)]}' { "log": [ { "commit": "6672f76a5a1878d42264c1deba8f1ab52b4618d9", "tree": "77396eefed3548183c1f0c3d1dc38f034d8fc429", "parents": [ "73285082745045bcd64333c1fbaa88f8490f2626" ], "author": { "name": "Simon Horman", "email": "horms@verge.net.au", "time": "Tue May 08 00:28:22 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Tue May 08 11:15:07 2007 -0700" }, "message": "kdump/kexec: calculate note size at compile time\n\nCurrently the size of the per-cpu region reserved to save crash notes is\nset by the per-architecture value MAX_NOTE_BYTES. Which in turn is\ncurrently set to 1024 on all supported architectures.\n\nWhile testing ia64 I recently discovered that this value is in fact too\nsmall. The particular setup I was using actually needs 1172 bytes. This\nlead to very tedious failure mode where the tail of one elf note would\noverwrite the head of another if they ended up being alocated sequentially\nby kmalloc, which was often the case.\n\nIt seems to me that a far better approach is to caclculate the size that\nthe area needs to be. This patch does just that.\n\nIf a simpler stop-gap patch for ia64 to be squeezed into 2.6.21(.X) is\nneeded then this should be as easy as making MAX_NOTE_BYTES larger in\narch/asm-ia64/kexec.h. Perhaps 2048 would be a good choice. However, I\nthink that the approach in this patch is a much more robust idea.\n\nAcked-by: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Simon Horman \u003chorms@verge.net.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "b6f35b4966e0ae59cec45e5292b100698d12dc5f", "tree": "5ff3b2832c1c3eebbef699df710068ab106ab2db", "parents": [ "47585d8f5dea10dea49c948f1fb13ef7632409c7" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Wed Jul 05 14:39:43 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Aug 17 16:41:10 2006 +1000" }, "message": "[POWERPC] Make crash.c work on 32-bit and 64-bit\n\nTo compile kexec on 32-bit we need a few more bits and pieces. Rather\nthan add empty definitions, we can make crash.c work on 32-bit, with\nonly a couple of kludges.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c0ce7d0886cf0c2579c604eac41a7e125bc0e96d", "tree": "f31448371b4295e98753f1551178fdddb8d18b0e", "parents": [ "2cd90bc8fba8720ef7f3fdfd1e0c1a5397a18271" ], "author": { "name": "David Wilder", "email": "dwilder@us.ibm.com", "time": "Fri Jun 23 15:29:34 2006 -0700" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Jun 28 15:18:52 2006 +1000" }, "message": "[POWERPC] Add the use of the firmware soft-reset-nmi to kdump.\n\nWith this patch, kdump uses the firmware soft-reset NMI for two purposes:\n1) Initiate the kdump (take a crash dump) by issuing a soft-reset.\n2) Break a CPU out of a deadlock condition that is detected during kdump\nprocessing.\n\nWhen a soft-reset is initiated each CPU will enter\nsystem_reset_exception() and set its corresponding bit in the global\nbit-array cpus_in_sr then call die(). When die() finds the CPU\u0027s bit set\nin cpu_in_sr crash_kexec() is called to initiate a crash dump. The first\nCPU to enter crash_kexec() is called the \"crashing CPU\". All other CPUs\nare \"secondary CPUs\". The secondary CPU\u0027s pass through to\ncrash_kexec_secondary() and sleep. The crashing CPU waits for all CPUs\nto enter via soft-reset then boots the kdump kernel (see\ncrash_soft_reset_check())\n\nWhen the system crashes due to a panic or exception, crash_kexec() is\ncalled by panic() or die(). The crashing CPU sends an IPI to all other\nCPUs to notify them of the pending shutdown. If a CPU is in a deadlock\nor hung state with interrupts disabled, the IPI will not be delivered.\nThe result being, that the kdump kernel is not booted. This problem is\nsolved with the use of a firmware generated soft-reset. After the\ncrashing_cpu has issued the IPI, it waits for 10 sec for all CPUs to\nenter crash_ipi_callback(). A CPU signifies its entry to\ncrash_ipi_callback() by setting its corresponding bit in the\ncpus_in_crash bit array. After 10 sec, if one or more CPUs have not set\ntheir bit in cpus_in_crash we assume that the CPU(s) is deadlocked. The\noperator is then prompted to generate a soft-reset to break the\ndeadlock. Each CPU enters the soft reset handler as described above.\n\nTwo conditions must be handled at this point:\n1) The system crashed because the operator generated a soft-reset. See\n2) The system had crashed before the soft-reset was generated ( in the\ncase of a Panic or oops).\n\nThe first CPU to enter crash_kexec() uses the state of the kexec_lock to\ndetermine this state. If kexec_lock is already held then condition 2 is\ntrue and crash_kexec_secondary() is called, else; this CPU is flagged as\nthe crashing CPU, the kexec_lock is acquired and crash_kexec() proceeds\nas described above.\n\nEach additional CPUs responding to the soft-reset will pass through\ncrash_kexec() to kexec_secondary(). All secondary CPUs call\ncrash_ipi_callback() readying them self\u0027s for the shutdown. When ready\nthey clear their bit in cpus_in_sr. The crashing CPU waits in\nkexec_secondary() until all other CPUs have cleared their bits in\ncpus_in_sr. The kexec kernel boot is then started.\n\nSigned-off-by: Haren Myneni \u003charen@us.ibm.com\u003e\nSigned-off-by: David Wilder \u003cdwilder@us.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "aa98c50dcb5d5b85d2a4c26d54fa1e3c31c11e4b", "tree": "4c0747cace834f0fe75244ccd69c6ff8fc2827cb", "parents": [ "c30a4df3f134f511e8cdf828da429e333753daab" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Fri Jun 23 18:17:32 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Jun 28 11:59:47 2006 +1000" }, "message": "[POWERPC] Make kexec_setup() a regular initcall\n\nThere\u0027s no reason kexec_setup() needs to be called explicitly from\nsetup_system(), it can just be a regular initcall.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "35dd54326e857f1648c7cc1028e8d5e1dbe04992", "tree": "dea05effa25bc751d1b777a0b80f0f46d3ff0b8e", "parents": [ "473104134b35ce1c3ca77b738c561d6c215adc1b" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 18 11:16:11 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 19 15:02:18 2006 +1000" }, "message": "[PATCH] powerpc: Move crashkernel\u003d handling into the kernel.\n\nThis was missing a quilt ref.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2babf5c2ec2f2d5de3e38d20f7df7fd815fd10c9", "tree": "9ecda21067fe36f36fbefae87141150b62c39acd", "parents": [ "846f77b08c8301682ded5ce127c56397327a60d0" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Wed May 17 18:00:46 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 19 15:02:15 2006 +1000" }, "message": "[PATCH] powerpc: Unify mem\u003d handling\n\nWe currently do mem\u003d handling in three seperate places. And as benh pointed out\nI wrote two of them. Now that we parse command line parameters earlier we can\nclean this mess up.\n\nMoving the parsing out of prom_init means the device tree might be allocated\nabove the memory limit. If that happens we\u0027d have to move it. As it happens\nwe already have logic to do that for kdump, so just genericise it.\n\nThis also means we might have reserved regions above the memory limit, if we\ndo the bootmem allocator will blow up, so we have to modify\nlmb_enforce_memory_limit() to truncate the reserves as well.\n\nTested on P5 LPAR, iSeries, F50, 44p. Tested moving device tree on P5 and\n44p and F50.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2ec5e3a867d63d04932e11c6097f63760d9be3fe", "tree": "bb3ea7b72232ec127a104c37cce0f17881ad5d28", "parents": [ "62287fbb54b4af71dc5a4918350f81a4cd467788" ], "author": { "name": "Michael Matz", "email": "matz@suse.de", "time": "Tue Mar 07 21:55:48 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Mar 08 14:15:04 2006 -0800" }, "message": "[PATCH] fix kexec asm\n\nWhile testing kexec and kdump we hit problems where the new kernel would\nfreeze or instantly reboot. The easiest way to trigger it was to kexec a\nkernel compiled for CONFIG_M586 on an athlon cpu. Compiling for CONFIG_MK7\ninstead would work fine.\n\nThe patch fixes a few problems with the kexec inline asm.\n\nSigned-off-by: Chris Mason \u003cmason@suse.com\u003e\nAcked-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "39931e41becd6abeb2014747369d8b6392f5dbac", "tree": "1da728fbefb60f930df6357dcb8f9b1034d6b6d4", "parents": [ "9ac49d22138348198f729f07371ffb11991368e6" ], "author": { "name": "Albert Herranz", "email": "albert_herranz@yahoo.es", "time": "Wed Feb 01 03:05:57 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Feb 01 08:53:19 2006 -0800" }, "message": "[PATCH] powerpc: fix for kexec ppc32\n\n- kexec.h is included from assembly code, thus C code must be properly\n protected.\n\n- (embedded) ppc32 systems use machine_kexec_simple whose declaration\n vanished during a recent powerpc merge change.\n\nSigned-off-by: Albert Herranz \u003calbert_herranz@yahoo.es\u003e\nCc: \u003cfastboot@osdl.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "8385a6a3acfbb4b68150c25cfe9084b6c4f501cf", "tree": "9985155527550b8025603515e4ba96d5c24e6072", "parents": [ "9216ad8cb7849d88b6d8e9f097718de4f25d4121" ], "author": { "name": "Haren Myneni", "email": "haren@us.ibm.com", "time": "Fri Jan 13 19:15:36 2006 -0800" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Sun Jan 15 13:14:42 2006 +1100" }, "message": "[PATCH] powerpc: Fix kdump copy regs and dynamic allocate per-cpu crash notes\n\n- This contains the arch specific changes for the following the\nkdump generic fixes which were already accepted in the upstream.\n . Capturing CPU registers (for the case of \u0027panic\u0027 and invoking\nthe dump using \u0027sysrq-trigger\u0027) from a function (stack frame) which will\nbe not be available during the kdump boot. Hence, might result in\ninvalid stack trace.\n . Dynamically allocating per cpu ELF notes section instead of\nstatically for NR_CPUS.\n\n- Fix the compiler warning in prom_init.c.\n\nSigned-off-by: Haren Myneni \u003charen@us.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "bf2083050d7879390ddedbd64e431cf167854a34", "tree": "036c435ff4dc6b5fad694b9ccd92faacd92c08a2", "parents": [ "f05b690248bc928b85cc19ea85a5c6268bb3acaf" ], "author": { "name": "akpm@osdl.org", "email": "akpm@osdl.org", "time": "Mon Jan 09 20:51:45 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Jan 10 08:01:27 2006 -0800" }, "message": "[PATCH] Kdump: powerpc and s390 build failure fix\n\n\r)\n\nFrom: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\n\ncrash_setup_regs() is an architecture dependent function which is called in\narchitecture independent section. So every architecture supporting kexec\nshould at least provide a dummy definition of crash_setup_regs() even if\ncrash dumping is not implemented yet, to avoid build failures.\n\nSigned-off-by: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "cc57165874e938ef684d71ba7d36e7088b551489", "tree": "e1847f5547a7a426214e9ef0719eab908ee305d7", "parents": [ "82409411571ad89d271dc46f7fa26149fad9efdf" ], "author": { "name": "Vivek Goyal", "email": "vgoyal@in.ibm.com", "time": "Mon Jan 09 20:51:41 2006 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Tue Jan 10 08:01:26 2006 -0800" }, "message": "[PATCH] kdump: dynamic per cpu allocation of memory for saving cpu registers\n\n- In case of system crash, current state of cpu registers is saved in memory\n in elf note format. So far memory for storing elf notes was being allocated\n statically for NR_CPUS.\n\n- This patch introduces dynamic allocation of memory for storing elf notes.\n It uses alloc_percpu() interface. This should lead to better memory usage.\n\n- Introduced based on Andi Kleen\u0027s and Eric W. Biederman\u0027s suggestions.\n\n- This patch also moves memory allocation for elf notes from architecture\n dependent portion to architecture independent portion. Now crash_notes is\n architecture independent. The whole idea is that size of memory to be\n allocated per cpu (MAX_NOTE_BYTES) can be architecture dependent and\n allocation of this memory can be architecture independent.\n\nSigned-off-by: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "88ced0314938814e1772b4d0d7ab20c52e4472b6", "tree": "3e06de882c0bf5706ee7a8994e06eb8c9ed3feca", "parents": [ "e1333803c3a8fb167ba67ffc5540dbb53fa7deb3" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Dec 16 22:43:46 2005 +0100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 15:13:08 2006 +1100" }, "message": "[PATCH] powerpc: sanitize header files for user space includes\n\ninclude/asm-ppc/ had #ifdef __KERNEL__ in all header files that\nare not meant for use by user space, include/asm-powerpc does\nnot have this yet.\n\nThis patch gets us a lot closer there. There are a few cases\nwhere I was not sure, so I left them out. I have verified\nthat no CONFIG_* symbols are used outside of __KERNEL__\nany more and that there are no obvious compile errors when\nincluding any of the headers in user space libraries.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "54c32021eb6feafc32e90104e960b38301521b7b", "tree": "7aff1e11966569f4b8dce57d6f70eb4d1e72bec2", "parents": [ "dcee30361d25ea83499a99f921f9a56b4a1a79e7" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Sun Dec 04 18:39:51 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 14:52:35 2006 +1100" }, "message": "[PATCH] powerpc: Add arch-dependent copy_oldmem_page\n\nSigned-off-by: Haren Myneni \u003charen@us.ibm.com\u003e\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "cc53291521701f9c7c7265bbb3c140563174d8b2", "tree": "9fa71dbb7d853f983c9033fc629edcd118c67858", "parents": [ "758438a7b8da593c9116e95cc7fdff6e9e0b0c40" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Sun Dec 04 18:39:43 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 14:52:28 2006 +1100" }, "message": "[PATCH] powerpc: Add arch dependent basic infrastructure for Kdump.\n\nImplementing the machine_crash_shutdown which will be called by\ncrash_kexec (called in case of a panic, sysrq etc.). Disable the\ninterrupts, shootdown cpus using debugger IPI and collect regs\nfor all CPUs.\n\nelfcorehdr\u003d specifies the location of elf core header stored by\nthe crashed kernel. This command line option will be passed by\nthe kexec-tools to capture kernel.\n\nsavemaxmem\u003d specifies the actual memory size that the first kernel\nhas and this value will be used for dumping in the capture kernel.\nThis command line option will be passed by the kexec-tools to\ncapture kernel.\n\nSigned-off-by: Haren Myneni \u003charen@us.ibm.com\u003e\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "3d1229d6ae92ed1994f4411b8493327ef8f4b76f", "tree": "a50dbd6b52f1d653897147c48c57db28cefb69c7", "parents": [ "afcc2472d80569059b5fe71fcb67e618b9d83fa8" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Mon Nov 14 23:35:00 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 14:48:52 2006 +1100" }, "message": "[PATCH] powerpc: Merge kexec\n\nThis patch merges, to some extent, the PPC32 and PPC64 kexec implementations.\n\nWe adopt the PPC32 approach of having ppc_md callbacks for the kexec functions.\nThe current PPC64 implementation becomes the \"default\" implementation for PPC64\nwhich platforms can select if they need no special treatment.\n\nI\u0027ve added these default callbacks to pseries/maple/cell/powermac, this means\niSeries no longer supports kexec - but it never worked anyway.\n\nI\u0027ve renamed PPC32\u0027s machine_kexec_simple to default_machine_kexec, inline with\nPPC64. Judging by the comments it might be better named machine_kexec_non_of,\nor something, but at the moment it\u0027s the only implementation for PPC32 so it\u0027s\nthe \"default\".\n\nKexec requires machine_shutdown(), which is in machine_kexec.c on PPC32, but we\nalready have in setup-common.c on powerpc. All this does is call\nppc_md.nvram_sync, which only powermac implements, so instead make\nmachine_shutdown a ppc_md member and have it call core99_nvram_sync directly\non powermac.\n\nI\u0027ve also stuck relocate_kernel.S into misc_32.S for powerpc.\n\nBuilt for ARCH\u003dppc, and 32 \u0026 64 bit ARCH\u003dpowerpc, with KEXEC\u003dy/n. Booted on\nP5 LPAR and successfully kexec\u0027ed.\n\nShould apply on top of 493f25ef4087395891c99fcfe2c72e62e293e89f.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "593e537b93193d1696809817533ce5ad510445b1", "tree": "b2045132486c3d39950899e3a9b574188df55ce6", "parents": [ "c5e24354efae9f962e0e369d875d45f47e0bb9aa" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Sat Nov 12 00:06:06 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Nov 14 16:34:06 2005 +1100" }, "message": "[PATCH] powerpc: Export htab start/end via device tree\n\nThe userspace kexec-tools need to know the location of the htab on non-lpar\nmachines, as well as the end of the kernel. Export via the device tree.\n\nNB. This patch has been updated to use \"linux,x\" property names. You may\nneed to update your kexec-tools to match.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e1df870d546f4d033030615aa3d01c0341c1ef1f", "tree": "67621ddb359f503ba4a01f777313740ab479a849", "parents": [ "a1218720321d778134914cf90ef54cf0d1d8477c" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu Nov 03 15:35:45 2005 +1100" }, "committer": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu Nov 03 15:35:45 2005 +1100" }, "message": "powerpc: Merge asm-ppc/kexec.h and asm-ppc64/kexec.h\n\nMerge include/asm-ppc/kexec.h and include/asm-ppc64/kexec.h.\n\nThe only thing that\u0027s really changed is that we now allocate crash_notes\nproperly on PPC32. It\u0027s address is exported via sysfs, so it\u0027s not correct\nfor it to be a pointer.\n\nI\u0027ve also removed some of the \"we don\u0027t use this\" comments, because they\u0027re\nwrong (or perhaps were referring only to arch code).\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\n" } ] }