > > [1] https://sourceware.org/bugzilla/show_bug.cgi?id=28794#c2 > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer - glaub...@debian.org > `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: Kernel build fails
> So, in order to fix this issue, someone needs to update the kernel to work with the new binutils version. That's what the patch does, I did a full Debian kernel build, but I'm having trouble explaining myself with my left hand. On Thu, Jan 20, 2022, 09:41 Mike wrote: > It did boot, that much I tested before rejoining my couch adventure. > > On Thu, Jan 20, 2022, 09:28 John Paul Adrian Glaubitz < > glaub...@physik.fu-berlin.de> wrote: > >> Hi Mike! >> >> On 1/20/22 08:48, Mike wrote: >> > I sent a mail fixing this a few days ago, but I didn't finalize >> > the patch according to suggestions, my finger broke. >> >> This is actually a more complex issue, in particular the assembler was >> changed on PowerPC when handling the .machine directive, see [1]. >> >> So, in order to fix this issue, someone needs to update the kernel to >> work with the new binutils version. >> >> Adrian >> >> > [1] https://sourceware.org/bugzilla/show_bug.cgi?id=28794#c2 >> >> -- >> .''`. John Paul Adrian Glaubitz >> : :' : Debian Developer - glaub...@debian.org >> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de >> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 >> >>
Re: Kernel build fails
Wtf, I'll work it out in a smoke.
On Thu, Jan 20, 2022, 10:05 John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:
> Hi Mike!
>
> On 1/20/22 09:46, Mike wrote:
> >> So, in order to fix this issue, someone needs to update the kernel to
> work with the new binutils version.
> >
> > That's what the patch does, I did a full Debian kernel build, but I'm
> having trouble explaining myself with my left hand.
>
> I still think that your patch is incomplete as the "dsall" instruction is
> not the only
> instruction that is affected by this change. See the kernel build logs
> below.
>
> From [1]:
>
> {standard input}: Assembler messages:
> {standard input}:1353: Error: unrecognized opcode: `ptesync'
>
> and [2]:
>
> {standard input}: Assembler messages:
> {standard input}:2146: Error: unrecognized opcode: `stbcix'
>
> > [1]
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=powerpc&ver=5.15.15-1&stamp=1642579068&raw=0
> > [2]
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=ppc64&ver=5.15.15-1&stamp=1642578946&raw=0
>
> Adrian
>
> --
> .''`. John Paul Adrian Glaubitz
> : :' : Debian Developer - glaub...@debian.org
> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>
>
Re: Kernel build fails
'stbcix' has to come from arch/powerpc/include/asm/io.h - and is already
behind an ifdefs, though try changing it to #ifdef CONFIG_PPC64? I'm
building now.
Michael
On Thu, Jan 20, 2022, 10:05 John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:
> Hi Mike!
>
> On 1/20/22 09:46, Mike wrote:
> >> So, in order to fix this issue, someone needs to update the kernel to
> work with the new binutils version.
> >
> > That's what the patch does, I did a full Debian kernel build, but I'm
> having trouble explaining myself with my left hand.
>
> I still think that your patch is incomplete as the "dsall" instruction is
> not the only
> instruction that is affected by this change. See the kernel build logs
> below.
>
> From [1]:
>
> {standard input}: Assembler messages:
> {standard input}:1353: Error: unrecognized opcode: `ptesync'
>
> and [2]:
>
> {standard input}: Assembler messages:
> {standard input}:2146: Error: unrecognized opcode: `stbcix'
>
> > [1]
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=powerpc&ver=5.15.15-1&stamp=1642579068&raw=0
> > [2]
> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=ppc64&ver=5.15.15-1&stamp=1642578946&raw=0
>
> Adrian
>
> --
> .''`. John Paul Adrian Glaubitz
> : :' : Debian Developer - glaub...@debian.org
> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>
>
Re: Kernel build fails
Waiting for git to do it's thing, but do we need a voodoo priest(es)
here? The attached patch is building.
-Michael
On Fri, 21 Jan 2022 at 20:23, Mike wrote:
>
> 'stbcix' has to come from arch/powerpc/include/asm/io.h - and is already
> behind an ifdefs, though try changing it to #ifdef CONFIG_PPC64? I'm building
> now.
>
> Michael
>
> On Thu, Jan 20, 2022, 10:05 John Paul Adrian Glaubitz
> wrote:
>>
>> Hi Mike!
>>
>> On 1/20/22 09:46, Mike wrote:
>> >> So, in order to fix this issue, someone needs to update the kernel to
>> >> work with the new binutils version.
>> >
>> > That's what the patch does, I did a full Debian kernel build, but I'm
>> > having trouble explaining myself with my left hand.
>>
>> I still think that your patch is incomplete as the "dsall" instruction is
>> not the only
>> instruction that is affected by this change. See the kernel build logs below.
>>
>> From [1]:
>>
>> {standard input}: Assembler messages:
>> {standard input}:1353: Error: unrecognized opcode: `ptesync'
>>
>> and [2]:
>>
>> {standard input}: Assembler messages:
>> {standard input}:2146: Error: unrecognized opcode: `stbcix'
>>
>> > [1]
>> > https://buildd.debian.org/status/fetch.php?pkg=linux&arch=powerpc&ver=5.15.15-1&stamp=1642579068&raw=0
>> > [2]
>> > https://buildd.debian.org/status/fetch.php?pkg=linux&arch=ppc64&ver=5.15.15-1&stamp=1642578946&raw=0
>>
>> Adrian
>>
>> --
>> .''`. John Paul Adrian Glaubitz
>> : :' : Debian Developer - glaub...@debian.org
>> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>>
commit 8a6358dfdf06a35d2aa9cf171b8444a716ad0c2a
Author: threader
Date: Fri Jan 21 20:36:33 2022 +0100
arch: powerpc: fix building, 'dssall' in mmu_context.c is an altivec instruction, so build that accordingly if such is the case.
'ptesync' is a PPC64 instruction, so dont go there for PPC32. And apparently ifdefined __powerpc64__ isnt enough in all configurations(?) and 'stbcix' sticks its head up.
Signed-off-by: Micahel B Heltne
diff --git a/arch/powerpc/include/asm/io.h b/arch/powerpc/include/asm/io.h
index beba4979bff9..cb09701ed114 100644
--- a/arch/powerpc/include/asm/io.h
+++ b/arch/powerpc/include/asm/io.h
@@ -334,7 +334,7 @@ static inline void __raw_writel(unsigned int v, volatile void __iomem *addr)
}
#define __raw_writel __raw_writel
-#ifdef __powerpc64__
+#ifdef CONFIG_PPC64
static inline unsigned long __raw_readq(const volatile void __iomem *addr)
{
return *(volatile unsigned long __force *)PCI_FIX_ADDR(addr);
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index 86f49e3e7cf5..becbe2898e10 100644
--- a/arch/powerpc/lib/sstep.c
+++ b/arch/powerpc/lib/sstep.c
@@ -3267,9 +3267,11 @@ void emulate_update_regs(struct pt_regs *regs, struct instruction_op *op)
case BARRIER_LWSYNC:
asm volatile("lwsync" : : : "memory");
break;
+#ifdef CONFIG_PPC64
case BARRIER_PTESYNC:
asm volatile("ptesync" : : : "memory");
break;
+#endif
}
break;
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index df8172da2301..2f87e7731599 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -4,6 +4,9 @@
#
ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
+ifeq ($(CONFIG_ALTIVEC),y)
+CFLAGS_mmu_context.o += $(call cc-option, -maltivec, -mabi=altivec)
+endif
obj-y:= fault.o mem.o pgtable.o mmap.o maccess.o pageattr.o \
init_$(BITS).o pgtable_$(BITS).o \
diff --git a/arch/powerpc/mm/pageattr.c b/arch/powerpc/mm/pageattr.c
index edea388e9d3f..ccd04a386e28 100644
--- a/arch/powerpc/mm/pageattr.c
+++ b/arch/powerpc/mm/pageattr.c
@@ -54,11 +54,11 @@ static int change_page_attr(pte_t *ptep, unsigned long addr, void *data)
}
pte_update(&init_mm, addr, ptep, ~0UL, pte_val(pte), 0);
-
+#ifdef CONFIG_PPC64
/* See ptesync comment in radix__set_pte_at() */
if (radix_enabled())
asm volatile("ptesync": : :"memory");
-
+#endif
flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
spin_unlock(&init_mm.page_table_lock);
Re: Bug#885245: linux: FTBFS on powerpcspe: sstep.c: ptesync unrecognized
That will be wrong for PPC spe, as its got differing and conflicting
instructions compared with other CPUs in its class. The attached patch
built and is finishing up modules now .
Michael
On Sat, Jan 22, 2022, 01:00 Cameron MacPherson
wrote:
> to get the latest kernel to compile on a powermac g5 takes removing the
> power8 instructions from sstep.c it will compile with -mpower8 but then
> wont boot if you dont get rid of the instructions
>
> On Thu, Jan 20, 2022, 1:11 AM John Paul Adrian Glaubitz <
> glaub...@physik.fu-berlin.de> wrote:
>
>> Hi!
>>
>> On 12/26/17 04:32, Aaron M. Ucko wrote:
>> > Builds of linux for powerpcspe (admittedly not a release architecture)
>> > have been failing lately:
>> >
>> > {standard input}: Assembler messages:
>> > {standard input}:5854: Error: unrecognized opcode: `ptesync'
>> > /<>/scripts/Makefile.build:319: recipe for target
>> 'arch/powerpc/lib/sstep.o' failed
>> > make[6]: *** [arch/powerpc/lib/sstep.o] Error 1
>>
>> Funny, it seems that this particular issue is now uncovering on powerpc
>> [1] and
>> ppc64 [2] as well after this change [3] in the GNU Assembler.
>>
>> Adrian
>>
>> > [1]
>> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=powerpc&ver=5.15.15-1&stamp=1642579068&raw=0
>> > [2]
>> https://buildd.debian.org/status/fetch.php?pkg=linux&arch=ppc64&ver=5.15.15-1&stamp=1642578946&raw=0
>> > [3]
>> https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=b25f942e18d6ecd7ec3e2d2e9930eb4f996c258a
>>
>> --
>> .''`. John Paul Adrian Glaubitz
>> : :' : Debian Developer - glaub...@debian.org
>> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>>
>>
commit 8a6358dfdf06a35d2aa9cf171b8444a716ad0c2a
Author: threader
Date: Fri Jan 21 20:36:33 2022 +0100
arch: powerpc: fix building, 'dssall' in mmu_context.c is an altivec instruction, so build that accordingly if such is the case.
'ptesync' is a PPC64 instruction, so dont go there for PPC32. And apparently ifdefined __powerpc64__ isnt enough in all configurations(?) and 'stbcix' sticks its head up.
Signed-off-by: Micahel B Heltne
diff --git a/arch/powerpc/include/asm/io.h b/arch/powerpc/include/asm/io.h
index beba4979bff9..cb09701ed114 100644
--- a/arch/powerpc/include/asm/io.h
+++ b/arch/powerpc/include/asm/io.h
@@ -334,7 +334,7 @@ static inline void __raw_writel(unsigned int v, volatile void __iomem *addr)
}
#define __raw_writel __raw_writel
-#ifdef __powerpc64__
+#ifdef CONFIG_PPC64
static inline unsigned long __raw_readq(const volatile void __iomem *addr)
{
return *(volatile unsigned long __force *)PCI_FIX_ADDR(addr);
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index 86f49e3e7cf5..becbe2898e10 100644
--- a/arch/powerpc/lib/sstep.c
+++ b/arch/powerpc/lib/sstep.c
@@ -3267,9 +3267,11 @@ void emulate_update_regs(struct pt_regs *regs, struct instruction_op *op)
case BARRIER_LWSYNC:
asm volatile("lwsync" : : : "memory");
break;
+#ifdef CONFIG_PPC64
case BARRIER_PTESYNC:
asm volatile("ptesync" : : : "memory");
break;
+#endif
}
break;
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index df8172da2301..2f87e7731599 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -4,6 +4,9 @@
#
ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
+ifeq ($(CONFIG_ALTIVEC),y)
+CFLAGS_mmu_context.o += $(call cc-option, -maltivec, -mabi=altivec)
+endif
obj-y:= fault.o mem.o pgtable.o mmap.o maccess.o pageattr.o \
init_$(BITS).o pgtable_$(BITS).o \
diff --git a/arch/powerpc/mm/pageattr.c b/arch/powerpc/mm/pageattr.c
index edea388e9d3f..ccd04a386e28 100644
--- a/arch/powerpc/mm/pageattr.c
+++ b/arch/powerpc/mm/pageattr.c
@@ -54,11 +54,11 @@ static int change_page_attr(pte_t *ptep, unsigned long addr, void *data)
}
pte_update(&init_mm, addr, ptep, ~0UL, pte_val(pte), 0);
-
+#ifdef CONFIG_PPC64
/* See ptesync comment in radix__set_pte_at() */
if (radix_enabled())
asm volatile("ptesync": : :"memory");
-
+#endif
flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
spin_unlock(&init_mm.page_table_lock);
Re: Kernel build fails
I think I need to add a || CONFIG_POWER6_CPU to that 'stbcix' condition.. -Mike On Sat, Jan 22, 2022, 09:42 John Paul Adrian Glaubitz < glaub...@physik.fu-berlin.de> wrote: > Hello Mike! > > On 1/21/22 23:07, Mike wrote: > > Waiting for git to do it's thing, but do we need a voodoo priest(es) > > here? The attached patch is building. > > I will have a look at this issue next week myself. We need to make sure > that it not only fixes 32-bit PowerPC but also 64-bit PowerPC big-endian. > > Also, test builds of Debian the packages are necessary as well. > > Adrian > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer - glaub...@debian.org > `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: Kernel build fails
It should be just CONFIG_POWER6_CPU unless ppc64 also require this?iv -Mike On Sat, Jan 22, 2022, 10:36 Mike wrote: > I think I need to add a || CONFIG_POWER6_CPU to that 'stbcix' condition.. > > -Mike > > > > On Sat, Jan 22, 2022, 09:42 John Paul Adrian Glaubitz < > glaub...@physik.fu-berlin.de> wrote: > >> Hello Mike! >> >> On 1/21/22 23:07, Mike wrote: >> > Waiting for git to do it's thing, but do we need a voodoo priest(es) >> > here? The attached patch is building. >> >> I will have a look at this issue next week myself. We need to make sure >> that it not only fixes 32-bit PowerPC but also 64-bit PowerPC big-endian. >> >> Also, test builds of Debian the packages are necessary as well. >> >> Adrian >> >> -- >> .''`. John Paul Adrian Glaubitz >> : :' : Debian Developer - glaub...@debian.org >> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de >> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 >> >>
Re: Kernel build fails
I've sent in a patch with my remaining concerns attached to the mailing
lists and set up a linux-next branch finally.
https://github.com/threader/linux/tree/master-build-ppc
In regards to asm/io.h:
---
Will there come a mail saying this broke the PPC6'ish based CPU
someone made in their garage? And lwesync is a valid PPC32
instruction, should i just follow the example above where
BARRIER_LWESYNC is PPC64 only?
---
-Michael
On Sat, Jan 22, 2022, 10:48 Mike wrote:
> It should be just CONFIG_POWER6_CPU unless ppc64 also require this?iv
>
> -Mike
>
> On Sat, Jan 22, 2022, 10:36 Mike wrote:
>
>> I think I need to add a || CONFIG_POWER6_CPU to that 'stbcix' condition..
>>
>> -Mike
>>
>>
>>
>> On Sat, Jan 22, 2022, 09:42 John Paul Adrian Glaubitz <
>> glaub...@physik.fu-berlin.de> wrote:
>>
>>> Hello Mike!
>>>
>>> On 1/21/22 23:07, Mike wrote:
>>> > Waiting for git to do it's thing, but do we need a voodoo priest(es)
>>> > here? The attached patch is building.
>>>
>>> I will have a look at this issue next week myself. We need to make sure
>>> that it not only fixes 32-bit PowerPC but also 64-bit PowerPC big-endian.
>>>
>>> Also, test builds of Debian the packages are necessary as well.
>>>
>>> Adrian
>>>
>>> --
>>> .''`. John Paul Adrian Glaubitz
>>> : :' : Debian Developer - glaub...@debian.org
>>> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>>> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>>>
>>>
From 226efa05733457bb5c483f30aab6d5c6a304422c Mon Sep 17 00:00:00 2001
From: threader
Date: Sun, 23 Jan 2022 14:17:10 +0100
Subject: [PATCH] arch: powerpc: fix building after binutils changes. 'dssall'
in mmu_context.c is an altivec instruction, build that accordingly. 'ptesync'
is a PPC64 instruction, so dont go there for if not. And apparently ifdef
__powerpc64__ isnt enough in all configurations and 'stbcix' and friends, all
POWER6 instructions hopefully not needed by CONFIG_PPC64 in general, wanted
to play.
Signed-off-by: Micahel B Heltne
---
arch/powerpc/include/asm/io.h | 7 ---
arch/powerpc/lib/sstep.c | 4 +++-
arch/powerpc/mm/Makefile | 3 +++
arch/powerpc/mm/pageattr.c| 4 ++--
4 files changed, 12 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/include/asm/io.h b/arch/powerpc/include/asm/io.h
index beba4979bff939..d3a9c91cd06a8b 100644
--- a/arch/powerpc/include/asm/io.h
+++ b/arch/powerpc/include/asm/io.h
@@ -334,7 +334,7 @@ static inline void __raw_writel(unsigned int v, volatile void __iomem *addr)
}
#define __raw_writel __raw_writel
-#ifdef __powerpc64__
+#ifdef CONFIG_PPC64
static inline unsigned long __raw_readq(const volatile void __iomem *addr)
{
return *(volatile unsigned long __force *)PCI_FIX_ADDR(addr);
@@ -352,7 +352,8 @@ static inline void __raw_writeq_be(unsigned long v, volatile void __iomem *addr)
__raw_writeq((__force unsigned long)cpu_to_be64(v), addr);
}
#define __raw_writeq_be __raw_writeq_be
-
+#endif
+#ifdef CONFIG_POWER6_CPU
/*
* Real mode versions of the above. Those instructions are only supposed
* to be used in hypervisor real mode as per the architecture spec.
@@ -417,7 +418,7 @@ static inline u64 __raw_rm_readq(volatile void __iomem *paddr)
: "=r" (ret) : "r" (paddr) : "memory");
return ret;
}
-#endif /* __powerpc64__ */
+#endif /* CONFIG_POWER6_CPU */
/*
*
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index a94b0cd0bdc5ca..4ffd6791b03ec0 100644
--- a/arch/powerpc/lib/sstep.c
+++ b/arch/powerpc/lib/sstep.c
@@ -1465,7 +1465,7 @@ int analyse_instr(struct instruction_op *op, const struct pt_regs *regs,
switch ((word >> 1) & 0x3ff) {
case 598: /* sync */
op->type = BARRIER + BARRIER_SYNC;
-#ifdef __powerpc64__
+#ifdef CONFIG_PPC64
switch ((word >> 21) & 3) {
case 1: /* lwsync */
op->type = BARRIER + BARRIER_LWSYNC;
@@ -3267,9 +3267,11 @@ void emulate_update_regs(struct pt_regs *regs, struct instruction_op *op)
case BARRIER_LWSYNC:
asm volatile("lwsync" : : : "memory");
break;
+#ifdef CONFIG_PPC64
case BARRIER_PTESYNC:
asm volatile("ptesync" : : : "memory");
break;
+#endif
}
break;
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index df8172da2301b7..2f87e77315997a 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -4,6 +4,9 @@
#
ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC)
+ifeq ($(CONFIG_ALTIVEC),y)
+CFLAGS_mmu_context.o += $(call cc-option, -maltivec, -mabi=altivec)
+endif
obj-y:= fault.o mem.o pgtable
Re: Kernel build fails
I remembered something about 'lwsync' i read about it in 2013
suddenly, and came to the following proposal to deal with remove the
PPC64 requirement for BARRIER_LWSYNC, but i've literally really just
started poking about the kernel side of powerpc and reading around.
But i do remember the penalty in using 'sync' in place of 'lwsync'
https://gcc.gnu.org/legacy-ml/gcc-patches/2006-11/msg01238.html
https://gcc.gnu.org/legacy-ml/gcc-patches/2012-07/msg01062.html
So something like this
diff --git
a/media/thread/12fd50d6-d14c-42af-ad1d-a595e5f080cd/dev/linux-main/linux/arch/powerpc/lib/sstep.c
b/home/thread/dev/linus/linux/arch/powerpc/lib/sstep.c
index 4ffd679..39861d3 100644
---
a/media/thread/12fd50d6-d14c-42af-ad1d-a595e5f080cd/dev/linux-main/linux/arch/powerpc/lib/sstep.c
+++ b/home/thread/dev/linus/linux/arch/powerpc/lib/sstep.c
@@ -3265,7 +3265,11 @@ void emulate_update_regs(struct pt_regs *regs,
struct instruction_op *op)
eieio();
break;
case BARRIER_LWSYNC:
+#if defined (CONFIG_40x || CONFIG_44x || CONFIG_E500 ||
CONFIG_PPA8548 || CONFIG_TQM8548 || CONFIG_MPC8540_ADS ||
CONFIG_PPC_BOOK3S_603)
+ asm volatile("sync" : : : "memory");
+#else
asm volatile("lwsync" : : : "memory");
+#endif
break;
#ifdef CONFIG_PPC64
case BARRIER_PTESYNC:
On Sun, 23 Jan 2022 at 14:50, Mike wrote:
>
> I've sent in a patch with my remaining concerns attached to the mailing lists
> and set up a linux-next branch finally.
> https://github.com/threader/linux/tree/master-build-ppc
>
> In regards to asm/io.h:
> ---
> Will there come a mail saying this broke the PPC6'ish based CPU
> someone made in their garage? And lwesync is a valid PPC32
> instruction, should i just follow the example above where
> BARRIER_LWESYNC is PPC64 only?
> ---
>
> -Michael
> On Sat, Jan 22, 2022, 10:48 Mike wrote:
>>
>> It should be just CONFIG_POWER6_CPU unless ppc64 also require this?iv
>>
>> -Mike
>>
>> On Sat, Jan 22, 2022, 10:36 Mike wrote:
>>>
>>> I think I need to add a || CONFIG_POWER6_CPU to that 'stbcix' condition..
>>>
>>> -Mike
>>>
>>>
>>>
>>> On Sat, Jan 22, 2022, 09:42 John Paul Adrian Glaubitz
>>> wrote:
>>>>
>>>> Hello Mike!
>>>>
>>>> On 1/21/22 23:07, Mike wrote:
>>>> > Waiting for git to do it's thing, but do we need a voodoo priest(es)
>>>> > here? The attached patch is building.
>>>>
>>>> I will have a look at this issue next week myself. We need to make sure
>>>> that it not only fixes 32-bit PowerPC but also 64-bit PowerPC big-endian.
>>>>
>>>> Also, test builds of Debian the packages are necessary as well.
>>>>
>>>> Adrian
>>>>
>>>> --
>>>> .''`. John Paul Adrian Glaubitz
>>>> : :' : Debian Developer - glaub...@debian.org
>>>> `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>>>> `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
>>>>
Re: Linux kernel: powerpc: KVM guest can trigger host crash on Power8
I just made the huge mistake of hibernating and resuming, I'm going trough the process of rescue and all, thankfully I had a 2016 cd in the drive. I'll read up once the sheer panic settles. -Michael On Wed, Jan 26, 2022, 21:22 John Paul Adrian Glaubitz < glaub...@physik.fu-berlin.de> wrote: > Hi Michael! > > On 1/13/22 01:17, John Paul Adrian Glaubitz wrote: > > On 1/9/22 23:17, John Paul Adrian Glaubitz wrote: > >> On 1/7/22 12:20, John Paul Adrian Glaubitz wrote: > Can you separately test with (on the host): > > # echo 0 > /sys/module/kvm_hv/parameters/dynamic_mt_modes > >>> > >>> I'm trying to turn off "dynamic_mt_modes" first and see if that makes > any difference. > >>> > >>> I will report back. > >> > >> So far the machine is running stable now and the VM built gcc-9 without > >> crashing the host. I will continue to monitor the machine and report > back > >> if it crashes, but it looks like this could be it. > > > > So, it seems that setting "dynamic_mt_modes" actually did the trick, the > host is no longer > > crashing. However, I have observed on two occasions now that the build > VM is just suddenly > > off as if someone has shut it down using the "force-off" option in the > virt-manager user > > interface. > > Just as a heads-up. Ever since I set > > echo 0 > /sys/module/kvm_hv/parameters/dynamic_mt_modes > > on the host machine, I never saw the crash again. So the issue seems to be > related to the > dynamic_mt_modes feature. > > Thanks, > Adrian > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer - glaub...@debian.org > `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: [BUGS] PowerPC 32bit - Hibernate/resume is dangerous - Xserver crashes after updating in rescue mode arguing with appletouch it seems
Hm, that log was a little more complete then what startx said about appletouch on screen. On Thu, 27 Jan 2022 at 17:59, Mike wrote: > > So I gave myself a heart attack powering on. And found another headache. I've > got no logs from the resume process but I thought maybe I should replicate it > in qemu. This at least seems to have corrupted libpam.so in my case. It > booted normally to desktop with the exception of network indicator showing > 2gbps up and down, disconnected, when it came time to resume the workload the > WM crashed first when I launched a new terminal, then it became clear disk > was inaccessible as I didn't get past login in shell, infact not past > username Rebooted, PMU seemed like it was corrupted and was greated by a > panic. Thankfully I had a 2016 Debian in the CD drive and sorted that by > reinstalling libpam0g and updated, that's when the Xserver appletouch decided > my mouse didn't have enough buttons. I'll get back with the proper logs in a > bit,I just literally sat down to look at another issue. > > > -Michael [ 218.471] X.Org X Server 1.20.14 X Protocol Version 11, Revision 0 [ 218.503] Build Operating System: linux Debian [ 218.514] Current Operating System: Linux PowerBook-G4 5.15.0-2-powerpc #1 Debian 5.15.5-2 (2021-12-18) ppc [ 218.514] Kernel command line: root=UUID=6fcbe74d-622e-4fd3-a2b5-7c6d6d056816 ro [ 218.536] Build Date: 11 January 2022 02:21:08PM [ 218.547] xorg-server 2:1.20.14-1 (https://www.debian.org/support) [ 218.558] Current version of pixman: 0.40.0 [ 218.579] Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. [ 218.579] Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. [ 218.622] (==) Log file: "/var/log/Xorg.0.log", Time: Thu Jan 27 17:40:53 2022 [ 218.633] (==) Using system config directory "/usr/share/X11/xorg.conf.d" [ 218.634] (==) No Layout section. Using the first Screen section. [ 218.634] (**) |-->Screen "Default screen" (0) [ 218.634] (**) | |-->Monitor "" [ 218.635] (**) | |-->Device "Radeon" [ 218.635] (==) No monitor specified for screen "Default screen". Using a default monitor configuration. [ 218.636] (==) Automatically adding devices [ 218.636] (==) Automatically enabling devices [ 218.636] (==) Automatically adding GPU devices [ 218.636] (==) Max clients allowed: 256, resource mask: 0x1f [ 218.636] (WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist. [ 218.636] Entry deleted from font path. [ 218.636] (==) FontPath set to: /usr/share/fonts/X11/misc, /usr/share/fonts/X11/100dpi/:unscaled, /usr/share/fonts/X11/75dpi/:unscaled, /usr/share/fonts/X11/Type1, /usr/share/fonts/X11/100dpi, /usr/share/fonts/X11/75dpi, built-ins [ 218.636] (==) ModulePath set to "/usr/lib/xorg/modules" [ 218.636] (II) The server relies on udev to provide the list of input devices. If no devices become available, reconfigure udev or disable AutoAddDevices. [ 218.636] (II) Loader magic: 0x9f0738 [ 218.636] (II) Module ABI versions: [ 218.636] X.Org ANSI C Emulation: 0.4 [ 218.636] X.Org Video Driver: 24.1 [ 218.636] X.Org XInput driver : 24.1 [ 218.636] X.Org Server Extension : 10.0 [ 218.640] (++) using VT number 1 [ 218.654] (II) systemd-logind: took control of session /org/freedesktop/login1/session/_32 [ 218.656] (II) xfree86: Adding drm device (/dev/dri/card0) [ 218.661] (II) systemd-logind: got fd for /dev/dri/card0 226:0 fd 11 paused 0 [ 218.672] (--) PCI:*(0@0:16:0) 1002:4e50:1002:4e50 rev 0, Mem @ 0xb800/134217728, 0xb000/65536, I/O @ 0x0400/256, BIOS @ 0x/131072 [ 218.672] (II) "glx" will be loaded by default. [ 218.672] (II) LoadModule: "dri" [ 218.672] (II) Module "dri" already built-in [ 218.673] (II) LoadModule: "glx" [ 218.673] (II) Loading /usr/lib/xorg/modules/extensions/libglx.so [ 218.677] (II) Module glx: vendor="X.Org Foundation" [ 218.677] compiled for 1.20.14, module version = 1.0.0 [ 218.677] ABI class: X.Org Server Extension, version 10.0 [ 218.677] (II) LoadModule: "radeon" [ 218.677] (II) Loading /usr/lib/xorg/modules/drivers/radeon_drv.so [ 218.686] (II) Module radeon: vendor="X.Org Foundation" [ 218.686] compiled for 1.20.9, module version = 19.1.0 [ 218.686] Module class: X.Org Video Driver [ 218.686] ABI class: X.Org Video Driver, version 24.1 [ 218.686] (II) RADEON: Driver for ATI/AMD Radeon chipsets: ATI Radeon Mobility X600 (M24), ATI FireMV 2400, ATI Radeon Mobility X300 (M24), ATI FireGL M24 GL, ATI Radeon X600 (RV380), ATI FireG
Re: [BUGS] PowerPC 32bit - Hibernate/resume is dangerous - Xserver crashes after updating in rescue mode arguing with appletouch it seems
So, udev and systemd are fighting? On Thu, 27 Jan 2022 at 20:40, Mike wrote: > > Hm, that log was a little more complete then what startx said about > appletouch on screen. > > On Thu, 27 Jan 2022 at 17:59, Mike wrote: > > > > So I gave myself a heart attack powering on. And found another headache. > > I've got no logs from the resume process but I thought maybe I should > > replicate it in qemu. This at least seems to have corrupted libpam.so in my > > case. It booted normally to desktop with the exception of network indicator > > showing 2gbps up and down, disconnected, when it came time to resume the > > workload the WM crashed first when I launched a new terminal, then it > > became clear disk was inaccessible as I didn't get past login in shell, > > infact not past username Rebooted, PMU seemed like it was corrupted and > > was greated by a panic. Thankfully I had a 2016 Debian in the CD drive and > > sorted that by reinstalling libpam0g and updated, that's when the Xserver > > appletouch decided my mouse didn't have enough buttons. I'll get back with > > the proper logs in a bit,I just literally sat down to look at another issue. > > > > > > -Michael
Re: [BUGS] PowerPC 32bit - Hibernate/resume is dangerous - Xserver crashes after updating in rescue mode arguing with appletouch it seems
System is rather corrupted, ldconfig complained about libXcomposite and libgtk-3.0 and I'm now in X11 after reinstalling but I'll need to trigger a reinstall of all packages :( and I've no idea how other partitions are doing. On Thu, Jan 27, 2022, 20:43 Mike wrote: > So, udev and systemd are fighting? > > On Thu, 27 Jan 2022 at 20:40, Mike wrote: > > > > Hm, that log was a little more complete then what startx said about > > appletouch on screen. > > > > On Thu, 27 Jan 2022 at 17:59, Mike wrote: > > > > > > So I gave myself a heart attack powering on. And found another > headache. I've got no logs from the resume process but I thought maybe I > should replicate it in qemu. This at least seems to have corrupted > libpam.so in my case. It booted normally to desktop with the exception of > network indicator showing 2gbps up and down, disconnected, when it came > time to resume the workload the WM crashed first when I launched a new > terminal, then it became clear disk was inaccessible as I didn't get past > login in shell, infact not past username Rebooted, PMU seemed like it > was corrupted and was greated by a panic. Thankfully I had a 2016 Debian in > the CD drive and sorted that by reinstalling libpam0g and updated, that's > when the Xserver appletouch decided my mouse didn't have enough buttons. > I'll get back with the proper logs in a bit,I just literally sat down to > look at another issue. > > > > > > > > > -Michael >
Re: Kernel build fails
This works : https://github.com/threader/linux/commit/226efa05733457bb5c483f30aab6d5c6a304422c , but in doing so I was allerted to 'lwsync' not being enabled on 32bit, and so will have to check out the assembly after enabled. So it works but I have deep rooted questions as always. -Mike On Sat, Jan 29, 2022, 22:16 John Paul Adrian Glaubitz < glaub...@physik.fu-berlin.de> wrote: > Hello Elimar! > > On 1/29/22 14:56, Elimar Riesebieter wrote: > >> I assume the latest binutils (2.37.50.20220106-2) don't work > >> properly. Anyone made the same experience? > > > > No fix with binutils 2.37.90.20220123-2 > > That's not surprising because the bug must be fixed on the kernel side, > not binutils. See the explanation here [1]. > > Adrian > > > [1] https://sourceware.org/bugzilla/show_bug.cgi?id=28794 > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer - glaub...@debian.org > `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: Kernel build fails
The behavior of binutils is probably correct. 'dssall' needs altivec and that was disabled for arch/powerpc/lib an mm etc -Mike On Sun, Jan 30, 2022, 12:56 John Paul Adrian Glaubitz < glaub...@physik.fu-berlin.de> wrote: > Hello Mike! > > On 1/30/22 12:11, Mike wrote: > > This works : > https://github.com/threader/linux/commit/226efa05733457bb5c483f30aab6d5c6a304422c > , > > but in doing so I was allerted to 'lwsync' not being enabled on 32bit, > and so will have to check > > out the assembly after enabled. So it works but I have deep rooted > questions as always. > > I would just wait until the maintainers of the Linux PowerPC port come up > with their own build fix, > they just haven't seen this issue yet because they're not using a > pre-release version of binutils. > > Adrian > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer - glaub...@debian.org > `. `' Freie Universitaet Berlin - glaub...@physik.fu-berlin.de > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: Debian freezing on PowerMac G4
You are going to have to be a bit more specific, what gpu? You are probably supposed to be using xserver-xorg-video-radeon or xserver-xorg-video-ati unless you happen to have a very new GPU in there. https://www.x.org/wiki/radeon/ https://github.com/freedesktop/xorg-xf86-video-ati Michael On Sat, 5 Feb 2022 at 23:50, Herminio Hernandez Jr. wrote: > > Hey, > > Took out my PowerMac G4 and installed Debian again. The install went smooth > with the exception of the grub install. Got through that and it booted > perfectly. I installed the nonfree firmware and the amd graphics package. > When I rebooted the system froze. In the past I would put the kernel setting > Radeon.agpmode=-1 and that would resolve it. I dropped to grub edit mode and > added that like and it still froze. There something else I need to do? > > Herminio > > Sent from my iPhone
Re: Fwd: X-Windows on PPC in Debian SID
Remember to regenerate the crypto keys so these aren't the same. Cheers.
On Wed, 31 Jan 2024, 04:07 Stan Johnson, wrote:
> Just FYI to close this thread.
>
> In an attempt to duplicate the problem on a G4 Cube, I noticed that
> there were no issues with "apt-get upgrade" followed by "apt-get
> dist-upgrade". So I just copied the Cube's rootfs to the G4 PowerMac
> MDD, configured the hostname and network, restored the correct SMP
> kernel and modules, and everything is working.
>
> -Stan
>
> -
>
> On 1/22/24 8:24 AM, Stan Johnson wrote:
> > Hi Adrian,
> >
> > Please see the last paragrraph of the message; it may provide a clue for
> > ppc (and probably ppc64).
> >
> > Mr. Simon McVittie seems to be saying that debian-devel isn't the right
> > place for these kinds of issues, since Debian 8 was the last supported
> > version of Debian for ppc. Perhaps the bug report he submitted will
> > address the problem.
> >
> > -Stan
> >
> > Forwarded Message
> > Subject: Re: X-Windows on PPC in Debian SID
> > Resent-Date: Mon, 22 Jan 2024 10:49:25 + (UTC)
> > Resent-From: debian-de...@lists.debian.org
> > Date: Mon, 22 Jan 2024 10:49:01 +
> > From: Simon McVittie
> > To: debian-de...@lists.debian.org
> >
> > On Sun, 21 Jan 2024 at 17:14:11 -0700, Stan Johnson wrote:
> >> The bottom line is that there appears to be a dependency issue in Debian
> >> SID at the moment
> >
> > You can't *necessarily* draw this conclusion from a failure to
> > upgrade. You are using powerpc, which is a "ports" architecture that is
> > not really part of the Debian release process any more:
> >
> > The last supported release for 32-bit PowerPC is Debian 8 ("jessie").
> > — https://www.debian.org/ports/powerpc/
> >
> > For reference, Debian 8 "jessie" reached end-of-life in 2018.
> >
> > powerpc enthusiasts continue to compile packages from the unstable (sid)
> > rolling release on powerpc, but "ports" architectures are not supported
> by
> > the Debian project as a whole. The mailing list for the big-endian
> powerpc
> > and ppc64 ports (and the little-endian ppc64el architecture, which *is*
> > supported) is debian-powerpc.
> >
> > It is common for potentially large categories of packages to be
> temporarily
> > uninstallable in unstable, particularly in "ports" architectures, and you
> > cannot expect upgrades to go smoothly at all times. I would personally
> > suggest using an interactive apt user interface like aptitude to get a
> > better idea of what depends on what and why.
> >
> >> that makes wdm (and other X-Windows packages such as
> >> the Xorg server) dependent on systemd, even if systemd is already
> >> installed, regardless of whether systemd is being used as the init
> >
> > systemd is the default init system in Debian, and also provides the
> > default implementation of several other important systemd services
> > like logind. If you have chosen not to use systemd, you can expect
> > that you will have to take steps to select other non-default packages
> > (for example dbus-x11 instead of dbus-user-session, and libpam-elogind
> > instead of libpam-systemd). apt will not necessarily be able to do this
> > automatically.
> >
> > Trying this on amd64, it appears that the problem you encountering is
> > probably that libelogind, elogind's partial replacement for libsystemd,
> > does not appear to provide all of the functions required by the current
> > versions of important packages like procps: procps currently requires
> > libsystemd0 (>= 254), and libelogind only provides a replacement for
> > version 252. I've reported this libelogind limitation as a bug in
> > elogind.
> >
> > smcv
> >
>
>
Re: eMac can't boot recent kernels
I am also interested in this grub migration guide. My trusty powerbook is on yaboot, but if grub can't boot OSX or MorphOS I'll need to chain these as i bet yaboot boots grub just fine. Could you link to that script for booting osx under grub? I'll be taking a deep dive into grub anyway I've decided. Two or three of my old workhorse PCs that support some early EFI standard or draft now don't display the grub menu, but works regardless. On Tue, 13 Aug 2024, 13:20 Ed Robbins, wrote: > Thank you Adrian. > > It looks as though this also affects boot in grub as it's a kernel > assertion? > > It seems also that the culprit kernel config option was found in the > thread you linked to. Is some further investigation required to > discover why that config option causes an issue? > > Aside: > I trawled the mailing list archives last night and found the guide to > booting OS X using an openfirmware script when grub is installed [1]. > I still find OS X useful mostly for testing the hardware. > > What is the process to switch from yaboot to grub? Just install grub > and uninstall yaboot? I couldn't find much info on this, though I know > I did it previously on my ibook. > > Thanks, > Ed > > [1] https://lists.debian.org/debian-powerpc/2023/06/msg9.html > > On Tue, 13 Aug 2024 at 06:49, John Paul Adrian Glaubitz > wrote: > > > > Hi, > > > > On Mon, 2024-08-12 at 21:09 +0100, Ed Robbins wrote: > > > ERROR: claim of 0x170 in range 0xc200 0x1000 failed > > > Claim error, can't allocate kernel memory > > > > This is a known issue (see [1]), but I have not had time yet to work on > it. > > > > Adrian > > > > > [1] https://lists.debian.org/debian-powerpc/2024/07/msg1.html > >
Re: Trouble installing on iBook G4
What about installing from a qemu or debbootstrap from a usb/chroot and do some OF trickery? On Sun, 22 Sept 2024, 06:28 Ken Cunningham, wrote: > So the key points are: > > 1. you have to use the special ISO, and no other ISO than the latest > special ISO should be used. Right now, I think you're still waiting to hear > which ISO that is. > 2. Do NOT believe any other website you might find or YouTube channel you > might find that claims to tell you how to install debian on PPC. I have not > yet seen a single one of them that has accurate information on it. > 3. use GRUB. Use all the default settings during the ISO installation. > Forget about triple booting into nine different operating systems. Give > debian the whole disk. > 4. the latest kernel is broken. Don't use it. You have to use the previous > working kernel instead, and not upgrade the kernel until the issue gets > sorted out. > 5. you have to manually install the firmware needed to support the > hardware. It's not installed by default as it is not free, so debian won't > bundle it. > 6. use a plugged-in ethernet connection. WIFI works, eventually, after a > lot of screwing around, but don't bother with it off the start. > 7. getting from the terminal display (text screen) to a whole GUI > graphical interface takes a while. There will be blood. Not all video cards > will work right. You will be sure you have set everything up right, but it > just won't work. And then you will discover some setting that wasn't right, > and it will finally work. And then you should never touch it again :> . > 8. be careful with large updates / upgrades. You can quite easily get > yourself into a situation where you have broken everything, and will see no > good path back to a working state. > > > Ken > > > > > On Sep 21, 2024, at 7:54 PM, Cedar Maxwell wrote: > > Hello Adrian, > > > Yes, I tried booting into the older kernel from GRUB. In fact, attempting > to load the latest kernel from the repositories causes the system to crash > back to Open Firmware, as discussed previously. > > > However, you mentioned that all further discussion should be based on a > known working image. Does the image I used fit this description? > On 9/20/24 01:06, John Paul Adrian Glaubitz wrote: > > I went ahead and tried this one: > https://cdimage.debian.org/cdimage/ports/current/ > > -- > Sincerely, > Cedar Maxwell > > >
Re: Trouble installing on iBook G4
It is very much the problem i will be trying to fix when https://github.com/threader/AltanOS/tree/main/src/ppc/linux_ppc32_deb is in order, it'll take some time to work out everything and clean up all and build from a buildroot or debian build. I'm going to visit the debian-installer soon anyway, it's a matter of planning and execution, im having enough fun toying with thoughts atm, but these problems with booting ppc etc. have been on my mind since i heard about grun for ppc. I see i compiled 'kexecboot' https://github.com/kexecboot/kexecboot but i must have forgotten to upload it or something, i need to verify my XNU stuff before i do much more really. On Sun, 22 Sept 2024, 09:27 Mike, wrote: > What about installing from a qemu or debbootstrap from a usb/chroot and do > some OF trickery? > > On Sun, 22 Sept 2024, 06:28 Ken Cunningham, < > ken.cunningham.web...@gmail.com> wrote: > >> So the key points are: >> >> 1. you have to use the special ISO, and no other ISO than the latest >> special ISO should be used. Right now, I think you're still waiting to hear >> which ISO that is. >> 2. Do NOT believe any other website you might find or YouTube channel you >> might find that claims to tell you how to install debian on PPC. I have not >> yet seen a single one of them that has accurate information on it. >> 3. use GRUB. Use all the default settings during the ISO installation. >> Forget about triple booting into nine different operating systems. Give >> debian the whole disk. >> 4. the latest kernel is broken. Don't use it. You have to use the >> previous working kernel instead, and not upgrade the kernel until the issue >> gets sorted out. >> 5. you have to manually install the firmware needed to support the >> hardware. It's not installed by default as it is not free, so debian won't >> bundle it. >> 6. use a plugged-in ethernet connection. WIFI works, eventually, after a >> lot of screwing around, but don't bother with it off the start. >> 7. getting from the terminal display (text screen) to a whole GUI >> graphical interface takes a while. There will be blood. Not all video cards >> will work right. You will be sure you have set everything up right, but it >> just won't work. And then you will discover some setting that wasn't right, >> and it will finally work. And then you should never touch it again :> . >> 8. be careful with large updates / upgrades. You can quite easily get >> yourself into a situation where you have broken everything, and will see no >> good path back to a working state. >> >> >> Ken >> >> >> >> >> On Sep 21, 2024, at 7:54 PM, Cedar Maxwell wrote: >> >> Hello Adrian, >> >> >> Yes, I tried booting into the older kernel from GRUB. In fact, >> attempting to load the latest kernel from the repositories causes the >> system to crash back to Open Firmware, as discussed previously. >> >> >> However, you mentioned that all further discussion should be based on a >> known working image. Does the image I used fit this description? >> On 9/20/24 01:06, John Paul Adrian Glaubitz wrote: >> >> I went ahead and tried this one: >> https://cdimage.debian.org/cdimage/ports/current/ >> >> -- >> Sincerely, >> Cedar Maxwell >> >> >>
Re: Trouble installing on iBook G4
Keeping an eye on everything 'boot' if anyone has any suggestions or thought's and notes of my notes, please let me know, im apparently trying to solve all the problems i can imagine; https://github.com/threader/AltanOS https://github.com/threader/xnu_gcc_libc_etc_darwin Im especially interested in yaboot configs and grub.confs or whatever, i need both Debian/Netbsd/OSX and would we even bother winnt-ppc much? Still my notes here are dizzying and im chillig and nailing down notes, i mean, no one can kill an open source thought note? No? I basically decided i need to handle booting on ppc anyway, my debian and setup is ancient, and i dont want to jump without gaving ' the foggiest idea' what im doing ' ( i love that commit msg and im going to frame it on my wall or something ;) On Fri, 20 Sept 2024, 08:06 John Paul Adrian Glaubitz, < glaub...@physik.fu-berlin.de> wrote: > Hi, > > On Thu, 2024-09-19 at 22:49 -0500, Cedar Maxwell wrote: > > Great, which one is the proper image? > > Sorry, I was too busy and forgot looking for the recommended image. > > > I went ahead and tried this one: > https://cdimage.debian.org/cdimage/ports/current/ > > > > The Ethernet worked during the installation, but afterwards Debian > doesn't detect it at all. > > You might be missing firmware here that you have to install. > > Did you try booting into an older kernel from GRUB? I would assume that > the installer > installed both the older kernel version from CD as well as the latest > kernel from > the repositories. > > Adrian > > -- > .''`. John Paul Adrian Glaubitz > : :' : Debian Developer > `. `' Physicist > `-GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913 > >
Re: Installing on PowerBook G4 - grub failure
Yo! I'm lost trying to make mental notes of my actual notes, but im om it and wont give 'much rest' till im halfway there, my best suggestion is to use OpenFirmware 'truckery' and boot grub to boot the proper ISO or env? On Tue, 24 Sep 2024, 13:42 Mark Cave-Ayland, wrote: > On 24/09/2024 10:21, Steffen Grunewald wrote: > > > On Tue, 2024-09-24 at 09:43:39 +0100, Mark Cave-Ayland wrote: > >> > >> FWIW I did my last install on a G4 Mac Mini using a USB stick running > >> Adrian's latest (at the time) image. I used dd to copy the netinst > image to > >> a USB stick, used "show-devs" to locate the USB drive, and then tweaked > the > >> cd alias to point to it: > >> > >>devalias cd /pci@f200 > >>boot cd:,\\:tbxi (I think?) > > > > Worth an attempt, but results in > > > > load-size=0 adler32=1 > > > > LOAD_SIZE is too small > > > > (same respronse if I try other "file" names) > > > > There are hints in some forum threads that the A1095 indeed may be one of > > few (perhaps the only?) G4 models that cannot boot from USB. > > BootROM version is 4.8.6f0, fwiw. > > That's frustrating. Can you confirm which partition contains the main > filesystem by > testing each number from 0 to 9 e.g. > > dir usb0:0,\ > dir usb0:1,\ > dir usb0:2,\ > ... > ... > dir usb0:9,\ > > > ATB, > > Mark. > >
Request: Success reports of using quik to boot an OldWorld Mac
I'll start first: Success: 7300 Failed: 7200 beige g3
[Fwd: It works! (Is: Replacing kernel on boot floppy) (Was Re: Problem installing debian on G3 beige)]
Hi, Thanks to Manea for posting the information below a few weeks ago, everything works great. Now I want to replace the kernel on the floppy, but I get a penguin while it reads, then a red X over the little penguin. Which file should I use? arch/ppc/boot/zvmlinux arch/ppc/boot/zImage arch/ppc/coffboot/vmlinux.gz arch/ppc/coffboot/vmlinux.coff arch/ppc/coffboot/zImage arch/ppc/coffboot/vmlinux.elf arch/ppc/coffboot/miboot.image vmlinux I hope it's not the last one, since it is 2 MB and won't fit on the floppy :( Original Message Subject: It works! (Was Re: Problem installing debian on G3 beige) Resent-Date: Sun, 07 Jan 2001 02:31:04 -0800 Resent-From: debian-powerpc@lists.debian.org Date: Sun, 7 Jan 2001 11:30:22 +0100 From: Manea Mirko <[EMAIL PROTECTED]> To: Ethan Benson <[EMAIL PROTECTED]> CC: debian-powerpc@lists.debian.org Hi! thanks to the help of Ethan I succeeded in creating a boot floppy. The steps are the following: 1. get boot-floppy-hfs.img and dd it to a floppy; 2. test if you can boot (it will prompt for a ramdisk). I had to use an old boot-floppy-hfs.img (with kernel 2.2.17) because latest (with 2.2.18pre21, I think) locks the keyboard; 3. get boot-floppies sources (from http://packages.debian.org/stable/admin/boot-floppies.html) and untar it; 4. cd to boot-floppies-2.2.17/powerpc-specials/miBoot; 5. with a hex editor (like khexedit or beav) edit System.bin; 6. near the end (~00012d00) you'll find: root=0200 load_ramdisk=1 prompt_ramdisk=1 adb_buttons=100,111 video=atyfb:vmode:14,cmode:16,mclk:63 where root=XXYY, ie XX major and YY minor number of the device to mount as root (/dev/sda2 -> root=0802, /dev/sda3 -> root=0803, look in /dev/); 7. set load_ramdisk=0 and prompt_ramdisk=0; 8. insert boot-floppy-hfs.img and do a hmount /dev/fd0; 9. hcopy System.bin : 10. humount; Now you've got a boot floppy which mounts your root filesytem :) For me it works! I've just finished installing the base system, but when the installer asked for type of package selection (simple or advanced) I selected simple and then it did a reboot (clean shutdown). Maybe there is some problems in the installer... but now debian boots and I'm going to use dselect and apt.. Thanks to everybody, Mirko On Sun, Jan 07, 2001 at 12:37:00AM -0900, Ethan Benson wrote: > On Sun, Jan 07, 2001 at 10:23:40AM +0100, Manea Mirko wrote: > > On Sun, Jan 07, 2001 at 12:03:10AM -0900, Ethan Benson wrote: > > > On Sun, Jan 07, 2001 at 09:43:09AM +0100, Manea Mirko wrote: > > > > > > > > Do you know if I can use it to mount my root partition (/dev/sda3) > > > > instead > > > > of a ramdisk? > > > > > > yes you should be able to, however i don't know if miboot supports any > > > kind of interaction, i don't believe it does. so you have to do one > > > of two things: > > > > > > make sure you compile the kernel with the correct embedded default > > > root device, this will happen automatically if you compile on the > > > target system, the Makefile will embedd the current root device into > > > the kernel (rdev is missing on powerpc for some reason) > > > > > > edit miboot.bin to hand the correct root= argument to the kernel, you > > > may have to do this already as it might have a initrd arguments. you > > > must do this with a hexeditor, its not as bad as it sounds, you just > > > need to find the spot where the kernel arguments are and change them. > > > (i know freaky, miboot is not what you would call a general purpose > > > bootloader) > > > > > > > sorry, but did you mean System.bin ? I cannot find miboot.bin in the > > boot-floppies-2.2.17 sources. > > ah yes, sorry. > > -- > Ethan Benson > http://www.alaska.net/~erbenson/ -- Best Regards, Mirko Manea <[EMAIL PROTECTED]> URL: http://www.mami.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Which file is kernel image?
Which file should I use? arch/ppc/boot/zvmlinux arch/ppc/boot/zImage arch/ppc/coffboot/vmlinux.gz arch/ppc/coffboot/vmlinux.coff arch/ppc/coffboot/zImage arch/ppc/coffboot/vmlinux.elf arch/ppc/coffboot/miboot.image vmlinux I hope it's not the last one, since it is 2 MB and won't fit on the miBoot floppy :( Mike
test
Is the server dead or my subscription?
[Fwd: It works! (Is: Replacing kernel on boot floppy) (Was Re: Problem installing debian on G3 beige)]
Hi, Thanks to Manea for posting the information below a few weeks ago, everything works great. Now I want to replace the kernel on the floppy, but I get a penguin while it reads, then a red X over the little penguin. Which file should I use? arch/ppc/boot/zvmlinux arch/ppc/boot/zImage arch/ppc/coffboot/vmlinux.gz arch/ppc/coffboot/vmlinux.coff arch/ppc/coffboot/zImage arch/ppc/coffboot/vmlinux.elf arch/ppc/coffboot/miboot.image vmlinux I hope it's not the last one, since it is 2 MB and won't fit on the floppy :( Original Message Subject: It works! (Was Re: Problem installing debian on G3 beige) Resent-Date: Sun, 07 Jan 2001 02:31:04 -0800 Resent-From: debian-powerpc@lists.debian.org Date: Sun, 7 Jan 2001 11:30:22 +0100 From: Manea Mirko <[EMAIL PROTECTED]> To: Ethan Benson <[EMAIL PROTECTED]> CC: debian-powerpc@lists.debian.org Hi! thanks to the help of Ethan I succeeded in creating a boot floppy. The steps are the following: 1. get boot-floppy-hfs.img and dd it to a floppy; 2. test if you can boot (it will prompt for a ramdisk). I had to use an old boot-floppy-hfs.img (with kernel 2.2.17) because latest (with 2.2.18pre21, I think) locks the keyboard; 3. get boot-floppies sources (from http://packages.debian.org/stable/admin/boot-floppies.html) and untar it; 4. cd to boot-floppies-2.2.17/powerpc-specials/miBoot; 5. with a hex editor (like khexedit or beav) edit System.bin; 6. near the end (~00012d00) you'll find: root=0200 load_ramdisk=1 prompt_ramdisk=1 adb_buttons=100,111 video=atyfb:vmode:14,cmode:16,mclk:63 where root=XXYY, ie XX major and YY minor number of the device to mount as root (/dev/sda2 -> root=0802, /dev/sda3 -> root=0803, look in /dev/); 7. set load_ramdisk=0 and prompt_ramdisk=0; 8. insert boot-floppy-hfs.img and do a hmount /dev/fd0; 9. hcopy System.bin : 10. humount; Now you've got a boot floppy which mounts your root filesytem :) For me it works! I've just finished installing the base system, but when the installer asked for type of package selection (simple or advanced) I selected simple and then it did a reboot (clean shutdown). Maybe there is some problems in the installer... but now debian boots and I'm going to use dselect and apt.. Thanks to everybody, Mirko On Sun, Jan 07, 2001 at 12:37:00AM -0900, Ethan Benson wrote: > On Sun, Jan 07, 2001 at 10:23:40AM +0100, Manea Mirko wrote: > > On Sun, Jan 07, 2001 at 12:03:10AM -0900, Ethan Benson wrote: > > > On Sun, Jan 07, 2001 at 09:43:09AM +0100, Manea Mirko wrote: > > > > > > > > Do you know if I can use it to mount my root partition (/dev/sda3) > > > > instead > > > > of a ramdisk? > > > > > > yes you should be able to, however i don't know if miboot supports any > > > kind of interaction, i don't believe it does. so you have to do one > > > of two things: > > > > > > make sure you compile the kernel with the correct embedded default > > > root device, this will happen automatically if you compile on the > > > target system, the Makefile will embedd the current root device into > > > the kernel (rdev is missing on powerpc for some reason) > > > > > > edit miboot.bin to hand the correct root= argument to the kernel, you > > > may have to do this already as it might have a initrd arguments. you > > > must do this with a hexeditor, its not as bad as it sounds, you just > > > need to find the spot where the kernel arguments are and change them. > > > (i know freaky, miboot is not what you would call a general purpose > > > bootloader) > > > > > > > sorry, but did you mean System.bin ? I cannot find miboot.bin in the > > boot-floppies-2.2.17 sources. > > ah yes, sorry. > > -- > Ethan Benson > http://www.alaska.net/~erbenson/ -- Best Regards, Mirko Manea <[EMAIL PROTECTED]> URL: http://www.mami.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: fdisk kernel panic
Bruce McIntyre wrote: > > I am continuing to have frustration with my internal hard disk (7300, > Quantum Fireball 2.1) I decided to this time mash Apple OS > altogether, and so gave mac-fdisk the 'i' command to write a new > patition map. I was suspecting that the apple patch and driver 43 > patitions were causing the scsi error 8000 that I was getting > with my previous installs. The initialisation went fine (fdisk > prompts you to enter the block size of the disk, which I got from > first giving it the 'p' command---it returns a size 1 block less than > that given by dmesg). I wrote in some partitions. 32M Root, 32M swap, > 100M var, 500M home and the rest usr. when I gave mac-fdisk the 'w' > command (to write the new patition table) it freaked out and gave me > something like (copied by hand from screen): > Kernel access of bad area pc c0206808 lr c0205a70 address 4 task mac-fdisk 44 > it then panicked and all consoles froze as it attempted reboot (which > never eventuated). > > Now I cannot reinstall a working Mac system on the disk to start the > debian tranistion--- it formats ok with the apple Drive Setup > utillity, but refuses to make the disk bootable with the startup disk > control panel (system writes to it ok in the install program). > while the startup control panel indicates it is the boot disk, at > boot i dont get anything, only grey... not even the disk icon. > to boot again from cd, i have to erase the pram, and then reboot with > c and d held down. after a minute or two of indecision it boots, and > mounts the drive fine, with the system on it and still indicating > that it is the boot disk. > The drive is on the mesh bus at id 1. (I have been trying different > ids to get debian without scsi errors to no avail) > Is my computer cursed? > Probably not. I have a 7300 and it's a good machine, but it's hard drive is starting to go bad on one sector. I'm going to have to get a new drive. You might try that too. Here's some other things to try: Remove and reinstall the memory modules and cpu board Unplug each connector and reinstall them; like the scsi cables, power cables, etc) Check the power output of the power supply with a voltage meter (buy a power extender and cut and strip the cable to attach the meter clips) Sorry, don't know which wires to check... Kicking it a couple times might help... ;) Hope this helps, Mike
Re: OF setting for Quik booting in a 7200
Leandro Dutra wrote: > > Could I please have your help? I have three Macs which I intend to > use with Debian. I'm experienced in IBM PC-compatibles installations but > never succeeded in a RISC one - have already tried a Digital Alpha and a > beige G3. > > Now I have a 7200/75 from which I eliminated any HFS traces thru the > i command in mac-fdisk. I booted it with diskettes (the ones with kernel > 2.2.17) and performed the installation up to almost the end. Then I've > tryed nvsetenv but it failed not finding /dev/nvram. > > Any ideas? Also, if there are docs besides the ones in > http://debian.org./ and the CD I would like to know. > > TIA! > > --_ > / \ Leandro Guimarães Faria Corsetti Dutra +55 (11) 246 96 07 resl > \ / Amdocs Brasil Ltda, Sao Paulo +55 (11) 3040 8913 coml > X http://terravista.pt./Enseada/1989/ mailto:[EMAIL PROTECTED] > / \ Campanha fita ASCII, contra correio HTMLBRASIL I've spent the last month trying to get QUIK to work on a 7200, if you want to get it working in a reasonable time, make a 50MB hfs partition install MacOS 8.5 and bootX. It'll make the easiest boot loader I've ever worked with, 'cept for lilo. Mike
Re: OF setting for Quik booting in a 7200
Ethan Benson wrote: > > what kernel? the only problems a friend of mine has had is 2.2.15 and > 2.2.16 14 and 17 work great with quik. you need to compile with > CONFIG_BOOTX=n though. (perhaps not on 2.2.18, im (rather he is) not > going to test it though) > Kernel 2.2.18pre21 from the base2_2.tgz as installed by debian. OF wouldn't even load the first stage of Quik. I gave it the right path, put it kept failing "cannot open ..." This was on a Beige g3 and 7200 75Mhz. I haven't tried the 90 or 120 Mhz 7200s I have. Although, 50 MB less and 10 seconds more and I get linux through BootX. Does bootX have trouble with MacOS 8.0? It hangs with a black screen when you try to load kernel during extention loading. If you wait until the desktop is loaded, it works fine. 8.5 doesn't have this problem though.
Re: help, please
Jeremiah Merkl wrote: That's not completely true. Worst choice would be somewhere near a 5400/180, or perhaps a 7200/75. :) In order to get Debian (or any other flavour) to run on an oldworld Mac, there's a few more hoops you need to jump through. True, I have several 7200s 75 90 and 120 Mhz. the 75 wouldn't work very well with BootX, but the 90 and 120 are fine. If could be the machine, I'll have to do more testing. First off, if you can live with an extra 200 meg partition to hold MacOS, then you can use BootX to start linux...if you don't like that, then you've I can get MacOS 8.5 into 35MB with networking. Doesn't leave room for kernels and bootX and kernels, so that's why I use a 50MB partition. Just do a custom MacOS install with: base system cd/dvd driver built-in ethernet appleshare opentransport mouse control panel memory control panel Let it install, reboot and turn off virtual memory. Now you have room for bootx and friends. got to go with quik, which can be harder to deal with. I'm currently using bootx on an beige-G3 for simplicity. As often as I actually need to reboot the blasted thing, I don't really care, and quik just plain scares me. :) I've learned a lot about Open Firmware from my Quik experiences, and was able to get it to work on a 7300. I failed with a 7200 and beige g3. YMMV Basically...for the initial install, you need to use bootx anyways, AFAIK. Download it, plus the boot/root disks, and the base-xxx tarball. The rest of the install directions are either on the debian website, or this list's archives. You can boot from hfs floppy, but if you're going to use bootx anyway, you might as well install from it too. I've left the debian kernel, and ramdisk.tar.gz on my systems in case I need to repair my current installation. I've found that the stripped MacOS partition is handy anyways, cuz if you manage to complete screw up your networking, you can at least reboot, download packages/patches to the MacOS drive, and then startup in Linux again afterwards. Handier than you might think. True, but it depends on how familiar you are with Linux/Unix. Mike
Re: help
Ethan Benson wrote: 2) debian CDs are not and won't be bootable on OldWorld hardware in the forseeable future. (until someone writes an Oldworld bootloader that does not require proprietary Apple drivers) Why? We have miBoot. Won't that work on a CD? OK, call me ignorant, but it would be cool of installing linux on a Mac was as easy as MacOS is to get started. Pop in MacOS 8.0 in a m68k to a 6500 and you're set, 8.5 or higher for newer macs. I've been using linux on x86 for a couple years, and at the very worst, you need a floppy, and an ethernet/modem to install the base system. miBoot works on hard drives, right? Maybe you can point me at some docs... Mike
External Modems
Hi, I have a bunch of external mac modems that use the printer/modem port on a mac. I'm wondering if I can use any of them for a fax/modem server. Are they supported by linux? Do I just need to plug it in and type "AT" to test or does it use some other protocol? Here is a partial list of what I have: Global Village - Teleport Platinum A812 Global Village - Teleport 65 Fax/modem A824 Diamond - SupraExpress 56e Sp SUP2110 TIA Mike
Re: 7200/90 -- next step
Matthew Kirkwood wrote: > > On Sun, 11 Feb 2001, Andrew Sharp wrote: > > > Stuffit Expander can be downloaded free off the internet. I forget > > the URL, but it is easily found with google search ~:^) > > And if, as I explained, I can't get TCP working, what do > I try? > What exactly happens when you try to setup tcp/ip? You go to the control panel and what happens? What extentions do you have? What MacOS version? That's the kinda stuff you need to tell us in your first post. This'll give you a final answer faster. Mike
Re: External Modems
Leandro Dutra wrote: > > > From: Mike Fedyk [mailto:[EMAIL PROTECTED] > > Are they supported by Linux? Do I just need to plug it in > > and type "AT" to test > > or does it use some other protocol? > > Yes, all modems are Hayes compatible. Fax/modem software will > usually stick to the standard AT Hayes commands. > Great Do you know if I'll be able to chain any modems together? Using one modem port for two modems... Is that possible? I saw an UPS that chained it's connection to a bunch of macs. Maybe that's just for one way broadcast... TIA Mike
Re: moving from linuxppc - newbie needs help please
sobrien wrote: > will be able to use my exisiting partitions (i don't see why not) and if > i will still be able to use BootX, or will i have to use yaboot? i > really want to be able to use apt, i hate redhat's package manager. > download the linux file, and put it in "linux kernels". get ramdisk.tar.gz and put it in "system folder" boot from bootx with linux kernel and ramdisk get the base2_2.tgz, and know how to get to it from linux Don't know about installing over linuxppc > my equipment > RevB iMac 233 > 64M ram > 2 gigs for Mac 8.5 > 2 for debian (right now linuxppc) > > i am downloading the .iso now, but the download has failed twice. can > anyone suggest the best mirror? will i be ok with just the first binary > disc? > don't get the iso, mirror locally, or just download packages as needed. you can mirror and then build iso from that with kit. Mike
FTE problems on console of Pmac g3 beige
I get colorful gibberish instead of a working fte window. Is there any way to fix this? It has an ATI Mach64. Mike
IRC Channels for PPC Linux
Hi, I'm looking for a couple channels to visit where we PPC guys hang out. I've been to #debian or irc.debian.org and #linux on irc.undernet.org but there aren't many PPC users there if any. Where do you go? Mike
Re: FTE problems on console of Pmac g3 beige
Michel Dänzer wrote: > > Mike Fedyk wrote: > > > > I get colorful gibberish instead of a working fte window. > > > > Is there any way to fix this? It has an ATI Mach64. > > Please share a bit of needed information (fbset -i should give most). > Here you go mode "1024x768-75" # D: 78.358 MHz, H: 59.724 kHz, V: 74.655 Hz geometry 1024 768 1024 2032 8 timings 12762 160 32 28 1 96 3 hsync high vsync high accel true rgba 8/0,8/0,8/0,0/0 endmode Frame buffer device information: Name: ATY Mach64 Address : 0x8280 Size: 2093056 Type: PACKED PIXELS Visual : PSEUDOCOLOR XPanStep: 8 YPanStep: 1 YWrapStep : 0 LineLength : 1024 MMIO Address: 0x827ff800 MMIO Size : 2048 Accelerator : ATI Mach64GT It's a Beige G3 266 2.2.18 and 2.2.19pre2 have trouble Mike
Re: Pushing up daisies...
Rob Andrews wrote: > > In newsgate.debian.powerpc, Leandro Dutra wrote: > > Maybe I'm naive, but why then your disk drive and CD would spin? > > Aren't they spinned by OpenFirware, or just by having power applied to = > > them? > > AFAIK, they're only spinning up because they have power. > > > I wouldn't think you could alter the entry point for Open Firmware, but > > that you could alter the initialization process after Open Firmware has > > started. > > Could very well be completely wrong, judging from what I read here: > http://bananajr6000.apple.com/1275/proposals/Closed/Withdrawn/315-it.txt > "Common Hardware Reference Platform (CHRP)" Does that mean that the open specs from IBM have hardware made from them, or is this something else? Mike
kernel: mace: lost a status word
I'm getting this on a 7200/90, I haven't checked to see if it's on my other 7200/90 and /120s. I am getting this on 2.2.17 that came with debian. Is this a driver bug or hardware failure? TIA Mike
Re: FTE problems on console of Pmac g3 beige
Mike Fedyk wrote: > > I get colorful gibberish instead of a working fte window. > > Is there any way to fix this? It has an ATI Mach64. > > Mike Oh, well, I'm using jed, and trying out a couple other editors. Something had to make me give the others a try :)
"CONFIG_BOOTX_TEXT=y" in default debian ppc kernel
Look in /boot/vmlinux-2.2.17, you'll see. Ethan, what kind of machines need CONFIG_BOOTX_TEXT=n with quik? What benefit do you get by enabling this option? I've compiled kernels for my g3 with CONFIG_BOOTX_TEXT=n and using bootx without trouble. I've heard that this option does more than just give you more messages. Am I right? Mike
Re: OF setting for Quik booting in a 7200
Ethan Benson wrote: > > On Tue, Feb 06, 2001 at 02:10:03PM -0800, Mike Fedyk wrote: > > > > I've spent the last month trying to get QUIK to work on a 7200, if you want > > to > > get it working in a reasonable time, make a 50MB hfs partition install > > MacOS 8.5 > > and bootX. It'll make the easiest boot loader I've ever worked with, 'cept > > for > > lilo. > > what kernel? the only problems a friend of mine has had is 2.2.15 and > 2.2.16 14 and 17 work great with quik. you need to compile with > CONFIG_BOOTX=n though. (perhaps not on 2.2.18, im (rather he is) not > going to test it though) > All attempts have been made on a stock debian ppc kernel, which I have recently found has CONFIG_BOOTX_TEXT=y :( I'll give it another try next in the week after next, and try setting the boot-device using the SCSI ID. ... Hmm, on one of my 7200s /dev/sda is on mesh id 0 Oh, well. I'll compile a 2.2.18 or 19pre2 and test it with quik. Mike
Re: OF setting for Quik booting in a 7200
Andrew Sharp wrote: > > Gee, quik works just fine on my 7200. But the OF is so lame that it doesn't > really matter whether quik works or not. The two reliable methods I use for > the 7200 are a boot floppy (the one from the CD with the editing changes > suggested by another lister to boot using my hard drive's root partition) or > using BootX from MacOS, which very easily allows me to use several different > kernels. The later was harder this time around because I had to install > macos from scratch, and all I have is this 7.5.2 CD, and you don't get much > when install that! > > Hmm, I can fit MacOS 8.5 in less than 35MB with a custom install. 50MB partition, bootx and virtual memory disabled in MacOS and I'm set. I keep BootX, kernel and debian ramdisk on a netatalk server and copy it over to each comp I'm installing linux on. > > http://www.aladdinnsys.com/expander/ > > Use their FTP site for hassle free downloading. Netscape wanted to slaughter > the download file otherwise no matter what I did. > Just use wget... > One weird thing is that booting with quik on the 7500/7600 was always hit or > miss, 90% miss actually, because the lame OF just didn't understand that the > drive takes time to get ready. Even the various workarounds suggested on the > nvsetenv man page and on this list didn't work. > Did you try this? nvsetenv boot-command "begin ['] boot catch 1000 ms cr again" It works well on my 7300, which BTW is the only machine that I've had QUIK work on. It's hard drive is going bad, so the week after next I'll work on it and try quik again. > So I made a special boot floppy as mentioned above. Now quik works every time! > > I can't use the floppy on several machines because I need to make custom kernels and they won't fit on the floppy with the other utils. I'm sticking with BootX for a while now, because I don't want to switch to Quik and have it bite me in the ass when I need to use the latest kernel on my firewall... Mike
Re: OF setting for Quik booting in a 7200
Ethan Benson wrote: > > > > On Tue, Feb 06, 2001 at 02:10:03PM -0800, Mike Fedyk wrote: > > ... Hmm, on one of my 7200s /dev/sda is on mesh id 0 > > the standard apple setup is the hard disk is id 0 cdrom id 3. > > > Oh, well. I'll compile a 2.2.18 or 19pre2 and test it with quik. > > if you use 2.2.19 use pre13 or later, earlier versions didn't compile > on powerpc. pre13 has all the needed patches to fix all the powerpc > issues that cropped up. > Why was I able to compile 2.2.19pre2 and run 10 bonnie benchmarks at 256MB each? This failed on 2.2.18. pre3 and up had a lot of patches added that I didn't want to have to deal with. Specifically, I needed to fix the "VM: do_try_to_free_pages" problem in 2.2.18. Andrea's VM global was reported to fix it, and that was first in 2.2.19pre2. I'm also working on getting FreeS/Wan on this comp and they suggest 2.2.17, so I'd want the closest version to that I can get, and will handle the load. Maybe quik had trouble with them, but I'm using bootx. Mike
Re: "CONFIG_BOOTX_TEXT=y" in default debian ppc kernel
Takehiko Abe wrote: > > Ethan Benson wrote: > > > > > Ethan, what kind of machines need CONFIG_BOOTX_TEXT=n with quik? > > > > > > At least the 7200, I believe. > > > > correct, i talked to ben about this recently the problem is the way > > this particular feature trys to access the OF framebuffer, on > > oldworlds with non-existant or very broken OF video drivers this code > > would crash OF. > > This seems to be exactly what happened to my system (kernel 2.2.18/ > oldworld)--it didn't boot without the monitor attached. Upon reading What model?
Re: "CONFIG_BOOTX_TEXT=y" in default debian ppc kernel
Ethan Benson wrote: > > On Thu, Feb 15, 2001 at 09:21:48PM -0500, Daniel Jacobowitz wrote: > > On Thu, Feb 15, 2001 at 06:15:31PM -0800, Mike Fedyk wrote: > > > Look in /boot/vmlinux-2.2.17, you'll see. > > > > > > Ethan, what kind of machines need CONFIG_BOOTX_TEXT=n with quik? > > > > At least the 7200, I believe. > > correct, i talked to ben about this recently the problem is the way > this particular feature trys to access the OF framebuffer, on > oldworlds with non-existant or very broken OF video drivers this code > would crash OF. he told me this should be fixed in recent kernels but > i have not tested this. (i don't intend to since the only oldworld i ... > BOOTX_TEXT is a misnomor though, it really doesn't have much to do > with BootX at all. its also not required for either BootX or > non-BootX booting. > Why do we need BOOTX_TEXT in the debian kernel? Is the installer using the debugger? Are these early messages also reported do klogd? If so, I can only see six lines that show up before I get a display on my mac. System.map ... symbols ... device tree ... How does this really help? Does the maintainer use xmon to test anything before it is released? If so, I could understand him wanting to use the exact kernel he had his debugger on... This is obviously creating problems. What's the big reason to have this enabled? Mike
"PCI quirks" in kernel for ppc
Does this help for ppc? The help talks about BIOS which I know is only on x86. Does this code include anything that helps a non x86 comp? Mike
Re: "CONFIG_BOOTX_TEXT=y" in default debian ppc kernel
Daniel Jacobowitz wrote: > > Please, check the current release before you start discussions like > this! The current kernel packages have not had this enabled in months. > > Dan Is this the same one that wouldn't boot without disabling the "Please enter the root disk" prompt? I'm glad I mirrored before the "good" one was replaced. Is was based on 2.2.17. I'm glad that newer ones don't have this enabled anymore. I'll have to give the new one a try. Mike
Re: PowerMac G3 not booting after installation
jeramy b smith wrote: > > "Mohamed Abdulbaset" <[EMAIL PROTECTED]> wrote: > > > > greetings, > > > > i have downloaded the 3 binary CD images for the potato version and write > > them my self to CD's i boot from the firest CD and i went through > > installation steps till end but my > > system not rebooting from the harddisk. > > my computer is Power Mac G3, > > > > PLS. Assist me. > > Beige or Blue&White? I have some experience with the beige > booting directly via quik. > Really? I have a Beige g3 that I haven't been able to boot with Quik. Can you list what you had to do? Every time OF couldn't "read the disk" I don't remember the exact message. Mike
Re: PowerMac G3 not booting after installation
Ethan Benson wrote: > > On Mon, Feb 19, 2001 at 05:25:36PM -0800, Mike Fedyk wrote: > > > Really? > > http://penguinppc.org is said box. > > > I have a Beige g3 that I haven't been able to boot with Quik. > > > > Can you list what you had to do? > > set the load-base to 60 (remember OLDWORLD ONLY do NOT do this to > a newworld) > > we also applied the OF patches from darwin, you should be able to set > these from linux from a text file which i can send you if you want > it. (i can send you the entire OF configuration also) Yes, I'd like that very much. Are there patches for other OF versions? I seem to remember seeing something that would patch OF 1.0.5. > > > Every time OF couldn't "read the disk" I don't remember the exact message. > > you probably need the darwin patches, OF can't read disks very well > without them on this machine. > Hmm... are there any web sites that list quirks with OF with different mac computers? > -- > Ethan Benson Mike Fedyk
Re: getting kernel 2.4.2-pre3 to work
Ethan Benson wrote: > > On Tue, Feb 20, 2001 at 12:57:38PM -0800, Grant Miller wrote: > > $ mkdir linux-pmac-stable > > $ cd linux-pmac-stable > > $ rsync -avz --delete ppc.linuxcare.com::linux-pmac-devel . > > penguinppc.org also carries up to date trees. Which tree for 2.4 is the best to run a working system? It looks like the external trees have been closed for 2.2 on ppc. Am I right? > > you don't need to move the source to /usr/src/linux it can be > anywhere. /usr/local/src is really a better place. > > reiserfs is totally broken on big endian archetectures, which means no > reiserfs for powerpc or sparc. (it also used to be 64 bit dirty which > ruled out alpha, that might be fixed though.) see penguinppc.org for > jeffm's work on rewriting reiserfs to be portable. > -- > Ethan Benson I was looking at the reiserfs web site yesterday, and it said that there were reports of their FS working on x86 and ppc. Was there an old version that didn't have endian problems? Is this also true for the patch for 2.2 kernels? Mike
Re: getting kernel 2.4.2-pre3 to work
Cort Dougan wrote: > > } but for 2.4 i imagine paulus tree is best. the bk tree is bleeding > > Quite the opposite. The 2.4 and 2.2 BK kernels are quite stable. The [snip] > If you want stable 2.4 or 2.2, use the BK trees. > OK, let's get this straight. Which is it? Mike
Re: Apache
> Lorenzo De Vito wrote: > > Do you know where I can find Apache (complete) ?? What do you mean exactly? You can apt-get install apache apache-doc and get the latest version of apache in debian. You can go to www.apache.org to download the source. A search on google.com will give you faster results than asking the mailing list... You might try that next time. Mike
Floppy connections in Macs
Hi, I took apart my mac to change the hard drive, and I don't think I connected the floppy correctly. Doesn't the floppy drive require a power connection, or does it get it's power from the data cable? Mike
Re: BootX and the Stuffit Fiasco.
Ethan Benson wrote: > > because the self extracting archive in and of itself must be encoded > since the MacOS uses a exectuable format that cannot survive on > anything but the HFS filesystem. since its encoded you need stuffit > expander to extract the self extracting archive. What about the apple single format? IIRC, it includes the Res fork that MacOS needs to execute. Is MacOS able to execute these files, or does it have to be converted to appledouble format first? > IMO BootX should be recursivly MacBinarized, then put into a .tar.gz > archive. then any version of stuffit can extract it, along with Free > tools available for MacOS. > What are the obstacles to do this? > -- > Ethan Benson > http://www.alaska.net/~erbenson/ > Mike
BootX + zvmlinux
Is there a way to get BootX to work with compressed kernels? BTW, zvmlinux is from `make zImage` and in arch/ppc/boot
Re: BootX + zvmlinux
Benjamin Herrenschmidt wrote: > > >Is there a way to get BootX to work with compressed kernels? > > > >BTW, zvmlinux is from `make zImage` and in arch/ppc/boot > > I've not looked at that bit of source for long, but it's possible that > the special miboot image works with BootX. You can also try compressing a > plain vmlinux with gzip -f9. The zImage won't work as it's embedded > decompressor will not work when called from bootx. > > ben. Sorry, no good. It wouldn't even show on the menu. :( So, BootX doesn't "execute" the kernel, is that why the embedded decompressor doesn't work? Probably just copy the image to memory and a stack pointer? Mike
Re: Debian PC and PPC
Lorenzo De Vito wrote: > > I'm a new user of Linux on Powerpc , which kind differences I can find with > Linux on Intel platforms, all packages I've used on PC I can use on my G3, > how can I find compatible software? If I want update my Kernel I must use a > specific for LinuxPPC ? Excuse me for my english and thank you. Most of what you want to find is source code. PPClinux is such a small userbase compared to i386 that you won't find very many binaries from commercial companies. Even the source may not work if they have endian problems with their code. In other words, it won't compile for sparc m68k or arm. At the moment, you can download 2.2.18 directly from kernel.org and compile. 2.4 needs much work and it takes a while for the updated code to get to linus. If you look in the archive, we have already talked about a 2.4 tree available. You will find that a lot of x86 hardware works on the PPC. Ethernet cards, scsi, ide, etc except for video cards. They have more trouble if you can get them to work at all. You're used to 16 interrupts on x86, I have 64 on my g3 :) Mike
Re: Sustaining RedHat [was: BootX and the Stuffit Fiasco.]
Ethan Benson wrote: > > On Fri, Feb 23, 2001 at 11:05:38AM +0100, Michael Schmitz wrote: > > > > getting a free (as in speech) StuffIt packager for Debian are practically > > nil as you pointed out rather convincingly. > > it will happen either when: > > 1) hell freezes over. > 2) redhat (or someone) buys aladdin and GPLs all thier code. How is RedHat going to buy any company if the trends keep going the way they are? Redhat always ends up having more things "wrong" with it, maybe because it will help convince people to buy support from them... The community is always kicking them in the leg, how are they going to make a profit with bad publicity all of the time? Though, I agree that some of their decisions (a "value-added" gcc, just to have a newer version...) have been bad, I wonder if they will be around in a few years. If so, we'll be at "Linux 15" or so... How will buying aladdin help redhat anyway? Mike
Re: Debian on nubus-ppc
Michael Schmitz wrote: > > > Dont know if any of you on this list would be interested, > > but I have debian and a modified 2.4 kernel installed on my > > old nubus machine (7100/66AV). Runs great with stable and > > testing. If anyone is interested I can post the full > > instructions. > > Either post them, or send me a copy by PM. I might need to install at > least a basic system on a Nubus box to play with the 539Cx SCSI driver. > > Michael Yes! I have a couple 5300 laptops that I need to get working with linux. They start slow as hell on MacOS. Does this mean that there is nubus support in stock 2.4.x? Mike
Email client and conversion from Netscape Mail on win32
Hi, I am not on the debian-user list, so please cc me as well as the list... I have been using netscape mail for several years on winblows, and have over 50 folders and even more filters. I have been getting about 10,000 messages a month, and now that I'm a LKML reader, it's gone even higher. So, threading support is very important to me. So is mail filtering, weather it's with a separate package or not doesn't matter. Now, here's the problem. I want to convert my email activities over to linux. So, I may as well drop netscape at the same time, if I can. I would like to be able to read html mail also. I've looked at a couple in the last half hour, and here's what I've come across in potato-r2. elm+me: this looks pretty good, and it seems to have html viewing. Though, I didn't see any threading features. mutt: didn't look very much, so don't know. postilion: interesting, it seems to support several folders, but I didn't take the time to really figure out how to work with the three rows. balsa: looks workable, but not for my email load. I'm sure I missed a lot about them, maybe you can point some out. TIA, Mike
Re: Database port
. > 2. Our expertise is MD databases, we need DebianPPC help. > When you seek help, please ask specific questions, and not broad "why can't I login" type questions. That will give you a better response. > 4. Within the MD-Linux source code, the result of taking Linux "wheat" > and MD "water" and "baking" that into "bread", the proprietary MD source > will be neither recognizable nor usable. > > 5. MD-Linux will be an Open Source product under the same license as > Debian. > > 7. MD-Linux Scientific is envisioned as a non-profit that will have > education and research as its major objective and its "distribution" > will be geared toward educational institutions giving them the source > code that will enable them to teach the principle of multi-dimensional > databases. What part? You said above that it is proprietary and no source code would be available and unusable if viewed. > > 8. I am a systems architect, an interface designer/ergonomist. I am > not a programmer. > > Henry Keultjes > MD-Linux Scientific > Mansfield Ohio > Voice 419-525- > Your system should be an addon package set, just like kde or gnome or most any large project. You can take an existing window manager and modify it so that it shows your logo in the titlebar of each window, or somesuch. You would have to release any changes you have made to the code, but not the logo or explicit text... read the license. With the window manager in place, you will be able to give a much more "integrated" look and feel. Your packages should be compiled to the new stable versions within 6 months to a year, because after that most support will be forgotten, or unavailable. md-linux doesn't describe your goal very well, as far as I can see. Maybe a new name should be chosen. NOTE: I am not part of the debian team. Mike Fedyk
Re: Database port
Henry Keultjes wrote:enry Keultjes wrote: > > Mike: > > 1. There is no win32 front end. Current front-ends are green screen. > > 2. What is the standard Linux file system? You will have to admit that > there are many choices, thus no standard. MD-Linux will take one of > those and make it more capable so that it does allow the application > from the Linux side and the existing MD applications to run as a single > set of applications. linux supports several different file systems, and even more recently, but most will agree that ext2 is the "standard" file system at the moment, and has been for years. Are you proposing using a database engine as a file system? > > 3. "Porting" will be native. > Glad to hear it. > 4. You name a type of business or type of organization and there is > most likely an existing application. Each one of those will have > $millions of work and experience in it and "rejuvenating" those > applications is much easier than rewriting them in Linux. However, all > those applications are already available to Linux because the same > database that we will use to devise MD-Linux has already been ported to > Linux and it is a super product. It is just that I want to take it a > step further because it is a step further. Progress is inevitable. > Can you name this database? > 5. In the enterprise application arena where I have been for the last > twenty two years, the requirements are a lot more stringent than "just > want the computer to do work". We are talking about applications like > those that SAP, Baan, JD Edwards and Oracle Financials provide. > Are you planning on selling "enterprise" software to the typical phisician with one or two doctors? > 6. I did not want to get into specific of the GUI issue. However, the > option to use the command line will be an essential part of all the > applications. > Sounds good. > 7. You suggest that we continue with the Linux and Unix concepts. The > answer is most likely but we may find that there are places where we > should divert. That is progress. Many MD people hate what I am > proposing because it breaks the chains to the portability of > applications that has been adhered to for 35 years. Look at what Linus > did. He broke with the chains of the past. Better that we do the > breaking within Linux than letting BG3 break us. > The point I wanted to make was "don't reinvent the wheel". Case in point; using cron. Most people administrating a unix are familiar with cron, so you don't need to create another scheduler. Case in point; smtp, pop3 or imap. These are very capable email tools, and should probably be used in your system. In other words, don't reimplement standard tools just to lock in the users to your product. The user should like your product so much, and find it so useful that they wouldn't want to switch. > 8. Yes they will have existing X86 machines in their office and those > machines were obsolete before they arrived. They have been replacing > machines at a very rapid pitch. What's wrong with one more cycle and > than slowing it down. That's what PPC is all about, getting away from > the cat chasing its tail! > Linux is known for taking old machines and making them useful. You could at least make the client available on other platforms such as x86 m68k, or even open source, and be helpful to developers that want to port that code. Hopefully, it wouldn't have endian problems. > 9. I can't ever be thankful for choosing Debian unless we get the > project running on DEbian. It's a Catch 22! > Maintaining a debian system is much easier than redhat, and that is why you would be thankful. You already state that the DB engine is ported to linux, so it should work on debian without modification. > 10. It did not seem to make sense to ask more specific questions. What > I am looking for is some Debian geeks that are willing to make this > thing work. If then they have questions those geeks can pose more > intelligent questions than I can. > What exactly do you want the debian geeks to help you with? How will you be participating? What, besides more users from your products will we recieve out of this participation? > 11. We will use proprietary pieces to make an Open Source MD-Linux > product that meets the requirement of the Debian license - or we will > not do it at all. > What parts would be closed and open? > 12. Can't be an add-on package because that would not be progress. As > I stated above, Linux ports for the same database exist already and > those applications work fine for most people. The problem is that they > rely on win32 front-ends to make them sala
Re: Database port
Henry Keultjes wrote: > > Mike: > > Q: Are you proposing using a database engine as a file system? > > That's probably a good a way to describe it. However, remember that the > criteria is that both Linux and MD applications have to work. I believe > that this is plowing virgin ground. > > Q: Can you name this database? > > Pick (Raining Data) is the original and the rest are clones. What > exactly MD-Linux will use will be part of a confidentiality agreement. > Why would that be confidential? Wouldn't the database company want to have the publicity of being the backend in your product? > Q: Are you planning on selling "enterprise" software to the typical > physician with one or two doctors? > > Applications will be sold by commercial enterprises. MD-Linux does the > R&D to develop the product that those commercial enterprises can use. > OK, I think I'm getting the concept a little better. > Q: The user should like your product so much, and find it so useful that > they wouldn't want to switch. > > My sentiments exactly > My point is: I hope you see standards as useful instead of a hinderance. This may make your product acceptable where it might not have been in a larger setting. You cannot possibly make a product that will fit into everyone's problem and fix it. What if they want to use their current email server, or email client, you should make allowances for this type of situation. It cannot do anything but help the companies sell the product. > Q: Linux is known for taking old machines and making them useful. You > could at least make the client available on other platforms such as x86 > m68k, or even open source, and be helpful to developers that want to > port that code. Hopefully, it wouldn't have endian problems. > > Looking at this from a commercial perspective, given the right "offer" > prospects always exceed the capability to supply. Therefore support > issues and profit objectives dictate a single platform. MD-Linux does > not have to do those ports to prove the viability of the concept. > Then open sourcing the client shouldn't be a problem. Depending on where most of the processing will be done this can be made easier. For instance, if you have most of the business dicesions processed on the server, you can have a thinner client, and smaller code base that needs to be open source. > Q: Maintaining a debian system is much easier than redhat, and that is > why > you would be thankful. > > Yes I know that Debian is easier and yes I will be thankful . . . but > that requires that I get the "port" done. > . The there may be a problem if you plan to diverge from the way debian does things now. At least if you plan to keep in sync with it. If not, then you need to be able to divote the resources to maintain a distribution. This is why I'm trying to keep your product as an "addon" or "value add" to debian. > Q: You already state that the DB engine is ported to linux, so it should > work on debian without modification. > > No it will not, specifically because the existing ports are either Intel > or AIX. So either the OS is not right or the chip is not right. I > mentioned earlier that I have written an article "Perfect Pair - PowerPC > and Linux" that is supposed to go in the May issue of Linux Journal. > Very few people have this "narrow offering" approach that I have. > However, this approach has always worked well for me. Find a niche and > be better than anyone else. Finding a profitable niche in Intel, even > if it were possible, would not interest me. If you look at how Jack > Welch runs GE, his approach has been the same, only on a much much > bigger scale. However, I believe that a commercial company build on the > MD-Linux foundation can become a $10B company. > So, you plan on selling the hardware, is that right? I don't know about you, but I've known quite a few doctors that pinch pennies. By only supporting powerpc, you could be limiting your market, and thereby hindering the adoption of md-linux by your supervisors or dicesion makers in your organization. I can also see that having a wider list of hardware will make more cost in support, and maybe you don't want a doctor to go out and buy the latest consumer compaq and call support and complain that he can't get it to work. It can be a headache for support, but it can also be a revenue stream also. I can see both sides, but maybe this will show you another perspective. > Q: What exactly do you want the debian geeks to help you with? > > First I have to make them believers in the concept. Than, together with > the MD geeks they need to help me define the work and the
Re: Email client and conversion from Netscape Mail on win32
Sven LUTHER wrote: > > On Tue, Mar 06, 2001 at 05:33:34PM -0800, Mike Fedyk wrote: > > mutt: didn't look very much, so don't know. > > It is very nice, let's you do a lot of things, and is compatible with balsa > (which is just a frontend to libmutt anyway). > > That said, use procmail for mail filtering and such. and enable mutt's list > mode and other such goodies. I am reeiving almost as much mail as you (well > maybe not so much anymore ,since i unsubscribed to debian-devel and a few > other high volume lists) and it works well for me, also the ability to use it > over a ssh connection is rather nice. > > There is also a evolution package i have heard (even for unstable/ppc i think) > that should be rather nice. > > Also i can send you some procmail/mutt rules if you like, as they may be > rather obscure for starters. > Yes, I would like those procmail rules. I have been using exim as my smtp mailer, and I wonder what you guys think of using that for mail splitting. I haven't tried any others, but it seems to be able to hold up to large loads, and I've been using it for over a year. I'll probably use procmail anyway, since exim will work with it, and so will most other email servers. I'm switching to mutt right now. I've got to get off netscape, because it has crashed a couple times, and once just before I was going to reply to a long message. THAT scared me, that's for sure! Netscape uses mbox with a index file. I've used netscape over the network storing my profile on a samba server, and it really helps speed up the access. I've had it bring up a 70+ MB mail folder in a few seconds. The index file is named differently on win32 and unix. I don't know if they use compatible formats, but I suspect that they do. I just deleted the indexes and let netscape rebuild them itself. I'll take a look at gnus, but I have exactly 0 experience with emacs. I'm going to keep my search to a text based email client, because I don't like to have to use vnc to view my email from home... Mutt is great in an Xterm, and picture viewing is good too. I wonder if mutt can use links or netscape for html viewing... Anyone know? It's probably in the config file, or some symlinked "html-viewer" in a bin dir... Looked in /etc/alternatives, but no browser or www grep results... Thanks for all of the great messages
Re: initscripts
On Fri, Mar 09, 2001 at 12:46:24PM +0100, Damien GUIHAL wrote: > But now i'd like toput this instructions into an initscript to have them been > executed automatically... And I can't find any of the > /etc/rc.d/rc.(sysconfig, ...) script that I usually fond on my other > Mandrake7.2/i386), is anyone can tell me if these bootscripts exists and > where i can find them? The acutal scripts are in /etc/init.d, and the runlevels are /etc/rcN.d where N is the runlevel. Create a custom script in /etc/init.d and read the manual page for update-rc.d. It's something like update-rc.d scirpt defaults or similar. Make sure the script is executable, and look at the order of the other scripts, and put it after networking if your program needs it... Mike
Re: Email client and conversion from Netscape Mail on win32
Does mutt have any address book support? Autocompletion would be really nice too. Does mutt or gnus have that? Mike NOTE: please remove "mikef-linux-x86" from the last couple email messages from me to reply.
Re: running X on localhost with different user
On Tue, Mar 13, 2001 at 11:32:52PM -0800, Alan DuBoff wrote: > Alan DuBoff wrote: > > > I remember a friend mentioning that XFree86 requires something be put in > > .Xauthority, but I wasn't sure what it was. > > Ooops, that should be "XFree86 4.x requires...". > You can use the "xhost" command and enable access from the entire localhost. I looked into the X auth system a while ago, but forgot most of it. There is probably a better way... Mike
Re: Change swap after adding memory?
On Tue, Mar 13, 2001 at 06:09:31PM -0600, Kevin van Haaren wrote: > I'm getting ready to add 64mb to my 16mb machine. Currently it has a > swap of 48mb. Should I boost up the swap or just leave it? > Depends, is that 2.2 or 2.4? If 2.4, you should be ok, but keep an eye on the swap usage. > is it better to repartition to create a larger single swap or just > add a 2nd swap partition? > if you have multiple drives, by all means, put a swap on the second one. Though, a second partition on the drive that isn't next to the other swap will mean more head movement. I'd say, keep your partitions, and add a swap file. There is a priority setting you can put in /etc/fstab that will make the kernel use one swap before another. Do a search on google. Mike
Re: running X on localhost with different user
On Wed, Mar 14, 2001 at 12:16:56AM -0800, Alan DuBoff wrote: > Mike Fedyk wrote: > > > You can use the "xhost" command and enable access from the entire localhost. > > Yep, that is what I used to do on XFree86 3.3.x, and specifically I would > issue: > xhost +localhost > oh, I didn't know that wouldn't work with 4.x > Ah, I just figured it out... > > xhost +local:username > This looks really good. With 3.x you had to do some kind of key swap which seemed kinda funky. Though, I didn't take the time to really check it out. > does the trick...I guess XFree86 will only allow a specific user connect to X > for security reasons > Looks like a major improvement. > Sorry for the waste of bandwidth...:-/ > Yeah, we need to be able to get our mp3z and porn 0.001% faster! Mike
Re: Please help a complete newbie
On Wed, Mar 14, 2001 at 09:44:50PM +1100, Ken Simpson wrote: > I have heard great things about Debian PPC. Problem is I can't figure > out how to get the thing working! > > That is, I have d'led the 2.2 directory, and the rescue, root, > ramdisk, drivers and HFS boot disk, however I cannot figure out what > to do with them. > > My Beige (oldworld) G3 will not write a disk from the .bin images. > I made a floppy from the HFS disk image, but I just get a crossed out > symbol in a blank screen. > > Please help! > > I have a 2GB partition set aside that I can use as a Linux install > partition, and an 8GB disk for the actual install. > > I have read the dox over and over and cannot seem to work out what to do. > > Will someone please either point me to a step-by-step description for > a dummy, or give me one from their own experience??? > I don't know about the macos side, but do you have any friends with linux/dos that could make the floppies you need? Someone else could probably help with a MacOS only type solution. Mike
Re: non-us.debian.org having problems?
> One of my mirrors is on tux.creighton.edu. The entry for sources.list is > > deb http://tux.creighton.edu/debian potato updates/main > > if you want to trust that I haven't altered anything :) The entire tree > should be there, with the exception of the slink updates. how did you mirror it? rsync? wget? some http extraction mirror util? Mike
Re: Change swap after adding memory?
On Fri, Mar 16, 2001 at 06:14:50PM -0600, Kevin van Haaren wrote: > You can add the Aopen AON-325 10/100 ethernet PCI card as being > supported. Uses the Realtek RTL8139C chip. I just compilied a > kernel with the 8139too driver and it worked fine. Only 14 bucks too. > > No mac os support, but hey who cares? Don't be so sure, rtl8139 support is working great on a MacOS 8.5 backup comp (don't ask) I have. It transfers 100s of MBs a day at night. Mike
Ethernet card problems on PPC
Hi, I have been bitten with problems trying ethernet cards that work find in x86 but terribly on ppc. eepro100 - causes kernel to hard crash with no output pcnet32/lance - causes kernel to oops rtl8139/8139too - works for a couple hours, then silently (rtl8139) or reports interrupt error (8139too) tulip, and 3c9x both work reliably on ppc. I'd like to work with anyone that has patches for any of these problems, and help them test it. I don't know if I should post something like this to LKML, so I'm starting here. All of these tests were done on 7200s and g3 beige with bootx from macos8.5 to linux 2.2.17/18/18pre21 If you have a 7200 who's mace has "lost control word" on ip address set, trash it, or stop using that ethernet port. It will hold the address after 6x(up,down,up), but will fail under load. Mike
Ethernet card problems on PPC
Hi, I have been bitten with problems trying ethernet cards that work find in x86 but terribly on ppc. eepro100 - causes kernel to hard crash with no output pcnet32/lance - causes kernel to oops rtl8139/8139too - works for a couple hours, then silently (rtl8139) or reports interrupt error (8139too) tulip, and 3c9x both work reliably on ppc. I'd like to work with anyone that has patches for any of these problems, and help them test it. I don't know if I should post something like this to LKML, so I'm starting here. All of these tests were done on 7200s and g3 beige with bootx from macos8.5 to linux 2.2.17/18/18pre21 If you have a 7200 who's mace has "lost control word" on ip address set, trash it, or stop using that ethernet port. It will hold the address after 6x(up,down,up), but will fail under load. Mike
Re: Ethernet card problems on PPC
On Thu, Mar 29, 2001 at 06:47:46PM -0400, Peter Cordes wrote: > Most of the drivers in Linus's tree are old versions of Becker's drivers > (however, there seems to be more diversity now, esp. in 2.4). No idea why > they don't get updated. I need to run Don's newest version of the tulip Apparently, Becker won't submit patches to the drivers in Linus' kernel in small enough pieces, and just sends it all in one file... You can read "kernel traffic" to get summaries. Mike
Keyboard conversion (mac -> pc kvm switch)
Hi, I have an eight port kvm switch with only three ports used and three macs that I want to put on it. I found one once, but it was $100+ and since these are all oldworld machines, I'm looking for something cheap. I already have the converters to use the monitors, but I need something to convert adb --> (AT,PS/2). Can someone point me in the right direction... URLs would be appreciated. TIA, Mike
Keyboard converter - ADB to AT or PS/2
Hi, I have an eight port (at/serial,ps/2) kvm switch with only three ports used and three macs that I want to put on it. I found one once, but it was $100+ and since these are all oldworld machines, I'm looking for something cheap. I already have the converters to use the monitors, but I need something to convert adb --> (AT,PS/2). Can someone point me in the right direction... URLs would be appreciated. TIA, Mike
Re: Keyboard converter - ADB to AT or PS/2
On Mon, Apr 09, 2001 at 01:35:13AM -0700, Wilhelm *rafial* Fitzpatrick wrote: > >I found one once, but it was $100+ and since these are all oldworld > >machines, I'm looking for something cheap. I already have the converters to > >use the monitors, but I need something to convert adb --> (AT,PS/2). > > The only unit I have ever run across that does this was an adapter > made by Silicon Bus that was in the $100+ range. I actually bought > it way back when the 7100 was a zippy machine, so I could use an MS > natural keyboard with it, and it worked great, but it was definately > pricey. > > Perhaps you could find one used? > How about a mac KVM? One that outputs to at,ps/2? Or even just output to adb and use only *one* adb->at,ps/2. :) Any ideas on where to start looking? I'll look in the search engines too... Mike
Re: Booting to floppy
On Tue, Apr 10, 2001 at 05:58:59PM -0800, Ethan Benson wrote: > wish i had the requisite skills to try and make improvments to them > but i don't. > Hmm, what exactly would you need to know? 2.2.19 root disk prob would need kernel hacking skill Docs would need a good tech writer. I may be able to help. Anyone know what format the docs are in origionally? I haven't done any writing (except for a text editor) in linux, what programs should I learn first? What other issues need to be looked at? Booting is a large one. I know I'm forgetting a lot. Maybe you guys can help me remember... Mike
Re: Booting to floppy
On Tue, Apr 10, 2001 at 06:49:35PM -0800, Ethan Benson wrote: > On Tue, Apr 10, 2001 at 07:26:55PM -0700, Andrew Sharp wrote: > > > > Me neither, but what the user was insinuating is that he doesn't > > necessarily know the *right* answers, and wants to have a system > > where he can try many different combinations without having to start > > the install from the beginning each time. Myself, I would just grab > > the shell and muck with all the settings by hand until I found the > > one that works, but this isn't something that every person knows how > > to do, I guess. Hard to believe, really. > > well frankly you need to know the fscking network numbers before going > to setup a computer on the network, if you don't know what your > network numbers are that can hardly be blamed on debian. dammit Jim > im an installer not a psychic! but computers are just supposed to work! You plug them in and you get a pretty screen... ok ok ok enough of that. That is what dhcp is for, and we support that just fine. > > and you can redo the `setup the network' step as many times as you > want before rebooting the box, and console 2 is there so you can do a > quick test (i think ping is there). > But it's not part of the GUI!! You can't do anything without a GUI, right? heh, Am I making sence? no... Mike
Re: Exhaustive memory test for Macs?
On Wed, Apr 11, 2001 at 09:48:25PM +1000, Ross Hamilton wrote: > According to yesterday's Tidbits, the utility below will advise > whether your installed memory will pass muster with new firmware > updates from Apple (which have apparently refused to recognise some > DIMMs): > > <http://mactcp.org.nz/dimmfirstaid.sit> > > Probably this is just some packaged, G4-centric vendor checklist, > rather than the generic RAM-thrasher you envisage, but it's a tiny > download and might be worth a look if you can boot to macos 9.1 on your > machine. > > Sorry I can't offer a more informed opinion. It didn't run at all on my > iMac. > Has anyone else tried memtest? It compiles on my potato ppc box, 7200 and g3, but it runs under the OS, and only checks memory that is mlock able. I don't have an URL, but it comes up on the first page of a "memtest" search on google. Mike
SMP on UMAX S900/180 604e and processor from 8600/250
I know that you can under-clock the 250MHz processor and that should work, but I don't remember if it's a 604 or 603. Has anyone else tried something like this? Mike
Re: Booting to floppy
On Thu, Apr 12, 2001 at 05:08:42PM +0200, Sven LUTHER wrote: > On Thu, Apr 12, 2001 at 10:53:35AM -0400, Adam Di Carlo wrote: > > Sven LUTHER <[EMAIL PROTECTED]> writes: > > > > > I have installed (on i386) potato (r0 i think) from CD, and this is > > > exactly > > > how it works. it gets the linux file and the modules.tar.gz to get the > > > modules. > > > > No, that's wishful thinking. See choose_medium.c: > > > > char kernel_image_path[PATH_MAX+1] = "images-1.44/rescue.bin"; > > > > It really is not going to be all that hard to fix this for Woody -- > > the logic is simple. If we're installing the kernel from floppies, > > use the rescue disk; otherwise, use the vmlinuz file (or whatever) > > appropriate for your subarch or flavor. > > Yes, that would be really nice, ... > Why? What makes it so good to have the kernel installed differently from floppies and from network? The only thing I can see would be custom kernels after install. But then you'd have to change the way it gets the modules too. What am I missing? BTW, meant to send it to list before but hit r instead of L... Mike
Re: SMP on UMAX S900/180 604e and processor from 8600/250
On Thu, Apr 12, 2001 at 10:31:42AM -0700, Mike Fedyk wrote: > I know that you can under-clock the 250MHz processor and that should work, > but I don't remember if it's a 604 or 603. > > Has anyone else tried something like this? > > Mike I guess not... Maybe I should repost...? I'll wait another week. Mike
Re: problems realtek 8139 ?
On Sat, Apr 14, 2001 at 12:37:13PM +0200, Bruno Waes wrote: > are there any know problems with using realtek 8139 cards on powerpc (4400 > tanzania to be more exact) ? > > with kernel 2.2.12 it is initialized but when i ping a host i hardly get any > packets back, now with a 2.2.18pre21 kernel it kernel panics with a null > pointer ... > I asked the same question about a week ago, and people pointed me to newer drivers that haven't been updated in the kernel... I haven't tried it yet, bacause the tulip driver works great for me. Still need to try eepro100 and rtl8139 though... Mike
Re: Exhaustive memory test for Macs?
On Wed, Apr 11, 2001 at 04:35:35PM -0800, Ethan Benson wrote: > but its true the only way to really test virtually all the memory is > to write the memory tester as a OpenFirmware client and boot it > directly -- no OS at all. > What if you have two or more ram chips and rotate them between memory checks or every 20 checks. Do I have the concept, or is ram virtualized where any part could be locked and it wouldn't necisarily start from the beginning? Just a thought... If I'm right, then I'm great being up over 24hrs, if I'm wrong then I'm not... ;) Mike
