)]}' { "log": [ { "commit": "b3fcaaa8a6359e9ed623ed4c1d2d48c79eed4648", "tree": "d19456b1c7a928e643db783547459e53839a96d9", "parents": [ "5b0504c0d795d6b0a904ff861c043d7a231f67a4" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Mon Jul 14 19:25:57 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Jul 15 12:24:53 2008 +1000" }, "message": "powerpc: mman.h export fixups\n\nCommit ef3d3246a0d06be622867d21af25f997aeeb105f (\"powerpc/mm: Add Strong\nAccess Ordering support\") in the powerpc/{next,master} tree caused the\nfollowing in a powerpc allmodconfig build:\n\nusr/include/asm/mman.h requires linux/mm.h, which does not exist in exported headers\n\nWe should not use CONFIG_PPC64 in an unprotected (by __KERNEL__)\nsection of an exported include file and linux/mm.h is not exported. So\nprotect the whole section that is CONFIG_PPC64 with __KERNEL__ and put\nthe two introduced includes in there as well.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nAcked-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "930074b6b9c4895d20cdadba5aff97907e28728d", "tree": "3725eca121188f2e9c3b8bb4d4b8ba35e92640c7", "parents": [ "3fd44736db9a5bf33e4a216b9cd43c9cfd57c459", "2bf3016f89344d4cd8b2c96bbec2b642a2bde413" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Jul 15 11:54:57 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Jul 15 11:54:57 2008 +1000" }, "message": "Merge commit \u0027jwb/jwb-next\u0027\n" }, { "commit": "11c2d8174ed3dc4f1971564732689b4a39129702", "tree": "ac00daa548ea8ac24ae7a5c8062312e335ab9858", "parents": [ "cde274c0c789404df8ece3f9e7d6506caf0127e2", "bce7f793daec3e65ec5c5705d2457b81fe7b5725" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Jul 14 14:29:49 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Jul 14 14:29:49 2008 +1000" }, "message": "Merge commit \u0027origin/HEAD\u0027 into test-merge\n\nManual fixup of include/asm-powerpc/pgtable-ppc64.h" }, { "commit": "1bc54c03117b90716e0dedd7abb2a20405de65df", "tree": "8e82fd610abaff36f1e20b5aaaf7bdeaee883aac", "parents": [ "beae4c03c0fe69cf7d57518aa0572ad21730b8be" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Jul 08 15:54:40 2008 +1000" }, "committer": { "name": "Josh Boyer", "email": "jwboyer@linux.vnet.ibm.com", "time": "Wed Jul 09 13:36:17 2008 -0400" }, "message": "powerpc: rework 4xx PTE access and TLB miss\n\nThis is some preliminary work to improve TLB management on SW loaded\nTLB powerpc platforms. This introduce support for non-atomic PTE\noperations in pgtable-ppc32.h and removes write back to the PTE from\nthe TLB miss handlers. In addition, the DSI interrupt code no longer\ntries to fixup write permission, this is left to generic code, and\n_PAGE_HWWRITE is gone.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Josh Boyer \u003cjwboyer@linux.vnet.ibm.com\u003e\n" }, { "commit": "ef3d3246a0d06be622867d21af25f997aeeb105f", "tree": "9f0ae1913e0e637ec3aa104cc5e81557e5661b3c", "parents": [ "379070491e1e744a59e69e5bcf3765012d15ecb4" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@linux.vnet.ibm.com", "time": "Tue Jul 08 00:28:54 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:45 2008 +1000" }, "message": "powerpc/mm: Add Strong Access Ordering support\n\nAllow an application to enable Strong Access Ordering on specific pages of\nmemory on Power 7 hardware. Currently, power has a weaker memory model than\nx86. Implementing a stronger memory model allows an emulator to more\nefficiently translate x86 code into power code, resulting in faster code\nexecution.\n\nOn Power 7 hardware, storing 0b1110 in the WIMG bits of the hpte enables\nstrong access ordering mode for the memory page. This patchset allows a\nuser to specify which pages are thus enabled by passing a new protection\nbit through mmap() and mprotect(). I have defined PROT_SAO to be 0x10.\n\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "379070491e1e744a59e69e5bcf3765012d15ecb4", "tree": "32524cdac93dce578438d14529f2a5c7caca5092", "parents": [ "aba46c5027cb59d98052231b36efcbbde9c77a1d" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@linux.vnet.ibm.com", "time": "Tue Jul 08 00:28:53 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:45 2008 +1000" }, "message": "powerpc/mm: Add SAO Feature bit to the cputable\n\nAdd the CPU feature bit for the new Strong Access Ordering\nfacility of Power7\n\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Joel Schopp \u003cjschopp@austin.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "aba46c5027cb59d98052231b36efcbbde9c77a1d", "tree": "b69dd362c74db562789d14136294c5fcbb7632e7", "parents": [ "b845f313d78e4e259ec449909e3bbadf77b53a6d" ], "author": { "name": "Dave Kleikamp", "email": "shaggy@linux.vnet.ibm.com", "time": "Tue Jul 08 00:28:52 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:45 2008 +1000" }, "message": "powerpc/mm: Define flags for Strong Access Ordering\n\nThis patch defines:\n\n- PROT_SAO, which is passed into mmap() and mprotect() in the prot field\n- VM_SAO in vma-\u003evm_flags, and\n- _PAGE_SAO, the combination of WIMG bits in the pte that enables strong\naccess ordering for the page.\n\nSigned-off-by: Dave Kleikamp \u003cshaggy@linux.vnet.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "e5093ff05d36c64e8f36a9ddb26358256dc133ea", "tree": "06e27ffb2a70ddb293c56aaf6499e618fce761d1", "parents": [ "058c78f4ba89df7b2de82ac271452f09e2b8fa05" ], "author": { "name": "Srinivasa Ds", "email": "srinivasa@in.ibm.com", "time": "Tue Jul 08 00:22:27 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:44 2008 +1000" }, "message": "powerpc: Implement task_pt_regs() accessor\n\nThe task_pt_regs() macro allows access to the pt_regs of a given task.\n\nThis macro is not currently defined for the powerpc architecture, but\nwe need it for some upcoming utrace additions.\n\nSigned-off-by: Srinivasa DS \u003csrinivasa@in.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "3a4c6f0b1540811110a59112b4c83f55c229728c", "tree": "6d2cc36af1100103ceb649a6fbdf8b2a9b526947", "parents": [ "7e5f8105030038de94b44a74cd7b64dd000830fc" ], "author": { "name": "Mark Nelson", "email": "markn@au1.ibm.com", "time": "Sat Jul 05 05:05:45 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:44 2008 +1000" }, "message": "powerpc: move device_to_mask() to dma-mapping.h\n\nMove device_to_mask() to dma-mapping.h because we need to use it from\noutside dma_64.c in a later patch.\n\nSigned-off-by: Mark Nelson \u003cmarkn@au1.ibm.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "3affedc4e1ce837033b6c5e9289d2ce2f5a62d31", "tree": "83296af2c727e5b5f64b88b629dcf37f8e60e3f0", "parents": [ "c8692362db3db3a6f644e05a477161d967430aac" ], "author": { "name": "Mark Nelson", "email": "markn@au1.ibm.com", "time": "Sat Jul 05 05:05:42 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:43 2008 +1000" }, "message": "powerpc/dma: implement new dma_*map*_attrs() interfaces\n\nUpdate powerpc to use the new dma_*map*_attrs() interfaces. In doing so\nupdate struct dma_mapping_ops to accept a struct dma_attrs and propagate\nthese changes through to all users of the code (generic IOMMU and the\n64bit DMA code, and the iseries and ps3 platform code).\n\nThe old dma_*map_*() interfaces are reimplemented as calls to the\ncorresponding new interfaces.\n\nSigned-off-by: Mark Nelson \u003cmarkn@au1.ibm.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Geoff Levand \u003cgeoffrey.levand@am.sony.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "c8692362db3db3a6f644e05a477161d967430aac", "tree": "b240cfe972dda9768fc330458da5667a7c15864f", "parents": [ "fabb657005edbbcb0d13ee49a40f1f4b042a1d19" ], "author": { "name": "Mark Nelson", "email": "markn@au1.ibm.com", "time": "Sat Jul 05 05:05:41 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:43 2008 +1000" }, "message": "powerpc/dma: Add struct iommu_table argument to iommu_map_sg()\n\nMake iommu_map_sg take a struct iommu_table. It did so before commit\n740c3ce66700640a6e6136ff679b067e92125794 (iommu sg merging: ppc: make\niommu respect the segment size limits).\n\nThis stops the function looking in the archdata.dma_data for the iommu\ntable because in the future it will be called with a device that has\nno table there.\n\nThis also has the nice side effect of making iommu_map_sg() match the\nother map functions.\n\nSigned-off-by: Mark Nelson \u003cmarkn@au1.ibm.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "fabb657005edbbcb0d13ee49a40f1f4b042a1d19", "tree": "5f402e272ff2933c9e3fe9d6e0eeb10c166d7c2c", "parents": [ "88b90c96b787ecb5c72384b6873468f814cce650" ], "author": { "name": "Maxim Shchetynin", "email": "maxim@de.ibm.com", "time": "Sat Jul 05 05:05:39 2008 +1000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jul 09 16:30:42 2008 +1000" }, "message": "powerpc/spufs: add atomic busy_spus counter to struct cbe_spu_info\n\nAs nr_active counter includes also spus waiting for syscalls to return\nwe need a seperate counter that only counts spus that are currently running\non spu side. This counter shall be used by a cpufreq governor that targets\na frequency dependent from the number of running spus.\n\nSigned-off-by: Christian Krafft \u003ckrafft@de.ibm.com\u003e\nAcked-by: Jeremy Kerr \u003cjk@ozlabs.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "86df86424939d316b1f6cfac1b6204f0c7dee317", "tree": "447924f4ecf0ae4bd2ba1af513ff84af66e681b1", "parents": [ "7b51ba38d9baca207152d1c97ec793cfb673a6cd" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Tue Jul 08 15:58:16 2008 +1000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Jul 08 09:27:58 2008 -0700" }, "message": "Correct hash flushing from huge_ptep_set_wrprotect()\n\nAs Andy Whitcroft recently pointed out, the current powerpc version of\nhuge_ptep_set_wrprotect() has a bug. It just calls ptep_set_wrprotect()\nwhich in turn calls pte_update() then hpte_need_flush() with the \u0027huge\u0027\nargument set to 0. This will cause hpte_need_flush() to flush the wrong\nhash entries (of any). Andy\u0027s fix for this is already in the powerpc\ntree as commit 016b33c4958681c24056abed8ec95844a0da80a3.\n\nI have confirmed this is a real bug, not masked by some other\nsynchronization, with a new testcase for libhugetlbfs. A process write\na (MAP_PRIVATE) hugepage mapping, fork(), then alter the mapping and\nhave the child incorrectly see the second write.\n\nTherefore, this should be fixed for 2.6.26, and for the stable tree.\nHere is a suitable patch for 2.6.26, which I think will also be suitable\nfor the stable tree (neither of the headers in question has been changed\nmuch recently).\n\nIt is cut down slighlty from Andy\u0027s original version, in that it does\nnot include a 32-bit version of huge_ptep_set_wrprotect(). Currently,\nhugepages are not supported on any 32-bit powerpc platform. When they\nare, a suitable 32-bit version can be added - the only 32-bit hardware\nwhich supports hugepages does not use the conventional hashtable MMU and\nso will have different needs anyway.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nSigned-off-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "3c3f67eafad12d4ccabe491c6c8a50bf6e75b89a", "tree": "be4081be92d8a7c165281242384cb79cb8c2d06d", "parents": [ "92ecd1790b10e12015070e33a0f70493d51aca50" ], "author": { "name": "Nathan Fontenot", "email": "nfont@austin.ibm.com", "time": "Thu Jul 03 13:22:39 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jul 03 16:58:16 2008 +1000" }, "message": "powerpc/pseries: Update the device tree correctly for drconf memory add/remove\n\nThis updates the device tree manipulation routines so that memory\nadd/remove of lmbs represented under the\nibm,dynamic-reconfiguration-memory node of the device tree invokes the\nhotplug notifier chain.\n\nThis change is needed because of the change in the way memory is\nrepresented under the ibm,dynamic-reconfiguration-memory node. All lmbs\nare described in the ibm,dynamic-memory property instead of having a\nseparate node for each lmb as in previous device tree layouts. This\nrequires the update_node() routine to check for updates to the\nibm,dynamic-memory property and invoke the hotplug notifier chain.\n\nThis also updates the pseries hotplug notifier to be able to gather information\nfor lmbs represented under the ibm,dynamic-reconfiguration-memory node and\nhave the lmbs added/removed.\n\nSigned-off-by: Nathan Fontenot \u003cnfont@austin.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "138fc1ee06e58f12fc2b755e435ce15bb36a0471", "tree": "c553c9186d0d3d70b444680c0f500fc58ced61a4", "parents": [ "6a274c08f2f4dfac7167bbd849621f3a2b55d424" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jul 02 22:51:37 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jul 03 16:58:13 2008 +1000" }, "message": "powerpc: Remove old dump_task_* functions\n\nSince Roland\u0027s ptrace cleanup starting with commit\nf65255e8d51ecbc6c9eef20d39e0377d19b658ca (\"[POWERPC] Use user_regset\naccessors for FP regs\"), the dump_task_* functions are no longer being\nused.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2d1b2027626d5151fff8ef7c06ca8e7876a1a510", "tree": "6ecc861f4f45a5d26309bf12683aa0372358ef7c", "parents": [ "5888da18765ca9af7f10015263d8bc8e3057f128" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Jul 02 01:16:40 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jul 03 16:58:10 2008 +1000" }, "message": "powerpc: Fixup lwsync at runtime\n\nTo allow for a single kernel image on e500 v1/v2/mc we need to fixup lwsync\nat runtime. On e500v1/v2 lwsync causes an illop so we need to patch up\nthe code. We default to \u0027sync\u0027 since that is always safe and if the cpu\nis capable we will replace \u0027sync\u0027 with \u0027lwsync\u0027.\n\nWe introduce CPU_FTR_LWSYNC as a way to determine at runtime if this is\nneeded. This flag could be moved elsewhere since we dont really use it\nfor the normal CPU_FTR purpose.\n\nFinally we only store the relative offset in the fixup section to keep it\nas small as possible rather than using a full fixup_entry.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e17a2565bf61204cb925c3f77c3c7f6a09eb2fbe", "tree": "f82df5037a68fa9ded94d81115dde005057ad923", "parents": [ "db7f37de2c8346c33cf9279fa9d8e8316e8d821c" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Tue Jul 01 17:00:39 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jul 03 16:58:08 2008 +1000" }, "message": "powerpc: Fix compile warning in init_thread\n\nCurrently we get this warning:\narch/powerpc/kernel/init_task.c:33: warning: missing braces around initializer\narch/powerpc/kernel/init_task.c:33: warning: (near initialization for \u0027init_task.thread.fpr[0]\u0027)\n\nThis fixes it.\n\nNoticed by Stephen Rothwell.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "db7f37de2c8346c33cf9279fa9d8e8316e8d821c", "tree": "0a49c9aeb1c5c071d5766fca1fbfc6c5f2c3d0ff", "parents": [ "f3e909c2750eb20536bacacc867dc9047b70546a" ], "author": { "name": "Tony Breeds", "email": "tony@bakeyournoodle.com", "time": "Tue Jul 01 11:30:06 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jul 03 16:58:07 2008 +1000" }, "message": "powerpc: Fix building of arch/powerpc/mm/mem.o when MEMORY_HOTPLUG\u003dy and SPARSEMEM\u003dn\n\nCurrently the kernel fails to build with the above config options with:\n CC arch/powerpc/mm/mem.o\narch/powerpc/mm/mem.c: In function \u0027arch_add_memory\u0027:\narch/powerpc/mm/mem.c:130: error: implicit declaration of function \u0027create_section_mapping\u0027\n\nThis explicitly includes asm/sparsemem.h in arch/powerpc/mm/mem.c and\nmoves the guards in include/asm-powerpc/sparsemem.h to protect the\nSPARSEMEM specific portions only.\n\nSigned-off-by: Tony Breeds \u003ctony@bakeyournoodle.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "f3e909c2750eb20536bacacc867dc9047b70546a", "tree": "97bf54aafb93313d54aaeaa05eb99d1a9ab48c3c", "parents": [ "436db693c4f4d02d375780a1f0b2ebdca2451cd6" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Tue Jul 01 14:01:39 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 14:47:09 2008 +1000" }, "message": "powerpc: Update for VSX core file and ptrace\n\nThis correctly hooks the VSX dump into Roland McGrath core file\ninfrastructure. It adds the VSX dump information as an additional elf\nnote in the core file (after talking more to the tool chain/gdb guys).\nThis also ensures the formats are consistent between signals, ptrace\nand core files.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "a91a03ee31a5c29a934708b7cf37bb8da516d016", "tree": "bb2c25456759a865a64ae5326846301e7962802b", "parents": [ "89b5810f6ed4b2d42415e5ec656ab6b148cd2bde" ], "author": { "name": "Eric B Munson", "email": "ebmunson@us.ibm.com", "time": "Tue Jul 01 02:12:13 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 14:47:02 2008 +1000" }, "message": "powerpc: Keep 3 high personality bytes across exec\n\nCurrently when a 32 bit process is exec\u0027d on a powerpc 64 bit host the\nvalue in the top three bytes of the personality is clobbered. patch\nadds a check in the SET_PERSONALITY macro that will carry all the\nvalues in the top three bytes across the exec.\n\nThese three bytes currently carry flags to disable address randomisation,\nlimit the address space, force zeroing of an mmapped page, etc. Should an\napplication set any of these bits they will be maintained and honoured on\nhomogeneous environment but discarded and ignored on a heterogeneous\nenvironment. So if an application requires all mmapped pages to be initialised\nto zero and a wrapper is used to setup the personality and exec the target,\nthese flags will remain set on an all 32 or all 64 bit envrionment, but they\nwill be lost in the exec on a mixed 32/64 bit environment. Losing these bits\nmeans that the same application would behave differently in different\nenvironments. Tested on a POWER5+ machine with 64bit kernel and a mixed\n64/32 bit user space.\n\nSigned-off-by: Eric B Munson \u003cebmunson@us.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "89b5810f6ed4b2d42415e5ec656ab6b148cd2bde", "tree": "e3116100d5be38f2ebfaa078cd53033d1f612059", "parents": [ "fcbc5a976b1cafe2e866871c86d239d57503bfd5" ], "author": { "name": "Bart Van Assche", "email": "bart.vanassche@gmail.com", "time": "Sat Jun 28 16:51:35 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:29:00 2008 +1000" }, "message": "powerpc: Make sure that include/asm-powerpc/spinlock.h does not trigger compilation warnings\n\nWhen compiling kernel modules for ppc that include \u003clinux/spinlock.h\u003e,\ngcc prints a warning message every time it encounters a function\ndeclaration where the inline keyword appears after the return type.\nThis makes sure that the order of the inline keyword and the return\ntype is as gcc expects it. Additionally, the __inline__ keyword is\nreplaced by inline, as checkpatch expects.\n\nSigned-off-by: Bart Van Assche \u003cbart.vanassche@gmail.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "016b33c4958681c24056abed8ec95844a0da80a3", "tree": "67ba81cb6bc980119fab4eee15f09488e5777012", "parents": [ "03d70617b8a789c3721afaafde06fcbba7c7ebf1" ], "author": { "name": "Andy Whitcroft", "email": "apw@shadowen.org", "time": "Thu Jun 26 19:55:58 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:56 2008 +1000" }, "message": "powerpc: Add 64 bit version of huge_ptep_set_wrprotect\n\nThe implementation of huge_ptep_set_wrprotect() directly calls\nptep_set_wrprotect() to mark a hugepte write protected. However this\ncall is not appropriate on ppc64 kernels as this is a small page only\nimplementation. This can lead to the hash not being flushed correctly\nwhen a mapping is being converted to COW, allowing processes to continue\nusing the original copy.\n\nCurrently huge_ptep_set_wrprotect() unconditionally calls\nptep_set_wrprotect(). This is fine on ppc32 kernels as this call is\ngeneric. On 64 bit this is implemented as:\n\n\tpte_update(mm, addr, ptep, _PAGE_RW, 0);\n\nOn ppc64 this last parameter is the page size and is passed directly on\nto hpte_need_flush():\n\n\thpte_need_flush(mm, addr, ptep, old, huge);\n\nAnd this directly affects the page size we pass to flush_hash_page():\n\n\tflush_hash_page(vaddr, rpte, psize, ssize, 0);\n\nAs this changes the way the hash is calculated we will flush the wrong\npages, potentially leaving live hashes to the original page.\n\nMove the definition of huge_ptep_set_wrprotect() to the 32/64 bit specific\nheaders.\n\nSigned-off-by: Andy Whitcroft \u003capw@shadowen.org\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "ce48b2100785e5ca629fb3aa8e3b50aca808f692", "tree": "63532ff7cc68b18ca4902bd10e03fcbaaf01cade", "parents": [ "72ffff5b1792b0fa4d40a8e2f3276fff999820ec" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jun 25 14:07:18 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:50 2008 +1000" }, "message": "powerpc: Add VSX context save/restore, ptrace and signal support\n\nThis patch extends the floating point save and restore code to use the\nVSX load/stores when VSX is available. This will make FP context\nsave/restore marginally slower on FP only code, when VSX is available,\nas it has to load/store 128bits rather than just 64bits.\n\nMixing FP, VMX and VSX code will get constant architected state.\n\nThe signals interface is extended to enable access to VSR 0-31\ndoubleword 1 after discussions with tool chain maintainers. Backward\ncompatibility is maintained.\n\nThe ptrace interface is also extended to allow access to VSR 0-31 full\nregisters.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "72ffff5b1792b0fa4d40a8e2f3276fff999820ec", "tree": "50dbf11b043350b98fcccd9f838b0f31f49997a5", "parents": [ "b962ce9d26fd6677e6720949642420ceb029a102" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jun 25 14:07:18 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:48 2008 +1000" }, "message": "powerpc: Add VSX assembler code macros\n\nThis adds the macros for the VSX load/store instruction as most\nbinutils are not going to support this for a while.\n\nAlso add VSX register save/restore macros and vsr[0-63] register definitions.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "b962ce9d26fd6677e6720949642420ceb029a102", "tree": "dd05752d5f48204b3b87f8330be043b9c39745a8", "parents": [ "c6e6771b87d4e339d27f1383c8a808ae9b4ee5b8" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jun 25 14:07:18 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:47 2008 +1000" }, "message": "powerpc: Add VSX CPU feature\n\nAdd a VSX CPU feature. Also add code to detect if VSX is available\nfrom the device tree.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Joel Schopp \u003cjschopp@austin.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c6e6771b87d4e339d27f1383c8a808ae9b4ee5b8", "tree": "1900b7350ec685c3a31f2233fd88a57e34725b5c", "parents": [ "6f3d8e6947ec98e358514fc0f7b2e37fe88a21bb" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jun 25 14:07:18 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:46 2008 +1000" }, "message": "powerpc: Introduce VSX thread_struct and CONFIG_VSX\n\nThe layout of the new VSR registers and how they overlap on top of the\nlegacy FPR and VR registers is:\n\n VSR doubleword 0 VSR doubleword 1\n ----------------------------------------------------------------\n VSR[0] | FPR[0] | |\n ----------------------------------------------------------------\n VSR[1] | FPR[1] | |\n ----------------------------------------------------------------\n | ... | |\n | ... | |\n ----------------------------------------------------------------\n VSR[30] | FPR[30] | |\n ----------------------------------------------------------------\n VSR[31] | FPR[31] | |\n ----------------------------------------------------------------\n VSR[32] | VR[0] |\n ----------------------------------------------------------------\n VSR[33] | VR[1] |\n ----------------------------------------------------------------\n | ... |\n | ... |\n ----------------------------------------------------------------\n VSR[62] | VR[30] |\n ----------------------------------------------------------------\n VSR[63] | VR[31] |\n ----------------------------------------------------------------\n\nVSX has 64 128bit registers. The first 32 regs overlap with the FP\nregisters and hence extend them with and additional 64 bits. The\nsecond 32 regs overlap with the VMX registers.\n\nThis commit introduces the thread_struct changes required to reflect\nthis register layout. Ptrace and signals code is updated so that the\nfloating point registers are correctly accessed from the thread_struct\nwhen CONFIG_VSX is enabled.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "9c75a31c3525a127f70b919856e32be3d8b03755", "tree": "3beee76d3dd4a55af868b5edfbef80c911819fed", "parents": [ "9e7511861c4f8d35852a3721c5bcd92661cb4c9f" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Thu Jun 26 17:07:48 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:43 2008 +1000" }, "message": "powerpc: Add macros to access floating point registers in thread_struct.\n\nWe are going to change where the floating point registers are stored\nin the thread_struct, so in preparation add some macros to access the\nfloating point registers. Update all code to use these new macros.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "fac23fe4be23259a8eaa9bad822f5b14dd07d15c", "tree": "23979268a5b893d676bba2bdb514307d5417ea56", "parents": [ "c1137c37a9370f510dc022e6e95fff036ccb90be" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:54 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:28 2008 +1000" }, "message": "powerpc: Introduce infrastructure for feature sections with alternatives\n\nThe current feature section logic only supports nop\u0027ing out code, this means\nif you want to choose at runtime between instruction sequences, one or both\ncases will have to execute the nop\u0027ed out contents of the other section, eg:\n\nBEGIN_FTR_SECTION\n\tor\t1,1,1\nEND_FTR_SECTION_IFSET(FOO)\nBEGIN_FTR_SECTION\n\tor\t2,2,2\nEND_FTR_SECTION_IFCLR(FOO)\n\nand the resulting code will be either,\n\n\tor\t1,1,1\n\tnop\n\nor,\n\tnop\n\tor\t2,2,2\n\nFor small code segments this is fine, but for larger code blocks and in\nperformance criticial code segments, it would be nice to avoid the nops.\nThis commit starts to implement logic to allow the following:\n\nBEGIN_FTR_SECTION\n\tor\t1,1,1\nFTR_SECTION_ELSE\n\tor\t2,2,2\nALT_FTR_SECTION_END_IFSET(FOO)\n\nand the resulting code will be:\n\n\tor\t1,1,1\nor,\n\tor\t2,2,2\n\nWe achieve this by extending the existing FTR macros. The current feature\nsection semantic just becomes a special case, ie. if the else case is empty\nwe nop out the default case.\n\nThe key limitation is that the size of the else case must be less than or\nequal to the size of the default case. If the else case is smaller the\nremainder of the section is nop\u0027ed.\n\nWe let the linker put the else case code in with the rest of the text,\nso that relative branches from the else case are more likley to link,\nthis has the disadvantage that we can\u0027t free the unused else cases.\n\nThis commit introduces the required macro and linker script changes, but\ndoes not enable the patching of the alternative sections.\n\nWe also need to update two hand-made section entries in reg.h and timex.h\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c1137c37a9370f510dc022e6e95fff036ccb90be", "tree": "96fd89dd4daa06178ade06558f6ab489791be667", "parents": [ "c5157e587b33b1185cf73207fc53760cf1351430" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:48 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:26 2008 +1000" }, "message": "powerpc: Consolidate feature fixup macros for 64/32 bit\n\nCurrently we have three versions of MAKE_FTR_SECTION_ENTRY(), the macro that\ngenerates a feature section entry. There is 64bit version, a 32bit version\nand version for 32bit code built with a 64bit kernel.\n\nRather than triplicating (?) the MAKE_FTR_SECTION_ENTRY() logic, we can\nmove the 64bit/32bit differences into separate macros, and then only have\none version of MAKE_FTR_SECTION_ENTRY().\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c5157e587b33b1185cf73207fc53760cf1351430", "tree": "996a4e76bae11001a8e97141c617c159dde37ec8", "parents": [ "51c52e86694f19e84600a40f6156889feafd8ae9" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:39 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:25 2008 +1000" }, "message": "powerpc: Consolidate CPU and firmware feature fixup macros\n\nThe CPU and firmware feature fixup macros are currently spread across\nthree files, firmware.h, cputable.h and asm-compat.h. Consolidate them\ninto their own file, feature-fixups.h\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "b7bcda631e87eb3466d0baa9885650ba7d7ed89d", "tree": "3a04957c5f1b22e2d4405ccdaa980e6e513ef388", "parents": [ "ae0dc73625f9b0e636ccd130e394c9b654a062fb" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:35 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:23 2008 +1000" }, "message": "powerpc: Add PPC_NOP_INSTR, a hash define for the preferred nop instruction\n\nA bunch of code has hard-coded the value for a \"nop\" instruction, it\nwould be nice to have a #define for it.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "411781a290b0d0a31fd73826b3ee110f1e3cc3b6", "tree": "4a5eb482d99916c17f19a013b52cf808161f32ea", "parents": [ "07630a37beefe8e4401c602f04e3e5bcbba50b31" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:29 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:21 2008 +1000" }, "message": "powerpc: Add new code patching routines\n\nThis commit adds some new routines for patching code, which will be used\nin a following commit.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "07630a37beefe8e4401c602f04e3e5bcbba50b31", "tree": "7e7c7b76abe634eeebb1b161ebfbd18638af2149", "parents": [ "053a858efa46c9ab86363b271374ec02ad2af753" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:28 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:20 2008 +1000" }, "message": "powerpc: Add ppc_function_entry() which gets the entry point for a function\n\nBecause function pointers point to different things on 32-bit vs 64-bit,\nadd a macro that deals with dereferencing the OPD on 64-bit. The soon to\nbe merged ftrace wants this, as well as other code I am working on.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e7a57273c6407bb6903fbaddec8c2119bf318617", "tree": "4d5b787c286e1d0ae64f757ef1f68fedd9743a3c", "parents": [ "aaddd3eacaeaef3503035750b3f21ac2bfe97cbf" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:22 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:19 2008 +1000" }, "message": "powerpc: Allow create_branch() to return errors\n\nCurrently create_branch() creates a branch instruction for you, and\npatches it into the call site. In some circumstances it would be nice\nto be able to create the instruction and patch it later, and also some\ncode might want to check for errors in the branch creation before\ndoing the patching. A future commit will change create_branch() to\ncheck for errors.\n\nFor callers that don\u0027t care, replace create_branch() with\npatch_branch(), which just creates the branch and patches it directly.\n\nWhile we\u0027re touching all the callers, change to using unsigned int *,\nas this seems to match usage better. That allows (and requires) us to\nremove the volatile in the definition of vector in powermac/smp.c and\nmpc86xx_smp.c, that\u0027s correct because now that we\u0027re passing vector as\nan unsigned int * the compiler knows that it\u0027s value might change\nacross the patch_branch() call.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nAcked-by: Jon Loeliger \u003cjdl@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "aaddd3eacaeaef3503035750b3f21ac2bfe97cbf", "tree": "2bac0bedc2185edec916679d218e1c8b89276453", "parents": [ "73f38fe1b563a9d23ffacbda7b51decf41b0c49c" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Tue Jun 24 11:32:21 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:18 2008 +1000" }, "message": "powerpc: Move code patching code into arch/powerpc/lib/code-patching.c\n\nWe currently have a few routines for patching code in asm/system.h, because\nthey didn\u0027t fit anywhere else. I\u0027d like to clean them up a little and add\nsome more, so first move them into a dedicated C file - they don\u0027t need to\nbe inlined.\n\nWhile we\u0027re moving the code, drop create_function_call(), it\u0027s intended\ncaller never got merged and will be replaced in future with something\ndifferent.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "178f8d78fdeaace8b22a759a059b74f39caf23a4", "tree": "b2c7561f04f86944aa76206495fe02a3f5aa8e6d", "parents": [ "d1cdcf2241723d6596d56abb8777b5ef79b4a878" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Jun 24 03:48:28 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:15 2008 +1000" }, "message": "powerpc: asm/elf.h: Reduce userspace header\n\nThis makes asm/elf.h export less non-userspace stuff to userspace.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "d1cdcf2241723d6596d56abb8777b5ef79b4a878", "tree": "05c9fe12af0027c517b08f4c6c205615d1979554", "parents": [ "9b4a8dd2e9f8af206eb39e3d99c442b0d6158953" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Tue Jun 24 03:48:21 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:28:09 2008 +1000" }, "message": "powerpc: Don\u0027t export asm/asm-compat.h to userspace\n\nasm/asm-compat.h doesn\u0027t seem to be intended for userspace usage.\n\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "3a8247cc2c856930f34eafce33f6a039227ee175", "tree": "aa8599cdf09893f1150a2bc137878d8b8a661780", "parents": [ "e952e6c4d6635b36c212c056a9427bd93460178c" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Jun 18 15:29:12 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Jul 01 11:27:57 2008 +1000" }, "message": "powerpc: Only demote individual slices rather than whole process\n\nAt present, if we have a kernel with a 64kB page size, and some\nprocess maps something that has to be mapped with 4kB pages (such as a\ncache-inhibited mapping on POWER5+, or the eHCA infiniband queue-pair\npages), we change the process to use 4kB pages everywhere. This hurts\nthe performance of HPC programs that access eHCA from userspace.\n\nWith this patch, the kernel will only demote the slice(s) containing\nthe eHCA or cache-inhibited mappings, leaving the remaining slices\nable to use 64kB hardware pages.\n\nThis also changes the slice_get_unmapped_area code so that it is\nwilling to place a 64k-page mapping into (or across) a 4k-page slice\nif there is no better alternative, i.e. if the program specified\nMAP_FIXED or if there is not sufficient space available in slices that\nare either empty or already have 64k-page mappings in them.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "e952e6c4d6635b36c212c056a9427bd93460178c", "tree": "1791007d187ffea2e66a0150977dc5f3cc3b8c53", "parents": [ "e2d755051129c61a9f67bf752d0b7cc82b115338" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Jun 18 10:47:26 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:31:11 2008 +1000" }, "message": "powerpc: Add cputable entry for POWER7\n\nAdd a cputable entry for the POWER7 processor.\n\nAlso tell firmware that we know about POWER7.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Joel Schopp \u003cjschopp@austin.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e2d755051129c61a9f67bf752d0b7cc82b115338", "tree": "d3b150352393244018e26469383a9f859517b757", "parents": [ "316a405841cc03723d401b098d188aa4e3daa69b" ], "author": { "name": "Scott Wood", "email": "scottwood@freescale.com", "time": "Wed Jun 18 02:59:59 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:31:07 2008 +1000" }, "message": "powerpc: Fix copy-and-paste error in clrsetbits_le16\n\nThis was pointed out by Detlev Zundel when this code was being\nadded to U-boot.\n\nSigned-off-by: Scott Wood \u003cscottwood@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "316a405841cc03723d401b098d188aa4e3daa69b", "tree": "476be4fad16c690d1650890757aacc3095e85129", "parents": [ "7c5c4325d2d911fe54db3bc14149bfa558ae0acb" ], "author": { "name": "Becky Bruce", "email": "becky.bruce@freescale.com", "time": "Sat Jun 14 09:41:43 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:31:05 2008 +1000" }, "message": "powerpc: Get rid of bitfields in ppc_bat struct\n\nWhile working on the 36-bit physical support, I noticed that there\nwas exactly one line of code that actually referenced the bitfields.\nSo I got rid of them and redefined ppc_bat as a struct of 2 u32\u0027s:\nbatu and batl. I also got rid of the previous union that held the\nbitfield structs and a word representation of the batu/l values.\n\nThis seems like a nicer solution than adding in a bunch of\nnew bitfields to support extended bat addressing that would never\nget used, and just leaving the struct as-is would have been\nincomplete in the face of large physical addressing.\n\nSigned-off-by: Becky Bruce \u003cbecky.bruce@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7c5c4325d2d911fe54db3bc14149bfa558ae0acb", "tree": "97c02e9d03bbbb1fff89e9e44eef1e53762e823e", "parents": [ "bf2737f74ffd36f43c6ed11cd2c0826e1d59ad48" ], "author": { "name": "Becky Bruce", "email": "becky.bruce@freescale.com", "time": "Sat Jun 14 09:41:42 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:31:03 2008 +1000" }, "message": "powerpc: Change BAT code to use phys_addr_t\n\nCurrently, the physical address is an unsigned long, but it should\nbe phys_addr_t in set_bat, [v/p]_mapped_by_bat. Also, create a\nmacro that can convert a large physical address into the correct\nformat for programming the BAT registers.\n\nSigned-off-by: Becky Bruce \u003cbecky.bruce@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "bf2737f74ffd36f43c6ed11cd2c0826e1d59ad48", "tree": "e42e618207c22141cb17c77dfba99473d865d5de", "parents": [ "36c35be332faa8ba1aaa8a237e72b9460117129d" ], "author": { "name": "Becky Bruce", "email": "becky.bruce@freescale.com", "time": "Sat Jun 14 09:12:44 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:31:01 2008 +1000" }, "message": "powerpc: Silly spelling fix in pgtable-ppc32\n\nSigned-off-by: Becky Bruce \u003cbecky.bruce@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "aee10c6145199f872bc51f811b115a4b686b2426", "tree": "138ff9e8eb0929992d0c97cc4da67e7a939b5071", "parents": [ "41743a4e34f0777f51c1cf0675b91508ba143050" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Thu Jun 12 19:14:34 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:30:55 2008 +1000" }, "message": "powerpc: Provide dummy crash_shutdown_register\n\nWhen kexec is disabled, the crash_shutdown_{un,}register\nfunctions are not available in the kernel.\nThis provides dummy inline functions for those so that\nthe callers don\u0027t have to worry about it.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "41743a4e34f0777f51c1cf0675b91508ba143050", "tree": "9e63c8c8d70169e6c8be699167234db33b102a1f", "parents": [ "ff1f4ee94c3c4480b9cee95da2d19668262636c6" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Jun 11 15:37:10 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:30:53 2008 +1000" }, "message": "powerpc: Free a PTE bit on ppc64 with 64K pages\n\nThis frees a PTE bit when using 64K pages on ppc64. This is done\nby getting rid of the separate _PAGE_HASHPTE bit. Instead, we just test\nif any of the 16 sub-page bits is set. For non-combo pages (ie. real\n64K pages), we set SUB0 and the location encoding in that field.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "ff1f4ee94c3c4480b9cee95da2d19668262636c6", "tree": "004fd1b38a03ddd32a4e06134d67faa2d3d5019c", "parents": [ "9a24729d8aeef967eac7af71c6a69edc83d06558" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Wed Jun 11 08:31:00 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:30:51 2008 +1000" }, "message": "powerpc: Implement OF PCI address accessors stubs for CONFIG_PCI\u003dn\n\nTo avoid \"#ifdef CONFIG_PCI\" in the drivers we should provide stubs in\nplace of OF PCI address accessors.\n\nWithout these stubs build breaks for drivers not strictly requiring PCI,\nfor example CONFIG_FB_OF\u003dy without CONFIG_PCI:\n\n LD .tmp_vmlinux1\ndrivers/built-in.o: In function `offb_map_reg\u0027:\noffb.c:(.text+0x6e7c): undefined reference to `of_get_pci_address\u0027\n\nOF PCI IRQ accessors require pci_dev argument, so drivers using PCI\nIRQs should depend on CONFIG_PCI anyway.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "74f0609526afddd88bef40b651da24f3167b10b2", "tree": "be6d04cbc81d37957f484b4085113765c9903e4c", "parents": [ "e9a4b6a3f6592862a67837e80aad3f50468857a6" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Thu May 22 00:12:31 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 22:30:25 2008 +1000" }, "message": "powerpc: Optimise smp_wmb on 64-bit processors\n\nFor 64-bit processors, lwsync is the recommended method of store/store\nordering on caching enabled memory. For those subarchs which have\nlwsync, use it rather than eieio for smp_wmb.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e9a4b6a3f6592862a67837e80aad3f50468857a6", "tree": "087420cf91ac922c7e99757c4490a605ec7b10f2", "parents": [ "441dbb500b4344a8e3125e5aabab4f9dbf72514f", "1702b52092e9a6d05398d3f9581ddc050ef00d06" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 10:16:50 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 10:16:50 2008 +1000" }, "message": "Merge branch \u0027linux-2.6\u0027\n" }, { "commit": "441dbb500b4344a8e3125e5aabab4f9dbf72514f", "tree": "3bb8fb1a88db3f5d03062a4887312f25c2e8041d", "parents": [ "dee805532a30183c4462305b4614e58bd83a7bc1", "5ce4b59653b2c2053cd9a011918ac1e4747f24cc" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 09:57:05 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 30 09:57:05 2008 +1000" }, "message": "Merge branch \u0027next\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/jwboyer/powerpc-4xx\n" }, { "commit": "b660398101cd0622325480a67ac88bb4d33d553a", "tree": "3e527c721c71a948727fcd1ad52ac240ca61f43d", "parents": [ "543cf4cb3fe6f6cae3651ba918b9c56200b257d0" ], "author": { "name": "David Woodhouse", "email": "david.woodhouse@intel.com", "time": "Fri Jun 27 14:39:42 2008 +0100" }, "committer": { "name": "Sam Ravnborg", "email": "sam@ravnborg.org", "time": "Fri Jun 27 23:13:54 2008 +0200" }, "message": "kbuild: fix a.out.h export to userspace with O\u003d build.\n\nWe need to check for existence of the a.out.h header in the source tree,\nnot the object tree, if we want it to get the right answer with O\u003d.\n\nSigned-off-by: David Woodhouse \u003cdavid.woodhouse@intel.com\u003e\nSigned-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\n" }, { "commit": "d14b3dd6190af7ce4f88be68f8df828af6d44584", "tree": "510889655bcdeca927bffc5a22421e6b5d942b38", "parents": [ "2308c954f55eae9f165a5a59bbdb0a0c586afe5a" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Thu Jun 12 03:42:14 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 26 01:49:09 2008 -0500" }, "message": "powerpc/QE: use arch_initcall to probe QUICC Engine GPIOs\n\nIt was discussed that global arch_initcall() is preferred way to probe\nQE GPIOs, so let\u0027s use it.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "d3c511ac1d72661fcf5f9c6f0ab60b889199a542", "tree": "ebcf7de8219c73461e043d6b70d5b91d869677d2", "parents": [ "0b2a2e5b7747f1f63bd86ca22b5c6097da5b2137" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 12 08:12:49 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 26 01:49:06 2008 -0500" }, "message": "powerpc/cpm: Remove !CONFIG_PPC_CPM_NEW_BINDING code\n\nNow that arch/ppc is gone we always define CONFIG_PPC_CPM_NEW_BINDING so\nwe can remove all the code associated with !CONFIG_PPC_CPM_NEW_BINDING.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "aba11fc50c925bbd6fb25d54eae2f86277a3b107", "tree": "52c85a46371ca4ef05042600507ecd5bdcfa2842", "parents": [ "fc4033b2f8b1482022bff3d05505a1b1631bb6de" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 19 09:40:31 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 26 01:49:03 2008 -0500" }, "message": "powerpc/e500mc: flush L2 on NAP for e500mc\n\nIf we have an L2CSR register (e500mc) we need to flush the L2 before going\nto nap. We use the HW flush mechanism provided in that register.\n\nThe code reuses the CPU_FTR_604_PERF_MON bit as it is no longer used by\nany code in the kernel. Additionally we didn\u0027t reuse the exist L2CR\nfeature bit as this is intended for the 7xxx L2CR register and L2CSR\nis part of the new Freescale \"Book-E\" registers.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "fc4033b2f8b1482022bff3d05505a1b1631bb6de", "tree": "c84b275968011911d8c96acebe89aa2dd92323bf", "parents": [ "3dfa8773674e16f95f70a0e631e80c69390d04d7" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Jun 18 16:26:52 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu Jun 26 01:48:56 2008 -0500" }, "message": "powerpc/85xx: add DOZE/NAP support for e500 core\n\nThe e500 core enter DOZE/NAP power-saving modes when the core go to\ncpu_idle routine.\n\nThe power management default running mode is DOZE, If the user\n\necho 1 \u003e /proc/sys/kernel/powersave-nap\n\nthe system will change to NAP running mode.\n\nSigned-off-by: Dave Liu \u003cdaveliu@freescale.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "3dfa8773674e16f95f70a0e631e80c69390d04d7", "tree": "95e8989bbc8373e61f69ca2ac4c98ffd3c709bd9", "parents": [ "bccaea8fe287454d70f5b2546910561e9f884053" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Mon Jun 16 09:41:32 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Jun 18 16:17:56 2008 -0500" }, "message": "powerpc/booke: Add support for new e500mc core\n\nThe new e500mc core from Freescale is based on the e500v2 but with the\nfollowing changes:\n\n* Supports only the Enhanced Debug Architecture (DSRR0/1, etc)\n* Floating Point\n* No SPE\n* Supports lwsync\n* Doorbell Exceptions\n* Hypervisor\n* Cache line size is now 64-bytes (e500v1/v2 have a 32-byte cache line)\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "5ce4b59653b2c2053cd9a011918ac1e4747f24cc", "tree": "9dd8ba43094c80c31459caa43d2014668617c2e2", "parents": [ "d2146cb274e494dd14e50f5b4cdeb7a5e5af5421" ], "author": { "name": "Josh Boyer", "email": "jwboyer@linux.vnet.ibm.com", "time": "Tue Jun 17 19:01:38 2008 -0400" }, "committer": { "name": "Josh Boyer", "email": "jwboyer@linux.vnet.ibm.com", "time": "Tue Jun 17 19:01:38 2008 -0400" }, "message": "powerpc/4xx: Workaround for PPC440EPx/GRx PCI_28 Errata\n\nThe 440EPx/GRx chips don\u0027t support PCI MRM commands. Drivers determine this\nby looking for a zero value in the PCI cache line size register. However,\nsome drivers write to this register upon initialization. This can cause\nMRMs to be used on these chips, which may cause deadlocks on PLB4.\n\nThe workaround implemented here introduces a new indirect_type flag, called\nPPC_INDIRECT_TYPE_BROKEN_MRM. This is set in the pci_controller structure in\nthe pci fixup function for 4xx PCI bridges by determining if the bridge is\ncompatible with 440EPx/GRx. The flag is checked in the indirect_write_config\nfunction, and forces any writes to the PCI_CACHE_LINE_SIZE register to be\nzero, which will disable MRMs for these chips.\n\nA similar workaround has been tested by AMCC on various PCI cards, such as\nthe Silicon Image ATA card and Intel E1000 GIGE card. Hangs were seen with\nthe Silicon Image card, and MRMs were seen on the bus with a PCI analyzer.\nWith the workaround in place, the card functioned properly and only Memory\nReads were seen on the bus with the analyzer.\n\nAcked-by: Stefan Roese \u003csr@denx.de\u003e\nSigned-off-by: Josh Boyer \u003cjwboyer@linux.vnet.ibm.com\u003e\n" }, { "commit": "bccaea8fe287454d70f5b2546910561e9f884053", "tree": "965aaf6cb1e0cbd8c932ef471c1122f40e623285", "parents": [ "fec6a82282cc38397ba1c4a7b5b99d70eea06532" ], "author": { "name": "Jerone Young", "email": "jyoung5@us.ibm.com", "time": "Fri Jun 06 14:09:05 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Mon Jun 16 09:56:18 2008 -0500" }, "message": "powerpc/booke: Fix definitions for dbcr[1-2] and dbsr registers\n\nThis takes values from the PowerPC ISA BookIII-E specifications that are\nfor DBCR0. Many of these values are different from those currently\nspecified, which are for the ppc405. Also added some bookE definitions\nfor DBCR1 \u0026 DBCR2.\n\n[ galak@kernel.crashing.org: Added aliases to 40x DBCR0 to match Book-E,\n Added enhanced debug DBCR0/DBSR _CIRPT and _CRET defines and DBSR\n IRPT and RET. ]\n\nSigned-off-by: Jerone Young \u003cjyoung5@us.ibm.com\u003e\nAcked-by: Josh Boyer \u003cjwboyer@linux.vnet.ibm.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "305c73687157d677bee6f2a5dbee438d844a028c", "tree": "be91e711074389f3238ca44446b3de6878840521", "parents": [ "8e01520c06c65a1a376059199fc24d4f3d606991" ], "author": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Wed Jun 11 03:47:45 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 16 15:00:47 2008 +1000" }, "message": "[POWERPC] Build fix for drivers/macintosh/mediabay.c\n\nThis fixes the following build error with CONFIG_BLK_DEV_IDE_PMAC\u003dn:\n\n\u003c-- snip --\u003e\n\n...\n CC drivers/macintosh/mediabay.o\n/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/macintosh/mediabay.c: In function \u0027check_media_bay\u0027:\n/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/macintosh/mediabay.c:428: error: \u0027struct media_bay_info\u0027 has no member named \u0027cd_index\u0027\nmake[3]: *** [drivers/macintosh/mediabay.o] Error 1\n\n\u003c-- snip --\u003e\n\nReported-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "598056d5af8fef1dbe8f96f5c2b641a528184e5a", "tree": "4839b7a63cade4751527f94307a05687710311ad", "parents": [ "a9653cf540d407fb75deb3db65a1be6c81d53ee0" ], "author": { "name": "Nick Piggin", "email": "npiggin@suse.de", "time": "Thu May 22 00:10:56 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 16 15:00:20 2008 +1000" }, "message": "[POWERPC] Fix rmb to order cacheable vs. noncacheable\n\nlwsync is explicitly defined not to have any effect on the ordering of\naccesses to device memory, so it cannot be used for rmb(). sync appears\nto be the only barrier which fits the bill.\n\nSigned-off-by: Nick Piggin \u003cnpiggin@suse.de\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "a9653cf540d407fb75deb3db65a1be6c81d53ee0", "tree": "075fb79746f1d74443c9a9062e73c26a6266b05c", "parents": [ "e80ac32767d0f781ac195c472d500a7451d3729a", "066519068ad2fbe98c7f45552b1f592903a9c8c8" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 16 14:53:25 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 16 14:53:25 2008 +1000" }, "message": "Merge branch \u0027linux-2.6\u0027 into merge\n" }, { "commit": "1f64643aa5f5a17f1723f7ea0f17b7a3a8f632b3", "tree": "fbefce858013d3b53e48b5d3aa1b640025743241", "parents": [ "d84050f48ebba73994b93ccf61cea2364dac8d75" ], "author": { "name": "Luke Browning", "email": "lukebrowning@us.ibm.com", "time": "Thu Jun 05 17:30:25 2008 +0800" }, "committer": { "name": "Jeremy Kerr", "email": "jk@ozlabs.org", "time": "Mon Jun 16 14:35:00 2008 +1000" }, "message": "powerpc/spufs: remove class_0_dsisr from spu exception handling\n\nAccording to the CBEA, the SPU dsisr is not updated for class 0\nexceptions.\n\nspu_stopped() is testing the dsisr that was passed to it from the class\n0 exception handler, so we return a false positive here.\n\nThis patch cleans up the interrupt handler and erroneous tests in\nspu_stopped. It also removes the fields from the csa since it is not\nneeded to process class 0 events.\n\nSigned-off-by: Luke Browning \u003clukebrowning@us.ibm.com\u003e\nSigned-off-by: Jeremy Kerr \u003cjk@ozlabs.org\u003e\n" }, { "commit": "5848f16947026601b9cb4df694bb9f44dfefc354", "tree": "55fe5e351827d7acd9b634a9af85c01b3ea72bd7", "parents": [ "5093bb965a163fe288c3e5db0275165f86c895c2" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Wed Jun 11 16:32:48 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Jun 11 13:46:24 2008 -0500" }, "message": "powerpc/QE: qe_reset should be __init\n\nThis patch fixes following section mismatch:\n\nWARNING: arch/powerpc/sysdev/built-in.o(.text+0x11d8): Section mismatch in\nreference from the function qe_reset() to the function\n.init.text:cpm_muram_init()\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "5093bb965a163fe288c3e5db0275165f86c895c2", "tree": "77af9fbf28f45e277443be5abffb85feacc0097c", "parents": [ "b13e930906b313d787f4dd07fe78b74a3a8c22c4" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Fri May 23 20:39:06 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Tue Jun 10 11:11:21 2008 -0500" }, "message": "powerpc/QE: switch to the cpm_muram implementation\n\nThis is very trivial patch. We\u0027re transitioning to the cpm_muram_*\ncalls. That\u0027s it.\n\nLess trivial changes:\n- BD_SC_* defines were defined in the cpm.h and qe.h, so to avoid redefines\n we remove BD_SC from the qe.h and use cpm.h along with cpm_muram_*\n prototypes;\n- qe_muram_dump was unused and thus removed;\n- added some code to the cpm_common.c to support legacy QE bindings\n (data-only node name).\n- For convenience, define qe_* calls to cpm_*. So drivers need not to be\n changed.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "32def337aafee0bc65eb58d5b1b3617525eb7fb7", "tree": "22173ef1c86e1dd6de2eadee2600dbad1e029249", "parents": [ "9572653ee01a2134ae3a1d7aa29ce9d026a6afe9" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Mon May 19 21:47:05 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Tue Jun 10 11:11:10 2008 -0500" }, "message": "powerpc/QE: implement support for the GPIO LIB API\n\nThis is needed to access QE GPIOs via Linux GPIO API.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nAcked-By: Timur Tabi \u003ctimur@freescale.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "9572653ee01a2134ae3a1d7aa29ce9d026a6afe9", "tree": "c136d6fbcd38eff44473abc6f4d6262181b63c3e", "parents": [ "5e41486c408eb4206aee09303631427f57771691" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Fri May 23 20:38:58 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Tue Jun 10 10:39:18 2008 -0500" }, "message": "powerpc/QE: prepare QE PIO code for GPIO LIB support\n\n- split and export __par_io_config_pin() out of par_io_config_pin(), so we\n could use the prefixed version with GPIO LIB API;\n- rename struct port_regs to qe_pio_regs, and place it into qe.h;\n- rename #define NUM_OF_PINS to QE_PIO_PINS, and place it into qe.h.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nAcked-By: Timur Tabi \u003ctimur@freescale.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "5e41486c408eb4206aee09303631427f57771691", "tree": "04198c2fdffdb70816ba141aaa2f2b1d6d8e72aa", "parents": [ "83ff9dcf375c418ca3b98eb950711525ca1269e2" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Fri May 23 20:38:56 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Tue Jun 10 10:39:13 2008 -0500" }, "message": "powerpc/QE: add support for QE USB clocks routing\n\nThis patch adds a function to the qe_lib to setup QE USB clocks routing.\nTo setup clocks safely, cmxgcr register needs locking, so I just reused\nucc_lock since it was used only to protect cmxgcr.\n\nThe idea behind placing clocks routing functions into the qe_lib is that\nlater we\u0027ll hopefully switch to the generic Linux Clock API, thus, for\nexample, FHCI driver may be used for QE and CPM chips without nasty #ifdefs.\n\nThis patch also fixes QE_USB_RESTART_TX command definition in the qe.h.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nAcked-By: Timur Tabi \u003ctimur@freescale.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "83ff9dcf375c418ca3b98eb950711525ca1269e2", "tree": "b6dae1fee27667653dbb96d47c3042108ce9d4c0", "parents": [ "5399be7f4679251e8c4c6637fde240f7ac8efdb9" ], "author": { "name": "Anton Vorontsov", "email": "avorontsov@ru.mvista.com", "time": "Fri May 23 20:38:54 2008 +0400" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Tue Jun 10 10:38:50 2008 -0500" }, "message": "powerpc/sysdev: implement FSL GTM support\n\nGTM stands for General-purpose Timers Module and able to generate\ntimer{1,2,3,4} interrupts. These timers are used by the drivers that\nneed time precise interrupts (like for USB transactions scheduling for\nthe Freescale USB Host controller as found in some QE and CPM chips),\nor these timers could be used as wakeup events from the CPU deep-sleep\nmode.\n\nThings unimplemented:\n1. Cascaded (32 bit) timers (1-2, 3-4).\n This is straightforward to implement when needed, two timers should\n be marked as \"requested\" and configured as appropriate.\n2. Super-cascaded (64 bit) timers (1-2-3-4).\n This is also straightforward to implement when needed, all timers\n should be marked as \"requested\" and configured as appropriate.\n\nSigned-off-by: Anton Vorontsov \u003cavorontsov@ru.mvista.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\n" }, { "commit": "0f3d6bcd391b058c619fc30e8022e8a29fbf4bef", "tree": "ab5f0b8fb95afc01c7b1fd4c5254045ad30bbe4c", "parents": [ "19fc65b5251dfd90312ae0142cc8650cd273e6a6" ], "author": { "name": "Trent Piepho", "email": "tpiepho@freescale.com", "time": "Wed May 28 09:48:32 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 09 13:54:36 2008 +1000" }, "message": "powerpc: Improve (in|out)_[bl]eXX() asm code\n\nSince commit 4cb3cee03d558fd457cb58f56c80a2a09a66110c the code generated\nfor the in_beXX() and out_beXX() mmio functions has been sub-optimal.\n\nThe out_leXX() family of functions are created with the macro\nDEF_MMIO_OUT_LE() while the out_beXX() family are created with\nDEF_MMIO_OUT_BE(). In what was perhaps a bit too much macro use, both of\nthese macros are in turn created via the macro DEF_MMIO_OUT().\n\nFor the LE versions, eventually they boil down to an asm that will look\nsomething like this:\nasm(\"sync; stwbrx %1,0,%2\" : \"\u003dm\" (*addr) : \"r\" (val), \"r\" (addr));\n\nThe issue is that the \"stwbrx\" instruction only comes in an indexed, or\n\u0027x\u0027, version, in which the address is represented by the sum of two\nregisters (the \"0,%2\"). Unfortunately, gcc doesn\u0027t have a constraint for\nan indexed memory reference. The \"m\" constraint allows both indexed and\noffset, i.e. register plus constant, memory references and there is no\n\"stwbr\" version for offset references. \"m\" also allows updating addresses\nand there is no \u0027u\u0027 version of \"stwbrx\" like there is with \"stwux\".\n\nThe unused first operand to the asm is just to tell gcc that *addr is an\noutput of the asm. The address used is passed in a single register via the\nthird asm operand, and the index register is just hard coded as 0. This\nmeans gcc is forced to put the address in a single register and can\u0027t use\nindex addressing, e.g. if one has the data in register 9, a base address in\nregister 3 and an index in register 4, gcc must emit code like \"add 11,4,3;\nstwbrx 9,0,11\" instead of just \"stwbrx 9,4,3\". This costs an extra add\ninstruction and another register.\n\nFor gcc 4.0 and older, there doesn\u0027t appear to be anything that can be\ndone. But for 4.1 and newer, there is a \u0027Z\u0027 constraint. It does not allow\n\"updating\" addresses, but does allow both indexed and offset addresses.\nHowever, the only allowed constant offset is 0. We can then use the\nundocumented \u0027y\u0027 operand modifier, which causes gcc to convert \"0(reg)\"\ninto the equivilient \"0,reg\" format that can be used with stwbrx.\n\nThis brings us the to problem with the BE version. In this case, the \"stw\"\ninstruction does have both indexed and non-indexed versions. The final asm\nends up looking like this:\nasm(\"sync; stw%U0%X0 %1,%0\" : \"\u003dm\" (*addr) : \"r\" (val), \"r\" (addr));\n\nThe undocumented codes \"%U0\" and \"%0X\" will generate a \u0027u\u0027 if the memory\nreference should be an auto-updating one, and an \u0027x\u0027 if the memory\nreference is indexed, respectively. The third operand is unused, it\u0027s just\nthere because asm the code is reused from the LE version. However, gcc\ndoes not know this, and generates unnecessary code to stick addr in a\nregister! To use the example from the LE version, gcc will generate \"add\n11,4,3; stwx 9,4,3\". It is able to use the indexed address \"4,3\" for the\n\"stwx\", but still thinks it needs to put 4+3 into register 11, which will\nnever be used.\n\nThis also ends up happening a lot for the offset addressing mode, where\ncommon code like this: out_be32(\u0026device_registers-\u003esome_register, data);\nuses an instruction like \"stw 9, 42(3)\", where register 3 has the pointer\ndevice_registers and 42 is the offset of some_register in that structure.\ngcc will be forced to generate the unnecessary instruction \"addi 11, 3, 42\"\nto put the address into a single (unused) register.\n\nThe in_* versions end up having these exact same problems as well.\n\nSigned-off-by: Trent Piepho \u003ctpiepho@freescale.com\u003e\nCC: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCC: Andreas Schwab \u003cschwab@suse.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7c4f10b9003dc8423df07574ba197bbbe3bc382b", "tree": "be7dd0f555d4ae07a320dca922d1b8e96f517901", "parents": [ "cec0dd94cf2628d5ba184e725f02be061e7bb014" ], "author": { "name": "Rune Torgersen", "email": "runet@innovsys.com", "time": "Sat May 24 01:59:15 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 09 13:46:40 2008 +1000" }, "message": "powerpc: Check that TASK_SIZE does not overlap KERNEL_START\n\nMake sure CONFIG_TASK_SIZE does not overlap CONFIG_KERNEL_START\nThis could happen when overriding settings to get 1GB lowmem, and would lead\nto userland mysteriousely hanging.\n\nThis setting is only used by PPC32.\n\nSigned-off-by: Rune Torgersen \u003crunet@innovsys.com\u003e\nAcked-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "8a3e1c670e503ddd6f6c373b307f38b783ee3a50", "tree": "03094e8425b750d2693a271ebc89b49312e5476a", "parents": [ "e026892c85571e12f11abffde5a90bcc704d663e", "60d5019be8acef268f4676d229c490186d338fbc" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 09 12:19:41 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 09 12:19:41 2008 +1000" }, "message": "Merge branch \u0027merge\u0027\n\nConflicts:\n\n\tarch/powerpc/sysdev/fsl_soc.c\n" }, { "commit": "ce263d70e509287ee761f9bba519342f57b121ca", "tree": "923836b0b73e57f85914b2aa7335dc14a06f19c5", "parents": [ "e52b2af541bcb299212a63cfa3e3231618a415be" ], "author": { "name": "Hollis Blanchard", "email": "hollisb@us.ibm.com", "time": "Wed May 21 18:22:51 2008 -0500" }, "committer": { "name": "Avi Kivity", "email": "avi@qumranet.com", "time": "Fri Jun 06 21:22:09 2008 +0300" }, "message": "KVM: ppc: Remove duplicate function\n\nThis was left behind from some code movement.\n\nSigned-off-by: Hollis Blanchard \u003chollisb@us.ibm.com\u003e\nSigned-off-by: Avi Kivity \u003cavi@qumranet.com\u003e\n" }, { "commit": "bcf0b0880710409420a4e3b15dbf4b9a63542c0b", "tree": "ca1c813648ea075b772112237fde3ccfe3ddb776", "parents": [ "c054065bc10a7ee2bcf78b5bc95f4b4d9bdc923a" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Apr 30 03:49:55 2008 -0500" }, "committer": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Mon Jun 02 14:54:42 2008 -0500" }, "message": "[POWERPC] Move to runtime allocated exception stacks\n\nFor the additonal exception levels (critical, debug, machine check) on\n40x/book-e we were using \"static\" allocations of the stack in the\nassociated head.S.\n\nMove to a runtime allocation to make the code a bit easier to read as\nwe mimic how we handle IRQ stacks. Its also a bit easier to setup the\nstack with a \"dummy\" thread_info in C code.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nAcked-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "cfab3bdf8292edec19492c89520b1ad11279a648", "tree": "dbc3ed8bf7173b102bf844ed8121941b0b2303d8", "parents": [ "732bee4c859012edf05f3e09b53b68fc332a369d" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed May 28 10:18:17 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Sat May 31 17:08:28 2008 +1000" }, "message": "[POWERPC] Add \"memory\" clobber to MMIO accessors\n\nGcc might re-order MMIO accessors vs. surrounding consistent\nmemory accesses, which is a \"bad thing\", and could break drivers.\nThis fixes it by adding a \"memory\" clobber to the MMIO accessors,\nwhich should prevent gcc from doing that reordering.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "b786af117b360843349cf66165c4efa0217ca2a7", "tree": "713c515b49003fe2b593703820169ffc326ed4bb", "parents": [ "acf464817d5e7be9fb67aec4027dbee0ac9be17a" ], "author": { "name": "Stephen Neuendorffer", "email": "stephen.neuendorffer@xilinx.com", "time": "Wed May 07 04:29:17 2008 +1000" }, "committer": { "name": "Josh Boyer", "email": "jwboyer@linux.vnet.ibm.com", "time": "Thu May 29 07:06:56 2008 -0500" }, "message": "[POWERPC] Refactor DCR code\n\nPreviously, DCR support was configured at compile time to either use\nMMIO or native dcr instructions. Although this works for most\nplatforms, it fails on FPGA platforms:\n\n1) Systems may include more than one DCR bus.\n2) Systems may be native DCR capable and still use memory mapped DCR interface.\n\nThis patch provides runtime support based on the device trees for the\ncase where CONFIG_PPC_DCR_MMIO and CONFIG_PPC_DCR_NATIVE are both\nselected. Previously, this was a poorly defined configuration, which\nhappened to provide NATIVE support. The runtime selection is made\nbased on the dcr-controller having a \u0027dcr-access-method\u0027 attribute\nin the device tree. If only one of the above options is selected,\nthen the code uses #defines to select only the used code in order to\navoid introducing overhead in existing usage.\n\nSigned-off-by: Stephen Neuendorffer \u003cstephen.neuendorffer@xilinx.com\u003e\nSigned-off-by: Josh Boyer \u003cjwboyer@linux.vnet.ibm.com\u003e\n" }, { "commit": "acf464817d5e7be9fb67aec4027dbee0ac9be17a", "tree": "39fd061c0a91a11d3ae4ff19b7fe179325e39bbf", "parents": [ "80d267f9aee6f1b5df602b5a19fb7b4923f17db2", "8962cadbe7cbc4ed0fff94f56ebab505a10afd2e" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 23 16:53:23 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 23 16:53:23 2008 +1000" }, "message": "Merge branch \u0027merge\u0027 into powerpc-next\n" }, { "commit": "06a901c5621f85e07e00ac4816c7ca95620ee74a", "tree": "0f6a0d19d889e4209a41889ed868bc8af2a581eb", "parents": [ "d1e8d50d5a4bc7e40b119bf513804188ce2933d0" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Wed May 21 16:24:31 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 23 16:15:37 2008 +1000" }, "message": "[POWERPC] mpic: Fix use of uninitialized variable\n\nCompiling ppc64_defconfig with gcc 4.3 gives thes warnings:\n\narch/powerpc/sysdev/mpic.c: In function \u0027mpic_irq_get_priority\u0027:\narch/powerpc/sysdev/mpic.c:1351: warning: \u0027is_ipi\u0027 may be used uninitialized in this function\narch/powerpc/sysdev/mpic.c: In function \u0027mpic_irq_set_priority\u0027:\narch/powerpc/sysdev/mpic.c:1328: warning: \u0027is_ipi\u0027 may be used uninitialized in this function\n\nIt turns out that in the cases where is_ipi is uninitialized, another\nvariable (mpic) will be NULL and it is dereferenced. Protect against\nthis by returning if mpic is NULL in mpic_irq_set_priority, and removing\nmpic_irq_get_priority completely as it has no in tree callers.\n\nThis has the nice side effect of making the warning go away.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nAcked-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "475ca391b490a683d66bf19999a8a7a24913f139", "tree": "e7818dcd25169b9d9c9ed35723b3cafd2be455fd", "parents": [ "2ca7633dc73405ee775a2e9b7961b47a38bc882d" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Thu May 22 06:59:23 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 23 15:27:26 2008 +1000" }, "message": "[POWERPC] mpic: Deal with bogus NIRQ in Feature Reporting Register\n\nSome chips (like the SoCs from Freescale) report the wrong value in NIRQ\nand this causes issues if its doesn\u0027t match or exceed the value of\nirq_count.\n\nAdd a flag that board code can set to just use irq_count instead of\nFRR[NIRQ]. Eventually we\u0027ll add a device tree property with the number\nof sources.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2ca7633dc73405ee775a2e9b7961b47a38bc882d", "tree": "2a4e5340e7c3020421b1dafeb4fd1955798b0a42", "parents": [ "21451155d8858773ee764e9218de2ca0f4d6fc38" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Sun May 11 10:40:47 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 23 15:27:23 2008 +1000" }, "message": "[POWERPC] Tweak VDSO linker script to avoid upsetting old binutils\n\nThis works around bugs in older binutils\u0027 objcopy.\nThe placement of these sections does not really matter,\nbut it confused the buggy old BFD libraries.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "09e67ca2c523544e6b38aa570a5f62a0cf20b87b", "tree": "134d2ba98b567586dd34ffa7d7f8cfd8fe0de149", "parents": [ "7bc228b1ef71f395aeb89bdf81bf95556b08b374" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Fri May 16 11:57:45 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:22:28 2008 +1000" }, "message": "[POWERPC] Move of_device_get_modalias to drivers/of\n\nCommit 140b932f8cb6cced10b96860651a198b1b89cbb9 (\"Create modalias file\nin sysfs for of_platform bus\") needs this to avoid breaking the sparc\nbuilds.\n\nJust move the code and add whitespace around some binary operators.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nAcked-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7bc228b1ef71f395aeb89bdf81bf95556b08b374", "tree": "466d3bb7d8162c228361895884c2c4bb92de5388", "parents": [ "0b2cca804ee40495bc92449c3e22f3c3a3f2977a" ], "author": { "name": "Robert P. J. Day", "email": "rpjday@crashcourse.ca", "time": "Thu May 15 05:16:44 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:22:28 2008 +1000" }, "message": "[POWERPC] Get most of ioctl.h content from \u003casm-generic/ioctl.h\u003e\n\nNow that \u003casm-generic/ioctl.h\u003e allows overriding of the most commonly\nchanged macros, take advantage of that.\n\nSigned-off-by: Robert P. J. Day \u003crpjday@crashcourse.ca\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "0b2cca804ee40495bc92449c3e22f3c3a3f2977a", "tree": "201f219b4063b0a976bd7ea8e45cff0faa818602", "parents": [ "a560643e21e1ac10f7398b45111aecdd7f47a4a5" ], "author": { "name": "Scott Wood", "email": "scottwood@freescale.com", "time": "Tue Apr 29 01:38:13 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:22:28 2008 +1000" }, "message": "[POWERPC] Add 6xx-style HID0_SLEEP support.\n\nThis adds a function to put a 6xx/7xx/7xxx/83xx family CPU into sleep\nmode, and return after an interrupt has occurred. It expects to be\ncalled with interrupts disabled, and returns with interrupts disabled.\nInterrupts are enabled while the processor is asleep, but the interrupt\nthat wakes the processor is not handled; it is still pending when this\nfunction returns.\n\nSigned-off-by: Scott Wood \u003cscottwood@freescale.com\u003e\nSigned-off-by: Guennadi Liakhovetski \u003cg.liakhovetski@gmx.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "a560643e21e1ac10f7398b45111aecdd7f47a4a5", "tree": "32cd887f782c35c60cc2a925ab4c49f762851319", "parents": [ "fcff474ea5cb17ff015aa40e92ed86fede41f1e2" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 14:30:48 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:22:28 2008 +1000" }, "message": "[POWERPC] Defer processing of interrupts when the CPU wakes from sleep mode\n\nThis provides a way to defer processing of an interrupt that wakes the\nprocessor out of sleep mode. On 32-bit platforms that use an\ninterrupt to wake the processor, we have to have interrupts enabled in\nhardware at the point where we go to sleep, otherwise the processor\nwill never wake up. However, because interrupts are logically\ndisabled at this point, we don\u0027t want to process the interrupt\nstraight away.\n\nThis is handled by setting the _TLF_SLEEPING flag. When we get an\ninterrupt and _TLF_SLEEPING is set, we firstly clear the MSR_EE\n(external interrupt enable) bit in the saved MSR value, and secondly\nwe then return to the address in the link register, like we do for\n_TLF_NAPPING, but without actually handling the interrupt.\n\nNote that this is handled somewhat differently on powerbooks, so this\nnew code will only be used on non-Apple machines.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "fcff474ea5cb17ff015aa40e92ed86fede41f1e2", "tree": "a99c0e14daaf31cb078812fb2fbc6abadfcd738f", "parents": [ "541b2755c2ef7dd2242ac606c115daa11e43ef69", "f26a3988917913b3d11b2bd741601a2c64ab9204" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:13:42 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 16 23:13:42 2008 +1000" }, "message": "Merge branch \u0027linux-2.6\u0027 into powerpc-next\n" }, { "commit": "9c8387afdc93f90bf0241411d44e011d8d5b76df", "tree": "97611f22b5a627f5d6067aae0bff6c32247fb577", "parents": [ "64e4566f6d590fbb284da061b9b664c2486dd2de" ], "author": { "name": "Nate Case", "email": "ncase@xes-inc.com", "time": "Tue May 13 06:14:14 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu May 15 20:49:52 2008 +1000" }, "message": "[POWERPC] Fix uninitialized variable bug in copy_{to|from}_user\n\nCalls to copy_to_user() or copy_from_user() can fail when copying N\nbytes, where N is a constant less than 8, but not 1, 2, 4, or 8,\nbecause \u0027ret\u0027 is not initialized and is only set if the size is 1,\n2, 4 or 8, but is tested after the switch statement for any constant\nsize \u003c\u003d 8. This fixes it by initializing \u0027ret\u0027 to 1, causing the\ncode to fall through to the __copy_tofrom_user call for sizes other\nthan 1, 2, 4 or 8.\n\nSigned-off-by: Dave Scidmore \u003cdscidmore@xes-inc.com\u003e\nSigned-off-by: Nate Case \u003cncase@xes-inc.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "cec08e7a948326b01555be6311480aa08e637de2", "tree": "a08e1d91c9a0c5eaa4e81036a3d7f992b2de1745", "parents": [ "08fcf1d61193d7b7779aa6d7388535e26e064a0b" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Apr 30 15:41:48 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu May 15 20:49:25 2008 +1000" }, "message": "[POWERPC] vmemmap fixes to use smaller pages\n\nThis changes vmemmap to use a different region (region 0xf) of the\naddress space, and to configure the page size of that region\ndynamically at boot.\n\nThe problem with the current approach of always using 16M pages is that\nit\u0027s not well suited to machines that have small amounts of memory such\nas small partitions on pseries, or PS3\u0027s.\n\nIn fact, on the PS3, failure to allocate the 16M page backing vmmemmap\ntends to prevent hotplugging the HV\u0027s \"additional\" memory, thus limiting\nthe available memory even more, from my experience down to something\nlike 80M total, which makes it really not very useable.\n\nThe logic used by my match to choose the vmemmap page size is:\n\n - If 16M pages are available and there\u0027s 1G or more RAM at boot,\n use that size.\n - Else if 64K pages are available, use that\n - Else use 4K pages\n\nI\u0027ve tested on a POWER6 (16M pages) and on an iSeries POWER3 (4K pages)\nand it seems to work fine.\n\nNote that I intend to change the way we organize the kernel regions \u0026\nSLBs so the actual region will change from 0xf back to something else at\none point, as I simplify the SLB miss handler, but that will be for a\nlater patch.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "1c21a2937b1f342a8f5d580203c3396557d53b6e", "tree": "78ee69e632c12afb00647f7aade4bca9b4344905", "parents": [ "1c8950ff87de950a3b6ccfb26650fc0a56278836" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:19 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:59 2008 +1000" }, "message": "[POWERPC] Fix sparse warnings in arch/powerpc/kernel\n\nMake a few things static in lparcfg.c\nMake init and exit routines static in rtas_flash.c\nMake things static in rtas_pci.c\nMake some functions static in rtas.c\nMake fops static in rtas-proc.c\nRemove unneeded extern for do_gtod in smp.c\nMake clocksource_init() static in time.c\nMake last_tick_len and ticklen_to_xs static in time.c\nMove the declaration of the pvr per-cpu into smp.h\nMake kexec_smp_down() and kexec_stack static in machine_kexec_64.c\nDon\u0027t return void in arch_teardown_msi_irqs() in msi.c\nMove declaration of GregorianDay()into asm/time.h\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "1c8950ff87de950a3b6ccfb26650fc0a56278836", "tree": "6491e5a9148fd8e4d275faf725f490065b007715", "parents": [ "9f1067c2d98ac1c43f0c82892f5647774a6ac759" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:17 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:57 2008 +1000" }, "message": "[POWERPC] Make cpus_in_xmon static and remove extern mess from hvc_console.c\n\nThis is a little messier than I\u0027d like because xmon.h only exists\non powerpc and we can\u0027t have a static inline and an extern declaration\nvisible at the same time.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "af3b74df1b2fa4dbfb0534818167f6393f5ae7c7", "tree": "d6b5bdf18d2148382bd7166a43635b2699b05022", "parents": [ "7d9e793463d854073f5c0dedea991a5b63336d6a" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:15 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:54 2008 +1000" }, "message": "[POWERPC] Move xmon_irq() declaration into xmon.h\n\nThe typdef for irqreturn_t was moved into its own header a\nwhile back, so there\u0027s no reason we can\u0027t move xmon_irq()\ninto xmon.h now.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7d9e793463d854073f5c0dedea991a5b63336d6a", "tree": "6c85b3b97b46c2dc2b41487494e67a510ed9e4e7", "parents": [ "8fa29d22dcb2aec97f1a508cd3741c6e0b4a9434" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:13 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:52 2008 +1000" }, "message": "[POWERPC] Add a declaration for xmon()\n\nUsually we call xmon() via debugger(), so this could be static.\nSometimes when debugging it\u0027s nice to be able to call xmon()\ndirectly though, so add a declaration.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "572fb578de59efaaa8d197b0183db43b1128a06e", "tree": "9f0d07578b450efbdae7946114c07da17af5c7d0", "parents": [ "09de9ff872bca25ba717a7075c9ee49b1bdb56d2" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:08 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:47 2008 +1000" }, "message": "[POWERPC] Move declaration of tce variables into mmu-hash64.h\n\n... instead of having extern declarations in a .c file.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "5f25f06529ecb4b20efc7ba00de599f5b9f4b63c", "tree": "b41343c835446e16959faa1249424056e5340a3d", "parents": [ "683307da07afd3a7ee55269ee8cb70c8053b0923" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu May 08 14:27:07 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:44 2008 +1000" }, "message": "[POWERPC] Move declaration of init_bootmem_done into system.h\n\n... instead of having an extern declaration in a .c file.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "663276b7c6340e755ed62bed67a2b96f4fc3d513", "tree": "53e559feb737fce86ad4790f979d1a80a4564266", "parents": [ "7a10174eeafe737f3ccfcece5bdff749c3b044e0" ], "author": { "name": "Kumar Gala", "email": "galak@kernel.crashing.org", "time": "Wed Apr 30 20:44:53 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:37 2008 +1000" }, "message": "[POWERPC] Set lower flag bits in regs-\u003etrap to indicate debug level exception\n\nWe use the low bits of regs-\u003etrap as flag bits. We already indicate\ncritical and machine check level exceptions via this mechanism. Extend it\nto indicate debug level exceptions.\n\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7a10174eeafe737f3ccfcece5bdff749c3b044e0", "tree": "96d3139ed19153e3369683d6b71cd74852fd1ef5", "parents": [ "140b932f8cb6cced10b96860651a198b1b89cbb9" ], "author": { "name": "Roland McGrath", "email": "roland@redhat.com", "time": "Mon Apr 28 17:30:37 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 14 22:31:33 2008 +1000" }, "message": "[POWERPC] Define and use TLF_RESTORE_SIGMASK\n\nReplace TIF_RESTORE_SIGMASK with TLF_RESTORE_SIGMASK and define\nour own set_restore_sigmask() function. This saves the costly\nSMP-safe set_bit operation, which we do not need for the sigmask\nflag since TIF_SIGPENDING always has to be set too.\n\nSigned-off-by: Roland McGrath \u003croland@redhat.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2a5f2e3e6cd1ce9fb3f8b186b6bc9aa1f1497a92", "tree": "b2306840f227972a7c9d4a2b75e516fe81358ce8", "parents": [ "02539d71fa98d5737bb668b02286c76241e4bac9", "78be76476a34a77f0ea9db2f78ba46a2b0fd5ab5" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 09 20:12:06 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri May 09 20:12:06 2008 +1000" }, "message": "Merge branch \u0027for-2.6.26\u0027 of master.kernel.org:/pub/scm/linux/kernel/git/jwboyer/powerpc-4xx into merge\n" }, { "commit": "a96df496ed1496f3e52a9b3c860cf967aa48adda", "tree": "6cf14e3dcc3888181bc7dd074dfed658726fbcbf", "parents": [ "9185ef6787f1c8f1c06aa0cb3c7746fb4f101f50" ], "author": { "name": "Stefan Roese", "email": "sr@denx.de", "time": "Mon May 05 16:53:19 2008 +1000" }, "committer": { "name": "Josh Boyer", "email": "jwboyer@linux.vnet.ibm.com", "time": "Tue May 06 10:36:20 2008 -0500" }, "message": "[POWERPC] 4xx: Fix problem with new TLB storage attibute fields on 440x6 core\n\nThe new 440x6 core used on AMCC 460EX/GT introduces new storage attibure\nfields to the TLB2 word. Those are:\n\nBit 11 12 13 14 15\n WL1 IL1I IL1D IL2I IL2D\n\nWith these bits the cache (L1 and L2) can be configured in a more flexible\nway, instruction- and data-cache independently now. The \"old\" I and W bits\nare still available and setting these old bits will automically set these\nnew bits too (for backward compatibilty).\n\nThe current code does not clear these fields resulting in disabling the cache\nby chance. This patch now makes sure that these new bits are cleared when\nthe TLB2 word is written.\n\nSigned-off-by: Stefan Roese \u003csr@denx.de\u003e\nSigned-off-by: Josh Boyer \u003cjwboyer@linux.vnet.ibm.com\u003e\n" }, { "commit": "b41e5fffe8b81fc939067d8c1c195cc79115d5a3", "tree": "7839043d5c430029efdb023e69d6a717fa8ff86e", "parents": [ "af3ce514ade2fd0e18c5d078d138a6c1137a33df" ], "author": { "name": "Emil Medve", "email": "Emilian.Medve@Freescale.com", "time": "Sat May 03 06:34:04 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon May 05 16:47:14 2008 +1000" }, "message": "[POWERPC] devres: Add devm_ioremap_prot()\n\nWe provide an ioremap_flags, so this provides a corresponding\ndevm_ioremap_prot. The slight name difference is at Ben\nHerrenschmidt\u0027s request as he plans on changing ioremap_flags to\nioremap_prot in the future.\n\nSigned-off-by: Emil Medve \u003cEmilian.Medve@Freescale.com\u003e\nSigned-off-by: Kumar Gala \u003cgalak@kernel.crashing.org\u003e\nAcked-by: Tejun Heo \u003chtejun@gmail.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "de1028927ae3487e2e450dacf50fbf32042aee18", "tree": "615b7c5fa3526ccf11b63be88c8978180df0bee8", "parents": [ "f3d69e0507f84903059d456c5d19f10b2df3ac69" ], "author": { "name": "Luke Browning", "email": "lukebr@linux.vnet.ibm.com", "time": "Mon Apr 28 17:35:56 2008 +1000" }, "committer": { "name": "Jeremy Kerr", "email": "jk@ozlabs.org", "time": "Mon May 05 13:33:44 2008 +1000" }, "message": "[POWERPC] spufs: handle faults while the context switch pending flag is set\n\nCurrently, page fault handlers don\u0027t issue a mfc restart if the context\nswitch pending flag is set, which can leave us with a hanging DMA after\na context restore.\n\nThis patch introduces fault pending flag that is set by the fault\nhandler and read by the context switch code, so that the latter can add\nthe restart bit at the right spot, after it has successfuly saved the\nstate of the mfc control register.\n\nSigned-off-by: Luke Browning \u003clukebr@linux.vnet.ibm.com\u003e\nSigned-off-by: Jeremy Kerr \u003cjk@ozlabs.org\u003e\n" }, { "commit": "f3d69e0507f84903059d456c5d19f10b2df3ac69", "tree": "f8aa4062bc1a0939d7bdb3a634c01869f2bb32a2", "parents": [ "7a2142002f29a7b398c49da9bdec712dc57087c7" ], "author": { "name": "Luke Browning", "email": "lukebr@linux.vnet.ibm.com", "time": "Sun Apr 27 18:41:55 2008 +0000" }, "committer": { "name": "Jeremy Kerr", "email": "jk@ozlabs.org", "time": "Mon May 05 13:33:44 2008 +1000" }, "message": "[POWERPC] spufs: fix concurrent delivery of class 0 \u0026 1 exceptions\n\nSPU class 0 \u0026 1 exceptions may occur in parallel, so we may end up\noverwriting csa.dsisr.\n\nThis change adds dedicated fields for each class to the spu and the spu\ncontext so that fault data is not overwritten.\n\nSigned-off-by: Luke Browning \u003clukebr@linux.vnet.ibm.com\u003e\nSigned-off-by: Jeremy Kerr \u003cjk@ozlabs.org\u003e\n" }, { "commit": "45c5eb67da5a668abe79c23a7e64dbc87a600f90", "tree": "b2b9ff85ecb5b35d246d68da94b20cb9f9912bff", "parents": [ "3fe913e7c550a869e250d04c34410f7a6e263f7c" ], "author": { "name": "Hollis Blanchard", "email": "hollisb@us.ibm.com", "time": "Fri Apr 25 17:55:49 2008 -0500" }, "committer": { "name": "Avi Kivity", "email": "avi@qumranet.com", "time": "Sun May 04 14:44:44 2008 +0300" }, "message": "KVM: ppc: Handle guest idle by emulating MSR[WE] writes\n\nThis reduces host CPU usage when the guest is idle. However, the guest must\nset MSR[WE] in its idle loop, which Linux did not do until 2.6.26.\n\nSigned-off-by: Hollis Blanchard \u003chollisb@us.ibm.com\u003e\nSigned-off-by: Jerone Young \u003cjyoung5@us.ibm.com\u003e\nSigned-off-by: Avi Kivity \u003cavi@qumranet.com\u003e\n" }, { "commit": "d35c7b0e54a596c5a8134d75999b7f391a9c6550", "tree": "697bb89dbeccae28eb928b2589f500d747ed38ec", "parents": [ "2ddcca36c8bcfa251724fe342c8327451988be0d" ], "author": { "name": "Ulrich Drepper", "email": "drepper@redhat.com", "time": "Sat May 03 15:10:37 2008 -0400" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat May 03 13:50:33 2008 -0700" }, "message": "unified (weak) sys_pipe implementation\n\nThis replaces the duplicated arch-specific versions of \"sys_pipe()\" with\none unified implementation. This removes almost 250 lines of duplicated\ncode.\n\nIt\u0027s marked __weak, so that *if* an architecture wants to override the\ndefault implementation it can do so by simply having its own replacement\nversion, since many architectures use alternate calling conventions for\nthe \u0027pipe()\u0027 system call for legacy reasons (ie traditional UNIX\nimplementations often return the two file descriptors in registers)\n\nI still haven\u0027t changed the cris version even though Linus says the BKL\nisn\u0027t needed. The arch maintainer can easily do it if there are really\nno obstacles.\n\nSigned-off-by: Ulrich Drepper \u003cdrepper@redhat.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" } ], "next": "e098675635479e9267cf2b12fb969c463cf506ab" }