)]}'
{
  "log": [
    {
      "commit": "98075d245a5bc4aeebc2e9f16fa8b089a5c200ac",
      "tree": "5d248fa7ec872548e43ed2acb864d701adc65f42",
      "parents": [
        "c19b2930df0621500913c005c06978bd8933110b"
      ],
      "author": {
        "name": "Zoltan Menyhart",
        "email": "Zoltan.Menyhart@bull.net",
        "time": "Fri Apr 11 15:21:35 2008 -0700"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Fri Apr 11 15:21:35 2008 -0700"
      },
      "message": "[IA64] Fix NUMA configuration issue\n\nThere is a NUMA memory configuration issue in 2.6.24:\n\nA 2-node machine of ours has got the following memory layout:\n\nNode 0:\t0 - 2 Gbytes\nNode 0:\t4 - 8 Gbytes\nNode 1:\t8 - 16 Gbytes\nNode 0:\t16 - 18 Gbytes\n\n\"efi_memmap_init()\" merges the three last ranges into one.\n\n\"register_active_ranges()\" is called as follows:\n\nefi_memmap_walk(register_active_ranges, NULL);\n\ni.e. once for the 4 - 18 Gbytes range. It picks up the node\nnumber from the start address, and registers all the memory for\nthe node #0.\n\n\"register_active_ranges()\" should be called as follows to\nmake sure there is no merged address range at its entry:\n\nefi_memmap_walk(filter_memory, register_active_ranges);\n\n\"filter_memory()\" is similar to \"filter_rsvd_memory()\",\nbut the reserved memory ranges are not filtered out.\n\nSigned-off-by: Zoltan Menyhart \u003cZoltan.Menyhart@bull.net\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "cb3808532eeb1719667356157fac9222ccb2c4ff",
      "tree": "b86a8de1706db7e4b8968ddb587aa812d8da8084",
      "parents": [
        "5c3391f9f749023a49c64d607da4fb49263690eb"
      ],
      "author": {
        "name": "Bernhard Walle",
        "email": "bwalle@suse.de",
        "time": "Thu Oct 18 23:41:00 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:50 2007 -0700"
      },
      "message": "Use extended crashkernel command line on ia64\n\nThis patch adapts IA64 to use the generic parse_crashkernel() function instead\nof its own parsing for the crashkernel command line.\n\nBecause the total amount of System RAM must be known when calling this\nfunction, efi_memmap_init() is modified to return its accumulated total_memory\nvariable.\n\nAlso, the crashkernel handling is moved in an own function in\narch/ia64/kernel/setup.c to make the code more readable.\n\n[kamalesh@linux.vnet.ibm.com: build fix]\nSigned-off-by: Bernhard Walle \u003cbwalle@suse.de\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Kamalesh Babulal \u003ckamalesh@linux.vnet.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "a3f5c338b9f30f328276739d9589beae19254936",
      "tree": "5c197e9c6565382a548180bdfb57ee5315d9fc60",
      "parents": [
        "be521466feb3bb1cd89de82a2b1d080e9ebd3cb6"
      ],
      "author": {
        "name": "Zou Nan hai",
        "email": "nanhai.zou@intel.com",
        "time": "Tue Mar 20 13:41:57 2007 -0700"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Tue Mar 20 13:41:57 2007 -0700"
      },
      "message": "[IA64] min_low_pfn and max_low_pfn calculation fix\n\nWe have seen bad_pte_print when testing crashdump on an SN machine in\nrecent 2.6.20 kernel.  There are tons of bad pte print (pfn \u003c max_low_pfn)\nreports when the crash kernel boots up, all those reported bad pages\nare inside initmem range; That is because if the crash kernel code and\ndata happens to be at the beginning of the 1st node. build_node_maps in\ndiscontig.c will bypass reserved regions with filter_rsvd_memory. Since\nmin_low_pfn is calculated in build_node_map, so in this case, min_low_pfn\nwill be greater than kernel code and data.\n\nBecause pages inside initmem are freed and reused later, we saw\npfn_valid check fail on those pages.\n\nI think this theoretically happen on a normal kernel. When I check\nmin_low_pfn and max_low_pfn calculation in contig.c and discontig.c.\nI found more issues than this.\n\n1. min_low_pfn and max_low_pfn calculation is inconsistent between\ncontig.c and discontig.c,\nmin_low_pfn is calculated as the first page number of boot memmap in\ncontig.c (Why? Though this may work at the most of the time, I don\u0027t\nthink it is the right logic). It is calculated as the lowest physical\nmemory page number bypass reserved regions in discontig.c.\nmax_low_pfn is calculated include reserved regions in contig.c. It is\ncalculated exclude reserved regions in discontig.c.\n\n2. If kernel code and data region is happen to be at the begin or the\nend of physical memory, when min_low_pfn and max_low_pfn calculation is\nbypassed kernel code and data, pages in initmem will report bad.\n\n3. initrd is also in reserved regions, if it is at the begin or at the\nend of physical memory, kernel will refuse to reuse the memory. Because\nthe virt_addr_valid check in free_initrd_mem.\n\nSo it is better to fix and clean up those issues.\nCalculate min_low_pfn and max_low_pfn in a consistent way.\n\nSigned-off-by:\tZou Nan hai \u003cnanhai.zou@intel.com\u003e\nAcked-by: Jay Lan \u003cjlan@sgi.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "cee87af2a5f75713b98d3e65e43872e547122cd5",
      "tree": "1b5e4778d66cab374e333b4a327d28b0e037ab3f",
      "parents": [
        "41d5e5d73ecef4ef56b7b4cde962929a712689b4"
      ],
      "author": {
        "name": "Magnus Damm",
        "email": "magnus@valinux.co.jp",
        "time": "Tue Mar 06 02:34:26 2007 -0800"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Tue Mar 06 14:50:33 2007 -0800"
      },
      "message": "[IA64] kexec: Use EFI_LOADER_DATA for ELF core header\n\nThe address where the ELF core header is stored is passed to the secondary\nkernel as a kernel command line option.  The memory area for this header is\nalso marked as a separate EFI memory descriptor on ia64.\n\nThe separate EFI memory descriptor is at the moment of the type\nEFI_UNUSABLE_MEMORY.  With such a type the secondary kernel skips over the\nentire memory granule (config option, 16M or 64M) when detecting memory.\nIf we are lucky we will just lose some memory, but if we happen to have\ndata in the same granule (such as an initramfs image), then this data will\nnever get mapped and the kernel bombs out when trying to access it.\n\nSo this is an attempt to fix this by changing the EFI memory descriptor\ntype into EFI_LOADER_DATA.  This type is the same type used for the kernel\ndata and for initramfs.  In the secondary kernel we then handle the ELF\ncore header data the same way as we handle the initramfs image.\n\nThis patch contains the kernel changes to make this happen.  Pretty\nstraightforward, we reserve the area in reserve_memory().  The address for\nthe area comes from the kernel command line and the size comes from the\nspecialized EFI parsing function vmcore_find_descriptor_size().\n\nThe kexec-tools-testing code for this can be found here:\nhttp://lists.osdl.org/pipermail/fastboot/2007-February/005983.html\n\nSigned-off-by: Magnus Damm \u003cmagnus@valinux.co.jp\u003e\nCc: Simon Horman \u003chorms@verge.net.au\u003e\nCc: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "139b830477ccdca21b68c40f9a83ec327e65eb56",
      "tree": "0aab2140315579525dfef89189b9bea5033af2ba",
      "parents": [
        "d1598e05faa11d9f04e0a226122dd57674fb1dab"
      ],
      "author": {
        "name": "Bob Picco",
        "email": "bob.picco@hp.com",
        "time": "Tue Jan 30 02:11:09 2007 -0800"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Mon Feb 05 15:07:47 2007 -0800"
      },
      "message": "[IA64] register memory ranges in a consistent manner\n\nWhile pursuing and unrelated issue with 64Mb granules I noticed a problem\nrelated to inconsistent use of add_active_range.  There doesn\u0027t appear any\nreason to me why FLATMEM versus DISCONTIG_MEM should register memory to\nadd_active_range with different code.  So I\u0027ve changed the code into a\ncommon implementation.\n\nThe other subtle issue fixed by this patch was calling add_active_range in\ncount_node_pages before granule aligning is performed.  We were lucky with\n16MB granules but not so with 64MB granules.  count_node_pages has reserved\nregions filtered out and as a consequence linked kernel text and data\naren\u0027t covered by calls to count_node_pages.  So linked kernel regions\nwasn\u0027t reported to add_active_regions.  This resulted in free_initmem\ncausing numerous bad_page reports.  This won\u0027t occur with this patch\nbecause now all known memory regions are reported by\nregister_active_ranges.\n\nAcked-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nAcked-by: Simon Horman \u003chorms@verge.net.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "a79561134f38de12dce14ed72138f38e55ef53fc",
      "tree": "abe109dbe85e5b0085ba9b9a7eed7cc623d67eec",
      "parents": [
        "620034c84d1d939717bdfbe02c51a3fee43541c3"
      ],
      "author": {
        "name": "Zou Nan hai",
        "email": "nanhai.zou@intel.com",
        "time": "Thu Dec 07 09:51:35 2006 -0800"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Thu Dec 07 09:51:35 2006 -0800"
      },
      "message": "[IA64] IA64 Kexec/kdump\n\nChanges and updates.\n\n1. Remove fake rendz path and related code according to discuss with Khalid Aziz.\n2. fc.i offset fix in relocate_kernel.S.\n3. iospic shutdown code eoi and mask race fix from Fujitsu.\n4. Warm boot hook in machine_kexec to SN SAL code from Jack Steiner.\n5. Send slave to SAL slave loop patch from Jay Lan.\n6. Kdump on non-recoverable MCA event patch from Jay Lan\n7. Use CTL_UNNUMBERED in kdump_on_init sysctl.\n\nSigned-off-by: Zou Nan hai \u003cnanhai.zou@intel.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "05e0caad3b7bd0d0fbeff980bca22f186241a501",
      "tree": "d213789aca5bf91b74bbf5946d428590e3e368b1",
      "parents": [
        "5cb248abf5ab65ab543b2d5fc16c738b28031fc0"
      ],
      "author": {
        "name": "Mel Gorman",
        "email": "mel@csn.ul.ie",
        "time": "Wed Sep 27 01:49:54 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Wed Sep 27 08:26:11 2006 -0700"
      },
      "message": "[PATCH] Have ia64 use add_active_range() and free_area_init_nodes\n\nSize zones and holes in an architecture independent manner for ia64.\n\n[bob.picco@hp.com: fix ia64 FLATMEM+VIRTUAL_MEM_MAP]\nSigned-off-by: Mel Gorman \u003cmel@csn.ul.ie\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nCc: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nCc: Andy Whitcroft \u003capw@shadowen.org\u003e\nCc: Andi Kleen \u003cak@muc.de\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: \"Keith Mannthey\" \u003ckmannth@gmail.com\u003e\nCc: \"Luck, Tony\" \u003ctony.luck@intel.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "e44e41d0c832ebbda7311a1fe43584d844026357",
      "tree": "05a8ac85029b041db8defef35bf808d431bb3528",
      "parents": [
        "921eea1cdf6ce7f0db88e4579474a04b1fb0fe6d"
      ],
      "author": {
        "name": "Bob Picco",
        "email": "bob.picco@hp.com",
        "time": "Wed Jun 28 12:55:43 2006 -0400"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Thu Aug 03 10:13:23 2006 -0700"
      },
      "message": "[IA64] fix show_mem for VIRTUAL_MEM_MAP+FLATMEM\n\ncontig.c (FLATMEM) requires the same optimization as in discontig.c for show_mem\nwhen VIRTUAL_MEM_MAP is in use. Otherwise FLATMEM has softlockup timeouts.\nThis was boot tested for memory configuration: SPARSEMEM,\nDISCONTIG+VIRTUAL_MEM_MAP, FLATMEM, FLATMEM+VIRTUAL_MEM_MAP and\nFLATMEM+VIRTUAL_MEM_MAP with largest memory gap less than LARGE_GAP by\nusing boot parameter \"mem\u003d\".\n\nThis was boot tested and \"echo m \u003e/proc/sysrq-trigger\" output evaluated for\n: FLATMEM, FLATMEM+VIRTUAL_MEM_MAP, DISCONTIGMEM+VIRTUAL_MEM_MAP and\nSPARSEMEM.\n\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "62c4f0a2d5a188f73a94f2cb8ea0dba3e7cf0a7f",
      "tree": "e85ca2d0dd43f90dccf758338764c3caa55f333f",
      "parents": [
        "089f26d5e31b7bf42a9a8fefec08b30cd27f4b0e"
      ],
      "author": {
        "name": "David Woodhouse",
        "email": "dwmw2@infradead.org",
        "time": "Wed Apr 26 12:56:16 2006 +0100"
      },
      "committer": {
        "name": "David Woodhouse",
        "email": "dwmw2@infradead.org",
        "time": "Wed Apr 26 12:56:16 2006 +0100"
      },
      "message": "Don\u0027t include linux/config.h from anywhere else in include/\n\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\n"
    },
    {
      "commit": "c87ff94333642ce3db4fc0857ad1f723cb42c1dd",
      "tree": "9577010b5ca20b692904c4728b305a9cbde5373e",
      "parents": [
        "556902cd2d2cfdc54fe1f1d7f3ac5e2eb276ac09",
        "1e5c936e3cc2ab745f66c0ae296c42f892ffd42a"
      ],
      "author": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Fri Oct 28 14:32:56 2005 -0700"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Fri Oct 28 14:32:56 2005 -0700"
      },
      "message": "Pull sparsemem-v5 into release branch\n"
    },
    {
      "commit": "1be7d9935b9c7fb9bd5964bfaf3ac543381277db",
      "tree": "fe39c6d9e19fb890812318f5f3b318de6c3afc57",
      "parents": [
        "c678796cab4b5288ad578802a54cb1480ae20a08"
      ],
      "author": {
        "name": "Bob Picco",
        "email": "bob.picco@hp.com",
        "time": "Tue Oct 04 15:13:50 2005 -0400"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Tue Oct 04 13:21:13 2005 -0700"
      },
      "message": "[PATCH] V5 ia64 SPARSEMEM - conditional changes for SPARSEMEM\n\nThis patch introduces the conditional changes required for the three\nmemory models.  With [patch 1/4] there are three memory models; FLATMEM,\nDISCONTIG and SPARSEMEM.  Also a new arch include file sparemem.h is\nintroduced for defining SPARSEMEM parameters.\n\nSigned-off-by: Bob Picco \u003cbob.picco@hp.com\u003e\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "d8c97d5f3aa348272df2ccb4e224b1cf9a1eb6d7",
      "tree": "cb71557f43084027559762f58e51f2df5d5e5c46",
      "parents": [
        "4706df3d3c42af802597d82c8b1542c3d52eab23"
      ],
      "author": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Thu Sep 08 12:39:59 2005 -0700"
      },
      "committer": {
        "name": "Tony Luck",
        "email": "tony.luck@intel.com",
        "time": "Thu Sep 08 12:39:59 2005 -0700"
      },
      "message": "[IA64] simplified efi memory map parsing\n\nNew version leaves the original memory map unmodified.\nAlso saves any granule trimmings for use by the uncached\nmemory allocator.\n\nInspired by Khalid Aziz (various traces of his patch still\nremain).  Fixes to uncached_build_memmap() and sn2 testing\nby Martin Hicks.\n\nSigned-off-by: Tony Luck \u003ctony.luck@intel.com\u003e\n"
    },
    {
      "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2",
      "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d",
      "parents": [],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Apr 16 15:20:36 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Apr 16 15:20:36 2005 -0700"
      },
      "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n"
    }
  ]
}