Hi,
On 4/9/19 6:19 PM, John Paul Adrian Glaubitz wrote:
I just uploaded updated installation images for the following
Debian Ports architectures:
* alpha
* hppa
* ia64
* m68k
* powerpc
* ppc64
* sh4
* sparc64
I uploaded both CD images [1] as well as netboot images [2].
Please test those images and report back over the mailing list for
the corresponding architecture.
Thanks!
I tested this in Hatari Falcon emulation (16Mhz 030, 14+128MB RAM,
LILO loads kernel to ST-RAM).
Using "initcall_blacklist=dh_init" makes kernel boot 10x (9.5 minutes)
faster to initrd.
-> Could Diffie-Hellman algorithm be module instead of builtin:
[ 7.510000] calling dh_init+0x0/0x10 @ 1
[ 66.670000] random: crng init done
[ 571.460000] initcall dh_init+0x0/0x10 returned 0 after 550720670 usecs
?
On m68k (which doesn't have crypto HW / instructions), somebody
requiring tortuously slow encryption for initrd seems somewhat
unlikely...
Could you also add to debian-installer package the System.map file
for the included 4.19 kernel, so that I can debug different Oopses
I get when running /init?
Note: If I use init=/bin/sh, initrd boots often up fine, assuming
I have some hard disk attached. If not, kernel Oopses.
- Eero
Some of the crashes I see are "udevadm" crash during bootup:
-------------------------------------------------
[ 30.980000] Run /init as init process
[ 55.800000] *** LINE 1111 *** FORMAT=0
[ 55.810000] Current process id is 48
[ 55.820000] BAD KERNEL TRAP: 00000000
[ 55.830000] Modules linked in:
[ 55.840000] PC: [<00002a2c>] user_inthandler+0x4/0x20
[ 55.850000] SR: 2d08 SP: (ptrval) a2: 00140000
[ 55.860000] d0: c016281c d1: 0000b550 d2: 00000f1c d3: c0040f44
[ 55.870000] d4: c0020000 d5: c001ed40 a0: c0020000 a1: c00430bc
[ 55.880000] Process udevadm (pid: 48, task=(ptrval))
[ 55.890000] Frame format=0
[ 55.900000] Stack from 05a57ff8:
[ 55.900000] 0208c000 b6460114
[ 55.910000] Call Trace:
[ 55.920000] Code: 0000 2b24 508f 60ff ffff ff64 42a7 4878 <ffff> 2f00
48e7 7ce0 200f 0280 ffff e000 2440 2452 e9ef 010a 0032 0440 0038 2f0f
-------------------------------------------------
Sometimes I see also this, and in this case shell also starts
afterwards:
-------------------------------------------------
[ 30.980000] Run /init as init process
[ 59.770000] random: crng init done
[ 60.790000] Data read fault at 0xeffd917c in Super Data (pc=0x2a28)
[ 60.800000] BAD KERNEL BUSERR
[ 60.810000] Oops: 00000000
[ 60.820000] Modules linked in:
[ 60.830000] PC: [<00002a28>] user_inthandler+0x0/0x20
[ 60.850000] SR: 2c00 SP: 2f44f623 a2: c001f4e8
[ 60.860000] d0: 00000000 d1: c001f790 d2: 00000001 d3: 00000001
[ 60.870000] d4: c001f120 d5: c001ed40 a0: c015e000 a1: c001db88
[ 60.880000] Process logger (pid: 51, task=ba408265)
[ 60.900000] Frame format=B ssw=0345 isc=2f00 isb=48e7 daddr=effd917c
dobuf=c001f4e8
[ 60.910000] baddr=effd917c dibuf=effd917c ver=0
[ 60.920000] Stack from 05ae5ff8:
[ 60.920000] 0208c000 b1560110
[ 60.930000] Call Trace:
[ 60.940000] Code: 2f00 4eb9 0000 2b24 508f 60ff ffff ff64 <42a7> 4878
ffff 2f00 48e7 7ce0 200f 0280 ffff e000 2440 2452 e9ef 010a 0032 0440
-------------------------------------------------
Sometimes I get Oops from:
-------------------------------------------------
[ 33.880000] Data read fault at 0xc0023d60 in Super Data (pc=0x29f4)
[ 33.890000] BAD KERNEL BUSERR
[ 33.910000] Oops: 00000000
[ 33.920000] Modules linked in:
[ 33.930000] PC: [<000029f4>] auto_inthandler+0x0/0x28
[ 33.940000] SR: 2400 SP: (ptrval) a2: c001f4e8
[ 33.950000] d0: 00000000 d1: 00000029 d2: c0020000 d3: 00000022
[ 33.970000] d4: 00000000 d5: 00000001 a0: c015df40 a1: c001f508
[ 33.990000] Process mount (pid: 37, task=(ptrval))
[ 34.000000] Frame format=B ssw=0345 isc=2f00 isb=48e7 daddr=c0023d60
dobuf=c0155060
[ 34.010000] baddr=c0023d60 dibuf=c0023d60 ver=0
[ 34.020000] Stack from 05a5bff8:
[ 34.020000] 0208c000 68b40070
[ 34.030000] Call Trace:
[ 34.040000] Code: 0005 61ff 0003 1630 508f 588f 60a2 0000 <42a7> 4878
ffff 2f00 48e7 7ce0 200f 0280 ffff e000 2440 2452 e9ef 010a 0032 0440
[ 34.050000] Disabling lock debugging due to kernel taint
-------------------------------------------------
And shell doesn't start.
If I enable 030 CPU cache emulation, kernel often hangs to:
-------------------------------------------------
[ 62.630000] Run /init as init process
[ 64.110000] Instruction fault at 0x00000000
[ 64.120000] BAD KERNEL BUSERR
[ 64.130000] Oops: 00000000
[ 64.140000] Modules linked in:
[ 64.150000] PC: [<00000000>] (null)
[ 64.160000] SR: 2400 SP: (ptrval) a2: 00c76f90
[ 64.170000] d0: 00000000 d1: 00000004 d2: 00000000 d3: 00000004
[ 64.180000] d4: 05a32400 d5: 00050862 a0: 00000000 a1: 003d77c4
[ 64.200000] Process mount (pid: 31, task=(ptrval))
[ 64.210000] Frame format=B ssw=5046 isc=2800 isb=ffff daddr=00000000
dobuf=000508e0
[ 64.220000] baddr=00000000 dibuf=00000000 ver=0
[ 64.230000] Stack from 00c79eb4:
[ 64.230000] 000508e0 00000004 00000000 00000000 3401fffe
05a32400 d00582f8 00000000
[ 64.230000] 003d77c4 000002f8 d0053d8c 00c79efc 0005099e
003d77c4 00c79ef8 00000000
[ 64.230000] 003d77c4 00000000 ef954c98 000509f2 003d77c4
003d77c4 0005297c 003d77c4
[ 64.230000] 00812000 00050364 003d77c4 00002b44 00000004
067fc7bd 00002a20 00000004
[ 64.230000] 00c79f38 d00564e8 067fc7bd 3401fffe 05a32400
d00582f8 00812d08 008d9a40
[ 64.230000] 00c76f90 00000000 ffffffff 00000000 2204002c
93540070 00001000 d00592f8
[ 64.280000] Call Trace: [<000508e0>] __handle_irq_event_percpu+0x3e/0xe0
[ 64.290000] [<0005099e>] handle_irq_event_percpu+0x1c/0x52
[ 64.300000] [<000509f2>] handle_irq_event+0x1e/0x30
[ 64.310000] [<0005297c>] handle_simple_irq+0x50/0x54
[ 64.320000] [<00050364>] generic_handle_irq+0x18/0x22
[ 64.340000] [<00002b44>] do_IRQ+0x20/0x32
[ 64.350000] [<00002a20>] auto_irqhandler_fixup+0x4/0xc
[ 64.360000] [<00001000>] kernel_pg_dir+0x0/0x1000
[ 64.380000] [<000f00ac>] copy_mount_options+0x64/0xba
[ 64.400000] [<000ca298>] kfree+0x0/0x126
[ 64.420000] [<000f0b5e>] ksys_mount+0x4e/0x82
[ 64.430000] [<00008002>] pcmcia_copy_tuple+0x22/0x76
[ 64.440000] [<000f0bac>] sys_mount+0x1a/0x20
[ 64.450000] [<00008002>] pcmcia_copy_tuple+0x22/0x76
[ 64.460000] [<00002950>] syscall+0x8/0xc
[ 64.470000] [<00008002>] pcmcia_copy_tuple+0x22/0x76
[ 64.480000] [<00008002>] pcmcia_copy_tuple+0x22/0x76
-------------------------------------------------
=> PCMCIA???
Known issues:
* alpha
- No known issues
* hppa
- No known issues
* ia64
- The kernel might be unstable during install
(try passing "hardened_usercopy=off" at the boot prompt,
e.g. "install hardened_usercopy=off"
- The debootstrap process might fail due to a missing vim
package (this should be fixed with the next CD release)
* m68k
- The drivers for IDE hardware may have to be loaded manually, e.g.:
+ modprobe pata_gayle
+ modprobe pata_falcon
- There is no automatic installation of the bootloader yet,
users have to boot the kernel and initrd manually after
installation.
* powerpc/ppc64
- The bootloader installation still defaults to Yaboot and
is being switched to GRUB for IBM POWER and PowerMacs,
please report any issues you find.
* sh4
- There is no CD image, installation has to be performed
using the debian-installer images.
- Currently supported targets are Renesas SH-7785LCR and
SH-7751R, the latter should be identical to qemu-sh4.
- debian-installer for sh4 is completely untested, please
test with QEMU and report back.
* sparc64
- Installation over a serial console is currently broken
on sparc64 due to a bug in the rootskel package
(can also be considered a kernel bug), see: #926539.
Use any image before 2019-04-06 as workaround.
Important:
When reporting issues or bugs, please include the syslog
from your installation in the bug report/mail. The log
file can be access when switching to another console using
<Alt>+<Cursor> for normal installations or <Ctrl>+<a>+<n>
when installing over a serial console.
Thanks,
Adrian
[1] https://cdimage.debian.org/cdimage/ports/2019-04-09/
[2] https://cdimage.debian.org/cdimage/ports/debian-installer/
