usefdt boot mode on 2-socket/1-core-each G5 PowerMac7, 2: visible progress (using vt, not sc), new visible stopping point
Mark Millard
marklmi at yahoo.com
Sat Apr 13 07:50:12 UTC 2019
On 2019-Apr-12, at 21:51, Mark Millard <marklmi at yahoo.com> wrote:
> [More details on where the failure stage is.]
>
> On 2019-Apr-12, at 20:49, Mark Millard <marklmi at yahoo.com> wrote:
>
>> [I experiment with more modern tools and such for targeting
>> powerpc64. Also with non-default modern modes of operation,
>> such as usefdt mode. I had to patch the openfirmware->fdt
>> translation to allow a 2-socket/1-core-each G4 PowerMac3,6
>> to boot normally, with the ethernet present (and operable).]
>>
>> Turns out that I had been using:
>>
>> kern.vty=sc
>>
>> and for some reason this stops visible output just after
>> "Kernel entry at . . .". (I've no clue if there is hidden
>> activity vs. it just stopped there.) But with:
>>
>> kern.vty=vt
>>
>> the PowerMac7,2 visibly gets farther, though it fails
>> to finish booting in usefdt mode.
>>
>> The new visible stopping point is (typed from a picture
>> of boot -v output for usefdt mode):
>>
>> atapci1: <ServerWorks K2 SATA150 controller> at device 12.1 on pci8
>> pcib1: failed to reserve resource for pcib8
>> pcib8: failed to allocate initial I/O port window (0-0xffffffff,0x10)
>> atapci1: 0x10 bytes of rid 0x28 res 4 failed (0, 0xffffffffffffffff).
>> atapci1: unable to map interrupt
>> device_attach: atapci1 attach returned 6
>> ofwbus0: <vsp> mem 0xf9000000-0xf9ffffff irq 43 type vsp compat u3-vsp (no driver attached)
>> cryptosoft0: <software crypto> on nexus0
>> crypto: assign cryptosoft0 driver id 0, flags 0x6000000
>>
>> (I'll not repeat the long list of crypto registers/alg/flags/maxoplen lines.)
>>
>> Later material below suggests that the above is not the
>> cause of stopping: a non-usefdt boot finishes but also
>> has such text.
>>
>> Of course some material scrolled off screen. So from
>> attempting booting again, but without -v , there is
>> before the above:
>>
>> atapci0: <ServerWorks K2 SATA150 controller> mem 0x80600000-0x80601fff irq 33554432 at device 12.0 on pci8
>> pcib1: failed to reserve resource for pcib8
>> atapci0: 0x10 bytes of rid 0x20 res 4 failed (0, 0xffffffffffffffff).
>> ata2: <ATA channel> at channel 0 on atapci0
>> ata3: <ATA channel> at channel 1 on atapci0
>> ata4: <ATA channel> at channel 2 on atapci0
>> ata5: <ATA channel> at channel 3 on atapci0
>>
>>
>> A non-usefdt mode (so normal openfirmware mode) boot works
>> and shows similar output:
>>
>> atapci0: <ServerWorks K2 SATA150 controller> mem 0x80600000-0x80601fff irq 33554432 at device 12.0 on pci8
>> pcib1: failed to reserve resource for pcib8
>> atapci0: 0x10 bytes of rid 0x20 res 4 failed (0, 0xffffffffffffffff).
>> ata2: <ATA channel> at channel 0 on atapci0
>> ata3: <ATA channel> at channel 1 on atapci0
>> ata4: <ATA channel> at channel 2 on atapci0
>> ata5: <ATA channel> at channel 3 on atapci0
>> atapci1: <ServerWorks K2 SATA150 controller> at device 12.1 on pci8
>> pcib1: failed to reserve resource for pcib8
>> atapci1: 0x10 bytes of rid 0x20 res 4 failed (0, 0xffffffffffffffff).
>> atapci1: unable to map interrupt
>> device_attach: atapci1 attach returned 6
>> cryptosoft0: <software crypto> on nexus0
>>
>> This suggests that such text is normal and the hangup is
>> for some other reason. It just suggests a time frame.
>>
>> After that for non-usefdt mode is:
>>
>> Timecounter "timebase" frequency 33333333 Hz quality 0
>> Event timer "decrementer" frequency 33333333 Hz quality 1000
>> Timecounters tick every 1.000 msec
>>
>> and so on. Such does not show up for usefdt mode.
>
>
>
>
> subsystem 3800000
> configure_first(0)... done.
> module_register_init(&ata_moduledata)... done.
> fbd_evh_init(0)... done.
> module_register_init(&cam_moduledata)... done.
> configure(0)... ofwbus0: <Open Firmware Device Tree> on nexus0
> . . .
> cryptosoft0: <software crypto> on nexus0
> done.
> cn_drvinit(0)... done.
> mpt_postattach(0)... done.
> cn_drvinit(0)... done.
> vmem_start_callout(0)... done.
> vm_lowmem_evh_init(0)... done.
> mountroot_evh_init(0)... done.
> ifaddr_event_ext_evh_init(0)... done.
> configure_final(0)...
>
> (That is the end of the output.)
>
> So during:
>
> SI_SUB_CONFIGURE = 0x3800000, /* Configure devices */
>
> for:
>
> static void
> configure_final(void *dummy)
> {
>
> /*
> * Now that we're guaranteed to have a PIC driver (or we'll never
> * have one), program it with all the previously setup interrupts.
> */
> powerpc_enable_intr();
>
> /* Enable external interrupts. */
> mtmsr(mfmsr() | PSL_EE);
>
> cninit_finish();
> cold = 0;
> }
>
>
With printf's added:
printf("After powerpc_enable_intr; before mtmsr masking in PSL_EE\n");
/* Enable external interrupts. */
mtmsr(mfmsr() | PSL_EE);
printf("After mtmsr masking in PSL_EE; before cninit_finish\n");
the first printf shows up and the second does not.
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
More information about the freebsd-ppc
mailing list