Skip to contentSkip to conversation

Boot failure on ia64.

BISECT Robin Holt 3 participants 8 messages Jun 24, 2008

lore
Share & tools
lore

Robin Holt

author
#0 lore raw

I bisected to this commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

3463a93def55c309f3c0d0a8aaf216be3be42d64 is first bad commit commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

Author: Alex Chiang <[email protected]>
Date:   Wed Jun 11 17:29:27 2008 -0600

[IA64] Update check_sal_cache_flush to use platform_send_ipi()
...

This fails to boot on any sn2 ia64 with the sn2_defconfig.

Here is the output from that boot.

fs0:\efi\SuSE> elilo net0:holt/v1 root=/dev/sda7 console=ttySG0
ELILO
Uncompressing Linux... done
Linux version 2.6.26-rc5-00223-g3463a93 (holt@attica) (gcc version 4.1.2 20070115 (prerelease) (SUSE Linux)) #14 SMP Tue Jun 24 07:27:34 CDT 2008 EFI v1.10 by INTEL: SALsystab=0x6002c25f10 ACPI 2.0=0x6002c26000
console [sn_sal0] enabled
ACPI: RSDP 6002C26000, 0024 (r2 SGI)
ACPI: XSDT 6002C297F0, 0044 (r1 SGI XSDTSN2 10001 7C)
ACPI: APIC 6002C26870, 032C (r1 SGI APICSN2 10001 1)
ACPI: SRAT 6002C26BB0, 06B0 (r1 SGI SRATSN2 10001 1)
ACPI: SLIT 6002C27270, 012C (r1 SGI SLITSN2 10001 1)
ACPI: FACP 6002C27400, 00F4 (r3 SGI FACPSN2 30001 1)
ACPI: DSDT 6002C2AAF0, 0024 (r2 SGI DSDTSN2 20001 AAC)
ACPI: FACS 6002C273B0, 0040
Number of logical nodes in system = 16
Number of memory chunks in system = 16
SAL 3.2: SGI SN2 version 1.50
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12

Unable to handle kernel NULL pointer dereference (address 00000000000044b8)
swapper[0]: Oops 8813272891392 [1]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 00001010084a2010 ifs : 8000000000000491 ip : [<a000000100087020>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at sn2_send_IPI+0x80/0x240
unat: 0000000000000000 pfs : 0000000000000491 rsc : 0000000000000003
rnat: 000000000000afc8 bsps: 000000000001003e pr : 65691ba55aa68599
ldrs: 0000000000000000 ccv : 0000000000ff03ff fpsr: 0009804c8a70433f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100942870 b6 : 00000000ff5423b0 b7 : e000000001fffc00 f6 : 1003e0000000000000000 f7 : 1003e0000000000000001 f8 : 1003e0000000000000000 f9 : 1003e0000000000000000 f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : 00000000000044b8 r3 : 0000000000000000 r8 : 0000000000000000 r9 : 0000000000000000 r10 : ffffffffffff6298 r11 : 0000000000000000 r12 : a000000100adfc30 r13 : a000000100ad0000 r14 : 0000000000000000 r15 : e000006003106298 r16 : e000006003110000 r17 : a000000100d0dce8 r18 : a000000100d0dce8 r19 : a000000100d0dce8 r20 : 0000000000000000 r21 : ffffffffffff0420 r22 : 0000000000000800 r23 : 0000000000000007 r24 : e0000060030b0000 r25 : 000000000004ffff r26 : a00000010097c460 r27 : e0000060030b0010 r28 : e0000060030b0000 r29 : e0000060030b0020 r30 : 0000000000000000 r31 : 00000000000007ff

Unable to handle kernel NULL pointer dereference (address 0000000000000000)
swapper[0]: Oops 8813272891392 [2]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 0000101008022018 ifs : 800000000000038c ip : [<a000000100175b30>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at kmem_cache_alloc+0x70/0x180
unat: 0000000000000000 pfs : 0000000000000610 rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr : 65691ba55aa69aa5
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70033f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100040bc0 b6 : a000000100040e00 b7 : a00000010000b730 f6 : 1003e45b3373c16c02344 f7 : 1003e9e3779b97f4a7c16 f8 : 1003e0a00000010001426 f9 : 10006c7fffffffd73ea5c f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : a000000100bae950 r3 : a000000100bac860 r8 : 0000000000000000 r9 : 0000000000000000 r10 : a000000100ad0c54 r11 : 0000000000000000 r12 : a000000100adf100 r13 : a000000100ad0000 r14 : 0000000000000014 r15 : a000000100adf190 r16 : a000000100adf198 r17 : a000000100ca1480 r18 : a000000100adf17c r19 : a000000100adf170 r20 : 0000000000000000 r21 : 0000000000000000 r22 : a000000100adf170 r23 : a000000100adf174 r24 : 000000000000000c r25 : a000000100adf180 r26 : a000000100adf174 r27 : 0000000000000000 r28 : 0000000000000000 r29 : a000000100adf178 r30 : 000000007fffffff r31 : 000000000000000c

#1 lore raw
Quoted 11 lines · Robin Holt

I bisected to this commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

3463a93def55c309f3c0d0a8aaf216be3be42d64 is first bad commit commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

Author: Alex Chiang <[email protected]>
Date:   Wed Jun 11 17:29:27 2008 -0600

[IA64] Update check_sal_cache_flush to use platform_send_ipi()
...

This fails to boot on any sn2 ia64 with the sn2_defconfig.

how about CONFIG_IA64_GENERIC?

#2 lore raw
In reply to Robin Holt

Hi Robin,

* Robin Holt <[email protected]>:

Quoted 73 lines

I bisected to this commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

3463a93def55c309f3c0d0a8aaf216be3be42d64 is first bad commit commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

Author: Alex Chiang <[email protected]>
Date:   Wed Jun 11 17:29:27 2008 -0600

[IA64] Update check_sal_cache_flush to use platform_send_ipi()
...

This fails to boot on any sn2 ia64 with the sn2_defconfig.

Here is the output from that boot.

fs0:\efi\SuSE> elilo net0:holt/v1 root=/dev/sda7 console=ttySG0
ELILO
Uncompressing Linux... done
Linux version 2.6.26-rc5-00223-g3463a93 (holt@attica) (gcc version 4.1.2 20070115 (prerelease) (SUSE Linux)) #14 SMP Tue Jun 24 07:27:34 CDT 2008 EFI v1.10 by INTEL: SALsystab=0x6002c25f10 ACPI 2.0=0x6002c26000
console [sn_sal0] enabled
ACPI: RSDP 6002C26000, 0024 (r2 SGI)
ACPI: XSDT 6002C297F0, 0044 (r1 SGI XSDTSN2 10001 7C)
ACPI: APIC 6002C26870, 032C (r1 SGI APICSN2 10001 1)
ACPI: SRAT 6002C26BB0, 06B0 (r1 SGI SRATSN2 10001 1)
ACPI: SLIT 6002C27270, 012C (r1 SGI SLITSN2 10001 1)
ACPI: FACP 6002C27400, 00F4 (r3 SGI FACPSN2 30001 1)
ACPI: DSDT 6002C2AAF0, 0024 (r2 SGI DSDTSN2 20001 AAC)
ACPI: FACS 6002C273B0, 0040
Number of logical nodes in system = 16
Number of memory chunks in system = 16
SAL 3.2: SGI SN2 version 1.50
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12

Unable to handle kernel NULL pointer dereference (address 00000000000044b8)
swapper[0]: Oops 8813272891392 [1]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 00001010084a2010 ifs : 8000000000000491 ip : [<a000000100087020>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at sn2_send_IPI+0x80/0x240
unat: 0000000000000000 pfs : 0000000000000491 rsc : 0000000000000003
rnat: 000000000000afc8 bsps: 000000000001003e pr : 65691ba55aa68599
ldrs: 0000000000000000 ccv : 0000000000ff03ff fpsr: 0009804c8a70433f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100942870 b6 : 00000000ff5423b0 b7 : e000000001fffc00 f6 : 1003e0000000000000000 f7 : 1003e0000000000000001 f8 : 1003e0000000000000000 f9 : 1003e0000000000000000 f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : 00000000000044b8 r3 : 0000000000000000 r8 : 0000000000000000 r9 : 0000000000000000 r10 : ffffffffffff6298 r11 : 0000000000000000 r12 : a000000100adfc30 r13 : a000000100ad0000 r14 : 0000000000000000 r15 : e000006003106298 r16 : e000006003110000 r17 : a000000100d0dce8 r18 : a000000100d0dce8 r19 : a000000100d0dce8 r20 : 0000000000000000 r21 : ffffffffffff0420 r22 : 0000000000000800 r23 : 0000000000000007 r24 : e0000060030b0000 r25 : 000000000004ffff r26 : a00000010097c460 r27 : e0000060030b0010 r28 : e0000060030b0000 r29 : e0000060030b0020 r30 : 0000000000000000 r31 : 00000000000007ff

Unable to handle kernel NULL pointer dereference (address 0000000000000000)
swapper[0]: Oops 8813272891392 [2]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 0000101008022018 ifs : 800000000000038c ip : [<a000000100175b30>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at kmem_cache_alloc+0x70/0x180
unat: 0000000000000000 pfs : 0000000000000610 rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr : 65691ba55aa69aa5
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70033f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100040bc0 b6 : a000000100040e00 b7 : a00000010000b730 f6 : 1003e45b3373c16c02344 f7 : 1003e9e3779b97f4a7c16 f8 : 1003e0a00000010001426 f9 : 10006c7fffffffd73ea5c f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : a000000100bae950 r3 : a000000100bac860

Quoted 8 lines

r8 : 0000000000000000 r9 : 0000000000000000 r10 : a000000100ad0c54 r11 : 0000000000000000 r12 : a000000100adf100 r13 : a000000100ad0000 r14 : 0000000000000014 r15 : a000000100adf190 r16 : a000000100adf198 r17 : a000000100ca1480 r18 : a000000100adf17c r19 : a000000100adf170 r20 : 0000000000000000 r21 : 0000000000000000 r22 : a000000100adf170 r23 : a000000100adf174 r24 : 000000000000000c r25 : a000000100adf180 r26 : a000000100adf174 r27 : 0000000000000000 r28 : 0000000000000000 r29 : a000000100adf178 r30 : 000000007fffffff r31 : 000000000000000c

Here's the disassembly of sn2_send_IPI:

(gdb) disass sn2_send_IPI
Dump of assembler code for function sn2_send_IPI:

0xa000000100633f80 <sn2_send_IPI+0>:    [MMI]       alloc r39=ar.pfs,17,9,0
0xa000000100633f81 <sn2_send_IPI+1>:                adds r12=-160,r12
0xa000000100633f82 <sn2_send_IPI+2>:                mov r38=b0
0xa000000100633f90 <sn2_send_IPI+16>:   [MMI]       addl r19=-1557544,r1
0xa000000100633f91 <sn2_send_IPI+17>:               mov r40=r1
0xa000000100633f92 <sn2_send_IPI+18>:               sxt4 r20=r32
0xa000000100633fa0 <sn2_send_IPI+32>:   [MMI]       mov r21=-64480
0xa000000100633fa1 <sn2_send_IPI+33>:               nop.m 0x0
0xa000000100633fa2 <sn2_send_IPI+34>:               mov r10=-33312
0xa000000100633fb0 <sn2_send_IPI+48>:   [MMI]       nop.m 0x0;;
0xa000000100633fb1 <sn2_send_IPI+49>:               ld8 r16=[r21]
0xa000000100633fb2 <sn2_send_IPI+50>:               shladd r8=r20,2,r0
0xa000000100633fc0 <sn2_send_IPI+64>:   [MII]       mov r18=r19
0xa000000100633fc1 <sn2_send_IPI+65>:               adds r41=48,r12;;
0xa000000100633fc2 <sn2_send_IPI+66>:               add r17=r8,r18;;
0xa000000100633fd0 <sn2_send_IPI+80>:   [MII]       ld4.acq r11=[r17]
0xa000000100633fd1 <sn2_send_IPI+81>:               nop.i 0x0;;
0xa000000100633fd2 <sn2_send_IPI+82>:               sxt4 r37=r11;;
0xa000000100633fe0 <sn2_send_IPI+96>:   [MMI]       add r15=r10,r16;;
0xa000000100633fe1 <sn2_send_IPI+97>:               ld8 r9=[r15]
0xa000000100633fe2 <sn2_send_IPI+98>:               nop.i 0x0;;
0xa000000100633ff0 <sn2_send_IPI+112>:  [MII]       nop.m 0x0
0xa000000100633ff1 <sn2_send_IPI+113>:              add r3=r8,r9;;
0xa000000100633ff2 <sn2_send_IPI+114>:              addl r2=17592,r3;;
0xa000000100634000 <sn2_send_IPI+128>:  [MMI]       ld2 r3=[r2];;

Looks like we're dying on this access above ^^

0xa000000100634001 <sn2_send_IPI+129>:              nop.m 0x0
0xa000000100634002 <sn2_send_IPI+130>:              sxt2 r14=r3;;
0xa000000100634010 <sn2_send_IPI+144>:  [MIB]       mov r32=r14
0xa000000100634011 <sn2_send_IPI+145>:              cmp4.eq p7,p6=-1,r14

My guess something bad is happening when we try this:

	nasid = cpuid_to_nasid(cpuid);

And include/asm-ia64/sn/sn_cpuid.h says:

#define cpuid_to_nasid(cpuid) (sn_nodepda->phys_cpuid[cpuid].nasid)

Are we calling sn2_send_IPI too early? Do we have to do some sort of special initialization before sn_nodepda is valid? It all *looks* like we should be fine because we do

	cpu_init()
	platform_cpu_init()
	sn_cpu_init()

Before calling check_sal_cache_flush()... Very curious.

Can you try the debug patch included below?

Thanks.

/ac

arch/ia64/sn/kernel/setup.c +5 -0
index bb1d249..a6a0be5 100644
@@ -627,13 +627,18 @@ void __cpuinit sn_cpu_init(void)
 			nodepdaindr[i]->phys_cpuid[cpuid].nasid = nasid;
 			nodepdaindr[i]->phys_cpuid[cpuid].slice = slice;
 			nodepdaindr[i]->phys_cpuid[cpuid].subnode = subnode;
+			printk(KERN_INFO "nodepdaindr[%d]->phys_cpuid[%d] - nasid %d slice %d subnode %d\n", i, cpuid, nasid, slice, subnode);
 		}
 	}
 
 	cnode = nasid_to_cnodeid(nasid);
 
+	printk(KERN_INFO "cnode %d\n", cnode);
+
 	sn_nodepda = nodepdaindr[cnode];
 
+	printk(KERN_INFO "sn_nodepda 0x%p\n", sn_nodepda);
+
 	pda->led_address =
 	    (typeof(pda->led_address)) (LED0 + (slice << LED_CPU_SHIFT));
 	pda->led_state = LED_ALWAYS_SET;
author
#3 lore raw

I have not tried your patch yet. Actually just read the email. Jack Steiner did point out that booting with force_pal_cache_flush on the command line will get it to boot.

I will try your patch shortly and send you output.

Thanks,
Robin

Quoted 168 lines · Alex Chiang

Hi Robin,

* Robin Holt <[email protected]>:

Quoted 73 lines

I bisected to this commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

3463a93def55c309f3c0d0a8aaf216be3be42d64 is first bad commit commit 3463a93def55c309f3c0d0a8aaf216be3be42d64

Author: Alex Chiang <[email protected]>
Date:   Wed Jun 11 17:29:27 2008 -0600

[IA64] Update check_sal_cache_flush to use platform_send_ipi()
...

This fails to boot on any sn2 ia64 with the sn2_defconfig.

Here is the output from that boot.

fs0:\efi\SuSE> elilo net0:holt/v1 root=/dev/sda7 console=ttySG0
ELILO
Uncompressing Linux... done
Linux version 2.6.26-rc5-00223-g3463a93 (holt@attica) (gcc version 4.1.2 20070115 (prerelease) (SUSE Linux)) #14 SMP Tue Jun 24 07:27:34 CDT 2008 EFI v1.10 by INTEL: SALsystab=0x6002c25f10 ACPI 2.0=0x6002c26000
console [sn_sal0] enabled
ACPI: RSDP 6002C26000, 0024 (r2 SGI)
ACPI: XSDT 6002C297F0, 0044 (r1 SGI XSDTSN2 10001 7C)
ACPI: APIC 6002C26870, 032C (r1 SGI APICSN2 10001 1)
ACPI: SRAT 6002C26BB0, 06B0 (r1 SGI SRATSN2 10001 1)
ACPI: SLIT 6002C27270, 012C (r1 SGI SLITSN2 10001 1)
ACPI: FACP 6002C27400, 00F4 (r3 SGI FACPSN2 30001 1)
ACPI: DSDT 6002C2AAF0, 0024 (r2 SGI DSDTSN2 20001 AAC)
ACPI: FACS 6002C273B0, 0040
Number of logical nodes in system = 16
Number of memory chunks in system = 16
SAL 3.2: SGI SN2 version 1.50
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12

Unable to handle kernel NULL pointer dereference (address 00000000000044b8)
swapper[0]: Oops 8813272891392 [1]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 00001010084a2010 ifs : 8000000000000491 ip : [<a000000100087020>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at sn2_send_IPI+0x80/0x240
unat: 0000000000000000 pfs : 0000000000000491 rsc : 0000000000000003
rnat: 000000000000afc8 bsps: 000000000001003e pr : 65691ba55aa68599
ldrs: 0000000000000000 ccv : 0000000000ff03ff fpsr: 0009804c8a70433f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100942870 b6 : 00000000ff5423b0 b7 : e000000001fffc00 f6 : 1003e0000000000000000 f7 : 1003e0000000000000001 f8 : 1003e0000000000000000 f9 : 1003e0000000000000000 f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : 00000000000044b8 r3 : 0000000000000000 r8 : 0000000000000000 r9 : 0000000000000000 r10 : ffffffffffff6298 r11 : 0000000000000000 r12 : a000000100adfc30 r13 : a000000100ad0000 r14 : 0000000000000000 r15 : e000006003106298 r16 : e000006003110000 r17 : a000000100d0dce8 r18 : a000000100d0dce8 r19 : a000000100d0dce8 r20 : 0000000000000000 r21 : ffffffffffff0420 r22 : 0000000000000800 r23 : 0000000000000007 r24 : e0000060030b0000 r25 : 000000000004ffff r26 : a00000010097c460 r27 : e0000060030b0010 r28 : e0000060030b0000 r29 : e0000060030b0020 r30 : 0000000000000000 r31 : 00000000000007ff

Unable to handle kernel NULL pointer dereference (address 0000000000000000)
swapper[0]: Oops 8813272891392 [2]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 0000101008022018 ifs : 800000000000038c ip : [<a000000100175b30>] Not tainted (2.6.26-rc5-00223-g3463a93)
ip is at kmem_cache_alloc+0x70/0x180
unat: 0000000000000000 pfs : 0000000000000610 rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr : 65691ba55aa69aa5
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70033f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100040bc0 b6 : a000000100040e00 b7 : a00000010000b730 f6 : 1003e45b3373c16c02344 f7 : 1003e9e3779b97f4a7c16 f8 : 1003e0a00000010001426 f9 : 10006c7fffffffd73ea5c f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e8bd10 r2 : a000000100bae950 r3 : a000000100bac860

Quoted 8 lines

r8 : 0000000000000000 r9 : 0000000000000000 r10 : a000000100ad0c54 r11 : 0000000000000000 r12 : a000000100adf100 r13 : a000000100ad0000 r14 : 0000000000000014 r15 : a000000100adf190 r16 : a000000100adf198 r17 : a000000100ca1480 r18 : a000000100adf17c r19 : a000000100adf170 r20 : 0000000000000000 r21 : 0000000000000000 r22 : a000000100adf170 r23 : a000000100adf174 r24 : 000000000000000c r25 : a000000100adf180 r26 : a000000100adf174 r27 : 0000000000000000 r28 : 0000000000000000 r29 : a000000100adf178 r30 : 000000007fffffff r31 : 000000000000000c

Here's the disassembly of sn2_send_IPI:

(gdb) disass sn2_send_IPI
Dump of assembler code for function sn2_send_IPI:

0xa000000100633f80 <sn2_send_IPI+0>:    [MMI]       alloc r39=ar.pfs,17,9,0
0xa000000100633f81 <sn2_send_IPI+1>:                adds r12=-160,r12
0xa000000100633f82 <sn2_send_IPI+2>:                mov r38°
0xa000000100633f90 <sn2_send_IPI+16>:   [MMI]       addl r19=-1557544,r1
0xa000000100633f91 <sn2_send_IPI+17>:               mov r40=r1
0xa000000100633f92 <sn2_send_IPI+18>:               sxt4 r20=r32
0xa000000100633fa0 <sn2_send_IPI+32>:   [MMI]       mov r21=-64480
0xa000000100633fa1 <sn2_send_IPI+33>:               nop.m 0x0
0xa000000100633fa2 <sn2_send_IPI+34>:               mov r10=-33312
0xa000000100633fb0 <sn2_send_IPI+48>:   [MMI]       nop.m 0x0;;
0xa000000100633fb1 <sn2_send_IPI+49>:               ld8 r16=[r21]
0xa000000100633fb2 <sn2_send_IPI+50>:               shladd r8=r20,2,r0
0xa000000100633fc0 <sn2_send_IPI+64>:   [MII]       mov r18=r19
0xa000000100633fc1 <sn2_send_IPI+65>:               adds r41H,r12;;
0xa000000100633fc2 <sn2_send_IPI+66>:               add r17=r8,r18;;
0xa000000100633fd0 <sn2_send_IPI+80>:   [MII]       ld4.acq r11=[r17]
0xa000000100633fd1 <sn2_send_IPI+81>:               nop.i 0x0;;
0xa000000100633fd2 <sn2_send_IPI+82>:               sxt4 r37=r11;;
0xa000000100633fe0 <sn2_send_IPI+96>:   [MMI]       add r15=r10,r16;;
0xa000000100633fe1 <sn2_send_IPI+97>:               ld8 r9=[r15]
0xa000000100633fe2 <sn2_send_IPI+98>:               nop.i 0x0;;
0xa000000100633ff0 <sn2_send_IPI+112>:  [MII]       nop.m 0x0
0xa000000100633ff1 <sn2_send_IPI+113>:              add r3=r8,r9;;
0xa000000100633ff2 <sn2_send_IPI+114>:              addl r2592,r3;;
0xa000000100634000 <sn2_send_IPI+128>:  [MMI]       ld2 r3=[r2];;

Looks like we're dying on this access above ^^

0xa000000100634001 <sn2_send_IPI+129>:              nop.m 0x0
0xa000000100634002 <sn2_send_IPI+130>:              sxt2 r14=r3;;
0xa000000100634010 <sn2_send_IPI+144>:  [MIB]       mov r32=r14
0xa000000100634011 <sn2_send_IPI+145>:              cmp4.eq p7,p6=-1,r14

My guess something bad is happening when we try this:

	nasid = cpuid_to_nasid(cpuid);

And include/asm-ia64/sn/sn_cpuid.h says:

#define cpuid_to_nasid(cpuid) (sn_nodepda->phys_cpuid[cpuid].nasid)

Are we calling sn2_send_IPI too early? Do we have to do some sort of special initialization before sn_nodepda is valid? It all *looks* like we should be fine because we do

	cpu_init()
	platform_cpu_init()
	sn_cpu_init()

Before calling check_sal_cache_flush()... Very curious.

Can you try the debug patch included below?

Thanks.

/ac

arch/ia64/sn/kernel/setup.c +5 -0
index bb1d249..a6a0be5 100644
@@ -627,13 +627,18 @@ void __cpuinit sn_cpu_init(void)
 			nodepdaindr[i]->phys_cpuid[cpuid].nasid = nasid;
 			nodepdaindr[i]->phys_cpuid[cpuid].slice = slice;
 			nodepdaindr[i]->phys_cpuid[cpuid].subnode = subnode;
+			printk(KERN_INFO "nodepdaindr[%d]->phys_cpuid[%d] - nasid %d slice %d subnode %d\n", i, cpuid, nasid, slice, subnode);
 		}
 	}
 
 	cnode = nasid_to_cnodeid(nasid);
 
+	printk(KERN_INFO "cnode %d\n", cnode);
+
 	sn_nodepda = nodepdaindr[cnode];
 
+	printk(KERN_INFO "sn_nodepda 0x%p\n", sn_nodepda);
+
 	pda->led_address >  	    (typeof(pda->led_address)) (LED0 + (slice << LED_CPU_SHIFT));
 	pda->led_state = LED_ALWAYS_SET;
#4 lore raw

* Robin Holt <[email protected]>:

Quoted 3 lines

I have not tried your patch yet. Actually just read the email. Jack Steiner did point out that booting with force_pal_cache_flush on the command line will get it to boot.

Yes, that command line arg will get your kernel to boot because it forces us to completely skip check_sal_cache_flush. It's probably not what you want to be using long term, though.

Quoted 1 line

I will try your patch shortly and send you output.

Thanks.

/ac

author
#5 lore raw
In reply to Alex Chiang

Here is the output. This is from a different boot, but it does look the same.

Robin

fs0:\efi\SuSE> elilo net0:holt/v1 root=/dev/sda7
ELILO
Uncompressing Linux... done
Initializing cgroup subsys cpuset
Linux version 2.6.26-rc7-holt-00051-g62786b9-dirty (holt@attica) (gcc version 4.1.2 20070115 (prerelease) (SUSE Linux)) #19 SMP Tue Jun 24 10:22:55 CDT 2008 EFI v1.10 by INTEL: SALsystab=0x6002c25f10 ACPI 2.0=0x6002c26000
console [sn_sal0] enabled
ACPI: RSDP 6002C26000, 0024 (r2 SGI)
ACPI: XSDT 6002C297F0, 0044 (r1 SGI XSDTSN2 10001 7C)
ACPI: APIC 6002C26870, 032C (r1 SGI APICSN2 10001 1)
ACPI: SRAT 6002C26BB0, 06B0 (r1 SGI SRATSN2 10001 1)
ACPI: SLIT 6002C27270, 012C (r1 SGI SLITSN2 10001 1)
ACPI: FACP 6002C27400, 00F4 (r3 SGI FACPSN2 30001 1)
ACPI: DSDT 6002C2AAF0, 0024 (r2 SGI DSDTSN2 20001 AAC)
ACPI: FACS 6002C273B0, 0040
Number of logical nodes in system = 16
Number of memory chunks in system = 16
SAL 3.2: SGI SN2 version 1.50
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12

Unable to handle kernel NULL pointer dereference (address 00000000000044b8)
swapper[0]: Oops 8813272891392 [1]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 00001010084a2010 ifs : 8000000000000491 ip : [<a000000100087020>] Not tainted (2.6.26-rc7-holt-00051-g62786b9-dirty)
ip is at sn2_send_IPI+0x80/0x240
unat: 0000000000000000 pfs : 0000000000000491 rsc : 0000000000000003
rnat: 000000000000afc8 bsps: 000000000001003e pr : 65691ba55aa68599
ldrs: 0000000000000000 ccv : 0000000000ff03ff fpsr: 0009804c8a70433f csd : 0000000000000000 ssd : 0000000000000000
b0 : a0000001009529b0 b6 : 00000000ff5423b0 b7 : e000000001fffc00 f6 : 1003e0000000000000000 f7 : 1003e0000000000000001 f8 : 1003e0000000000000000 f9 : 1003e0000000000000000 f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e9d010 r2 : 00000000000044b8 r3 : 0000000000000000 r8 : 0000000000000000 r9 : 0000000000000000 r10 : ffffffffffff6298 r11 : 0000000000000000 r12 : a000000100aefc30 r13 : a000000100ae0000 r14 : 0000000000000000 r15 : e000006003106298 r16 : e000006003110000 r17 : a000000100d1f3e8 r18 : a000000100d1f3e8 r19 : a000000100d1f3e8 r20 : 0000000000000000 r21 : ffffffffffff0420 r22 : 0000000000000800 r23 : 0000000000000007 r24 : e0000060030b0000 r25 : 000000000004ffff r26 : a00000010098d440 r27 : e0000060030b0010 r28 : e0000060030b0000 r29 : e0000060030b0020 r30 : 0000000000000000 r31 : 00000000000007ff

Unable to handle kernel NULL pointer dereference (address 0000000000000000)
swapper[0]: Oops 8813272891392 [2]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 0000101008022018 ifs : 800000000000038c ip : [<a000000100182e30>] Not tainted (2.6.26-rc7-holt-00051-g62786b9-dirty)
ip is at kmem_cache_alloc+0x70/0x180
unat: 0000000000000000 pfs : 0000000000000610 rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr : 65691ba55aa69aa5
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70033f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100040bc0 b6 : a000000100040e00 b7 : a00000010000b730 f6 : 1003e45b3373c16c02344 f7 : 1003e9e3779b97f4a7c16 f8 : 1003e0a00000010001426 f9 : 10006c7fffffffd73ea5c f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e9d010 r2 : a000000100bbe950 r3 : a000000100bbc860 r8 : 0000000000000000 r9 : 0000000000000000 r10 : a000000100ae0cf4 r11 : 0000000000000000 r12 : a000000100aef100 r13 : a000000100ae0000 r14 : 0000000000000014 r15 : a000000100aef190 r16 : a000000100aef198 r17 : a000000100cb3e50 r18 : a000000100aef17c r19 : a000000100aef170 r20 : 0000000000000000 r21 : 0000000000000000 r22 : a000000100aef170 r23 : a000000100aef174 r24 : 000000000000000c r25 : a000000100aef180 r26 : a000000100aef174 r27 : 0000000000000000 r28 : 0000000000000000 r29 : a000000100aef178 r30 : 000000007fffffff r31 : 000000000000000c

#6 lore raw

* Robin Holt <[email protected]>:

Quoted 2 lines

Here is the output. This is from a different boot, but it does look the same.

Hrm, that's odd. There's no debug output at all. Did you apply the patch?

Quoted 7 lines

Robin

fs0:\efi\SuSE> elilo net0:holt/v1 root=/dev/sda7
ELILO
Uncompressing Linux... done
Initializing cgroup subsys cpuset
Linux version 2.6.26-rc7-holt-00051-g62786b9-dirty (holt@attica) (gcc version 4.1.2 20070115 (prerelease) (SUSE Linux)) #19 SMP Tue Jun 24 10:22:55 CDT 2008

Ok, -dirty -- sorry, I believe you. :)

So that tells me we're not calling sn_cpu_init()? That's not what the code says should be happening...

The lack of output kinda makes sense, since the oops looks like it's coming from trying to access an uninitialized sn_nodepda, but I'm really confused as to why, since we should have initialized it before check_sal_cache_flush().

Anyone at SGI with more of a clue than me? (before I start digging in depth)

Thanks.

/ac

Quoted 64 lines

EFI v1.10 by INTEL: SALsystab=0x6002c25f10 ACPI 2.0=0x6002c26000 console [sn_sal0] enabled
ACPI: RSDP 6002C26000, 0024 (r2 SGI)
ACPI: XSDT 6002C297F0, 0044 (r1 SGI XSDTSN2 10001 7C)
ACPI: APIC 6002C26870, 032C (r1 SGI APICSN2 10001 1)
ACPI: SRAT 6002C26BB0, 06B0 (r1 SGI SRATSN2 10001 1)
ACPI: SLIT 6002C27270, 012C (r1 SGI SLITSN2 10001 1)
ACPI: FACP 6002C27400, 00F4 (r3 SGI FACPSN2 30001 1)
ACPI: DSDT 6002C2AAF0, 0024 (r2 SGI DSDTSN2 20001 AAC)
ACPI: FACS 6002C273B0, 0040
Number of logical nodes in system = 16
Number of memory chunks in system = 16
SAL 3.2: SGI SN2 version 1.50
SAL Platform features: ITC_Drift
SAL: AP wakeup using external interrupt vector 0x12

Unable to handle kernel NULL pointer dereference (address 00000000000044b8)
swapper[0]: Oops 8813272891392 [1]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 00001010084a2010 ifs : 8000000000000491 ip : [<a000000100087020>] Not tainted (2.6.26-rc7-holt-00051-g62786b9-dirty)
ip is at sn2_send_IPI+0x80/0x240
unat: 0000000000000000 pfs : 0000000000000491 rsc : 0000000000000003
rnat: 000000000000afc8 bsps: 000000000001003e pr : 65691ba55aa68599
ldrs: 0000000000000000 ccv : 0000000000ff03ff fpsr: 0009804c8a70433f csd : 0000000000000000 ssd : 0000000000000000
b0 : a0000001009529b0 b6 : 00000000ff5423b0 b7 : e000000001fffc00 f6 : 1003e0000000000000000 f7 : 1003e0000000000000001 f8 : 1003e0000000000000000 f9 : 1003e0000000000000000 f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e9d010 r2 : 00000000000044b8 r3 : 0000000000000000 r8 : 0000000000000000 r9 : 0000000000000000 r10 : ffffffffffff6298 r11 : 0000000000000000 r12 : a000000100aefc30 r13 : a000000100ae0000 r14 : 0000000000000000 r15 : e000006003106298 r16 : e000006003110000 r17 : a000000100d1f3e8 r18 : a000000100d1f3e8 r19 : a000000100d1f3e8 r20 : 0000000000000000 r21 : ffffffffffff0420 r22 : 0000000000000800 r23 : 0000000000000007 r24 : e0000060030b0000 r25 : 000000000004ffff r26 : a00000010098d440 r27 : e0000060030b0010 r28 : e0000060030b0000 r29 : e0000060030b0020 r30 : 0000000000000000 r31 : 00000000000007ff

Unable to handle kernel NULL pointer dereference (address 0000000000000000)
swapper[0]: Oops 8813272891392 [2]
Modules linked in:

Pid: 0, CPU 0, comm: swapper
psr : 0000101008022018 ifs : 800000000000038c ip : [<a000000100182e30>] Not tainted (2.6.26-rc7-holt-00051-g62786b9-dirty)
ip is at kmem_cache_alloc+0x70/0x180
unat: 0000000000000000 pfs : 0000000000000610 rsc : 0000000000000003
rnat: 0000000000000000 bsps: 0000000000000000 pr : 65691ba55aa69aa5
ldrs: 0000000000000000 ccv : 0000000000000000 fpsr: 0009804c8a70033f csd : 0000000000000000 ssd : 0000000000000000
b0 : a000000100040bc0 b6 : a000000100040e00 b7 : a00000010000b730 f6 : 1003e45b3373c16c02344 f7 : 1003e9e3779b97f4a7c16 f8 : 1003e0a00000010001426 f9 : 10006c7fffffffd73ea5c f10 : 100068fffffffff700000 f11 : 1003e0000000000000090 r1 : a000000100e9d010 r2 : a000000100bbe950 r3 : a000000100bbc860 r8 : 0000000000000000 r9 : 0000000000000000 r10 : a000000100ae0cf4 r11 : 0000000000000000 r12 : a000000100aef100 r13 : a000000100ae0000 r14 : 0000000000000014 r15 : a000000100aef190 r16 : a000000100aef198 r17 : a000000100cb3e50 r18 : a000000100aef17c r19 : a000000100aef170 r20 : 0000000000000000 r21 : 0000000000000000 r22 : a000000100aef170 r23 : a000000100aef174 r24 : 000000000000000c r25 : a000000100aef180 r26 : a000000100aef174 r27 : 0000000000000000 r28 : 0000000000000000 r29 : a000000100aef178 r30 : 000000007fffffff r31 : 000000000000000c

author
#7 lore raw
Quoted 6 lines · Alex Chiang

* Robin Holt <[email protected]>:

Quoted 2 lines

Here is the output. This is from a different boot, but it does look the same.

Hrm, that's odd. There's no debug output at all. Did you apply the patch?

...

Quoted 2 lines

Anyone at SGI with more of a clue than me? (before I start digging in depth)

Jes hit this at about the same time. He posted a patch and I verified it works on sn2. You should probably give it a try on some HP boxes as well.

Thanks,
Robin