2005-12-01 21:15:49

by Bharath Ramesh

[permalink] [raw]
Subject: Only one processor detected in 8-Way opteron in 32-bit mode

I have an 8-Way Opteron server. The machine has both 32-bit and 64-bit
installations. I find that I am unable to use more than one processor
in the 32-bit mode. Only one processor is detected while the other 7
arent. In 64-bit mode all the processors are detected. I am attaching
the dmesg from the currently running 2.6.11.10 kernel. I tried the
2.615-rc4 and I still see the same problem and dmesg is nearly the
same.

dmesg:

fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
BIOS-e820: 00000000ff780000 - 0000000100000000 (reserved)
BIOS-e820: 0000000100000000 - 0000000440000000 (usable)
16512MB HIGHMEM available.
896MB LOWMEM available.
found SMP MP-table at 000ff780
NX (Execute Disable) protection: active
On node 0 totalpages: 4456448
DMA zone: 4096 pages, LIFO batch:1
Normal zone: 225280 pages, LIFO batch:16
HighMem zone: 4227072 pages, LIFO batch:16
DMI 2.3 present.
ACPI: RSDP (v002 ACPIAM ) @ 0x000f6ca0
ACPI: XSDT (v001 A M I OEMXSDT 0x02000502 MSFT 0x00000097) @ 0xbfff0100
ACPI: FADT (v003 A M I OEMFACP 0x02000502 MSFT 0x00000097) @ 0xbfff0281
ACPI: MADT (v001 A M I OEMAPIC 0x02000502 MSFT 0x00000097) @ 0xbfff0380
ACPI: OEMB (v001 A M I OEMBIOS 0x02000502 MSFT 0x00000097) @ 0xbffff040
ACPI: SRAT (v001 A M I OEMSRAT 0x02000502 MSFT 0x00000097) @ 0xbfff3cc0
ACPI: DSDT (v001 H805V H805V120 0x00000000 INTL 0x02002026) @ 0x00000000
ACPI: Local APIC address 0xfee00000
ACPI: LAPIC (acpi_id[0x01] lapic_id[0x10] enabled)
Processor #16 15:5 APIC version 16
Processor #16 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x02] lapic_id[0x11] enabled)
Processor #17 15:5 APIC version 16
Processor #17 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x03] lapic_id[0x12] enabled)
Processor #18 15:5 APIC version 16
Processor #18 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x04] lapic_id[0x13] enabled)
Processor #19 15:5 APIC version 16
Processor #19 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x05] lapic_id[0x14] enabled)
Processor #20 15:5 APIC version 16
Processor #20 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x06] lapic_id[0x15] enabled)
Processor #21 15:5 APIC version 16
Processor #21 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x07] lapic_id[0x16] enabled)
Processor #22 15:5 APIC version 16
Processor #22 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x08] lapic_id[0x17] enabled)
Processor #23 15:5 APIC version 16
Processor #23 INVALID. (Max ID: 256).
ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
ACPI: IOAPIC (id[0x02] address[0xfcffe000] gsi_base[24])
IOAPIC[1]: apic_id 2, version 17, address 0xfcffe000, GSI 24-27
ACPI: IOAPIC (id[0x03] address[0xfcfff000] gsi_base[28])
IOAPIC[2]: apic_id 3, version 17, address 0xfcfff000, GSI 28-31
ACPI: IOAPIC (id[0x10] address[0xfbffe000] gsi_base[32])
IOAPIC[3]: Invalid apic_id 16, trying 3
IOAPIC[3]: apic_id 3 already used, trying 0
IOAPIC[3]: apic_id 0, version 17, address 0xfbffe000, GSI 32-35
ACPI: IOAPIC (id[0x11] address[0xfbfff000] gsi_base[36])
IOAPIC[4]: Invalid apic_id 17, trying 4
IOAPIC[4]: apic_id 4, version 17, address 0xfbfff000, GSI 36-39
ACPI: IOAPIC (id[0x12] address[0xfacfc000] gsi_base[40])
IOAPIC[5]: Invalid apic_id 18, trying 5
IOAPIC[5]: apic_id 5, version 17, address 0xfacfc000, GSI 40-43
ACPI: IOAPIC (id[0x13] address[0xfacfd000] gsi_base[44])
IOAPIC[6]: Invalid apic_id 19, trying 6
IOAPIC[6]: apic_id 6, version 17, address 0xfacfd000, GSI 44-47
ACPI: IOAPIC (id[0x14] address[0xfacfe000] gsi_base[48])
IOAPIC[7]: Invalid apic_id 20, trying 7
IOAPIC[7]: apic_id 7, version 17, address 0xfacfe000, GSI 48-51
ACPI: [APIC:0x01] ignored 1 entries of 9 found
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: IRQ0 used by override.
ACPI: IRQ2 used by override.
ACPI: IRQ9 used by override.
Enabling APIC mode: Flat. Using 8 I/O APICs
Using ACPI (MADT) for SMP configuration information
Allocating PCI resources starting at c0000000 (gap: c0000000:3f780000)
Built 1 zonelists
Kernel command line: ro root=/dev/hda3 vga=791 rhgb quiet
mapped APIC to ffffd000 (fee00000)
mapped IOAPIC to ffffc000 (fec00000)
mapped IOAPIC to ffffb000 (fcffe000)
mapped IOAPIC to ffffa000 (fcfff000)
mapped IOAPIC to ffff9000 (fbffe000)
mapped IOAPIC to ffff8000 (fbfff000)
mapped IOAPIC to ffff7000 (facfc000)
mapped IOAPIC to ffff6000 (facfd000)
mapped IOAPIC to ffff5000 (facfe000)
Initializing CPU#0
CPU 0 irqstacks, hard=c0409000 soft=c0405000
PID hash table entries: 4096 (order: 12, 65536 bytes)
Detected 2005.226 MHz processor.
Using tsc for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 16632192k/17825792k available (2064k kernel code, 143976k
reserved, 782k data, 220k init, 15859648k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay loop... 3940.35 BogoMIPS (lpj=1970176)
Security Framework v1.0.0 initialized
Capability LSM initialized
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
CPU: After generic identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: After vendor identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU: After all inits, caps: 078bfbff e1d3fbff 00000000 00000010
00000000 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
CPU0: AMD Opteron(tm) Processor 846 stepping 0a
per-CPU timeslice cutoff: 2925.70 usecs.
task migration cache decay timeout: 3 msecs.
weird, boot CPU (#0) not listed by the BIOS.
Total of 1 processors activated (3940.35 BogoMIPS).
ENABLING IO-APIC IRQs
..TIMER: vector=0x31 pin1=2 pin2=-1
Brought up 1 CPUs
CPU0 attaching sched-domain:
domain 0: span 1
groups: 1
checking if image is initramfs... it is


2005-12-01 23:27:41

by Keith Mannthey

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Looks like something odd is going on with your processor apic id's.

Lines like
Processor #20 INVALID. (Max ID: 256)
are bad...

Could you try the following patch with i386? It seems suspect to me
and could help identify the problem.

Your box has high processor apic ids and looks to be in flat apic mode
(versus clustered). I think the code is assuming the box is in
clustered apic mode?

Thanks,
Keith


Attachments:
(No filename) (407.00 B)
apicid.fix (344.00 B)
Download all attachments

2005-12-02 00:51:31

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Patched 2.6.15-rc4 with the patch. It still is the same.

dmesg:
Linux version 2.6.15-rc4 (root@drones) (gcc version 3.4.3 20050227
(Red Hat 3.4.3-22.1)) #1 SMP PREEMPT Thu Dec 1 19:37:03 EST 2005
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
BIOS-e820: 00000000ff780000 - 0000000100000000 (reserved)
BIOS-e820: 0000000100000000 - 0000000440000000 (usable)
16512MB HIGHMEM available.
896MB LOWMEM available.
found SMP MP-table at 000ff780
NX (Execute Disable) protection: active
On node 0 totalpages: 4456448
DMA zone: 4096 pages, LIFO batch:2
DMA32 zone: 0 pages, LIFO batch:2
Normal zone: 225280 pages, LIFO batch:64
HighMem zone: 4227072 pages, LIFO batch:64
DMI 2.3 present.
ACPI: RSDP (v002 ACPIAM ) @ 0x000f6ca0
ACPI: XSDT (v001 A M I OEMXSDT 0x02000502 MSFT 0x00000097) @ 0xbfff0100
ACPI: FADT (v003 A M I OEMFACP 0x02000502 MSFT 0x00000097) @ 0xbfff0281
ACPI: MADT (v001 A M I OEMAPIC 0x02000502 MSFT 0x00000097) @ 0xbfff0380
ACPI: OEMB (v001 A M I OEMBIOS 0x02000502 MSFT 0x00000097) @ 0xbffff040
ACPI: SRAT (v001 A M I OEMSRAT 0x02000502 MSFT 0x00000097) @ 0xbfff3cc0
ACPI: DSDT (v001 H805V H805V120 0x00000000 INTL 0x02002026) @ 0x00000000
ACPI: Local APIC address 0xfee00000
ACPI: LAPIC (acpi_id[0x01] lapic_id[0x10] enabled)
Processor #16 15:5 APIC version 16
Processor #16 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x02] lapic_id[0x11] enabled)
Processor #17 15:5 APIC version 16
Processor #17 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x03] lapic_id[0x12] enabled)
Processor #18 15:5 APIC version 16
Processor #18 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x04] lapic_id[0x13] enabled)
Processor #19 15:5 APIC version 16
Processor #19 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x05] lapic_id[0x14] enabled)
Processor #20 15:5 APIC version 16
Processor #20 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x06] lapic_id[0x15] enabled)
Processor #21 15:5 APIC version 16
Processor #21 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x07] lapic_id[0x16] enabled)
Processor #22 15:5 APIC version 16
Processor #22 INVALID. (Max ID: 256).
ACPI: LAPIC (acpi_id[0x08] lapic_id[0x17] enabled)
Processor #23 15:5 APIC version 16
Processor #23 INVALID. (Max ID: 256).
ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
ACPI: IOAPIC (id[0x02] address[0xfcffe000] gsi_base[24])
IOAPIC[1]: apic_id 2, version 17, address 0xfcffe000, GSI 24-27
ACPI: IOAPIC (id[0x03] address[0xfcfff000] gsi_base[28])
IOAPIC[2]: apic_id 3, version 17, address 0xfcfff000, GSI 28-31
ACPI: IOAPIC (id[0x10] address[0xfbffe000] gsi_base[32])
IOAPIC[3]: apic_id 16, version 17, address 0xfbffe000, GSI 32-35
ACPI: IOAPIC (id[0x11] address[0xfbfff000] gsi_base[36])
IOAPIC[4]: apic_id 17, version 17, address 0xfbfff000, GSI 36-39
ACPI: IOAPIC (id[0x12] address[0xfacfc000] gsi_base[40])
IOAPIC[5]: apic_id 18, version 17, address 0xfacfc000, GSI 40-43
ACPI: IOAPIC (id[0x13] address[0xfacfd000] gsi_base[44])
IOAPIC[6]: apic_id 19, version 17, address 0xfacfd000, GSI 44-47
ACPI: IOAPIC (id[0x14] address[0xfacfe000] gsi_base[48])
IOAPIC[7]: apic_id 20, version 17, address 0xfacfe000, GSI 48-51
ACPI: IOAPIC (id[0x15] address[0xfacff000] gsi_base[52])
IOAPIC[8]: apic_id 21, version 17, address 0xfacff000, GSI 52-55
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: IRQ0 used by override.
ACPI: IRQ2 used by override.
ACPI: IRQ9 used by override.
Enabling APIC mode: Flat. Using 9 I/O APICs
Using ACPI (MADT) for SMP configuration information
Allocating PCI resources starting at c4000000 (gap: c0000000:3f780000)
Built 1 zonelists
Kernel command line: ro root=/dev/hda3 vga=791 rhgb quiet
mapped APIC to ffffd000 (fee00000)
mapped IOAPIC to ffffc000 (fec00000)
mapped IOAPIC to ffffb000 (fcffe000)
mapped IOAPIC to ffffa000 (fcfff000)
mapped IOAPIC to ffff9000 (fbffe000)
mapped IOAPIC to ffff8000 (fbfff000)
mapped IOAPIC to ffff7000 (facfc000)
mapped IOAPIC to ffff6000 (facfd000)
mapped IOAPIC to ffff5000 (facfe000)
mapped IOAPIC to ffff4000 (facff000)
Initializing CPU#0
PID hash table entries: 4096 (order: 12, 65536 bytes)
Detected 2004.700 MHz processor.
Using tsc for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 16614040k/17825792k available (2161k kernel code, 161112k
reserved, 598k data, 224k init, 15859648k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay using timer specific routine.. 4017.36 BogoMIPS (lpj=8034729)
Security Framework v1.0.0 initialized
Mount-cache hash table entries: 512
CPU: After generic identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: After vendor identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU: After all inits, caps: 078bfbff e1d3fbff 00000000 00000010
00000000 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
mtrr: v2.0 (20020519)
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
CPU0: AMD Opteron(tm) Processor 846 stepping 0a
weird, boot CPU (#0) not listed by the BIOS.
Total of 1 processors activated (4017.36 BogoMIPS).
ENABLING IO-APIC IRQs
..TIMER: vector=0x31 apic1=0 pin1=2 apic2=0 pin2=0
Brought up 1 CPUs

Thanks,

Bharath

On 12/1/05, Keith Mannthey <[email protected]> wrote:
> Looks like something odd is going on with your processor apic id's.
>
> Lines like
> Processor #20 INVALID. (Max ID: 256)
> are bad...
>
> Could you try the following patch with i386? It seems suspect to me
> and could help identify the problem.
>
> Your box has high processor apic ids and looks to be in flat apic mode
> (versus clustered). I think the code is assuming the box is in
> clustered apic mode?
>
> Thanks,
> Keith
>
>
>

2005-12-02 01:12:13

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Could it be possible that AMD64 NUMA support is needed for Opteron
SMPs to work and Intel's APIC-based SMP wouldn't work on some boards.

lspci output:
00:01.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
00:01.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
00:02.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
00:02.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
00:06.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8111 PCI (rev 07)
00:07.0 ISA bridge: Advanced Micro Devices [AMD] AMD-8111 LPC (rev 05)
00:07.1 IDE interface: Advanced Micro Devices [AMD] AMD-8111 IDE (rev 03)
00:07.2 SMBus: Advanced Micro Devices [AMD] AMD-8111 SMBus 2.0 (rev 02)
00:07.3 Bridge: Advanced Micro Devices [AMD] AMD-8111 ACPI (rev 05)
00:18.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:19.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:19.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:19.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:19.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1a.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1a.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:1a.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1a.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1b.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1b.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:1b.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1b.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1c.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1c.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Address
Map
00:1c.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1c.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1d.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1d.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:1d.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1d.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1e.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1e.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:1e.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1e.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
00:1f.0 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] HyperTransport Technology Configuration
00:1f.1 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Address Map
00:1f.2 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] DRAM Controller
00:1f.3 Host bridge: Advanced Micro Devices [AMD] K8
[Athlon64/Opteron] Miscellaneous Control
01:00.0 USB Controller: Advanced Micro Devices [AMD] AMD-8111 USB (rev 0b)
01:00.1 USB Controller: Advanced Micro Devices [AMD] AMD-8111 USB (rev 0b)
01:04.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27)
01:05.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 10)
04:01.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
04:01.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
04:02.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
04:02.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
07:01.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
07:01.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
07:02.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
07:02.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
07:03.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
07:03.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
07:04.0 PCI bridge: Advanced Micro Devices [AMD] AMD-8131 PCI-X Bridge (rev 12)
07:04.1 PIC: Advanced Micro Devices [AMD] AMD-8131 PCI-X APIC (rev 01)
08:02.0 Ethernet controller: Intel Corp. 82546EB Gigabit Ethernet
Controller (Copper) (rev 01)
08:02.1 Ethernet controller: Intel Corp. 82546EB Gigabit Ethernet
Controller (Copper) (rev 01)
09:01.0 PCI bridge: Mellanox Technologies MT23108 PCI Bridge (rev a1)
09:02.0 Ethernet controller: Intel Corp. 82546EB Gigabit Ethernet
Controller (Copper) (rev 01)
09:02.1 Ethernet controller: Intel Corp. 82546EB Gigabit Ethernet Controller (Co
pper) (rev 01)
0a:00.0 InfiniBand: Mellanox Technologies MT23108 InfiniHost (rev a1)


Thanks,

Bharath

On 12/1/05, Bharath Ramesh <[email protected]> wrote:
> Patched 2.6.15-rc4 with the patch. It still is the same.
>
> dmesg:
> Linux version 2.6.15-rc4 (root@drones) (gcc version 3.4.3 20050227
> (Red Hat 3.4.3-22.1)) #1 SMP PREEMPT Thu Dec 1 19:37:03 EST 2005
> BIOS-provided physical RAM map:
> BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
> BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
> BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
> BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
> BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
> BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
> BIOS-e820: 00000000ff780000 - 0000000100000000 (reserved)
> BIOS-e820: 0000000100000000 - 0000000440000000 (usable)
> 16512MB HIGHMEM available.
> 896MB LOWMEM available.
> found SMP MP-table at 000ff780
> NX (Execute Disable) protection: active
> On node 0 totalpages: 4456448
> DMA zone: 4096 pages, LIFO batch:2
> DMA32 zone: 0 pages, LIFO batch:2
> Normal zone: 225280 pages, LIFO batch:64
> HighMem zone: 4227072 pages, LIFO batch:64
> DMI 2.3 present.
> ACPI: RSDP (v002 ACPIAM ) @ 0x000f6ca0
> ACPI: XSDT (v001 A M I OEMXSDT 0x02000502 MSFT 0x00000097) @ 0xbfff0100
> ACPI: FADT (v003 A M I OEMFACP 0x02000502 MSFT 0x00000097) @ 0xbfff0281
> ACPI: MADT (v001 A M I OEMAPIC 0x02000502 MSFT 0x00000097) @ 0xbfff0380
> ACPI: OEMB (v001 A M I OEMBIOS 0x02000502 MSFT 0x00000097) @ 0xbffff040
> ACPI: SRAT (v001 A M I OEMSRAT 0x02000502 MSFT 0x00000097) @ 0xbfff3cc0
> ACPI: DSDT (v001 H805V H805V120 0x00000000 INTL 0x02002026) @ 0x00000000
> ACPI: Local APIC address 0xfee00000
> ACPI: LAPIC (acpi_id[0x01] lapic_id[0x10] enabled)
> Processor #16 15:5 APIC version 16
> Processor #16 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x02] lapic_id[0x11] enabled)
> Processor #17 15:5 APIC version 16
> Processor #17 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x03] lapic_id[0x12] enabled)
> Processor #18 15:5 APIC version 16
> Processor #18 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x04] lapic_id[0x13] enabled)
> Processor #19 15:5 APIC version 16
> Processor #19 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x05] lapic_id[0x14] enabled)
> Processor #20 15:5 APIC version 16
> Processor #20 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x06] lapic_id[0x15] enabled)
> Processor #21 15:5 APIC version 16
> Processor #21 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x07] lapic_id[0x16] enabled)
> Processor #22 15:5 APIC version 16
> Processor #22 INVALID. (Max ID: 256).
> ACPI: LAPIC (acpi_id[0x08] lapic_id[0x17] enabled)
> Processor #23 15:5 APIC version 16
> Processor #23 INVALID. (Max ID: 256).
> ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
> IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
> ACPI: IOAPIC (id[0x02] address[0xfcffe000] gsi_base[24])
> IOAPIC[1]: apic_id 2, version 17, address 0xfcffe000, GSI 24-27
> ACPI: IOAPIC (id[0x03] address[0xfcfff000] gsi_base[28])
> IOAPIC[2]: apic_id 3, version 17, address 0xfcfff000, GSI 28-31
> ACPI: IOAPIC (id[0x10] address[0xfbffe000] gsi_base[32])
> IOAPIC[3]: apic_id 16, version 17, address 0xfbffe000, GSI 32-35
> ACPI: IOAPIC (id[0x11] address[0xfbfff000] gsi_base[36])
> IOAPIC[4]: apic_id 17, version 17, address 0xfbfff000, GSI 36-39
> ACPI: IOAPIC (id[0x12] address[0xfacfc000] gsi_base[40])
> IOAPIC[5]: apic_id 18, version 17, address 0xfacfc000, GSI 40-43
> ACPI: IOAPIC (id[0x13] address[0xfacfd000] gsi_base[44])
> IOAPIC[6]: apic_id 19, version 17, address 0xfacfd000, GSI 44-47
> ACPI: IOAPIC (id[0x14] address[0xfacfe000] gsi_base[48])
> IOAPIC[7]: apic_id 20, version 17, address 0xfacfe000, GSI 48-51
> ACPI: IOAPIC (id[0x15] address[0xfacff000] gsi_base[52])
> IOAPIC[8]: apic_id 21, version 17, address 0xfacff000, GSI 52-55
> ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
> ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
> ACPI: IRQ0 used by override.
> ACPI: IRQ2 used by override.
> ACPI: IRQ9 used by override.
> Enabling APIC mode: Flat. Using 9 I/O APICs
> Using ACPI (MADT) for SMP configuration information
> Allocating PCI resources starting at c4000000 (gap: c0000000:3f780000)
> Built 1 zonelists
> Kernel command line: ro root=/dev/hda3 vga=791 rhgb quiet
> mapped APIC to ffffd000 (fee00000)
> mapped IOAPIC to ffffc000 (fec00000)
> mapped IOAPIC to ffffb000 (fcffe000)
> mapped IOAPIC to ffffa000 (fcfff000)
> mapped IOAPIC to ffff9000 (fbffe000)
> mapped IOAPIC to ffff8000 (fbfff000)
> mapped IOAPIC to ffff7000 (facfc000)
> mapped IOAPIC to ffff6000 (facfd000)
> mapped IOAPIC to ffff5000 (facfe000)
> mapped IOAPIC to ffff4000 (facff000)
> Initializing CPU#0
> PID hash table entries: 4096 (order: 12, 65536 bytes)
> Detected 2004.700 MHz processor.
> Using tsc for high-res timesource
> Console: colour dummy device 80x25
> Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
> Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
> Memory: 16614040k/17825792k available (2161k kernel code, 161112k
> reserved, 598k data, 224k init, 15859648k highmem)
> Checking if this processor honours the WP bit even in supervisor mode... Ok.
> Calibrating delay using timer specific routine.. 4017.36 BogoMIPS (lpj=8034729)
> Security Framework v1.0.0 initialized
> Mount-cache hash table entries: 512
> CPU: After generic identify, caps: 078bfbff e1d3fbff 00000000 00000000
> 00000000 00000000 00000000
> CPU: After vendor identify, caps: 078bfbff e1d3fbff 00000000 00000000
> 00000000 00000000 00000000
> CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
> CPU: L2 Cache: 1024K (64 bytes/line)
> CPU: After all inits, caps: 078bfbff e1d3fbff 00000000 00000010
> 00000000 00000000 00000000
> Intel machine check architecture supported.
> Intel machine check reporting enabled on CPU#0.
> mtrr: v2.0 (20020519)
> Enabling fast FPU save and restore... done.
> Enabling unmasked SIMD FPU exception support... done.
> Checking 'hlt' instruction... OK.
> CPU0: AMD Opteron(tm) Processor 846 stepping 0a
> weird, boot CPU (#0) not listed by the BIOS.
> Total of 1 processors activated (4017.36 BogoMIPS).
> ENABLING IO-APIC IRQs
> ..TIMER: vector=0x31 apic1=0 pin1=2 apic2=0 pin2=0
> Brought up 1 CPUs
>
> Thanks,
>
> Bharath
>
> On 12/1/05, Keith Mannthey <[email protected]> wrote:
> > Looks like something odd is going on with your processor apic id's.
> >
> > Lines like
> > Processor #20 INVALID. (Max ID: 256)
> > are bad...
> >
> > Could you try the following patch with i386? It seems suspect to me
> > and could help identify the problem.
> >
> > Your box has high processor apic ids and looks to be in flat apic mode
> > (versus clustered). I think the code is assuming the box is in
> > clustered apic mode?
> >
> > Thanks,
> > Keith
> >
> >
> >
>

2005-12-02 01:42:46

by Keith Mannthey

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Did the smp kernel that came with your distro boot smp on i386?

An apic is an apic. I don't think there is a diffrent interrupt
controler (but maybe I am wrong) I can boot opteron SMP with i386
just fine. My guess is what you are seeing is specific to your box.
What sort of a box is this 8-way AMD Opteron.....

What i386 subarch are you selecting in your i386 build? Could you try
the Generic architecture? Maybe the box is in clustered apic mode.

Does booting with acpi=off help?

Can you send the dmesg from your x86_64 partition. It will show us
what the apic is doing in that arch.

Thanks,
Keith

2005-12-02 02:32:12

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

On 12/1/05, Keith Mannthey <[email protected]> wrote:
> Did the smp kernel that came with your distro boot smp on i386?

The distro's i386 SMP kernel detects only one processor.
>
> An apic is an apic. I don't think there is a diffrent interrupt
> controler (but maybe I am wrong) I can boot opteron SMP with i386
> just fine. My guess is what you are seeing is specific to your box.
> What sort of a box is this 8-way AMD Opteron.....

Its a custom built box, I dont have the mother board manual with me at
this instant. It uses a IWill 8-way motherboard.

>
> What i386 subarch are you selecting in your i386 build? Could you try
> the Generic architecture? Maybe the box is in clustered apic mode.
>
> Does booting with acpi=off help?
>

I dont have the permission to remotely login as root at this moment. I
will have access to the console in the morning. I can boot the machine
with acpi=off and try it.

> Can you send the dmesg from your x86_64 partition. It will show us
> what the apic is doing in that arch.

dmesg for x86_64 (runs 2.6.11.10):
Bootdata ok (command line is ro root=/dev/hda2 vga=791 rhgb quiet)
Linux version 2.6.11.10 (root@drones) (gcc version 3.4.3 20041212 (Red
Hat 3.4.3-9.EL4)) #4 SMP Thu May 19 13:28:55 EDT 2005
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
BIOS-e820: 00000000ff780000 - 0000000100000000 (reserved)
BIOS-e820: 0000000100000000 - 0000000440000000 (usable)
ACPI: RSDP (v002 ACPIAM ) @ 0x00000000000f6ca0
ACPI: XSDT (v001 A M I OEMXSDT 0x02000502 MSFT 0x00000097) @
0x00000000bfff0100
ACPI: FADT (v003 A M I OEMFACP 0x02000502 MSFT 0x00000097) @
0x00000000bfff0281
ACPI: MADT (v001 A M I OEMAPIC 0x02000502 MSFT 0x00000097) @
0x00000000bfff0380
ACPI: OEMB (v001 A M I OEMBIOS 0x02000502 MSFT 0x00000097) @
0x00000000bffff040
ACPI: SRAT (v001 A M I OEMSRAT 0x02000502 MSFT 0x00000097) @
0x00000000bfff3cc0
ACPI: DSDT (v001 H805V H805V120 0x00000000 INTL 0x02002026) @
0x0000000000000000
SRAT: PXM 0 -> APIC 0 -> Node 0
SRAT: PXM 1 -> APIC 1 -> Node 1
SRAT: PXM 2 -> APIC 2 -> Node 2
SRAT: PXM 3 -> APIC 3 -> Node 3
SRAT: PXM 4 -> APIC 4 -> Node 4
SRAT: PXM 5 -> APIC 5 -> Node 5
SRAT: PXM 6 -> APIC 6 -> Node 6
SRAT: PXM 7 -> APIC 7 -> Node 7
SRAT: Node 0 PXM 0 100000-7fffffff
SRAT: Node 1 PXM 1 80000000-bfffffff
SRAT: Node 2 PXM 2 140000000-1bfffffff
SRAT: Node 3 PXM 3 1c0000000-23fffffff
SRAT: Node 4 PXM 4 240000000-2bfffffff
SRAT: Node 5 PXM 5 2c0000000-33fffffff
SRAT: Node 6 PXM 6 340000000-3bfffffff
SRAT: Node 7 PXM 7 3c0000000-43fffffff
SRAT: Node 0 PXM 0 0-7fffffff
node 2 shift 24 addr 140000000 conflict 0
node 3 shift 25 addr 1fe000000 conflict 0
node 7 shift 26 addr 3fc000000 conflict 0
Bootmem setup node 0 0000000000000000-000000007fffffff
Bootmem setup node 1 0000000080000000-00000000bfffffff
Bootmem setup node 2 0000000140000000-00000001bfffffff
Bootmem setup node 3 00000001c0000000-000000023fffffff
Bootmem setup node 4 0000000240000000-00000002bfffffff
Bootmem setup node 5 00000002c0000000-000000033fffffff
Bootmem setup node 6 0000000340000000-00000003bfffffff
Bootmem setup node 7 00000003c0000000-000000043fffffff
On node 0 totalpages: 524287
DMA zone: 4096 pages, LIFO batch:1
Normal zone: 520191 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 1 totalpages: 262143
DMA zone: 0 pages, LIFO batch:1
Normal zone: 262143 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 2 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 3 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 4 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 5 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 6 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
On node 7 totalpages: 524287
DMA zone: 0 pages, LIFO batch:1
Normal zone: 524287 pages, LIFO batch:16
HighMem zone: 0 pages, LIFO batch:1
ACPI: Local APIC address 0xfee00000
ACPI: LAPIC (acpi_id[0x01] lapic_id[0x10] enabled)
Processor #16 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x02] lapic_id[0x11] enabled)
Processor #17 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x03] lapic_id[0x12] enabled)
Processor #18 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x04] lapic_id[0x13] enabled)
Processor #19 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x05] lapic_id[0x14] enabled)
Processor #20 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x06] lapic_id[0x15] enabled)
Processor #21 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x07] lapic_id[0x16] enabled)
Processor #22 15:5 APIC version 16
ACPI: LAPIC (acpi_id[0x08] lapic_id[0x17] enabled)
Processor #23 15:5 APIC version 16
ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
ACPI: IOAPIC (id[0x02] address[0xfcffe000] gsi_base[24])
IOAPIC[1]: apic_id 2, version 17, address 0xfcffe000, GSI 24-27
ACPI: IOAPIC (id[0x03] address[0xfcfff000] gsi_base[28])
IOAPIC[2]: apic_id 3, version 17, address 0xfcfff000, GSI 28-31
ACPI: IOAPIC (id[0x10] address[0xfbffe000] gsi_base[32])
IOAPIC[3]: apic_id 0, version 17, address 0xfbffe000, GSI 32-35
ACPI: IOAPIC (id[0x11] address[0xfbfff000] gsi_base[36])
IOAPIC[4]: apic_id 4, version 17, address 0xfbfff000, GSI 36-39
ACPI: IOAPIC (id[0x12] address[0xfacfc000] gsi_base[40])
IOAPIC[5]: apic_id 5, version 17, address 0xfacfc000, GSI 40-43
ACPI: IOAPIC (id[0x13] address[0xfacfd000] gsi_base[44])
IOAPIC[6]: apic_id 6, version 17, address 0xfacfd000, GSI 44-47
ACPI: IOAPIC (id[0x14] address[0xfacfe000] gsi_base[48])
IOAPIC[7]: apic_id 7, version 17, address 0xfacfe000, GSI 48-51
ACPI: IOAPIC (id[0x15] address[0xfacff000] gsi_base[52])
IOAPIC[8]: apic_id 8, version 17, address 0xfacff000, GSI 52-55
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
ACPI: IRQ0 used by override.
ACPI: IRQ2 used by override.
ACPI: IRQ9 used by override.
Setting APIC routing to flat
Using ACPI (MADT) for SMP configuration information
Checking aperture...
CPU 0: aperture @ 4000000 size 32 MB
Aperture from northbridge cpu 0 too small (32 MB)
No AGP bridge found
Your BIOS doesn't leave a aperture memory hole
Please enable the IOMMU option in the BIOS setup
This costs you 64 MB of RAM
Mapping aperture over 65536 KB of RAM @ 4000000
Built 8 zonelists
Kernel command line: ro root=/dev/hda2 vga=791 rhgb quiet console=tty0
Initializing CPU#0
PID hash table entries: 4096 (order: 12, 131072 bytes)
time.c: Using 1.193182 MHz PIT timer.
time.c: Detected 2004.577 MHz processor.
Console: colour dummy device 80x25
Dentry cache hash table entries: 2097152 (order: 12, 16777216 bytes)
Inode-cache hash table entries: 1048576 (order: 11, 8388608 bytes)
Memory: 15415088k/17825792k available (2083k kernel code, 0k reserved,
724k data, 208k init)
Calibrating delay loop... 3940.35 BogoMIPS (lpj=1970176)
Security Framework v1.0.0 initialized
Capability LSM initialized
Mount-cache hash table entries: 256 (order: 0, 4096 bytes)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 0(1) -> Node 0
Using local APIC NMI watchdog using perfctr0
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 0(1) -> Node 0
CPU0: AMD Opteron(tm) Processor 846 stepping 0a
per-CPU timeslice cutoff: 1023.98 usecs.
task migration cache decay timeout: 2 msecs.
Booting processor 1/17 rip 6000 rsp ffff8101c1c31f58
Initializing CPU#1
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 1(1) -> Node 1
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 2/18 rip 6000 rsp ffff8103bff81f58
Initializing CPU#2
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 2(1) -> Node 2
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 3/19 rip 6000 rsp ffff8103c3465f58
Initializing CPU#3
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 3(1) -> Node 3
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 4/20 rip 6000 rsp ffff8100bffb3f58
Initializing CPU#4
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 4(1) -> Node 4
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 5/21 rip 6000 rsp ffff81023ffa5f58
Initializing CPU#5
Calibrating delay loop... 3997.69 BogoMIPS (lpj=1998848)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 5(1) -> Node 5
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 6/22 rip 6000 rsp ffff8102bffa9f58
Initializing CPU#6
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 6(1) -> Node 6
AMD Opteron(tm) Processor 846 stepping 0a
Booting processor 7/23 rip 6000 rsp ffff81033ffb3f58
Initializing CPU#7
Calibrating delay loop... 4005.88 BogoMIPS (lpj=2002944)
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU 7(1) -> Node 7
AMD Opteron(tm) Processor 846 stepping 0a
Total of 8 processors activated (31973.37 BogoMIPS).
Using local APIC timer interrupts.
Detected 12.528 MHz APIC timer.
checking TSC synchronization across 8 CPUs: passed.
time.c: Using PIT/TSC based timekeeping.
Brought up 8 CPUs
CPU0 attaching sched-domain:
domain 0: span 01
groups: 01
domain 1: span ff
groups: 01 02 04 08 10 20 40 80
CPU1 attaching sched-domain:
domain 0: span 02
groups: 02
domain 1: span ff
groups: 02 04 08 10 20 40 80 01
CPU2 attaching sched-domain:
domain 0: span 04
groups: 04
domain 1: span ff
groups: 04 08 10 20 40 80 01 02
CPU3 attaching sched-domain:
domain 0: span 08
groups: 08
domain 1: span ff
groups: 08 10 20 40 80 01 02 04
CPU4 attaching sched-domain:
domain 0: span 10
groups: 10
domain 1: span ff
groups: 10 20 40 80 01 02 04 08
CPU5 attaching sched-domain:
domain 0: span 20
groups: 20
domain 1: span ff
groups: 20 40 80 01 02 04 08 10
CPU6 attaching sched-domain:
domain 0: span 40
groups: 40
domain 1: span ff
groups: 40 80 01 02 04 08 10 20
CPU7 attaching sched-domain:
domain 0: span 80
groups: 80
domain 1: span ff
groups: 80 01 02 04 08 10 20 40
checking if image is initramfs... it is


Thanks,

Bharath

2005-12-02 15:32:16

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

>
> Does booting with acpi=off help?

Booting with acpi=off doesnt help

dmesg with acpi=off:
Linux version 2.6.15-rc4 (root@drones) (gcc version 3.4.3 20050227
(Red Hat 3.4.3-22.1)) #1 SMP PREEMPT Thu Dec 1 19:37:03 EST 2005
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000bfff0000 (usable)
BIOS-e820: 00000000bfff0000 - 00000000bffff000 (ACPI data)
BIOS-e820: 00000000bffff000 - 00000000c0000000 (ACPI NVS)
BIOS-e820: 00000000ff780000 - 0000000100000000 (reserved)
BIOS-e820: 0000000100000000 - 0000000440000000 (usable)
16512MB HIGHMEM available.
896MB LOWMEM available.
found SMP MP-table at 000ff780
NX (Execute Disable) protection: active
On node 0 totalpages: 4456448
DMA zone: 4096 pages, LIFO batch:2
DMA32 zone: 0 pages, LIFO batch:2
Normal zone: 225280 pages, LIFO batch:64
HighMem zone: 4227072 pages, LIFO batch:64
DMI 2.3 present.
Intel MultiProcessor Specification v1.4
Virtual Wire compatibility mode.
OEM ID: IWILL Product ID: H805 APIC at: 0xFEE00000
Processor #16 15:5 APIC version 16
Processor #16 INVALID. (Max ID: 256).
Processor #17 15:5 APIC version 16
Processor #17 INVALID. (Max ID: 256).
Processor #18 15:5 APIC version 16
Processor #18 INVALID. (Max ID: 256).
Processor #19 15:5 APIC version 16
Processor #19 INVALID. (Max ID: 256).
Processor #20 15:5 APIC version 16
Processor #20 INVALID. (Max ID: 256).
Processor #21 15:5 APIC version 16
Processor #21 INVALID. (Max ID: 256).
Processor #22 15:5 APIC version 16
Processor #22 INVALID. (Max ID: 256).
Processor #23 15:5 APIC version 16
Processor #23 INVALID. (Max ID: 256).
I/O APIC #0 Version 17 at 0xFEC00000.
I/O APIC #1 Version 17 at 0xFCFFE000.
I/O APIC #2 Version 17 at 0xFCFFF000.
I/O APIC #3 Version 17 at 0xFBFFE000.
I/O APIC #4 Version 17 at 0xFBFFF000.
I/O APIC #5 Version 17 at 0xFACFC000.
I/O APIC #6 Version 17 at 0xFACFD000.
I/O APIC #7 Version 17 at 0xFACFE000.
I/O APIC #8 Version 17 at 0xFACFF000.
Enabling APIC mode: Flat. Using 9 I/O APICs
SMP mptable: no processors registered!
BIOS bug, MP table errors detected!...
... disabling SMP support. (tell your hw vendor)
Allocating PCI resources starting at c4000000 (gap: c0000000:3f780000)
Built 1 zonelists
Kernel command line: ro root=/dev/hda3 vga=791 acpi=off rhgb quiet
Found and enabled local APIC!
mapped APIC to ffffd000 (fee00000)
mapped IOAPIC to ffffc000 (0a955000)
mapped IOAPIC to ffffb000 (0a956000)
mapped IOAPIC to ffffa000 (0a957000)
mapped IOAPIC to ffff9000 (0a958000)
mapped IOAPIC to ffff8000 (0a959000)
mapped IOAPIC to ffff7000 (0a95a000)
mapped IOAPIC to ffff6000 (0a95b000)
mapped IOAPIC to ffff5000 (0a95c000)
mapped IOAPIC to ffff4000 (0a95d000)
Initializing CPU#0
PID hash table entries: 4096 (order: 12, 65536 bytes)
Detected 2004.813 MHz processor.
Using tsc for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 131072 (order: 7, 524288 bytes)
Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
Memory: 16614040k/17825792k available (2161k kernel code, 161148k
reserved, 598k data, 224k init, 15859648k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay using timer specific routine.. 4017.40 BogoMIPS (lpj=8034817)
Security Framework v1.0.0 initialized
Mount-cache hash table entries: 512
CPU: After generic identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: After vendor identify, caps: 078bfbff e1d3fbff 00000000 00000000
00000000 00000000 00000000
CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
CPU: L2 Cache: 1024K (64 bytes/line)
CPU: After all inits, caps: 078bfbff e1d3fbff 00000000 00000010
00000000 00000000 00000000
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
mtrr: v2.0 (20020519)
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Checking 'hlt' instruction... OK.
CPU0: AMD Opteron(tm) Processor 846 stepping 0a
SMP motherboard not detected.
Brought up 1 CPUs

Thanks,

Bharath

2005-12-02 22:07:50

by Keith Mannthey

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Welcome to hardware bring up. Ok I looked a little closer at the
story. In x86_64 the only check for valid apic is apicid < MAX_APICS
which make sense to me.

I386 has this concept of version. Where the valid check is diffrent

if (version >= 0x14)
return apicid < 0xff;
else
return apicid < 0xf;

Now the patch is sent before was a litle in haste. Your apic version
is 16. The valid check returns false because apicid is greater that
0xf (0x15). Most likey the bios entry for the apic version is wrong
but I don't know much about that or why it matters.

So what does this all mean. Legacy.....

Give this patch a whirl.

Thanks,
Keith


Attachments:
(No filename) (686.00 B)
apicid.fix (348.00 B)
Download all attachments

2005-12-03 02:06:12

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

I have attached the kernel config. I will enable debugging and rebuild
the kernel soon and send in the latest dmesg soon.

Thanks,

Bharath
On 12/2/05, Keith Mannthey <[email protected]> wrote:
> Hmm keep that patch for now. Can you attach your .config and turn on
> debugging?
>
> weird, boot CPU (#0) not listed by the BIOS.
>
> This messages implies perhaps your bios tables are setup in a some
> other way then i386 expects. Acpi is setting apicids to them and they
> are really there but the i386 boot up acpid mess is not setup
> correctly and it looks to be skipping booting the other processors for
> some reason.
>
> This code is all a bit of a mess.
>
> thanks,
> Keith
>


Attachments:
(No filename) (686.00 B)
config (29.67 kB)
Download all attachments

2005-12-03 17:52:01

by Andi Kleen

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Keith Mannthey <[email protected]> writes:

> Welcome to hardware bring up. Ok I looked a little closer at the
> story. In x86_64 the only check for valid apic is apicid < MAX_APICS
> which make sense to me.

It will still not work. He will need a variant of the physflat-i386
patch I posted some time ago. However it needs some cleanup
to be actually mergeable

My recommendation is a 64bit kernel.

-Andi

2005-12-04 14:24:11

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Can you point me to the thread in which you posted the patch so that I
can try it out. I am limited to a 32-bit kernel as the library I want
use requires a 32bit kernel. I need to try to get some performance
results.

Thanks,
Bharath

On 03 Dec 2005 15:21:14 -0700, Andi Kleen <[email protected]> wrote:
> Keith Mannthey <[email protected]> writes:
>
> > Welcome to hardware bring up. Ok I looked a little closer at the
> > story. In x86_64 the only check for valid apic is apicid < MAX_APICS
> > which make sense to me.
>
> It will still not work. He will need a variant of the physflat-i386
> patch I posted some time ago. However it needs some cleanup
> to be actually mergeable
>
> My recommendation is a 64bit kernel.
>
> -Andi
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>

2005-12-04 17:28:05

by Matti Aarnio

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

On Sun, Dec 04, 2005 at 09:24:08AM -0500, Bharath Ramesh wrote:
>
> Can you point me to the thread in which you posted the patch so that I
> can try it out. I am limited to a 32-bit kernel as the library I want
> use requires a 32bit kernel. I need to try to get some performance
> results.

With "libary" I presume you to mean something used by an application
program ?

If so, a 64-bit kernel on x86_64 will work just fine for your 32 bit
code. What you do need, however, are some compiler magic to construct
your binary -- in essence doing a cross-compilation on 64 bit platform
to 32 bit target.

The resulting application binary will run just fine.

You will need things like glibc installed both for 32 bit i386
and 64 bit x86_64, and possibly some others, but those should be
mostly coming in current distributions, anyway.

I use daily 32-bit Mozilla with several 3rd party plugins in my
64-bit machine. My kernel is fully 64-bit.


> Thanks,
> Bharath

/Matti Aarnio


> On 03 Dec 2005 15:21:14 -0700, Andi Kleen <[email protected]> wrote:
> > Keith Mannthey <[email protected]> writes:
> >
> > > Welcome to hardware bring up. Ok I looked a little closer at the
> > > story. In x86_64 the only check for valid apic is apicid < MAX_APICS
> > > which make sense to me.
> >
> > It will still not work. He will need a variant of the physflat-i386
> > patch I posted some time ago. However it needs some cleanup
> > to be actually mergeable
> >
> > My recommendation is a 64bit kernel.
> >
> > -Andi
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to [email protected]
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at http://www.tux.org/lkml/
> >
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/

2005-12-13 19:49:30

by Bill Davidsen

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

Bharath Ramesh wrote:
> I have attached the kernel config. I will enable debugging and rebuild
> the kernel soon and send in the latest dmesg soon.
>
> Thanks,
>
> Bharath

I'm not at all an expert on x86_64, but I notice the NUMA is not set in
this config, and I have a similar config from another post with DOES
work, and it has NUMA on (four way only, dual dual-core).

I have no idea if that can be useful, but if you're still out of ideas
you might try it.

--
-bill davidsen ([email protected])
"The secret to procrastination is to put things off until the
last possible moment - but no longer" -me

2005-12-14 02:02:47

by Bharath Ramesh

[permalink] [raw]
Subject: Re: Only one processor detected in 8-Way opteron in 32-bit mode

On 12/7/05, Bill Davidsen <[email protected]> wrote:
> Bharath Ramesh wrote:
> > I have attached the kernel config. I will enable debugging and rebuild
> > the kernel soon and send in the latest dmesg soon.
> >
> > Thanks,
> >
> > Bharath
>
> I'm not at all an expert on x86_64, but I notice the NUMA is not set in
> this config, and I have a similar config from another post with DOES
> work, and it has NUMA on (four way only, dual dual-core).

I am not sure about it but the NUMA option comes up only when the ARCH
type is set to x86_64. For some reason when I set the ARCH type to x86
then the NUMA option is never present. NUMA is only associated with
x86_64 but I am not sure how it would work if I am trying to run
x86_64 processor in x86 mode.

>
> I have no idea if that can be useful, but if you're still out of ideas
> you might try it.
>
> --
> -bill davidsen ([email protected])
> "The secret to procrastination is to put things off until the
> last possible moment - but no longer" -me
>
>