This patchset includes 3 parts. ------------------------------------- 1st part: Includes 13 upstream commits. They are recommended by Nigel Croxon since he back ported this feature to rhel7. This 13 back ports works for all ARCHs except of s390 since it always broke s390n kdump.
c2c1b08 fs/proc/vmcore.c: put if tests in the top of the while loop to reduce duplication 0fa73b8 include/linux/mm.h: add PAGE_ALIGNED() helper b27eb18 vmcore: clean up read_vmcore() f2bdacd vmcore: allocate buffer for ELF headers on page-size alignment 7f614cd vmcore: treat memory chunks referenced by PT_LOAD program header entries in page-size boundary in vmcore_list cef2ac3 vmalloc: make find_vm_area check in range e69e9d4a vmalloc: introduce remap_vmalloc_range_partial 087350c vmcore: allocate ELF note segment in the 2nd kernel vmalloc memory ef9e78f vmcore: allow user process to remap ELF note segment buffer 591ff71 vmcore: calculate vmcore file size from buffer size and total size of vmcore objects 8308697 vmcore: support mmap() on /proc/vmcore 5a74953 s390/kdump: Disable mmap for s390 0692ded fs/proc/vmcore.c:mmap_vmcore: skip non-ram pages reported by hypervisors ------------------------------------- 2nd part: Includes several s390 related memory commits. Since these 13 commits in 1st part worked on all ARCHs except of S390. Then I asked IBM S390 kernel developers Michael Holzheu <[email protected]>, Jan Willeke <[email protected]> to help, they suggested these 3 commits. With them it works for s390 kdump. 191a2fa s390/kdump: Allow copy_oldmem_page() copy to virtual memory" 9cb2181 vmcore: introduce remap_oldmem_pfn_range()" 23df79d s390/vmcore: implement remap_oldmem_pfn_range for s390" ------------------------------------- 3rd part: Includes 3 commits. They are back ported because 3 commits in 2nd part will fail x86 and ppc compiling. 997c136 fs/proc/vmcore.c: add hook to read_from_oldmem() to check for non-ram pages 816422a asm-generic, mm: pgtable: consolidate zero page helpers 238ec4e [S390] zero page cache synonyms Brew build: I made a brew build here and tested it on s390 and x86 machine, both works. https://brewweb.devel.redhat.com/taskinfo?taskID=9357037 Then Nigel took patches from my rh-git and made a brew build as below. Lisa from HP took a test and confirmed this works on HP machines and it improved the dump as expected. https://brewweb.devel.redhat.com/taskinfo?taskID=9380805 My rh-git for this feature is here: git://git.engineering.redhat.com/users/bhe/rhel6/.git mmap-new Baoquan He (19): [S390] zero page cache synonyms asm-generic, mm: pgtable: consolidate zero page helpers fs/proc/vmcore.c: add hook to read_from_oldmem() to check for non-ram pages fs/proc/vmcore.c: put if tests in the top of the while loop to reduce duplication include/linux/mm.h: add PAGE_ALIGNED() helper vmcore: clean up read_vmcore() vmcore: allocate buffer for ELF headers on page-size alignment vmcore: treat memory chunks referenced by PT_LOAD program header entries in page-size boundary in vmcore_list vmalloc: make find_vm_area check in range vmalloc: introduce remap_vmalloc_range_partial vmcore: allocate ELF note segment in the 2nd kernel vmalloc memory vmcore: allow user process to remap ELF note segment buffer vmcore: calculate vmcore file size from buffer size and total size of vmcore objects vmcore: support mmap() on /proc/vmcore s390/kdump: Disable mmap for s390 s390/kdump: Allow copy_oldmem_page() copy to virtual memory vmcore: introduce remap_oldmem_pfn_range() s390/vmcore: implement remap_oldmem_pfn_range for s390 fs/proc/vmcore.c:mmap_vmcore: skip non-ram pages reported by hypervisors arch/mips/include/asm/pgtable.h | 10 +- arch/s390/include/asm/pgtable.h | 14 +- arch/s390/kernel/crash_dump.c | 75 +++- arch/s390/mm/init.c | 49 ++- fs/proc/vmcore.c | 899 ++++++++++++++++++++++++++++++---------- include/asm-generic/pgtable.h | 26 ++ include/linux/crash_dump.h | 9 + include/linux/mm.h | 3 + include/linux/vmalloc.h | 4 + mm/memory.c | 14 - mm/vmalloc.c | 70 +++- 11 files changed, 901 insertions(+), 272 deletions(-) -- 2.1.0 -- 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/
