Skip to contentSkip to conversation

linux-2.6 jejb

parisc-linux Joel Soete 5 participants 7 messages Feb 6, 2004 – Feb 9, 2004 1 message not in the archive

lore
Share & tools
lore

Joel Soete

author
#0 lore raw

Hi Paul,

Quoted 5 lines
Quoted 3 lines

I am just curious to know the amount of memory on the Paul's b2k?

Mem: 505596k total

I only have 256Mb.

That was also a difference with c3k but here it is the only difference between our b2k systems. Unfortunately I don't have any spare part to increase my ram :(

Just in case Randolph has the opportunity to check the piminfo:

PROCESSOR PIM INFORMATION

----------------- Processor 0 HPMC Information ------------------

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

HPMC Chassis Codes = 2cbf0 2500b 2cbf5 2cbfc

General Registers 0 - 31

00-03   0000000000000000  000000001043e000  00000000000600d0  fffffff0f0000018
04-07   0000000000438000  0000000000439000  00000000fffffff0  00000000f0002f68
08-11   0000000000000027  00000000f0072848  000000000fff0000  000000001052c538
12-15   0000000000000000  00000000ffffffff  0000000000000000  00000000f0400004
16-19   00000000f00008c4  00000000f000017c  00000000f0000174  0000000000000000
20-23   0000000000000000  00000000fee003f8  00000000fee003fd  0000000000000000
24-27   0000000000000000  0000000000065e38  0000000000065e38  0000000000065e34
28-31   0000000000100000  0000000f4c7a3298  00000000004dc100  0000000000000060

Control Registers 0 - 31

00-03   0000000000000000  0000000000000000  0000000000000000  0000000000000000
04-07   0000000000000000  0000000000000000  0000000000000000  0000000000000000
08-11   0000000000000000  0000000000000000  0000000000000000  000000000000001f
12-15   0000000000000000  0000000000000000  0000000000000000  0000000000000000
16-19   0000000f4e0c5fee  0000000000000000  00000000001000e0  00000000702b0b30
20-23   0000000014340041  000000000f83e598  0000000000000008  0000000080000000
24-27   0000000000438000  0000000000438000  000000005fd50d56  000000007fdfdf2f
28-31   00000000ffbeffbe  00000000bfeb37dc  00000000104dc000  00000000104e0000

Space Registers 0 - 7

00-03   00000000          00000000          00000000          00000000
04-07   00000000          00000000          00000000          00000000
IIA Space                    = 0x0000000000000000
IIA Offset                   = 0x00000000001000e4
Check Type                   = 0x20000000
CPU State                    = 0x9e000004
Cache Check                  = 0x00000000
TLB Check                    = 0x00000000
Bus Check                    = 0x003010bb
Assists Check                = 0x00000000
Assist State                 = 0x00000000
Path Info                    = 0x00031800
System Responder Address     = 0xfffffffffed10200
System Requestor Address     = 0xfffffffffffa0000

Floating-Point Registers 0 - 31

00-03   0000001f00000000  ffffffffffffffff  ffffffffffffffff  ffffffffffffffff
04-07   00003d09ffffffff  00000078ffffffff  00000000001c9c38  ddffffffffffffff
08-11   ffffffffffffffff  bfffffffffffffff  ffffffffffffffff  ffffffffffffffff
12-15   ffffffffffffffff  ffffffffffffffff  ffffffffffffffff  ffffbfefffffffef
16-19   ffffffffffffffff  ffffffffffffffff  ffffffffffffffff  ffffffffffffffff
20-23   ffffffffffffffff  fbfffffffffffdff  ffffffffffffffff  ffffff7fffffffff
24-27   ffffffffffffffff  ffffffffffffffbf  ffffffffffffffff  ffffffffffffffff
28-31   ffffffffffffffff  dfffffffffffffff  ffffffffffffffff  ffffffffffffffff

'9000/785 B,C,J Workstation Unarchitected (per-CPU)', rev 1, 140 bytes:

Check Summary                = 0xcb81041000000000
Available Memory             = 0x0000000010000000
CPU Diagnose Register 2      = 0x0301000000802204
CPU Status Register 0        = 0x2420c20000000000
CPU Status Register 1        = 0x8080000000000000
SADD LOG                     = 0xf813ec001043e598
Read Short LOG               = 0xc18040fffee003fd
ERROR_STATUS                 = 0x0000000000000010
MEM_ADDR                     = 0x000001ff3fffffff
MEM_SYND                     = 0x0000000000000000
MEM_ADDR_CORR                = 0x000001ff3fffffff
MEM_SYND_CORR                = 0x0000000000000000
RUN_DATA_HIGH                = 0xc1bff0fffed08040
RUN_DATA_LOW                 = 0xc1bff0fffed08040
RUN_CTRL                     = 0x0000021c00001418
RUN_ADDR                     = 0xc1bff0fffed08040
System Responder Path        = 0x00ffffffffffffff

HPMC PIM Analysis Information:

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

'9000/785 B,C,J Workstation HPMC PIM Analysis (per-CPU)', rev 0, 1304 bytes:

A Data Miss Timeout occurred while CPU 0 was
requesting information.

Memory/IO Controller Error Analysis Information:

The Memory/IO Controller only observed the Broadcast Error. It did not log any additional information about the HPMC.

----------------- Processor 0 LPMC Information ------------------

Check Type                   = 0x00000000
I/D Cache Parity Info        = 0x00000000
Cache Check                  = 0x00000000
TLB Check                    = 0x00000000
Bus Check                    = 0x00000000
Assists Check                = 0x00000000
Assist State                 = 0x00000000
Path Info                    = 0x00000000
System Responder Address     = 0x0000000000000000
System Requestor Address     = 0x0000000000000000

----------------- Processor 0 TOC Information -------------------

General Registers 0 - 31

00-03   0000000000000000  fffffff0f009d000  000000f0f000b618  00000000000f4240
04-07   0000000c6d9b6071  0000000000000004  0ffffff0f0408aa8  0000000000000002
08-11   00fffff000408ab0  00000000f0072848  000000000fff0000  00000000104f04d0
12-15   0000000000000000  00000000ffffffff  0000000000000000  00000000f0400004
16-19   00000000f00008c4  00000000f000017c  00000000f0000174  fffffff0f007ee38
20-23   fffffff0f05f0068  0000000000000004  0000000000000009  fffffff0f003d428
24-27   0000000013cf8409  000000003d090000  0000000017d78400  fffffff0f0412000
28-31   0000000017d78400  0000000000000008  fffffff0f0408d68  fffffff0f007eff8

Control Registers 0 - 31

00-03   0000000000000000  0000000000000000  0000000000000000  0000000000000000
04-07   0000000000000000  0000000000000000  0000000000000000  0000000000000000
08-11   0000000000000000  0000000000000000  0000960000009600  0000000000000005
12-15   0000000000000000  0000000000000000  000000f0f0003800  0000000000000000
16-19   0000000c816e9083  00000000000000f0  000000f0f000b628  00000000020008b8
20-23   000000009a37c3c1  c000000002008d10  000000ff08098f08  8000000000000000
24-27   000000000042c000  000000000042c000  00000000fffffff7  00000000fffffff7
28-31   00000000fffffff7  00000000fffffff7  0000000010495880  000000001049a000

Space Registers 0 - 7

00-03   00000000          00000000          00000000          00000000
04-07   00000000          00000000          00000000          00000000
IIA Space                    = 0x00000000000000f0
IIA Offset                   = 0x000000f0f000b620
CPU State                    = 0x9e000001

Memory Error Log Information:

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

'9000/785 B,C,J Workstation Memory Error Log', rev 0, 64 bytes:

No memory errors logged

I/O Module Error Log Information:

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

'9000/785 B,C,J Workstation IO Error Log', rev 0, 228 bytes:

Rope Word1 Word2 Word3
------ ------------ ------------

   0    0x00000000   0x0e0cc009   0x00000000fed30048
   1    0x00000000   0x1e0cc009   0x00000000fed32048
   2    ----------   0x2e0cc009   ------------------
   3    ----------   0x3e0cc009   ------------------
   4    ----------   0x4e0cc009   ------------------
   5    ----------   0x5e0cc009   ------------------
   6    ----------   0x6e0cc009   ------------------
   7    ----------   0x7e0cc009   ------------------

Thanks a lot for attention,
Joel

-------------------------------------------------------------------------
Tiscali ADSL: 12 mois à 29,50 €/mois! L'Internet rapide, c'est pour tout le monde.
http://reg.tiscali.be/default.asp?lg=fr

#1 lore raw
Quoted 1 line

Just in case Randolph has the opportunity to check the piminfo:

did you do clearpim before the run? The values in the pim info don't make a lot of sense.... did the box HPMC? did it hang and then you hit the TOC button?

randolph

Signature · 3 lines
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/
author
#2 lore raw

Hi Randolph,

Quoted 1 line

did you do clearpim before the run?

Yes

Quoted 1 line

The values in the pim info don't make a lot of sense....

Not more sense with previous one I got, that's why it takes me so much time to find a workaround ;)

Quoted 1 line

did the box HPMC?

Yes that's what showing the small lcd on the box and seems to be confirm by hpmc time stamp ;)

Quoted 1 line

did it hang and then you hit the TOC button?

NO. Just after the last advise of palo (... If this is the last message...) appears very quickly the message on the lcd: hpmc init... (even if I put printk() at the very begining of init code in main.c nothing is printed on the serial console.

I also suspected a long time the compiler I used but recently I put this same kernel (scp) on the N4K-550 and it boot fine (but this system has 2Gb of ram)

And because 2.4 64bits runing fine on this system (b2k) and the pb should be located very early at the begining of the kernel start, I had the idea to compare head64.S and vmlinux.lds.S and find back this 2.5 patch and try to revert it. But I very don't undurstand what happen in head64 to hope to help more efficiently :( (sorry)

Thanks for your attention,
Joel

-------------------------------------------------------------------------
Tiscali ADSL: 12 mois à 29,50 €/mois! L'Internet rapide, c'est pour tout le monde.
http://reg.tiscali.be/default.asp?lg=fr

#3 lore raw
Quoted 5 lines

And because 2.4 64bits runing fine on this system (b2k) and the pb should be located very early at the begining of the kernel start, I had the idea to compare head64.S and vmlinux.lds.S and find back this 2.5 patch and try to revert it. But I very don't undurstand what happen in head64 to hope to help more efficiently :( (sorry)

Perhaps the problem is this....

in head64.S, we used to have:

        /* And the stack pointer too */
        load32          PA(init_task_union+TASK_SZ_ALGN),%sp
        [...]
        /* Save the rfi target address */
        std             %r11,  TASK_PT_GR11-TASK_SZ_ALGN(%sp)

Since the pt_regs struct used to be part of the init_task_union, we do a store against the PA of init_task_union, and all is well.

In the new code, init_task_union is a thread_info, and in order to store into the pt_regs we have to first dereference thread_info.task and then store into the resulting task pointer....

        /* And the stack pointer too */
        load32          PA(init_thread_union+THREAD_SZ_ALGN),%sp
        [...]
        /* Save the rfi target address */
        ldo             -THREAD_SZ_ALGN(%sp), %r1
        ldd             TI_TASK(%r1), %r1
        std             %r11,  TASK_PT_GR11(%r1)

so now we are storing into a VA of init_thread_union.task .... oops!

So, can you try this patch?

Index: arch/parisc/kernel/head64.S
===================================================================
RCS file: /var/cvs/linux-2.6/arch/parisc/kernel/head64.S,v retrieving revision 1.3

parisc/kernel/head64.S +6 -6
diff -u -p -r1.3 head64.S
@@ -165,9 +165,9 @@ common_stext:
 #endif /* CONFIG_SMP */
 
 	/* Save the rfi target address */
-	ldo		-THREAD_SZ_ALGN(%sp), %r1
-	ldd		TI_TASK(%r1), %r1
-	std		%r11,  TASK_PT_GR11(%r1)
+	ldd		TI_TASK-THREAD_SZ_ALGN(%sp), %r10
+	tophys_r1	%r10
+	std		%r11,  TASK_PT_GR11(%r10)
 
 #ifndef CONFIG_PDC_NARROW
 	/* Switch to wide mode; Superdome doesn't support narrow PDC

@@ -197,9 +197,9 @@ common_stext:
 
 stext_pdc_ret:
 	/* restore rfi target address*/
-	ldo		-THREAD_SZ_ALGN(%sp), %r1
-	ldd		TI_TASK(%r1), %r1
-	ldd		TASK_PT_GR11(%r1), %r11
+	ldd		TI_TASK-THREAD_SZ_ALGN(%sp), %r10
+	tophys_r1	%r10
+	ldd		TASK_PT_GR11(%r10), %r11
 
 	/* PARANOID: clear user scratch/user space SR's */
 	mtsp	%r0,%sr0

randolph

Signature · 3 lines
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/
author
#4 lore raw

Hi Randolph,

Quoted 1 line

So, can you try this patch?

Yes, yes, yessss it works fine ;)

Thanks agian for great help,
Joel

-------------------------------------------------------------------------
Tiscali ADSL: 12 mois à 29,50 €/mois! L'Internet rapide, c'est pour tout le monde.
http://reg.tiscali.be/default.asp?lg=fr

#5 lore raw
In reply to Joel Soete
Quoted 9 lines · Joel Soete

----------------- Processor 0 HPMC Information ------------------

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

HPMC Chassis Codes = 2cbf0 2500b 2cbf5 2cbfc

General Registers 0 - 31
00-03 0000000000000000 000000001043e000 00000000000600d0 fffffff0f0000018

What does 00000000000600d0 point at?
Is that the boot loader?

Quoted 2 lines

IIA Space = 0x0000000000000000
IIA Offset = 0x00000000001000e4

Is the CPU still in real mode?
ie haven't switched to virtual mode yet?

Quoted 10 lines
Check Type                   = 0x20000000
CPU State                    = 0x9e000004
Cache Check                  = 0x00000000
TLB Check                    = 0x00000000
Bus Check                    = 0x003010bb
Assists Check                = 0x00000000
Assist State                 = 0x00000000
Path Info                    = 0x00031800
System Responder Address     = 0xfffffffffed10200
System Requestor Address     = 0xfffffffffffa0000

CPU is trying to get data from 0xfed10200
What lives there?

Quoted 2 lines

A Data Miss Timeout occurred while CPU 0 was
requesting information.

Some device wasn't responding.

Quoted 2 lines

General Registers 0 - 31
00-03 0000000000000000 fffffff0f009d000 000000f0f000b618 00000000000f4240

looks like the box HPMC'd, and then TOC to recover. I think GR02 is a firmware address.
...

Quoted 11 lines
IIA Space                    = 0x00000000000000f0
IIA Offset                   = 0x000000f0f000b620
CPU State                    = 0x9e000001

I/O Module Error Log Information:

Timestamp =
Fri Feb 6 07:20:26 GMT 2004 (20:04:02:06:07:20:26)

Quoted 6 lines

'9000/785 B,C,J Workstation IO Error Log', rev 0, 228 bytes:

Rope Word1 Word2 Word3
------ ------------ ------------

   0    0x00000000   0x0e0cc009   0x00000000fed30048
   1    0x00000000   0x1e0cc009   0x00000000fed32048

Note this is NOT the same address range that the CPU was trying to talk to. And it looks like all the ropes are happy. ("cc" is the status or error word, IIRC)

If i had to guess, I would think Joel's box has a HW problem. But I really don't know what's going on.

hth,
grant

A message that is not in the archive (sent off-list or not yet received): [email protected]

#7 lore raw
Quoted 3 lines

What does 00000000000600d0 point at?
Is that the boot loader?

Yes, it's where the kernel would return to if it ever returned, so probably palo's made the call to the kernel at that point.