bugreport kernel panic on early stage, with HIGHMEM4G:

public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed

* bugreport kernel panic on early stage, with HIGHMEM4G:
@ 2008-01-07  5:47 Denys Fedoryshchenko
  2008-01-15 11:39 ` Ingo Molnar
  0 siblings, 1 reply; 5+ messages in thread
From: Denys Fedoryshchenko @ 2008-01-07  5:47 UTC (permalink / raw)
  To: linux-kernel

Hi

After physical memory upgrade from 3GB to 4GB (also it happens on 5GB) got 
kernel panic.

Because it is happening on early stage and my machine doesn't contain serial 
port, i had to take photo.
Kernel boots fine with 64GB highmem, no highmem, or highmem4G with limited 
memory by mem=3G. All dmesg attached.
Also i attach dmidecode and lspci -vvv output, probably it will be useful.


Photo (2.8MB, sorry, just original size from camera):
http://www.nuclearcat.com/files/panic-07012008/img_1232.jpg

dmesg without highmem
http://www.nuclearcat.com/files/panic-07012008/dmesg-nohighmem.txt

with highmem64G
http://www.nuclearcat.com/files/panic-07012008/dmesg-highmem64G.txt

with highmem4G limited by mem=3G
http://www.nuclearcat.com/files/panic-07012008/dmesg-highmem4G-memlim3G.txt
Kernel config for this specific boot:
http://www.nuclearcat.com/files/panic-07012008/config.txt

dmidecode output
http://www.nuclearcat.com/files/panic-07012008/dmidecode.txt

lspci output
http://www.nuclearcat.com/files/panic-07012008/lspci.txt

--
Denys Fedoryshchenko
Technical Manager
Virtual ISP S.A.L.


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: bugreport kernel panic on early stage, with HIGHMEM4G:
  2008-01-07  5:47 bugreport kernel panic on early stage, with HIGHMEM4G: Denys Fedoryshchenko
@ 2008-01-15 11:39 ` Ingo Molnar
  2008-01-15 13:13   ` Ingo Molnar
  2008-01-17 13:31   ` Denys Fedoryshchenko
  0 siblings, 2 replies; 5+ messages in thread
From: Ingo Molnar @ 2008-01-15 11:39 UTC (permalink / raw)
  To: Denys Fedoryshchenko
  Cc: linux-kernel, Andrew Morton, Thomas Gleixner, H. Peter Anvin,
	Mel Gorman


* Denys Fedoryshchenko <denys@visp.net.lb> wrote:

> Hi
> 
> After physical memory upgrade from 3GB to 4GB (also it happens on 5GB) 
> got kernel panic.
> 
> Because it is happening on early stage and my machine doesn't contain 
> serial port, i had to take photo. Kernel boots fine with 64GB highmem, 
> no highmem, or highmem4G with limited memory by mem=3G. All dmesg 
> attached. Also i attach dmidecode and lspci -vvv output, probably it 
> will be useful.

thanks for the detailed report, i think i know what's going on. Could 
you try the patch below, does it fix your problem?

this seems to be a SPARSEMEM bug which is present in v2.6.23 as well and 
has probably been present ever since SPARSEMEM was added to 32-bit x86.

There's a ~256MB hole in your e820 memory map (the pci aperture), which 
causes the last 4 sparsemem sections (each covering 64MB of RAM) to be 
not present - and they are thus missing from the sparsemem mem_map[] 
too. The highmem init code on the other hand assumes that all pages are 
in the mem_map[]:

 static void __init set_highmem_pages_init(int bad_ppro)
 {
        int pfn;
        for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
                add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);

the pfn_to_page() is unconditional and dereferences to a NULL-ish 
pointer which crashes your box. highend_pfn is what got miscalculated by 
256 MB, so set_highmem_pages_init() tried to reference a non-existing 
struct page - but it should still be robust enough against non-existent 
pages.

The patch below fixes this bug. Please also send a dmesg if you manage 
to boot the box up fine, i've added a few debug printouts to confirm 
this theory. (i'll figure out whether we need to clip highend_pfn as 
well - but this patch alone should be good enough to fix the crash on 
your box.)

	Ingo

------------->
Subject: x86: fix CONFIG_SPARSEMEM highmem init bug
From: Ingo Molnar <mingo@elte.hu>

fix CONFIG_SPARSEMEM highmem init bug.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/init_32.c |   43 ++++++++++++++++++++++++++++++++++++++++---
 mm/sparse.c           |    8 +++++++-
 2 files changed, 47 insertions(+), 4 deletions(-)

Index: linux/arch/x86/mm/init_32.c
===================================================================
--- linux.orig/arch/x86/mm/init_32.c
+++ linux/arch/x86/mm/init_32.c
@@ -321,11 +321,48 @@ extern void set_highmem_pages_init(int);
 static void __init set_highmem_pages_init(int bad_ppro)
 {
 	int pfn;
-	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
-		add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
+
+	printk("set_highmem_pages_init(bad_ppro:%d)\n", bad_ppro);
+	printk("sizeof(struct page):        %d\n", sizeof(struct page));
+	printk("sizeof(struct mem_section): %d\n", sizeof(struct mem_section));
+	printk("PFN_SECTION_SHIFT:          %d\n", PFN_SECTION_SHIFT);
+
+	printk("mem_map: %p\n", mem_map);
+	printk("  highstart_pfn: %9ld [page: %p]\n",
+		highstart_pfn, pfn_to_page(highstart_pfn));
+	printk("    highend_pfn: %9ld [page: %p]\n",
+		highend_pfn, pfn_to_page(highend_pfn));
+	printk("  highend_pfn-1: %9ld [page: %p]\n",
+		highend_pfn-1, pfn_to_page(highend_pfn-1));
+
+	printk("NR_MEM_SECTIONS: %ld\n", NR_MEM_SECTIONS);
+	printk("pfn_to_section_nr(highstart_pfn): %9ld\n",
+		pfn_to_section_nr(highstart_pfn));
+	printk("pfn_to_section_nr(highend_pfn):   %9ld\n",
+		pfn_to_section_nr(highend_pfn));
+	printk("pfn_to_section_nr(highend_pfn-1): %9ld\n",
+		pfn_to_section_nr(highend_pfn-1));
+
+	printk("totalhigh_pages: %9ld\n", totalhigh_pages);
+	printk(" totalram_pages: %9ld\n", totalram_pages);
+
+	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) {
+		if (pfn_valid(pfn))
+			add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
+		else {
+			if (WARN_ON_ONCE(1)) {
+				printk("bad pfn: %d\n", pfn);
+				break;
+			}
+		}
+	}
+	printk("totalhigh_pages: %9ld\n", totalhigh_pages);
+	printk(" totalram_pages: %9ld\n", totalram_pages);
+
+
 	totalram_pages += totalhigh_pages;
 }
-#endif /* CONFIG_FLATMEM */
+#endif /* CONFIG_NUMA */
 
 #else
 #define kmap_init() do { } while (0)
Index: linux/mm/sparse.c
===================================================================
--- linux.orig/mm/sparse.c
+++ linux/mm/sparse.c
@@ -295,9 +295,13 @@ void __init sparse_init(void)
 	struct page *map;
 	unsigned long *usemap;
 
+	printk("sparse_init()\n");
 	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
-		if (!present_section_nr(pnum))
+		printk("section %2ld: ", pnum);
+		if (!present_section_nr(pnum)) {
+			printk("!present\n");
 			continue;
+		}
 
 		map = sparse_early_mem_map_alloc(pnum);
 		if (!map)
@@ -309,6 +313,8 @@ void __init sparse_init(void)
 
 		sparse_init_one_section(__nr_to_section(pnum), pnum, map,
 								usemap);
+		printk("map: %p, usemap: %p [content: %08lx]\n",
+			map, usemap, __nr_to_section(pnum)->section_mem_map);
 	}
 }
 


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: bugreport kernel panic on early stage, with HIGHMEM4G:
  2008-01-15 11:39 ` Ingo Molnar
@ 2008-01-15 13:13   ` Ingo Molnar
  2008-01-17 12:26     ` Mel Gorman
  2008-01-17 13:31   ` Denys Fedoryshchenko
  1 sibling, 1 reply; 5+ messages in thread
From: Ingo Molnar @ 2008-01-15 13:13 UTC (permalink / raw)
  To: Denys Fedoryshchenko
  Cc: linux-kernel, Andrew Morton, Thomas Gleixner, H. Peter Anvin,
	Mel Gorman

* Ingo Molnar <mingo@elte.hu> wrote:

> thanks for the detailed report, i think i know what's going on. Could 
> you try the patch below, does it fix your problem?

find below the fix with a more complete changelog and with no debugging 
printouts.

	Ingo

-------------->
Subject: x86: fix boot crash on HIGHMEM4G && SPARSEMEM
From: Ingo Molnar <mingo@elte.hu>

Denys Fedoryshchenko reported a bootup crash when he upgraded
his system from 3GB to 4GB RAM:

   http://lkml.org/lkml/2008/1/7/9

the bug is due to HIGHMEM4G && SPARSEMEM kernels making pfn_to_page() to 
return an invalid pointer when the pfn is in a memory hole. The 256 MB 
PCI aperture at the end of RAM was not mapped by sparsemem, and hence 
the pfn was not valid. But set_highmem_pages_init() iterated this range 
without checking the pfn's validity first - crashing the bootup.

this bug was probably present in the sparsemem code ever since sparsemem 
has been introduced in v2.6.13. It was masked due to HIGHMEM64G using 
larger memory regions in sparsemem_32.h:

 #ifdef CONFIG_X86_PAE
 #define SECTION_SIZE_BITS       30
 #define MAX_PHYSADDR_BITS       36
 #define MAX_PHYSMEM_BITS        36
 #else
 #define SECTION_SIZE_BITS       26
 #define MAX_PHYSADDR_BITS       32
 #define MAX_PHYSMEM_BITS        32
 #endif

which creates 1GB sparsemem regions instead of 64MB sparsemem regions. 
So in practice we only ever created true sparsemem holes on x86 with 
HIGHMEM4G - but that was rarely used by distros.

( btw., we could probably save 2MB of mem_map[]s on X86_PAE if we reduced
  the sparsemem region size to 256 MB. )

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/mm/init_32.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

Index: linux/arch/x86/mm/init_32.c
===================================================================
--- linux.orig/arch/x86/mm/init_32.c
+++ linux/arch/x86/mm/init_32.c
@@ -321,8 +321,13 @@ extern void set_highmem_pages_init(int);
 static void __init set_highmem_pages_init(int bad_ppro)
 {
 	int pfn;
-	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
-		add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
+	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++) {
+		/*
+		 * Holes under sparsemem might not have no mem_map[]:
+		 */
+		if (pfn_valid(pfn))
+			add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
+	}
 	totalram_pages += totalhigh_pages;
 }
 #endif /* CONFIG_FLATMEM */

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: bugreport kernel panic on early stage, with HIGHMEM4G:
  2008-01-15 13:13   ` Ingo Molnar
@ 2008-01-17 12:26     ` Mel Gorman
  0 siblings, 0 replies; 5+ messages in thread
From: Mel Gorman @ 2008-01-17 12:26 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: Denys Fedoryshchenko, linux-kernel, Andrew Morton,
	Thomas Gleixner, H. Peter Anvin

On (15/01/08 14:13), Ingo Molnar didst pronounce:
> 
> * Ingo Molnar <mingo@elte.hu> wrote:
> 
> > thanks for the detailed report, i think i know what's going on. Could 
> > you try the patch below, does it fix your problem?
> 
> find below the fix with a more complete changelog and with no debugging 
> printouts.
> 

Looks good and the right thing to do. If you check the equivilant code for
DISCONTIG, it calls pfn_valid() searching for holes so it was expected.

Acked-by: Mel Gorman <mel@csn.ul.ie>

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: bugreport kernel panic on early stage, with HIGHMEM4G:
  2008-01-15 11:39 ` Ingo Molnar
  2008-01-15 13:13   ` Ingo Molnar
@ 2008-01-17 13:31   ` Denys Fedoryshchenko
  1 sibling, 0 replies; 5+ messages in thread
From: Denys Fedoryshchenko @ 2008-01-17 13:31 UTC (permalink / raw)
  To: Ingo Molnar
  Cc: linux-kernel, Andrew Morton, Thomas Gleixner, H. Peter Anvin,
	Mel Gorman

Patch fixes the problem.
Here is dmesg (i cut it, probably remaining part of it not required).
0]   Movable zone: 0 pages used for memmap
[    0.000000] DMI 2.4 present.
[    0.000000] ACPI: RSDP 000FE020, 0014 (r0 INTEL )
[    0.000000] ACPI: RSDT CFEFD038, 0050 (r1 INTEL  DG965SS       6B7       
1000013)
[    0.000000] ACPI: FACP CFEFC000, 0074 (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: DSDT CFEF7000, 41AA (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: FACS CFE9C000, 0040
[    0.000000] ACPI: APIC CFEF6000, 0078 (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: WDDT CFEF5000, 0040 (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: MCFG CFEF4000, 003C (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: ASF! CFEF3000, 00A6 (r32 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: HPET CFEF2000, 0038 (r1 INTEL  DG965SS       6B7 MSFT  
1000013)
[    0.000000] ACPI: SSDT CFE9A000, 020C (r1 INTEL     CpuPm      6B7 MSFT  
1000013)
[    0.000000] ACPI: SSDT CFE99000, 0175 (r1 INTEL   Cpu0Ist      6B7 MSFT  
1000013)
[    0.000000] ACPI: SSDT CFE98000, 0175 (r1 INTEL   Cpu1Ist      6B7 MSFT  
1000013)
[    0.000000] ACPI: SSDT CFE97000, 0175 (r1 INTEL   Cpu2Ist      6B7 MSFT  
1000013)
[    0.000000] ACPI: SSDT CFE96000, 0175 (r1 INTEL   Cpu3Ist      6B7 MSFT  
1000013)
[    0.000000] ACPI: PM-Timer IO Port: 0x408
[    0.000000] ACPI: Local APIC address 0xfee00000
[    0.000000] ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled)
[    0.000000] Processor #0 6:15 APIC version 20
[    0.000000] ACPI: LAPIC (acpi_id[0x02] lapic_id[0x01] enabled)
[    0.000000] Processor #1 6:15 APIC version 20
[    0.000000] ACPI: LAPIC (acpi_id[0x03] lapic_id[0x82] disabled)
[    0.000000] ACPI: LAPIC (acpi_id[0x04] lapic_id[0x83] disabled)
[    0.000000] ACPI: LAPIC_NMI (acpi_id[0x01] dfl dfl lint[0x1])
[    0.000000] ACPI: LAPIC_NMI (acpi_id[0x02] dfl dfl lint[0x1])
[    0.000000] ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_base[0])
[    0.000000] IOAPIC[0]: apic_id 2, version 32, address 0xfec00000, GSI 0-23
[    0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl)
[    0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level)
[    0.000000] ACPI: IRQ0 used by override.
[    0.000000] ACPI: IRQ2 used by override.
[    0.000000] ACPI: IRQ9 used by override.
[    0.000000] Enabling APIC mode:  Flat.  Using 1 I/O APICs
[    0.000000] ACPI: HPET id: 0x8086a201 base: 0xfed00000
[    0.000000] Using ACPI (MADT) for SMP configuration information
[    0.000000] Allocating PCI resources starting at d4000000 (gap: 
d0000000:2ff00000)
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total 
pages: 1040384
[    0.000000] Kernel command line: root=/dev/sdb2
[    0.000000] mapped APIC to ffffb000 (fee00000)
[    0.000000] mapped IOAPIC to ffffa000 (fec00000)
[    0.000000] Enabling fast FPU save and restore... done.
[    0.000000] Enabling unmasked SIMD FPU exception support... done.
[    0.000000] Initializing CPU#0
[    0.000000] PID hash table entries: 4096 (order: 12, 16384 bytes)
[    0.000000] Detected 2397.647 MHz processor.
[   23.770446] Console: colour VGA+ 80x25
[   23.770448] console [tty0] enabled
[   23.779138] Dentry cache hash table entries: 131072 (order: 7, 524288 
bytes)
[   23.779395] Inode-cache hash table entries: 65536 (order: 6, 262144 bytes)
[   23.788992] set_highmem_pages_init(bad_ppro:0)
[   23.789056] sizeof(struct page):        32
[   23.789118] sizeof(struct mem_section): 8
[   23.789180] PFN_SECTION_SHIFT:          14
[   23.789243] mem_map: 00000000
[   23.789304]   highstart_pfn:    229376 [page: c1700700]
[   23.789367]     highend_pfn:   1048576 [page: 02000000]
[   23.789431]   highend_pfn-1:   1048575 [page: 01ffffe0]
[   23.789494] NR_MEM_SECTIONS: 64
[   23.789555] pfn_to_section_nr(highstart_pfn):        14
[   23.789619] pfn_to_section_nr(highend_pfn):          64
[   23.789682] pfn_to_section_nr(highend_pfn-1):        63
[   23.789745] totalhigh_pages:         0
[   23.789807]  totalram_pages:    221519
[   23.924275] WARNING: at arch/x86/mm/init_32.c:353 set_highmem_pages_init()
[   23.924344] Pid: 0, comm: swapper Not tainted 2.6.24-rc8-git1 #1
[   23.924409]  [<c0104fac>] show_trace_log_lvl+0x1a/0x2f
[   23.924503]  [<c0105805>] show_trace+0x12/0x14
[   23.924591]  [<c0105bc1>] dump_stack+0x6c/0x72
[   23.924678]  [<c03ce93e>] mem_init+0x2a7/0x596
[   23.924768]  [<c03c38b6>] start_kernel+0x271/0x2fb
[   23.924858]  [<00000000>] _stext+0x3feff000/0x19
[   23.924945]  =======================
[   23.925007] bad pfn: 851968
[   23.925068] totalhigh_pages:    622129
[   23.925130]  totalram_pages:    221519
[   23.925194] Memory: 3373812k/4194304k available (1990k kernel code, 30976k 
reserved, 813k data, 208k init, 2488516k highmem)
[   23.925300] virtual kernel memory layout:
[   23.925301]     fixmap  : 0xffe15000 - 0xfffff000   (1960 kB)
[   23.925302]     pkmap   : 0xff800000 - 0xffc00000   (4096 kB)
[   23.925302]     vmalloc : 0xf8800000 - 0xff7fe000   ( 111 MB)
[   23.925303]     lowmem  : 0xc0000000 - 0xf8000000   ( 896 MB)
[   23.925304]       .init : 0xc03c3000 - 0xc03f7000   ( 208 kB)
[   23.925305]       .data : 0xc02f18c5 - 0xc03bcdec   ( 813 kB)
[   23.925305]       .text : 0xc0100000 - 0xc02f18c5   (1990 kB)
[   23.925805] Checking if this processor honours the WP bit even in 
supervisor mode... Ok.
[   23.926109] hpet clockevent registered
[   23.986100] Calibrating delay using timer specific routine.. 4797.85 
BogoMIPS (lpj=2398929)
[   23.986267] Mount-cache hash table entries: 512
[   23.986424] CPU: After generic identify, caps: bfebfbff 20100000 00000000 
00000000 0000e3bd 00000000 00000001 00000000
[   23.986429] monitor/mwait feature present.
[   23.987257] using mwait in idle threads.
[   23.987320] CPU: L1 I cache: 32K, L1 D cache: 32K
[   23.987408] CPU: L2 cache: 4096K
[   23.987470] CPU: Physical Processor ID: 0
[   23.987532] CPU: Processor Core ID: 0
[   23.987594] CPU: After all inits, caps: bfebfbff 20100000 00000000 
00003940 0000e3bd 00000000 00000001 00000000
[   23.987599] Intel machine check architecture supported.
[   23.987664] Intel machine check reporting enabled on CPU#0.
[   23.987729] Compat vDSO mapped to ffffe000.
[   23.987800] Checking 'hlt' instruction... OK.
[   23.991366] SMP alternatives: switching to UP code
[   23.991940] ACPI: Core revision 20070126
[   23.993505] Parsing all Control Methods:
[   23.993595] Table [DSDT](id 0001) - 610 Objects with 62 Devices 154 
Methods 38 Regions
[   23.993793] Parsing all Control Methods:
[   23.993879] Table [SSDT](id 0002) - 10 Objects with 0 Devices 4 Methods 0 
Regions
[   23.994055] Parsing all Control Methods:
[   23.994147] Table [SSDT](id 0003) - 5 Objects with 0 Devices 3 Methods 0 
Regions
[   23.994323] Parsing all Control Methods:
[   23.994408] Table [SSDT](id 0004) - 5 Objects with 0 Devices 3 Methods 0 
Regions
[   23.994584] Parsing all Control Methods:
[   23.994668] Table [SSDT](id 0005) - 5 Objects with 0 Devices 3 Methods 0 
Regions
[   23.994844] Parsing all Control Methods:
[   23.994930] Table [SSDT](id 0006) - 5 Objects with 0 Devices 3 Methods 0 
Regions
[   23.995088]  tbxface-0598 [00] tb_load_namespace     : ACPI Tables 
successfully acquired
[   23.996207] evxfevnt-0091 [00] enable                : Transition to ACPI 
mode successful
[   23.996377] CPU0: Intel(R) Core(TM)2 CPU          6600  @ 2.40GHz stepping 
06
[   23.996500] SMP alternatives: switching to SMP code
[   23.996894] Booting processor 1/1 eip 3000
[   24.007645] Initializing CPU#1
[   24.067080] Calibrating delay using timer specific routine.. 4795.20 
BogoMIPS (lpj=2397601)
[   24.067085] CPU: After generic identify, caps: bfebfbff 20100000 00000000 
00000000 0000e3bd 00000000 00000001 00000000
[   24.067089] monitor/mwait feature present.
[   24.067091] CPU: L1 I cache: 32K, L1 D cache: 32K
[   24.067092] CPU: L2 cache: 4096K
[   24.067093] CPU: Physical Processor ID: 0
[   24.067094] CPU: Processor Core ID: 1
[   24.067096] CPU: After all inits, caps: bfebfbff 20100000 00000000 
00003940 0000e3bd 00000000 00000001 00000000
[   24.067100] Intel machine check architecture supported.
[   24.067102] Intel machine check reporting enabled on CPU#1.
[   24.067636] CPU1: Intel(R) Core(TM)2 CPU          6600  @ 2.40GHz stepping 
06
[   24.068359] Total of 2 processors activated (9593.06 BogoMIPS).
[   24.068559] ENABLING IO-APIC IRQs
[   24.068778] ..TIMER: vector=0x31 apic1=0 pin1=2 apic2=-1 pin2=-1
[   24.180160] checking TSC synchronization [CPU#0 -> CPU#1]: passed.
[   24.200256] Brought up 2 CPUs
[   24.200332] CPU0 attaching sched-domain:
[   24.200334]  domain 0: span 00000003
[   24.200336]   groups: 00000001 00000002
[   24.200339] CPU1 attaching sched-domain:
[   24.200340]  domain 0: span 00000003
[   24.200341]   groups: 00000002 00000001
[   24.200493] net_namespace: 64 bytes
[   24.200849] NET: Registered protocol family 16
[   24.201014] ACPI: bus type pci registered
[   24.201138] PCI: BIOS Bug: MCFG area at f0000000 is not E820-reserved
[   24.201203] PCI: Not using MMCONFIG.
[   24.202626] PCI: Using configuration type 1
[   24.202688] Setting up standard PCI resources
[   24.209439] evgpeblk-0956 [00] ev_create_gpe_block   : GPE 00 to 1F [_GPE] 
4 regs on int 0x9
[   24.210162] evgpeblk-1052 [00] ev_initialize_gpe_bloc: Found 8 Wake, 
Enabled 4 Runtime GPEs in this block
[   24.210318] ACPI: EC: Look up EC in DSDT
[   24.211574] Completing Region/Field/Buffer/Package 
initialization:........................................................................
[   24.213966] Initialized 32/38 Regions 0/0 Fields 25/25 Buffers 15/35 
Packages (649 nodes)
[   24.214124] Initializing Device/Processor/Thermal objects by executing 
_INI methods:.
[   24.214272] Executed 1 _INI methods requiring 0 _STA executions (examined 
68 objects)
[   24.214443] ACPI: Interpreter enabled
[   24.214506] ACPI: (supports S0 S3 S5)
[   24.214648] ACPI: Using IOAPIC for interrupt routing
[   24.218082] ACPI: PCI Root Bridge [PCI0] (0000:00)
[   24.218839] PCI quirk: region 0400-047f claimed by ICH6 ACPI/GPIO/TCO
[   24.218906] PCI quirk: region 0500-053f claimed by ICH6 GPIO
[   24.219522] PCI: Transparent bridge - 0000:00:1e.0
[   24.219617] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT]
[   24.219888] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.P32_._PRT]
[   24.220094] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX0._PRT]
[   24.220180] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX1._PRT]
[   24.220267] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX2._PRT]
[   24.220353] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX3._PRT]
[   24.220438] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PEX4._PRT]
[   24.224894] ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 7 9 10 *11 12)
[   24.225280] ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 7 9 *10 11 12)
[   24.225662] ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 5 7 9 10 *11 12)
[   24.226043] ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 7 9 10 *11 12)
[   24.226428] ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 7 *9 10 11 12)
[   24.226810] ACPI: PCI Interrupt Link [LNKF] (IRQs 3 4 5 7 9 *10 11 12)
[   24.227191] ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 7 *9 10 11 12)
[   24.227574] ACPI: PCI Interrupt Link [LNKH] (IRQs 3 4 5 7 9 10 *11 12)
[   24.227980] Linux Plug and Play Support v0.97 (c) Adam Belay
[   24.228065] pnp: PnP ACPI init
[   24.228131] ACPI: bus type pnp registered
[   24.230140] pnp: PnP ACPI: found 11 devices
[   24.230203] ACPI: ACPI bus type pnp unregistered
[   24.230403] SCSI subsystem initialized
[   24.230493] libata version 3.00 loaded.
[   24.230556] PCI: Using ACPI for IRQ routing
[   24.230619] PCI: If a device doesn't work, try "pci=routeirq".  If it 
helps, post a report
[   24.235340] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
[   24.235515] hpet0: 3 64-bit timers, 14318180 Hz
[   24.236603] ACPI: RTC can wake from S4
[   24.237154] Time: tsc clocksource has been installed.
[   24.239432] system 00:01: iomem range 0xf0000000-0xf7ffffff has been 
reserved
[   24.239510] system 00:01: iomem range 0xfed13000-0xfed13fff has been 
reserved
[   24.239576] system 00:01: iomem range 0xfed14000-0xfed17fff has been 
reserved
[   24.239642] system 00:01: iomem range 0xfed18000-0xfed18fff has been 
reserved
[   24.239708] system 00:01: iomem range 0xfed19000-0xfed19fff has been 
reserved
[   24.239773] system 00:01: iomem range 0xfed1c000-0xfed1ffff has been 
reserved
[   24.239839] system 00:01: iomem range 0xfed20000-0xfed3ffff has been 
reserved
[   24.239905] system 00:01: iomem range 0xfed45000-0xfed99fff has been 
reserved
[   24.239971] system 00:01: iomem range 0xc0000-0xdffff could not be reserved
[   24.240036] system 00:01: iomem range 0xe0000-0xfffff could not be reserved
[   24.240106] system 00:06: ioport range 0x500-0x53f has been reserved
[   24.240170] system 00:06: ioport range 0x400-0x47f has been reserved
[   24.240235] system 00:06: ioport range 0x680-0x6ff has been reserved
[   24.270635] PCI: Failed to allocate mem resource #6:20000@e0000000 for 
0000:01:00.0
[   24.270735] PCI: Bridge: 0000:00:01.0
[   24.270797]   IO window: 2000-2fff
[   24.270861]   MEM window: e0000000-e1ffffff
[   24.270924]   PREFETCH window: d0000000-dfffffff
[   24.270988] PCI: Bridge: 0000:00:1c.0
[   24.271050]   IO window: disabled.
[   24.271114]   MEM window: disabled.
[   24.271177]   PREFETCH window: disabled.
[   24.271242] PCI: Bridge: 0000:00:1c.1
[   24.271305]   IO window: 1000-1fff
[   24.271369]   MEM window: e2000000-e20fffff
[   24.271435]   PREFETCH window: disabled.
[   24.271499] PCI: Bridge: 0000:00:1c.2
[   24.271561]   IO window: disabled.
[   24.271624]   MEM window: disabled.
[   24.271688]   PREFETCH window: disabled.
[   24.271752] PCI: Bridge: 0000:00:1c.3
[   24.271814]   IO window: disabled.
[   24.271878]   MEM window: disabled.
[   24.271941]   PREFETCH window: disabled.
[   24.272005] PCI: Bridge: 0000:00:1c.4
[   24.272067]   IO window: disabled.
[   24.272130]   MEM window: disabled.
[   24.272193]   PREFETCH window: disabled.
[   24.272258] PCI: Bridge: 0000:00:1e.0
[   24.272319]   IO window: disabled.
[   24.272383]   MEM window: disabled.
[   24.272447]   PREFETCH window: disabled.
[   24.272520] ACPI: PCI Interrupt 0000:00:01.0[A] -> GSI 16 (level, low) -> 
IRQ 16
[   24.272646] PCI: Setting latency timer of device 0000:00:01.0 to 64
[   24.272660] ACPI: PCI Interrupt 0000:00:1c.0[A] -> GSI 17 (level, low) -> 
IRQ 17
[   24.272786] PCI: Setting latency timer of device 0000:00:1c.0 to 64
[   24.272800] ACPI: PCI Interrupt 0000:00:1c.1[B] -> GSI 16 (level, low) -> 
IRQ 16
[   24.272925] PCI: Setting latency timer of device 0000:00:1c.1 to 64
[   24.272939] ACPI: PCI Interrupt 0000:00:1c.2[C] -> GSI 18 (level, low) -> 
IRQ 18
[   24.273065] PCI: Setting latency timer of device 0000:00:1c.2 to 64
[   24.273079] ACPI: PCI Interrupt 0000:00:1c.3[D] -> GSI 19 (level, low) -> 
IRQ 19
[   24.273204] PCI: Setting latency timer of device 0000:00:1c.3 to 64
[   24.273217] ACPI: PCI Interrupt 0000:00:1c.4[A] -> GSI 17 (level, low) -> 
IRQ 17
[   24.273343] PCI: Setting latency timer of device 0000:00:1c.4 to 64
[   24.273351] PCI: Setting latency timer of device 0000:00:1e.0 to 64
[   24.273370] NET: Registered protocol family 2
[   24.283667] IP route cache hash table entries: 32768 (order: 5, 131072 
bytes)
[   24.283894] TCP established hash table entries: 131072 (order: 8, 1048576 
bytes)
[   24.284300] TCP bind hash table entries: 65536 (order: 7, 786432 bytes)
[   24.284563] TCP: Hash tables configured (established 131072 bind 65536)
[   24.284634] TCP reno registered
[   24.288032] Machine check exception polling timer started.
[   24.288271] IA-32 Microcode Update Driver: v1.14a 
<tigran@aivazian.fsnet.co.uk>
[   24.288744] highmem bounce pool size: 64 pages
[   24.288809] Total HugeTLB memory allocated, 0
[   24.289015] Block layer SCSI generic (bsg) driver version 0.4 loaded 
(major 254)
[   24.289115] io scheduler noop registered
[   24.289188] io scheduler cfq registered (default)
[   24.289542] Boot video device is 0000:01:00.0
[   24.289613] PCI: Setting latency timer of device 0000:00:01.0 to 64



On Tue, 15 Jan 2008 12:39:47 +0100, Ingo Molnar wrote
> * Denys Fedoryshchenko <denys@visp.net.lb> wrote:
> 
> > Hi
> > 
> > After physical memory upgrade from 3GB to 4GB (also it happens on 5GB) 
> > got kernel panic.
> > 
> > Because it is happening on early stage and my machine doesn't contain 
> > serial port, i had to take photo. Kernel boots fine with 64GB highmem, 
> > no highmem, or highmem4G with limited memory by mem=3G. All dmesg 
> > attached. Also i attach dmidecode and lspci -vvv output, probably it 
> > will be useful.
> 
> thanks for the detailed report, i think i know what's going on. 
> Could you try the patch below, does it fix your problem?
> 
> this seems to be a SPARSEMEM bug which is present in v2.6.23 as well 
> and has probably been present ever since SPARSEMEM was added to 32-
> bit x86.
> 
> There's a ~256MB hole in your e820 memory map (the pci aperture),
>  which causes the last 4 sparsemem sections (each covering 64MB of 
> RAM) to be not present - and they are thus missing from the 
> sparsemem mem_map[] too. The highmem init code on the other hand 
> assumes that all pages are in the mem_map[]:
> 
>  static void __init set_highmem_pages_init(int bad_ppro)
>  {
>         int pfn;
>         for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
>                 add_one_highpage_init(pfn_to_page(pfn), pfn, 
> bad_ppro);
> 
> the pfn_to_page() is unconditional and dereferences to a NULL-ish 
> pointer which crashes your box. highend_pfn is what got 
> miscalculated by 256 MB, so set_highmem_pages_init() tried to 
> reference a non-existing struct page - but it should still be robust 
> enough against non-existent pages.
> 
> The patch below fixes this bug. Please also send a dmesg if you 
> manage to boot the box up fine, i've added a few debug printouts to 
> confirm this theory. (i'll figure out whether we need to clip 
> highend_pfn as well - but this patch alone should be good enough to 
> fix the crash on your box.)
> 
> 	Ingo
> 
> ------------->
> Subject: x86: fix CONFIG_SPARSEMEM highmem init bug
> From: Ingo Molnar <mingo@elte.hu>
> 
> fix CONFIG_SPARSEMEM highmem init bug.
> 
> Signed-off-by: Ingo Molnar <mingo@elte.hu>
> ---
>  arch/x86/mm/init_32.c |   43 
> ++++++++++++++++++++++++++++++++++++++++--- mm/sparse.c           |  
>   8 +++++++- 2 files changed, 47 insertions(+), 4 deletions(-)
> 
> Index: linux/arch/x86/mm/init_32.c
> ===================================================================
> --- linux.orig/arch/x86/mm/init_32.c
> +++ linux/arch/x86/mm/init_32.c
> @@ -321,11 +321,48 @@ extern void set_highmem_pages_init(int);
>  static void __init set_highmem_pages_init(int bad_ppro)
>  {
>  	int pfn;
> -	for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
> -		add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
> +
> +	printk("set_highmem_pages_init(bad_ppro:%d)\n", bad_ppro);
> +	printk("sizeof(struct page):        %d\n", sizeof(struct page));
> +	printk("sizeof(struct mem_section): %d\n", sizeof(struct 
> mem_section)); +	printk("PFN_SECTION_SHIFT:          %d\n",
>  PFN_SECTION_SHIFT); + +	printk("mem_map: %p\n", mem_map); +	
> printk("  highstart_pfn: %9ld [page: %p]\n", +		
highstart_pfn, 
> pfn_to_page(highstart_pfn)); +	printk("    highend_pfn: %9ld [page: 
> %p]\n", +		highend_pfn, pfn_to_page(highend_pfn)); +	
printk("  
> highend_pfn-1: %9ld [page: %p]\n", +		highend_pfn-1, 
> pfn_to_page(highend_pfn-1)); + +	printk("NR_MEM_SECTIONS: %ld\n",
>  NR_MEM_SECTIONS); +	printk("pfn_to_section_nr(highstart_pfn): 
> %9ld\n", +		pfn_to_section_nr(highstart_pfn)); +	
> printk("pfn_to_section_nr(highend_pfn):   %9ld\n", +	
> 	pfn_to_section_nr(highend_pfn)); +	
> printk("pfn_to_section_nr(highend_pfn-1): %9ld\n", +	
> 	pfn_to_section_nr(highend_pfn-1)); + +	printk("totalhigh_pages: 
> %9ld\n", totalhigh_pages); +	printk(" totalram_pages: %9ld\n", 
> totalram_pages); + +	for (pfn = highstart_pfn; pfn < highend_pfn; 
> pfn++) { +		if (pfn_valid(pfn)) +		
> 	add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro); +	
	else { +	
> 		if (WARN_ON_ONCE(1)) { +				
printk("bad pfn: %d\n", pfn); +				break;
> +			}
> +		}
> +	}
> +	printk("totalhigh_pages: %9ld\n", totalhigh_pages);
> +	printk(" totalram_pages: %9ld\n", totalram_pages);
> +
> +
>  	totalram_pages += totalhigh_pages;
>  }
> -#endif /* CONFIG_FLATMEM */
> +#endif /* CONFIG_NUMA */
> 
>  #else
>  #define kmap_init() do { } while (0)
> Index: linux/mm/sparse.c
> ===================================================================
> --- linux.orig/mm/sparse.c
> +++ linux/mm/sparse.c
> @@ -295,9 +295,13 @@ void __init sparse_init(void)
>  	struct page *map;
>  	unsigned long *usemap;
> 
> +	printk("sparse_init()\n");
>  	for (pnum = 0; pnum < NR_MEM_SECTIONS; pnum++) {
> -		if (!present_section_nr(pnum))
> +		printk("section %2ld: ", pnum);
> +		if (!present_section_nr(pnum)) {
> +			printk("!present\n");
>  			continue;
> +		}
> 
>  		map = sparse_early_mem_map_alloc(pnum);
>  		if (!map)
> @@ -309,6 +313,8 @@ void __init sparse_init(void)
> 
>  		sparse_init_one_section(__nr_to_section(pnum), pnum, map,
>  								usemap);
> +		printk("map: %p, usemap: %p [content: %08lx]\n",
> +			map, usemap, __nr_to_section(pnum)->section_mem_map);
>  	}
>  }


--
Denys Fedoryshchenko
Technical Manager
Virtual ISP S.A.L.


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2008-01-17 13:52 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-01-07  5:47 bugreport kernel panic on early stage, with HIGHMEM4G: Denys Fedoryshchenko
2008-01-15 11:39 ` Ingo Molnar
2008-01-15 13:13   ` Ingo Molnar
2008-01-17 12:26     ` Mel Gorman
2008-01-17 13:31   ` Denys Fedoryshchenko

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox