Re: [PATCH 4.14 023/159] mm/sparsemem: Allocate mem_section at runtime for CONFIG_SPARSEMEM_EXTREME=y

From: Dou Liyang
Date: Wed Feb 07 2018 - 04:25:20 EST


Hi All,

I met the makedumpfile failed in the upstream kernel which contained
this patch. Did I missed something else?

In fedora27 host:

[douly@localhost code]$ ./makedumpfile -d 31 --message-level 31 -x
vmlinux_4.15+ vmcore_4.15+_from_cp_command vmcore_4.15+

sadump: does not have partition header
sadump: read dump device as unknown format
sadump: unknown format
LOAD (0)
phys_start : 1000000
phys_end : 2a86000
virt_start : ffffffff81000000
virt_end : ffffffff82a86000
LOAD (1)
phys_start : 1000
phys_end : 9fc00
virt_start : ffff880000001000
virt_end : ffff88000009fc00
LOAD (2)
phys_start : 100000
phys_end : 13000000
virt_start : ffff880000100000
virt_end : ffff880013000000
LOAD (3)
phys_start : 33000000
phys_end : 7ffd7000
virt_start : ffff880033000000
virt_end : ffff88007ffd7000
Linux kdump
page_size : 4096

max_mapnr : 7ffd7

Buffer size for the cyclic mode: 131061
The kernel version is not supported.
The makedumpfile operation may be incomplete.

num of NODEs : 1


Memory type : SPARSEMEM_EX

mem_map (0)
mem_map : ffff88007ff26000
pfn_start : 0
pfn_end : 8000
mem_map (1)
mem_map : 0
pfn_start : 8000
pfn_end : 10000
mem_map (2)
mem_map : 0
pfn_start : 10000
pfn_end : 18000
mem_map (3)
mem_map : 0
pfn_start : 18000
pfn_end : 20000
mem_map (4)
mem_map : 0
pfn_start : 20000
pfn_end : 28000
mem_map (5)
mem_map : 0
pfn_start : 28000
pfn_end : 30000
mem_map (6)
mem_map : 0
pfn_start : 30000
pfn_end : 38000
mem_map (7)
mem_map : 0
pfn_start : 38000
pfn_end : 40000
mem_map (8)
mem_map : 0
pfn_start : 40000
pfn_end : 48000
mem_map (9)
mem_map : 0
pfn_start : 48000
pfn_end : 50000
mem_map (10)
mem_map : 0
pfn_start : 50000
pfn_end : 58000
mem_map (11)
mem_map : 0
pfn_start : 58000
pfn_end : 60000
mem_map (12)
mem_map : 0
pfn_start : 60000
pfn_end : 68000
mem_map (13)
mem_map : 0
pfn_start : 68000
pfn_end : 70000
mem_map (14)
mem_map : 0
pfn_start : 70000
pfn_end : 78000
mem_map (15)
mem_map : 0
pfn_start : 78000
pfn_end : 7ffd7
mmap() is available on the kernel.
Checking for memory holes : [100.0 %] | STEP [Checking for memory holes ] : 0.000060 seconds
__vtop4_x86_64: Can't get a valid pte.
readmem: Can't convert a virtual address(ffff88007ffd7000) to physical address.
readmem: type_addr: 0, addr:ffff88007ffd7000, size:32768
__exclude_unnecessary_pages: Can't read the buffer of struct page.
create_2nd_bitmap: Can't exclude unnecessary pages.
Checking for memory holes : [100.0 %] \ STEP [Checking for memory holes ] : 0.000010 seconds
Checking for memory holes : [100.0 %] - STEP [Checking for memory holes ] : 0.000004 seconds
__vtop4_x86_64: Can't get a valid pte.
readmem: Can't convert a virtual address(ffff88007ffd7000) to physical address.
readmem: type_addr: 0, addr:ffff88007ffd7000, size:32768
__exclude_unnecessary_pages: Can't read the buffer of struct page.
create_2nd_bitmap: Can't exclude unnecessary pages.

Thanks,
dou
At 01/09/2018 11:44 AM, Mike Galbraith wrote:
On Tue, 2018-01-09 at 03:13 +0300, Kirill A. Shutemov wrote:

Mike, could you test this? (On top of the rest of the fixes.)

homer:..crash/2018-01-09-04:25 # ll
total 1863604
-rw------- 1 root root 66255 Jan 9 04:25 dmesg.txt
-rw-r--r-- 1 root root 182 Jan 9 04:25 README.txt
-rw-r--r-- 1 root root 2818240 Jan 9 04:25 System.map-4.15.0.gb2cd1df-master
-rw------- 1 root root 1832914928 Jan 9 04:25 vmcore
-rw-r--r-- 1 root root 72514993 Jan 9 04:25 vmlinux-4.15.0.gb2cd1df-master.gz

Yup, all better.

Sorry for the mess.

(why, developers not installing shiny new bugs is a whole lot worse:)

From 100fd567754f1457be94732046aefca204c842d2 Mon Sep 17 00:00:00 2001
From: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Date: Tue, 9 Jan 2018 02:55:47 +0300
Subject: [PATCH] kdump: Write a correct address of mem_section into vmcoreinfo

Depending on configuration mem_section can now be an array or a pointer
to an array allocated dynamically. In most cases, we can continue to refer
to it as 'mem_section' regardless of what it is.

But there's one exception: '&mem_section' means "address of the array" if
mem_section is an array, but if mem_section is a pointer, it would mean
"address of the pointer".

We've stepped onto this in kdump code. VMCOREINFO_SYMBOL(mem_section)
writes down address of pointer into vmcoreinfo, not array as we wanted.

Let's introduce VMCOREINFO_ARRAY() that would handle the situation
correctly for both cases.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Fixes: 83e3c48729d9 ("mm/sparsemem: Allocate mem_section at runtime for CONFIG_SPARSEMEM_EXTREME=y")
---
include/linux/crash_core.h | 2 ++
kernel/crash_core.c | 2 +-
2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h
index 06097ef30449..83ae04950269 100644
--- a/include/linux/crash_core.h
+++ b/include/linux/crash_core.h
@@ -42,6 +42,8 @@ phys_addr_t paddr_vmcoreinfo_note(void);
vmcoreinfo_append_str("PAGESIZE=%ld\n", value)
#define VMCOREINFO_SYMBOL(name) \
vmcoreinfo_append_str("SYMBOL(%s)=%lx\n", #name, (unsigned long)&name)
+#define VMCOREINFO_ARRAY(name) \
+ vmcoreinfo_append_str("SYMBOL(%s)=%lx\n", #name, (unsigned long)name)
#define VMCOREINFO_SIZE(name) \
vmcoreinfo_append_str("SIZE(%s)=%lu\n", #name, \
(unsigned long)sizeof(name))
diff --git a/kernel/crash_core.c b/kernel/crash_core.c
index b3663896278e..d4122a837477 100644
--- a/kernel/crash_core.c
+++ b/kernel/crash_core.c
@@ -410,7 +410,7 @@ static int __init crash_save_vmcoreinfo_init(void)
VMCOREINFO_SYMBOL(contig_page_data);
#endif
#ifdef CONFIG_SPARSEMEM
- VMCOREINFO_SYMBOL(mem_section);
+ VMCOREINFO_ARRAY(mem_section);
VMCOREINFO_LENGTH(mem_section, NR_SECTION_ROOTS);
VMCOREINFO_STRUCT_SIZE(mem_section);
VMCOREINFO_OFFSET(mem_section, section_mem_map);

_______________________________________________
kexec mailing list
kexec@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/kexec