)]}' { "log": [ { "commit": "e86ac13b031cf71d8f40ff513e627aac80e6b765", "tree": "8f0027221533ef8b9ad028504f406505cd2ba205", "parents": [ "470d147428563aba9c2eb7c019383335249c6110" ], "author": { "name": "Mugunthan V N", "email": "mugunthanvnm@ti.com", "time": "Mon Mar 11 23:16:35 2013 +0000" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed Mar 13 04:38:19 2013 -0400" }, "message": "drivers: net: ethernet: cpsw: change cpts_active_slave to active_slave\n\nChange cpts_active_slave to active_slave so that the same DT property\ncan be used to ethtool and SIOCGMIIPHY.\n\nCC: Richard Cochran \u003crichardcochran@gmail.com\u003e\nSigned-off-by: Mugunthan V N \u003cmugunthanvnm@ti.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c6baa304b841673d3a55ea4fcf9a5cbf7a1674b", "tree": "a6f9970cebfb573a74a4abd06ff9caee41460e9a", "parents": [ "0cb77508252e2d0e00c5ec7e57b4be9b3f7eb24d", "cb16b91a449afd01b85ec4e59f30449d11c4acd7" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 11 07:54:29 2013 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 11 07:54:29 2013 -0700" }, "message": "Merge branch \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf fixes from Ingo Molnar:\n \"Misc minor fixes mostly related to tracing\"\n\n* \u0027perf-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n s390: Fix a header dependencies related build error\n tracing: update documentation of snapshot utility\n tracing: Do not return EINVAL in snapshot when not allocated\n tracing: Add help of snapshot feature when snapshot is empty\n ftrace: Update the kconfig for DYNAMIC_FTRACE\n" }, { "commit": "cb16b91a449afd01b85ec4e59f30449d11c4acd7", "tree": "614694aca044af054e2fe18f7017c6fbf0a8f703", "parents": [ "1abccd7419de9829bcdf9ab1f81d5f6cf74d55d3" ], "author": { "name": "Li Zefan", "email": "lizefan@huawei.com", "time": "Thu Mar 07 17:35:53 2013 +0800" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Mon Mar 11 10:43:35 2013 +0100" }, "message": "s390: Fix a header dependencies related build error\n\nCommit 877c685607925238e302cd3aa38788dca6c1b226\n(\"perf: Remove include of cgroup.h from perf_event.h\") caused\nthis build failure if PERF_EVENTS is enabled:\n\n In file included from arch/s390/include/asm/perf_event.h:9:0,\n from include/linux/perf_event.h:24,\n from kernel/events/ring_buffer.c:12:\n arch/s390/include/asm/cpu_mf.h: In function \u0027qctri\u0027:\n arch/s390/include/asm/cpu_mf.h:61:12: error: \u0027EINVAL\u0027 undeclared (first use in this function)\n\ncpu_mf.h had an implicit errno.h dependency, which was added\nindirectly via cgroups.h but not anymore. Add it explicitly.\n\nReported-by: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nTested-by: Fengguang Wu \u003cfengguang.wu@intel.com\u003e\nSigned-off-by: Li Zefan \u003clizefan@huawei.com\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nLink: http://lkml.kernel.org/r/51385F79.7000106@huawei.com\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n" }, { "commit": "2116bda6ad937d7acb6e2316fd9e65ad6ca01d42", "tree": "dc9b20f4bc76b335e8a395b4248f50ac2d3935dc", "parents": [ "cc4f02486c09977ccbe3ce2276aca5608a44ca00" ], "author": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:05:45 2013 +0100" }, "committer": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:08:04 2013 +0100" }, "message": "um: Use tty_port in SIGWINCH handler\n\nThe tty below tty_port might get destroyed by the tty layer\nwhile we hold a reference to it.\nSo we have to carry tty_port around...\n\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\n" }, { "commit": "cc4f02486c09977ccbe3ce2276aca5608a44ca00", "tree": "5b0729f494e535b3d5dbbe2e189ded60ae38cea3", "parents": [ "72383d43b223c410fc61d9e905690b9b9ba9d418" ], "author": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:03:42 2013 +0100" }, "committer": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:08:03 2013 +0100" }, "message": "um: Use tty_port_operations-\u003edestruct\n\nAs we setup the SIGWINCH handler in tty_port_operations-\u003eactivate\nit makes sense to tear down it in -\u003edestruct.\n\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\n" }, { "commit": "72383d43b223c410fc61d9e905690b9b9ba9d418", "tree": "12d6906f1c850e1f9f679d867e7d55f2ea423587", "parents": [ "fdfa4c952844fce881df8c76de9c7180cbe913ab" ], "author": { "name": "Sergei Trofimovich", "email": "slyfox@gentoo.org", "time": "Sun Dec 30 01:37:31 2012 +0300" }, "committer": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:08:03 2013 +0100" }, "message": "um: fix build failure due to mess-up of sig_info protorype\n\narch/um/os-Linux/signal.c:18:8: error: conflicting types for \u0027sig_info\u0027\nIn file included from /home/slyfox/linux-2.6/arch/um/os-Linux/signal.c:12:0:\narch/um/include/shared/as-layout.h:64:15: note: previous declaration of \u0027sig_info\u0027 was here\n\nSigned-off-by: Sergei Trofimovich \u003cslyfox@gentoo.org\u003e\nCC: Jeff Dike \u003cjdike@addtoit.com\u003e\nCC: Richard Weinberger \u003crichard@nod.at\u003e\nCC: \"Martin Pärtel\" \u003cmartin.partel@gmail.com\u003e\nCC: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCC: user-mode-linux-devel@lists.sourceforge.net\nCC: user-mode-linux-user@lists.sourceforge.net\nCC: linux-kernel@vger.kernel.org\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\n" }, { "commit": "fdfa4c952844fce881df8c76de9c7180cbe913ab", "tree": "491d44e21ede6f6bdefd249b4b276444d85d38a9", "parents": [ "55ea1cfab24e37a347f1beae9f0b6724cad18f6b" ], "author": { "name": "Sergei Trofimovich", "email": "slyfox@gentoo.org", "time": "Sun Dec 30 01:37:30 2012 +0300" }, "committer": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:08:03 2013 +0100" }, "message": "um: add missing declaration of \u0027getrlimit()\u0027 and friends\n\narch/um/os-Linux/start_up.c: In function \u0027check_coredump_limit\u0027:\narch/um/os-Linux/start_up.c:338:16: error: storage size of \u0027lim\u0027 isn\u0027t known\narch/um/os-Linux/start_up.c:339:2: error: implicit declaration of function \u0027getrlimit\u0027 [-Werror\u003dimplicit-function-declaration]\n\nSigned-off-by: Sergei Trofimovich \u003cslyfox@gentoo.org\u003e\nCC: Jeff Dike \u003cjdike@addtoit.com\u003e\nCC: Richard Weinberger \u003crichard@nod.at\u003e\nCC: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\nCC: user-mode-linux-devel@lists.sourceforge.net\nCC: user-mode-linux-user@lists.sourceforge.net\nCC: linux-kernel@vger.kernel.org\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\n" }, { "commit": "55ea1cfab24e37a347f1beae9f0b6724cad18f6b", "tree": "fceb7b8b631c112aa31fe099fa438ce958be5c47", "parents": [ "bc077320f8f6449197f4010d6774afab7bb998b2" ], "author": { "name": "Paul Chavent", "email": "Paul.Chavent@onera.fr", "time": "Thu Dec 06 16:25:05 2012 +0100" }, "committer": { "name": "Richard Weinberger", "email": "richard@nod.at", "time": "Mon Mar 11 10:08:02 2013 +0100" }, "message": "net : enable tx time stamping in the vde driver.\n\nThis new version moves the skb_tx_timestamp in the main uml\ndriver. This should avoid the need to call this function in each\ntransport (vde, slirp, tuntap, ...). It also add support for ethtool\nget_ts_info.\n\nSigned-off-by: Paul Chavent \u003cpaul.chavent@onera.fr\u003e\nAcked-by: Richard Cochran \u003crichardcochran@gmail.com\u003e\nSigned-off-by: Richard Weinberger \u003crichard@nod.at\u003e\n" }, { "commit": "72932611b4b05bbd89fafa369d564ac8e449809b", "tree": "10c55c8fef383cf72059f796b82a36efae058384", "parents": [ "8343bce195da8bb4d5a652ee085474a5cc62983f", "db04dc679bcc780ad6907943afe24a30de974a1b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 09 16:51:13 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sat Mar 09 16:51:13 2013 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace\n\nPull namespace bugfixes from Eric Biederman:\n \"This is three simple fixes against 3.9-rc1. I have tested each of\n these fixes and verified they work correctly.\n\n The userns oops in key_change_session_keyring and the BUG_ON triggered\n by proc_ns_follow_link were found by Dave Jones.\n\n I am including the enhancement for mount to only trigger requests of\n filesystem modules here instead of delaying this for the 3.10 merge\n window because it is both trivial and the kind of change that tends to\n bit-rot if left untouched for two months.\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:\n proc: Use nd_jump_link in proc_ns_follow_link\n fs: Limit sys_mount to only request filesystem modules (Part 2).\n fs: Limit sys_mount to only request filesystem modules.\n userns: Stop oopsing in key_change_session_keyring\n" }, { "commit": "59d8e5eb2bd5593d8220db0e25206cdfc42e83ea", "tree": "5a11524581ff040f2c335fc6a318ff54f100552c", "parents": [ "c4c5ff09872751eeb412e3e1af484955529426a1", "c3d6b628395fe6ec3442a83ddf02334c54867d43" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 15:05:42 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 15:05:42 2013 -0800" }, "message": "Merge branch \u0027akpm\u0027 (fixes from Andrew)\n\nMerge misc fixes from Andrew Morton.\n\n* emailed patches from Andrew Morton \u003cakpm@linux-foundation.org\u003e:\n alpha: boot: fix build breakage introduced by system.h disintegration\n memcg: initialize kmem-cache destroying work earlier\n Randy has moved\n ksm: fix m68k build: only NUMA needs pfn_to_nid\n dmi_scan: fix missing check for _DMI_ signature in smbios_present()\n Revert parts of \"hlist: drop the node parameter from iterators\"\n idr: remove WARN_ON_ONCE() on negative IDs\n mm/mempolicy.c: fix sp_node_init() argument ordering\n mm/mempolicy.c: fix wrong sp_node insertion\n ipc: don\u0027t allocate a copy larger than max\n ipc: fix potential oops when src msg \u003e 4k w/ MSG_COPY\n" }, { "commit": "c3d6b628395fe6ec3442a83ddf02334c54867d43", "tree": "e97868f4a90a0b2e7944c076078eae24cc69ff14", "parents": [ "15cf17d26e08ee95c2e392a3a71f55d32e99e971" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Fri Mar 08 12:43:37 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 15:05:34 2013 -0800" }, "message": "alpha: boot: fix build breakage introduced by system.h disintegration\n\nCommit ec2212088c42 (\"Disintegrate asm/system.h for Alpha\") removed the\nsystem.h include from boot/head.S, which puts the PAL_* asm constants\nout of scope.\n\nInclude \u003casm/pal.h\u003e so we can get building again.\n\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nCc: David Rusling \u003cdavid.rusling@linaro.org\u003e\nCc: David Howells \u003cdhowells@redhat.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "c4c5ff09872751eeb412e3e1af484955529426a1", "tree": "3eeaf398043ab6ff3fa7b1b43f30ba9c793dd48d", "parents": [ "92e840b84e75cd662d7e992711e1d073295c9b47", "87c319a2c3c2efd397281089b9cdce3050febeff" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 14:46:04 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 14:46:04 2013 -0800" }, "message": "Merge branch \u0027stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile\n\nPull tile architecture fixes from Chris Metcalf:\n \"This fixes the bug that Al Viro spotted with the compat llseek code.\n I also fixed the compat syscall definitions to use the new syscall\n define macros to properly sign-extend their arguments.\"\n\n* \u0027stable\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:\n tile: properly use COMPAT_SYSCALL_DEFINEx\n tile: work around bug in the generic sys_llseek\n" }, { "commit": "92e840b84e75cd662d7e992711e1d073295c9b47", "tree": "7264c6c54b74b6f592c8d6b762e20e53aa2eda8b", "parents": [ "eca968305c2a5fbc3445c22b8e8e809196d28893", "40f09f3cd69a356b4245794e46439674336413fb" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 14:45:17 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Fri Mar 08 14:45:17 2013 -0800" }, "message": "Merge tag \u0027metag-for-v3.9-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag\n\nPull metag bugfixes from James Hogan:\n \"A couple of fairly minor arch/metag integration fixes from v3.9-rc1:\n\n - remove SET_PERSONALITY(): use default definition like other arches\n now do.\n\n - inhibit NUMA balancing: like SH, NUMA is used for memories with\n different latencies. ARCH_WANT_VARIABLE_LOCALITY has been added\n for this purpose.\"\n\n* tag \u0027metag-for-v3.9-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag:\n metag: Inhibit NUMA balancing.\n metag: remove SET_PERSONALITY()\n" }, { "commit": "47b3bc907328db968bc9b43c41f48f8d1e140750", "tree": "00475e210f6b6c86554fad8a33a3baf7385bfbcc", "parents": [ "af2841cdd4cb35248e41f7427d996c8f6b563051", "cc67708891319dbdc9f29c04154833a67d23212c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 07 15:57:38 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 07 15:57:38 2013 -0800" }, "message": "Merge branch \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 fixes from Peter Anvin:\n \"Several boot fixes (MacBook, legacy EFI bootloaders), another\n please-don\u0027t-brick fix, and some minor stuff.\"\n\n* \u0027x86-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86: Do not try to sync identity map for non-mapped pages\n x86, doc: Be explicit about what the x86 struct boot_params requires\n x86: Don\u0027t clear efi_info even if the sentinel hits\n x86, mm: Make sure to find a 2M free block for the first mapped area\n x86: Fix 32-bit *_cpu_data initializers\n efivarfs: return accurate error code in efivarfs_fill_super()\n efivars: efivarfs_valid_name() should handle pstore syntax\n efi: be more paranoid about available space when creating variables\n iommu, x86: Add DMA remap fault reason\n x86, smpboot: Remove unused variable\n" }, { "commit": "19cf3edbf33375e81f9b41a087d3fad95858a4e3", "tree": "040d427ab7111cdee528536a2a6e6b1931f05817", "parents": [ "d345243629db38a6340bcb69f372329d35b8f650", "455bd4c430b0c0a361f38e8658a0d6cb469942b5" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 07 13:47:18 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Thu Mar 07 13:47:18 2013 -0800" }, "message": "Merge branch \u0027fixes\u0027 of git://git.linaro.org/people/rmk/linux-arm\n\nPull ARM fixes from Russell King:\n \"Mainly a group of fixes, the only exception is the wiring up of the\n kcmp syscall now that those patches went in during the last merge\n window.\"\n\n* \u0027fixes\u0027 of git://git.linaro.org/people/rmk/linux-arm:\n ARM: 7668/1: fix memset-related crashes caused by recent GCC (4.7.2) optimizations\n ARM: 7667/1: perf: Fix section mismatch on armpmu_init()\n ARM: 7666/1: decompressor: add -mno-single-pic-base for building the decompressor\n ARM: 7665/1: Wire up kcmp syscall\n ARM: 7664/1: perf: remove erroneous semicolon from event initialisation\n ARM: 7663/1: perf: fix ARMv7 EVTYPE_MASK to include NSH bit\n ARM: 7662/1: hw_breakpoint: reset debug logic on secondary CPUs in s2ram resume\n ARM: 7661/1: mm: perform explicit branch predictor maintenance when required\n ARM: 7660/1: tlb: add branch predictor maintenance operations\n ARM: 7659/1: mm: make mm-\u003econtext.id an atomic64_t variable\n ARM: 7658/1: mm: fix race updating mm-\u003econtext.id on ASID rollover\n ARM: 7657/1: head: fix swapper and idmap population with LPAE and big-endian\n ARM: 7655/1: smp_twd: make twd_local_timer_of_register() no-op for nosmp\n ARM: 7652/1: mm: fix missing use of \u0027asid\u0027 to get asid value from mm-\u003econtext.id\n ARM: 7642/1: netx: bump IRQ offset to 64\n" }, { "commit": "60f583d56aa515b896a9d94f860f52640c1e8a75", "tree": "b3e140c991bbdfe01eaa029d452784ef5242feb8", "parents": [ "3c4aff6b9a183b4f24eb7b8dd6c8a92cdba3bc75" ], "author": { "name": "Dave Hansen", "email": "dave@sr71.net", "time": "Thu Mar 07 08:31:51 2013 -0800" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Thu Mar 07 13:23:28 2013 -0800" }, "message": "x86: Do not try to sync identity map for non-mapped pages\n\nkernel_map_sync_memtype() is called from a variety of contexts. The\npat.c code that calls it seems to ensure that it is not called for\nnon-ram areas by checking via pat_pagerange_is_ram(). It is important\nthat it only be called on the actual identity map because there *IS*\nno map to sync for highmem pages, or for memory holes.\n\nThe ioremap.c uses are not as careful as those from pat.c, and call\nkernel_map_sync_memtype() on PCI space which is in the middle of the\nkernel identity map _range_, but is not actually mapped.\n\nThis patch adds a check to kernel_map_sync_memtype() which probably\nduplicates some of the checks already in pat.c. But, it is necessary\nfor the ioremap.c uses and shouldn\u0027t hurt other callers.\n\nI have reproduced this bug and this patch fixes it for me and the\noriginal bug reporter:\n\n\thttps://lkml.org/lkml/2013/2/5/396\n\nSigned-off-by: Dave Hansen \u003cdave@linux.vnet.ibm.com\u003e\nLink: http://lkml.kernel.org/r/20130307163151.D9B58C4E@kernel.stglabs.ibm.com\nSigned-off-by: Dave Hansen \u003cdave@sr71.net\u003e\nTested-by: Tetsuo Handa \u003cpenguin-kernel@i-love.sakura.ne.jp\u003e\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\n" }, { "commit": "455bd4c430b0c0a361f38e8658a0d6cb469942b5", "tree": "49b812d14d37c6df726cfb9ed3ed1eb1cd316e3c", "parents": [ "44d6b1fc3e3c6a3af8e599b724972e881c81e1c9" ], "author": { "name": "Ivan Djelic", "email": "ivan.djelic@parrot.com", "time": "Wed Mar 06 20:09:27 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Thu Mar 07 16:14:22 2013 +0000" }, "message": "ARM: 7668/1: fix memset-related crashes caused by recent GCC (4.7.2) optimizations\n\nRecent GCC versions (e.g. GCC-4.7.2) perform optimizations based on\nassumptions about the implementation of memset and similar functions.\nThe current ARM optimized memset code does not return the value of\nits first argument, as is usually expected from standard implementations.\n\nFor instance in the following function:\n\nvoid debug_mutex_lock_common(struct mutex *lock, struct mutex_waiter *waiter)\n{\n\tmemset(waiter, MUTEX_DEBUG_INIT, sizeof(*waiter));\n\twaiter-\u003emagic \u003d waiter;\n\tINIT_LIST_HEAD(\u0026waiter-\u003elist);\n}\n\ncompiled as:\n\n800554d0 \u003cdebug_mutex_lock_common\u003e:\n800554d0: e92d4008 push {r3, lr}\n800554d4: e1a00001 mov r0, r1\n800554d8: e3a02010 mov r2, #16 ; 0x10\n800554dc: e3a01011 mov r1, #17 ; 0x11\n800554e0: eb04426e bl 80165ea0 \u003cmemset\u003e\n800554e4: e1a03000 mov r3, r0\n800554e8: e583000c str r0, [r3, #12]\n800554ec: e5830000 str r0, [r3]\n800554f0: e5830004 str r0, [r3, #4]\n800554f4: e8bd8008 pop {r3, pc}\n\nGCC assumes memset returns the value of pointer \u0027waiter\u0027 in register r0; causing\nregister/memory corruptions.\n\nThis patch fixes the return value of the assembly version of memset.\nIt adds a \u0027mov\u0027 instruction and merges an additional load+store into\nexisting load/store instructions.\nFor ease of review, here is a breakdown of the patch into 4 simple steps:\n\nStep 1\n\u003d\u003d\u003d\u003d\u003d\u003d\nPerform the following substitutions:\nip -\u003e r8, then\nr0 -\u003e ip,\nand insert \u0027mov ip, r0\u0027 as the first statement of the function.\nAt this point, we have a memset() implementation returning the proper result,\nbut corrupting r8 on some paths (the ones that were using ip).\n\nStep 2\n\u003d\u003d\u003d\u003d\u003d\u003d\nMake sure r8 is saved and restored when (! CALGN(1)+0) \u003d\u003d 1:\n\nsave r8:\n- str lr, [sp, #-4]!\n+ stmfd sp!, {r8, lr}\n\nand restore r8 on both exit paths:\n- ldmeqfd sp!, {pc} @ Now \u003c64 bytes to go.\n+ ldmeqfd sp!, {r8, pc} @ Now \u003c64 bytes to go.\n(...)\n tst r2, #16\n stmneia ip!, {r1, r3, r8, lr}\n- ldr lr, [sp], #4\n+ ldmfd sp!, {r8, lr}\n\nStep 3\n\u003d\u003d\u003d\u003d\u003d\u003d\nMake sure r8 is saved and restored when (! CALGN(1)+0) \u003d\u003d 0:\n\nsave r8:\n- stmfd sp!, {r4-r7, lr}\n+ stmfd sp!, {r4-r8, lr}\n\nand restore r8 on both exit paths:\n bgt 3b\n- ldmeqfd sp!, {r4-r7, pc}\n+ ldmeqfd sp!, {r4-r8, pc}\n(...)\n tst r2, #16\n stmneia ip!, {r4-r7}\n- ldmfd sp!, {r4-r7, lr}\n+ ldmfd sp!, {r4-r8, lr}\n\nStep 4\n\u003d\u003d\u003d\u003d\u003d\u003d\nRewrite register list \"r4-r7, r8\" as \"r4-r8\".\n\nSigned-off-by: Ivan Djelic \u003civan.djelic@parrot.com\u003e\nReviewed-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nSigned-off-by: Dirk Behme \u003cdirk.behme@gmail.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "3c4aff6b9a183b4f24eb7b8dd6c8a92cdba3bc75", "tree": "1f5a62d998bf6c0e7ae66c9ecd2e95b7362c8875", "parents": [ "2e604c0f19dcdd433b3863ffc3da9bc0787ca765" ], "author": { "name": "Peter Jones", "email": "pjones@redhat.com", "time": "Wed Mar 06 13:00:23 2013 -0500" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Wed Mar 06 20:34:43 2013 -0800" }, "message": "x86, doc: Be explicit about what the x86 struct boot_params requires\n\nIf the sentinel triggers, we do not want the boot loader authors to\njust poke it and make the error go away, we want them to actually fix\nthe problem.\n\nThis should help avoid making the incorrect change in non-compliant\nbootloaders.\n\n[ hpa: dropped the Documentation/x86/boot.txt hunk pending\n clarifications ]\n\nSigned-off-by: Peter Jones \u003cpjones@redhat.com\u003e\nLink: http://lkml.kernel.org/r/1362592823-28967-1-git-send-email-pjones@redhat.com\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\n" }, { "commit": "2e604c0f19dcdd433b3863ffc3da9bc0787ca765", "tree": "cc1280c896e0db70cf53dda68f5b8c29651cbad6", "parents": [ "98e7a989979b185f49e86ddaed2ad6890299d9f0" ], "author": { "name": "Josh Boyer", "email": "jwboyer@gmail.com", "time": "Wed Mar 06 20:23:30 2013 -0800" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Wed Mar 06 20:23:30 2013 -0800" }, "message": "x86: Don\u0027t clear efi_info even if the sentinel hits\n\nWhen boot_params-\u003esentinel is set, all we really know is that some\nundefined set of fields in struct boot_params contain garbage. In the\nparticular case of efi_info, however, there is a private magic for\nthat substructure, so it is generally safe to leave it even if the\nbootloader is broken.\n\nkexec (for which we did the initial analysis) did not initialize this\nfield, but of course all the EFI bootloaders do, and most EFI\nbootloaders are broken in this respect (and should be fixed.)\n\nReported-by: Robin Holt \u003cholt@sgi.com\u003e\nLink: http://lkml.kernel.org/r/CA%2B5PVA51-FT14p4CRYKbicykugVb\u003dPiaEycdQ57CK2km_OQuRQ@mail.gmail.com\nTested-by: Josh Boyer \u003cjwboyer@gmail.com\u003e\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\n" }, { "commit": "98e7a989979b185f49e86ddaed2ad6890299d9f0", "tree": "52da45c05dff2ba6da79bcbf17ea464829ab8304", "parents": [ "015221fefbc93689dd47508a66326556adf2abcd" ], "author": { "name": "Yinghai Lu", "email": "yinghai@kernel.org", "time": "Wed Mar 06 20:18:21 2013 -0800" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Wed Mar 06 20:18:32 2013 -0800" }, "message": "x86, mm: Make sure to find a 2M free block for the first mapped area\n\nHenrik reported that his MacAir 3.1 would not boot with\n\n| commit 8d57470d8f859635deffe3919d7d4867b488b85a\n| Date: Fri Nov 16 19:38:58 2012 -0800\n|\n| x86, mm: setup page table in top-down\n\nIt turns out that we do not calculate the real_end properly:\nWe try to get 2M size with 4K alignment, and later will round down\nto 2M, so we will get less then 2M for first mapping, in extreme\ncase could be only 4K only. In Henrik\u0027s system it has (1M-32K) as\nlast usable rage is [mem 0x7f9db000-0x7fef8fff].\n\nThe problem is exposed when EFI booting have several holes and it\nwill force mapping to use PTE instead as we only map usable areas.\n\nTo fix it, just make it be 2M aligned, so we can be guaranteed to be\nable to use large pages to map it.\n\nReported-by: Henrik Rydberg \u003crydberg@euromail.se\u003e\nBisected-by: Henrik Rydberg \u003crydberg@euromail.se\u003e\nTested-by: Henrik Rydberg \u003crydberg@euromail.se\u003e\nSigned-off-by: Yinghai Lu \u003cyinghai@kernel.org\u003e\nLink: http://lkml.kernel.org/r/CAE9FiQX4nQ7_1kg5RL_vh56rmcSHXUi1ExrZX7CwED4NGMnHfg@mail.gmail.com\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\n" }, { "commit": "015221fefbc93689dd47508a66326556adf2abcd", "tree": "9ad7c56a7a2b900685807a6ec799b91064aa8bcd", "parents": [ "4ecccd9edd5eb4dd185486e6e593c671484691bc" ], "author": { "name": "Krzysztof Mazur", "email": "krzysiek@podlesie.net", "time": "Sun Mar 03 00:14:42 2013 +0100" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@zytor.com", "time": "Wed Mar 06 20:15:50 2013 -0800" }, "message": "x86: Fix 32-bit *_cpu_data initializers\n\nThe commit 27be457000211a6903968dfce06d5f73f051a217\n(\u0027x86 idle: remove 32-bit-only \"no-hlt\" parameter, hlt_works_ok\nflag\u0027) removed the hlt_works_ok flag from struct cpuinfo_x86, but\nboot_cpu_data and new_cpu_data initializers were not changed\ncausing setting f00f_bug flag, instead of fdiv_bug.\n\nIf CONFIG_X86_F00F_BUG is not set the f00f_bug flag is never\ncleared.\n\nTo avoid such problems in future C99-style initialization is now\nused.\n\nSigned-off-by: Krzysztof Mazur \u003ckrzysiek@podlesie.net\u003e\nAcked-by: Borislav Petkov \u003cbp@suse.de\u003e\nCc: len.brown@intel.com\nLink: http://lkml.kernel.org/r/1362266082-2227-1-git-send-email-krzysiek@podlesie.net\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nSigned-off-by: H. Peter Anvin \u003chpa@zytor.com\u003e\n" }, { "commit": "44d6b1fc3e3c6a3af8e599b724972e881c81e1c9", "tree": "54f1805f2c882246abb177c26fe294a8d04e1603", "parents": [ "b8083f86e8e9b23ad4d75d334c92b601947dfa91" ], "author": { "name": "Stephen Boyd", "email": "sboyd@codeaurora.org", "time": "Tue Mar 05 03:54:06 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Mar 06 23:45:43 2013 +0000" }, "message": "ARM: 7667/1: perf: Fix section mismatch on armpmu_init()\n\nWARNING: vmlinux.o(.text+0xfb80): Section mismatch in reference\nfrom the function armpmu_register() to the function\n.init.text:armpmu_init()\nThe function armpmu_register() references\nthe function __init armpmu_init().\nThis is often because armpmu_register lacks a __init\nannotation or the annotation of armpmu_init is wrong.\n\nJust drop the __init marking on armpmu_init() because\narmpmu_register() no longer has an __init marking.\n\nAcked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Stephen Boyd \u003csboyd@codeaurora.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "b8083f86e8e9b23ad4d75d334c92b601947dfa91", "tree": "5bdf85b31f631837afa1b35e9706789871fdd9c7", "parents": [ "3f7d1fe108dbaefd0c57a41753fc2c90b395f458" ], "author": { "name": "Jonathan Austin", "email": "Jonathan.Austin@arm.com", "time": "Mon Mar 04 15:17:36 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Wed Mar 06 23:45:42 2013 +0000" }, "message": "ARM: 7666/1: decompressor: add -mno-single-pic-base for building the decompressor\n\nBefore jumping to (position independent) C-code from the decompressor\u0027s\nassembler world we set-up the C environment. This setup currently does not\nset r9, which for arm-none-uclinux-uclibceabi toolchains is by default\nexpected to be the PIC offset base register (IE should point to the\nbeginning of the GOT).\n\nCurrently, therefore, in order to build working kernels that use the\ndecompressor it is necessary to use an arm-linux-gnueabi toolchain, or\nsimilar. uClinux toolchains cause a prefetch abort to occur at the beginning\nof the decompress_kernel function.\n\nThis patch allows uClinux toolchains to build bootable zImages by forcing\nthe -mno-single-pic-base option, which ensures that the location of the GOT\nis re-derived each time it is required, and r9 becomes free for use as a\ngeneral purpose register.\n\nThis has a small (4% in instruction terms) advantage over the alternative of\nsetting r9 to point to the GOT before calling into the C-world.\n\nSigned-off-by: Jonathan Austin \u003cjonathan.austin@arm.com\u003e\nAcked-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "576cfb404c9cab728e9462ea713f3422679d5cf7", "tree": "7774493ebeacef4bb8b8d6811b32cd1abb9f97fa", "parents": [ "6dbe51c251a327e012439c4772097a13df43c5b8" ], "author": { "name": "Borislav Petkov", "email": "bp@suse.de", "time": "Mon Mar 04 21:16:16 2013 +0100" }, "committer": { "name": "H. Peter Anvin", "email": "hpa@linux.intel.com", "time": "Tue Mar 05 15:26:45 2013 -0800" }, "message": "x86, smpboot: Remove unused variable\n\nThe cpuinfo_x86 ptr is unused now. Drop it. Got obsolete by 69fb3676df33\n(\"x86 idle: remove mwait_idle() and \"idle\u003dmwait\" cmdline param\")\nremoving its only user.\n\n[ hpa: fixes gcc warning ]\n\nSigned-off-by: Borislav Petkov \u003cbp@suse.de\u003e\nLink: http://lkml.kernel.org/r/1362428180-8865-2-git-send-email-bp@alien8.de\nCc: Len Brown \u003clen.brown@intel.com\u003e\nSigned-off-by: H. Peter Anvin \u003chpa@linux.intel.com\u003e\n" }, { "commit": "54c9b2253d34e8998e4bff9ac2d7a3ba0b861d52", "tree": "45830a7b21e9fe6a533772e8cb571c6cca7eb385", "parents": [ "fa759e9b0984748cf626aac59bca60bdab42c644" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Mon Mar 04 19:45:52 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:30 2013 +1100" }, "message": "powerpc: Set DSCR bit in FSCR setup\n\nWe support DSCR (Data Stream Control Register) so we should make sure we set it\nin the FSCR (Facility Status \u0026 Control Register) incase some firmwares don\u0027t\nset it. If we don\u0027t set this, we\u0027ll take a facility unavailable exception when\nusing the DSCR.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "fa759e9b0984748cf626aac59bca60bdab42c644", "tree": "62ffce01d34b73541c4a817b1b015144df9951d9", "parents": [ "57d231678ace658b3a73a0d144cfebbd4257bc0e" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Mon Mar 04 19:45:51 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:30 2013 +1100" }, "message": "powerpc: Add DSCR FSCR register bit definition\n\nThis sets the DSCR (Data Stream Control Register) in the FSCR (Facility Status\n\u0026 Control Register).\n\nAlso harmonise TAR (Target Address Register) FSCR bit definition too.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "57d231678ace658b3a73a0d144cfebbd4257bc0e", "tree": "6b644ca8c160e8e3415718545ae57eeaca620627", "parents": [ "8170a83f15eeca9a84ff895f1a89b58918425a3f" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Mon Mar 04 19:45:50 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:29 2013 +1100" }, "message": "powerpc: Fix setting FSCR for HV\u003d0 and on secondary CPUs\n\nCurrently we only set the FSCR (Facility Status and Control Register) when HV\u003d1\nbut this feature is available when HV\u003d0 also. This patch sets FSCR when HV\u003d0.\n\nAlso, we currently only set the FSCR on the master CPU. This patch also sets\nthe FSCR on secondary CPUs.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\ncc: Ian Munsie \u003cimunsie@au1.ibm.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "8170a83f15eeca9a84ff895f1a89b58918425a3f", "tree": "e66a5df57a378547e909fed131f2f73560c69fc4", "parents": [ "a74f350b5ce6d8dd1847aa1191ea177fff025567" ], "author": { "name": "Tony Breeds", "email": "tony@bakeyournoodle.com", "time": "Mon Mar 04 15:57:30 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:29 2013 +1100" }, "message": "powerpc: Wireup the kcmp syscall to sys_ni\n\nSince kmp takes 2 unsigned long args there should be a compat wrapper.\nSince one isn\u0027t provided I think it\u0027s safer just to hook this up to not\nimplemented. If we need it later we can do it properly then.\n\nSigned-off-by: Tony Breeds \u003ctony@bakeyournoodle.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "a74f350b5ce6d8dd1847aa1191ea177fff025567", "tree": "473fb35eeecceba75c039d472638ce214bbe065e", "parents": [ "6a404806dfceba9b154015705a9e647fa7749fd8" ], "author": { "name": "Akinobu Mita", "email": "akinobu.mita@gmail.com", "time": "Sat Mar 02 04:06:30 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:28 2013 +1100" }, "message": "powerpc: Remove unused BITOP_LE_SWIZZLE macro\n\nThe BITOP_LE_SWIZZLE macro was used in the little-endian bitops functions\nfor powerpc. But these functions were converted to generic bitops and\nthe BITOP_LE_SWIZZLE is not used anymore.\n\nSigned-off-by: Akinobu Mita \u003cakinobu.mita@gmail.com\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: linuxppc-dev@lists.ozlabs.org\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "6a404806dfceba9b154015705a9e647fa7749fd8", "tree": "773110aa64ab8777c7ca4ba56a5556a6262de7d1", "parents": [ "9276dfd27897a0b29d8b5814f39a1f82f56b6b6b" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Feb 27 10:45:52 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:28 2013 +1100" }, "message": "powerpc: Avoid link stack corruption in MMU on syscall entry path\n\nCurrently we use the link register to branch up high in the early MMU on\nsyscall entry path. Unfortunately, this trashes the link stack as the\naddress we are going to is not associated with the earlier mflr.\n\nThis patch simply converts us to used the count register (volatile over\nsyscalls anyway) instead. This is much better at predicting in this\nscenario and doesn\u0027t trash link stack causing a bunch of additional\nbranch mispredicts later. Benchmarking this on POWER8 saves a bunch of\ncycles on Anton\u0027s null syscall benchmark here:\n http://ozlabs.org/~anton/junkcode/null_syscall.c\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "6b6680c4ea3952af8ae76915cbca41245147741b", "tree": "3863f55ebb6a7f1c49830337c7a8aa52443c54a8", "parents": [ "27777890d0fae55d14ef18791fc7994faf1bc867" ], "author": { "name": "Chen Gang", "email": "gang.chen@asianux.com", "time": "Mon Feb 25 16:51:49 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:27 2013 +1100" }, "message": "powerpc/pseries/hvcserver: Fix strncpy buffer limit in location code\n\nthe dest buf len is 80 (HVCS_CLC_LENGTH + 1).\n the src buf len is PAGE_SIZE.\n if src buf string len is more than 80, it will cause issue.\n\nSigned-off-by: Chen Gang \u003cgang.chen@asianux.com\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "27777890d0fae55d14ef18791fc7994faf1bc867", "tree": "cfaaaa968dc7460c4e2d324622b53d22962935a8", "parents": [ "6dbe51c251a327e012439c4772097a13df43c5b8" ], "author": { "name": "Tony Breeds", "email": "tony@bakeyournoodle.com", "time": "Mon Feb 25 16:20:05 2013 +0000" }, "committer": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue Mar 05 16:56:26 2013 +1100" }, "message": "powerpc: Fix compile of sha1-powerpc-asm.S on 32-bit\n\nWhen building with CRYPTO_SHA1_PPC enabled we fail with:\n\npowerpc/crypto/sha1-powerpc-asm.S: Assembler messages:\npowerpc/crypto/sha1-powerpc-asm.S:116: Error: can\u0027t resolve `0\u0027 {*ABS* section} - `STACKFRAMESIZE\u0027 {*UND* section}\npowerpc/crypto/sha1-powerpc-asm.S:116: Error: expression too complex\npowerpc/crypto/sha1-powerpc-asm.S:178: Error: unsupported relocation against STACKFRAMESIZE\n\nUse INT_FRAME_SIZE instead of STACKFRAMESIZE.\n\nSigned-off-by: Tony Breeds \u003ctony@bakeyournoodle.com\u003e\nTested-by: Christian Kujau \u003clists@nerdbynature.de\u003e\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\n" }, { "commit": "87c319a2c3c2efd397281089b9cdce3050febeff", "tree": "762ae66c1cfe3e7f170ba01088273fcc429298a6", "parents": [ "5a114b98661e3aaa0ac085eb931584dce3b0ef9b" ], "author": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Mon Mar 04 13:37:32 2013 -0500" }, "committer": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Mon Mar 04 13:37:32 2013 -0500" }, "message": "tile: properly use COMPAT_SYSCALL_DEFINEx\n\nThis was pointed out by Al Viro. Using the correct wrappers\nproperly does sign extension as necessary on syscall arguments.\n\nSigned-off-by: Chris Metcalf \u003ccmetcalf@tilera.com\u003e\n" }, { "commit": "5a114b98661e3aaa0ac085eb931584dce3b0ef9b", "tree": "375a9103da9b40c1213d3bf78ed25f99c0df8455", "parents": [ "6dbe51c251a327e012439c4772097a13df43c5b8" ], "author": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Mon Mar 04 11:19:09 2013 -0500" }, "committer": { "name": "Chris Metcalf", "email": "cmetcalf@tilera.com", "time": "Mon Mar 04 11:19:09 2013 -0500" }, "message": "tile: work around bug in the generic sys_llseek\n\nsys_llseek should specify the high and low 32-bit seek values as \"unsigned\nint\" but instead it specifies \"unsigned long\". Since compat syscall\narguments are always sign-extended on tile, this means that a seek value\nof 0xffffffff will be incorrectly interpreted as a value of -1ULL.\n\nTo avoid the risk of breaking binary compatibility on architectures\nthat already use sys_llseek this way, we follow the same path as MIPS\nand provide a wrapper override.\n\nSigned-off-by: Chris Metcalf \u003ccmetcalf@tilera.com\u003e\nCc: stable@kernel.org [v3.6 onwards]\n" }, { "commit": "40f09f3cd69a356b4245794e46439674336413fb", "tree": "c07e8e3cdc8798add1e0455ddf2e75750baaea1c", "parents": [ "0010aeed7bfdc8d1048e524d285a69c11e0ac336" ], "author": { "name": "Paul Mundt", "email": "lethal@linux-sh.org", "time": "Mon Mar 04 10:12:35 2013 +0900" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Mon Mar 04 10:29:19 2013 +0000" }, "message": "metag: Inhibit NUMA balancing.\n\nThe metag NUMA implementation follows the SH model, using different nodes for\nmemories with different latencies. As such, we ensure that automated balancing\nbetween nodes is inhibited, by way of the new ARCH_WANT_VARIABLE_LOCALITY.\n\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "0010aeed7bfdc8d1048e524d285a69c11e0ac336", "tree": "538ddf0ecf7f077c16c615cea71750b75b81f149", "parents": [ "6dbe51c251a327e012439c4772097a13df43c5b8" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Feb 26 11:34:34 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Mon Mar 04 10:27:20 2013 +0000" }, "message": "metag: remove SET_PERSONALITY()\n\nCommit e72837e3e7bae3f182c4ac63c9424e86f1158dd0 (\"default\nSET_PERSONALITY() in linux/elf.h\").\n\nThe above commit moved the common definition of SET_PERSONALITY() in a\nbunch of the arch headers to linux/elf.h. Metag shares that common\ndefinition so remove it from arch/metag/include/asm/elf.h too.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "7f78e0351394052e1a6293e175825eb5c7869507", "tree": "76493af33d02bd3f411e69f95b0bcdfff50412b4", "parents": [ "ba0e3427b03c3d1550239779eca5c1c5a53a2152" ], "author": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Sat Mar 02 19:39:14 2013 -0800" }, "committer": { "name": "Eric W. Biederman", "email": "ebiederm@xmission.com", "time": "Sun Mar 03 19:36:31 2013 -0800" }, "message": "fs: Limit sys_mount to only request filesystem modules.\n\nModify the request_module to prefix the file system type with \"fs-\"\nand add aliases to all of the filesystems that can be built as modules\nto match.\n\nA common practice is to build all of the kernel code and leave code\nthat is not commonly needed as modules, with the result that many\nusers are exposed to any bug anywhere in the kernel.\n\nLooking for filesystems with a fs- prefix limits the pool of possible\nmodules that can be loaded by mount to just filesystems trivially\nmaking things safer with no real cost.\n\nUsing aliases means user space can control the policy of which\nfilesystem modules are auto-loaded by editing /etc/modprobe.d/*.conf\nwith blacklist and alias directives. Allowing simple, safe,\nwell understood work-arounds to known problematic software.\n\nThis also addresses a rare but unfortunate problem where the filesystem\nname is not the same as it\u0027s module name and module auto-loading\nwould not work. While writing this patch I saw a handful of such\ncases. The most significant being autofs that lives in the module\nautofs4.\n\nThis is relevant to user namespaces because we can reach the request\nmodule in get_fs_type() without having any special permissions, and\npeople get uncomfortable when a user specified string (in this case\nthe filesystem type) goes all of the way to request_module.\n\nAfter having looked at this issue I don\u0027t think there is any\nparticular reason to perform any filtering or permission checks beyond\nmaking it clear in the module request that we want a filesystem\nmodule. The common pattern in the kernel is to call request_module()\nwithout regards to the users permissions. In general all a filesystem\nmodule does once loaded is call register_filesystem() and go to sleep.\nWhich means there is not much attack surface exposed by loading a\nfilesytem module unless the filesystem is mounted. In a user\nnamespace filesystems are not mounted unless .fs_flags \u003d FS_USERNS_MOUNT,\nwhich most filesystems do not set today.\n\nAcked-by: Serge Hallyn \u003cserge.hallyn@canonical.com\u003e\nAcked-by: Kees Cook \u003ckeescook@chromium.org\u003e\nReported-by: Kees Cook \u003ckeescook@google.com\u003e\nSigned-off-by: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\n" }, { "commit": "3f7d1fe108dbaefd0c57a41753fc2c90b395f458", "tree": "898aba271372e74c2586adaa98d5b9ae787c9ae5", "parents": [ "e595ede6050b1ce982d74f7084f93715bcc32359" ], "author": { "name": "Cyrill Gorcunov", "email": "cyrill.gorcunov@gmail.com", "time": "Thu Feb 28 21:53:35 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:18 2013 +0000" }, "message": "ARM: 7665/1: Wire up kcmp syscall\n\nWire up kcmp syscall for ability to proceed checkpoint/restore\nprocedure on ARM platform.\n\nSigned-off-by: Alexander Kartashov \u003calekskartashov@parallels.com\u003e\nSigned-off-by: Cyrill Gorcunov \u003cgorcunov@openvz.org\u003e\nAcked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "e595ede6050b1ce982d74f7084f93715bcc32359", "tree": "418d4c421e94a3da4351a8a75855bfa4c5279526", "parents": [ "f2fe09b055e2549de41fb107b34c60bac4a1b0cf" ], "author": { "name": "Chen Gang", "email": "gang.chen@asianux.com", "time": "Thu Feb 28 17:51:29 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:17 2013 +0000" }, "message": "ARM: 7664/1: perf: remove erroneous semicolon from event initialisation\n\nCommit 9dcbf466559f (\"ARM: perf: simplify __hw_perf_event_init err\nhandling\") tidied up the error handling code for perf event\ninitialisation on ARM, but a copy-and-paste error left a dangling\nsemicolon at the end of an if statement.\n\nThis patch removes the broken semicolon, restoring the old group\nvalidation semantics.\n\nCc: Mark Rutland \u003cmark.rutland@arm.com\u003e\nAcked-by: Dirk Behme \u003cdirk.behme@gmail.com\u003e\nSigned-off-by: Chen Gang \u003cgang.chen@asianux.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "f2fe09b055e2549de41fb107b34c60bac4a1b0cf", "tree": "8a611c6c0f6e842a0d7cc1cff8b83f64374f0823", "parents": [ "1a8e611874da714ee7ef1e92e5160b38dc54959b" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:49:11 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:17 2013 +0000" }, "message": "ARM: 7663/1: perf: fix ARMv7 EVTYPE_MASK to include NSH bit\n\nMasked out PMXEVTYPER.NSH means that we can\u0027t enable profiling at PL2,\nregardless of the settings in the HDCR.\n\nThis patch fixes the broken mask.\n\nCc: \u003cstable@vger.kernel.org\u003e\nReported-by: Christoffer Dall \u003ccdall@cs.columbia.edu\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "1a8e611874da714ee7ef1e92e5160b38dc54959b", "tree": "bba75c3ac81ac9fafbcb033f144d1925a42f5854", "parents": [ "89c7e4b8bbb3d4fa52df5746a8ad38e610143651" ], "author": { "name": "Dietmar Eggemann", "email": "Dietmar.Eggemann@arm.com", "time": "Thu Feb 28 17:48:57 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:16 2013 +0000" }, "message": "ARM: 7662/1: hw_breakpoint: reset debug logic on secondary CPUs in s2ram resume\n\nWe must mask out the CPU_TASKS_FROZEN bit so that reset_ctrl_regs is\nalso called on a secondary CPU during s2ram resume, where only the boot\nCPU will receive the PM_EXIT notification.\n\nSigned-off-by: Dietmar Eggemann \u003cdietmar.eggemann@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "89c7e4b8bbb3d4fa52df5746a8ad38e610143651", "tree": "f6d3d835f5fca88fd4de01f166d782f16c2fe3c9", "parents": [ "862c588f062fe9339a180cf6429e4df1855c376a" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:48:40 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:16 2013 +0000" }, "message": "ARM: 7661/1: mm: perform explicit branch predictor maintenance when required\n\nThe ARM ARM requires branch predictor maintenance if, for a given ASID,\nthe instructions at a specific virtual address appear to change.\n\nFrom the kernel\u0027s point of view, that means:\n\n\t- Changing the kernel\u0027s view of memory (e.g. switching to the\n\t identity map)\n\t- ASID rollover (since ASIDs will be re-allocated to new tasks)\n\nThis patch adds explicit branch predictor maintenance when either of the\ntwo conditions above are met.\n\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "862c588f062fe9339a180cf6429e4df1855c376a", "tree": "20348a057421b00913563d103c1cd430d1a7bb6f", "parents": [ "8a4e3a9ead7e37ce1505602b564c15da09ac039f" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:48:11 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:15 2013 +0000" }, "message": "ARM: 7660/1: tlb: add branch predictor maintenance operations\n\nThe ARM architecture requires explicit branch predictor maintenance\nwhen updating an instruction stream for a given virtual address. In\nreality, this isn\u0027t so much of a burden because the branch predictor\nis flushed during the cache maintenance required to make the new\ninstructions visible to the I-side of the processor.\n\nHowever, there are still some cases where explicit flushing is required,\nso add a local_bp_flush_all operation to deal with this.\n\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "8a4e3a9ead7e37ce1505602b564c15da09ac039f", "tree": "5cc8ad188810649ea0739c769977451b825e4325", "parents": [ "37f47e3d62533c931b04cb409f2eb299e6342331" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:47:36 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:14 2013 +0000" }, "message": "ARM: 7659/1: mm: make mm-\u003econtext.id an atomic64_t variable\n\nmm-\u003econtext.id is updated under asid_lock when a new ASID is allocated\nto an mm_struct. However, it is also read without the lock when a task\nis being scheduled and checking whether or not the current ASID\ngeneration is up-to-date.\n\nIf two threads of the same process are being scheduled in parallel and\nthe bottom bits of the generation in their mm-\u003econtext.id match the\ncurrent generation (that is, the mm_struct has not been used for ~2^24\nrollovers) then the non-atomic, lockless access to mm-\u003econtext.id may\nyield the incorrect ASID.\n\nThis patch fixes this issue by making mm-\u003econtext.id and atomic64_t,\nensuring that the generation is always read consistently. For code that\nonly requires access to the ASID bits (e.g. TLB flushing by mm), then\nthe value is accessed directly, which GCC converts to an ldrb.\n\nCc: \u003cstable@vger.kernel.org\u003e # 3.8\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "37f47e3d62533c931b04cb409f2eb299e6342331", "tree": "346daa43b9edd75249d72a61c429a62cf662da4b", "parents": [ "d61947a164760ac520cb416768afdf38c33d60e7" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:47:20 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:13 2013 +0000" }, "message": "ARM: 7658/1: mm: fix race updating mm-\u003econtext.id on ASID rollover\n\nIf a thread triggers an ASID rollover, other threads of the same process\nmust be made to wait until the mm-\u003econtext.id for the shared mm_struct\nhas been updated to new generation and associated book-keeping (e.g.\nTLB invalidation) has ben performed.\n\nHowever, there is a *tiny* window where both mm-\u003econtext.id and the\nrelevant active_asids entry are updated to the new generation, but the\nTLB flush has not been performed, which could allow another thread to\nreturn to userspace with a dirty TLB, potentially leading to data\ncorruption. In reality this will never occur because one CPU would need\nto perform a context-switch in the time it takes another to do a couple\nof atomic test/set operations but we should plug the race anyway.\n\nThis patch moves the active_asids update until after the potential TLB\nflush on context-switch.\n\nCc: \u003cstable@vger.kernel.org\u003e # 3.8\nReviewed-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "d61947a164760ac520cb416768afdf38c33d60e7", "tree": "cc0fcb95a82816639184dc0c26fe0d1293679307", "parents": [ "904464b91eca8c665acea033489225af02eeb75a" ], "author": { "name": "Will Deacon", "email": "will.deacon@arm.com", "time": "Thu Feb 28 17:46:16 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:13 2013 +0000" }, "message": "ARM: 7657/1: head: fix swapper and idmap population with LPAE and big-endian\n\nThe LPAE page table format uses 64-bit descriptors, so we need to take\nendianness into account when populating the swapper and idmap tables\nduring early initialisation.\n\nThis patch ensures that we store the two words making up each page table\nentry in the correct order when running big-endian.\n\nCc: \u003cstable@vger.kernel.org\u003e\nAcked-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\nTested-by: Marc Zyngier \u003cmarc.zyngier@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "904464b91eca8c665acea033489225af02eeb75a", "tree": "06eb007212672dcc97de5b9d4d434754db25eab8", "parents": [ "78305c863074f809f09a96ea97b86d0f0c1c8399" ], "author": { "name": "Shawn Guo", "email": "shawn.guo@linaro.org", "time": "Fri Feb 22 14:01:42 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:12 2013 +0000" }, "message": "ARM: 7655/1: smp_twd: make twd_local_timer_of_register() no-op for nosmp\n\nWhen booting a SMP build kernel with nosmp on kernel cmdline, the\nfollowing fat warning will be hit.\n\n------------[ cut here ]------------\nWARNING: at arch/arm/kernel/smp_twd.c:345\ntwd_local_timer_of_register+0x7c/0x90()\ntwd_local_timer_of_register failed (-6)\nModules linked in:\nBacktrace:\n[\u003c80011f14\u003e] (dump_backtrace+0x0/0x10c) from [\u003c8044dd30\u003e]\n(dump_stack+0x18/0x1c)\n r7:805e9f58 r6:805ba84c r5:80539331 r4:00000159\n[\u003c8044dd18\u003e] (dump_stack+0x0/0x1c) from [\u003c80020fbc\u003e]\n(warn_slowpath_common+0x54/0x6c)\n[\u003c80020f68\u003e] (warn_slowpath_common+0x0/0x6c) from [\u003c80021078\u003e]\n(warn_slowpath_fmt+0x38/0x40)\n r9:412fc09a r8:8fffffff r7:ffffffff r6:00000001 r5:80633b8c\nr4:80b32da8\n[\u003c80021040\u003e] (warn_slowpath_fmt+0x0/0x40) from [\u003c805ba84]\n(twd_local_timer_of_register+0x7c/0x90)\n r3:fffffffa r2:8053934b\n[\u003c805ba7d0\u003e] (twd_local_timer_of_register+0x0/0x90) from [\u003c805c0bec\u003e]\n(imx6q_timer_init+0x18/0x4c)\n r5:80633800 r4:8053b701\n[\u003c805c0bd4\u003e] (imx6q_timer_init+0x0/0x4c) from [\u003c805ba4e8\u003e]\n(time_init+0x28/0x38)\n r5:80633800 r4:805dc0f4\n[\u003c805ba4c0\u003e] (time_init+0x0/0x38) from [\u003c805b6854\u003e]\n(start_kernel+0x1a0/0x310)\n[\u003c805b66b4\u003e] (start_kernel+0x0/0x310) from [\u003c10008044\u003e] (0x10008044)\n r8:1000406a r7:805f3f8c r6:805dc0c4 r5:805f0518 r4:10c5387d\n---[ end trace 1b75b31a2719ed1c ]---\n\nCheck (!is_smp() || !setup_max_cpus) in twd_local_timer_of_register()\nto make it be a no-op for the conditions, thus avoid above warning.\n\nReported-by: Dirk Behme \u003cdirk.behme@de.bosch.com\u003e\nSigned-off-by: Shawn Guo \u003cshawn.guo@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "78305c863074f809f09a96ea97b86d0f0c1c8399", "tree": "3e23da25ae712bfc165b9927efc1e642f972d6ea", "parents": [ "30a1b5ef0c953a7ba82169c46f92f52fa11ee15e" ], "author": { "name": "Ben Dooks", "email": "ben-linux@fluff.org", "time": "Fri Feb 15 13:30:58 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:12 2013 +0000" }, "message": "ARM: 7652/1: mm: fix missing use of \u0027asid\u0027 to get asid value from mm-\u003econtext.id\n\nFix missing use of the asid macro when getting the ASID from the mm-\u003econtext.id field.\n\nSigned-off-by: Ben Dooks \u003cben.dooks@codethink.co.uk\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "30a1b5ef0c953a7ba82169c46f92f52fa11ee15e", "tree": "1fc23566d74eae0209f907784d573fb5ae5661ea", "parents": [ "56a79b7b021bf1b08334e63c2c14b280e2dbf47a" ], "author": { "name": "Linus Walleij", "email": "linus.walleij@linaro.org", "time": "Fri Feb 08 23:02:33 2013 +0100" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 22:54:11 2013 +0000" }, "message": "ARM: 7642/1: netx: bump IRQ offset to 64\n\nThe Netx IRQs offset from zero, which is illegal, since Linux\nIRQ 0 is NO_IRQ.\n\nAcked-by: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nReviewed-by: Jamie Iles \u003cjamie@jamieiles.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n" }, { "commit": "8e8b180a5f1b237345f6e2d960bcceb8b6bc3793", "tree": "9f10c874b0d13cc4929ff2b68c65ae2d48d4fb9d", "parents": [ "56a79b7b021bf1b08334e63c2c14b280e2dbf47a", "45e27161c62216c163880d7aed751cb55a65c8e9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 14:22:53 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 14:22:53 2013 -0800" }, "message": "Merge tag \u0027stable/for-linus-3.9-rc1-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen\n\nPull Xen bug-fixes from Konrad Rzeszutek Wilk:\n - Update the Xen ACPI memory and CPU hotplug locking mechanism.\n - Fix PAT issues wherein various applications would not start\n - Fix handling of multiple MSI as AHCI now does it.\n - Fix ARM compile failures.\n\n* tag \u0027stable/for-linus-3.9-rc1-tag\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:\n xenbus: fix compile failure on ARM with Xen enabled\n xen/pci: We don\u0027t do multiple MSI\u0027s.\n xen/pat: Disable PAT using pat_enabled value.\n xen/acpi: xen cpu hotplug minor updates\n xen/acpi: xen memory hotplug minor updates\n" }, { "commit": "56a79b7b021bf1b08334e63c2c14b280e2dbf47a", "tree": "0419233e6194f4f12073c9284852885aa8984bec", "parents": [ "1c82315a12144cde732636e259d39e3ee81b3c5b", "dcf787f39162ce32ca325b3e784aba2d2444619a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 13:23:02 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 13:23:03 2013 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs\n\nPull more VFS bits from Al Viro:\n \"Unfortunately, it looks like xattr series will have to wait until the\n next cycle ;-/\n\n This pile contains 9p cleanups and fixes (races in v9fs_fid_add()\n etc), fixup for nommu breakage in shmem.c, several cleanups and a bit\n more file_inode() work\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:\n constify path_get/path_put and fs_struct.c stuff\n fix nommu breakage in shmem.c\n cache the value of file_inode() in struct file\n 9p: if v9fs_fid_lookup() gets to asking server, it\u0027d better have hashed dentry\n 9p: make sure -\u003elookup() adds fid to the right dentry\n 9p: untangle -\u003elookup() a bit\n 9p: double iput() in -\u003elookup() if d_materialise_unique() fails\n 9p: v9fs_fid_add() can\u0027t fail now\n v9fs: get rid of v9fs_dentry\n 9p: turn fid-\u003edlist into hlist\n 9p: don\u0027t bother with private lock in -\u003ed_fsdata; dentry-\u003ed_lock will do just fine\n more file_inode() open-coded instances\n selinux: opened file can\u0027t have NULL or negative -\u003ef_path.dentry\n\n(In the meantime, the hlist traversal macros have changed, so this\nrequired a semantic conflict fixup for the newly hlistified fid-\u003edlist)\n" }, { "commit": "530ede14cf63ef4a207b0003e14af2f750484180", "tree": "2849be2c8fd6d2d6a8a71447aebe5d7400317c8d", "parents": [ "6977c6fc77b864dc1851e100c8528c2f3ed5bd9b", "72a6b43e4bbb818ac1875a3a572cb2aa8c9af820" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:58:43 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:58:43 2013 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull second set of s390 patches from Martin Schwidefsky:\n \"The main part of this merge are Heikos uaccess patches. Together with\n commit 09884964335e (\"mm: do not grow the stack vma just because of an\n overrun on preceding vma\") the user string access is hopefully fixed\n for good.\n\n In addition some bug fixes and two cleanup patches.\"\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:\n s390/module: fix compile warning\n qdio: remove unused parameters\n s390/uaccess: fix kernel ds access for page table walk\n s390/uaccess: fix strncpy_from_user string length check\n input: disable i8042 PC Keyboard controller for s390\n s390/dis: Fix invalid array size\n s390/uaccess: remove pointless access_ok() checks\n s390/uaccess: fix strncpy_from_user/strnlen_user zero maxlen case\n s390/uaccess: shorten strncpy_from_user/strnlen_user\n s390/dasd: fix unresponsive device after all channel paths were lost\n s390/mm: ignore change bit for vmemmap\n s390/page table dumper: add support for change-recording override bit\n" }, { "commit": "6977c6fc77b864dc1851e100c8528c2f3ed5bd9b", "tree": "d43b5e8e33a01ebf7a4f356342c700e4f46ec8e3", "parents": [ "8fd5e7a2d9574b3cac1c9264ad1aed3b613ed6fe", "cf8e18ea2f8be26e8f9e44ed677d25f558ca301a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:57:38 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:57:38 2013 -0800" }, "message": "Merge branch \u0027fixes-for-3.9-latest\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux\n\nPull second round of PARISC updates from Helge Deller:\n \"The most important fix in this branch is the switch of io_setup,\n io_getevents and io_submit syscalls to use the available compat\n syscalls when running 32bit userspace on 64bit kernel. Other than\n that it\u0027s mostly removal of compile warnings.\"\n\n* \u0027fixes-for-3.9-latest\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:\n parisc: fix redefinition of SET_PERSONALITY\n parisc: do not install modules when installing kernel\n parisc: fix compile warnings triggered by atomic_sub(sizeof(),v)\n parisc: check return value of down_interruptible() in hp_sdc_rtc.c\n parisc: avoid unitialized variable warning in pa_memcpy()\n parisc: remove unused variable \u0027compat_val\u0027\n parisc: switch to compat_functions of io_setup, io_getevents and io_submit\n parisc: select ARCH_WANT_FRAME_POINTERS\n" }, { "commit": "8fd5e7a2d9574b3cac1c9264ad1aed3b613ed6fe", "tree": "5696f5d31c6c75b71bfc4852fb234b773e266cfe", "parents": [ "529e5fbcd8d3cb48cf824ac8fde91cc80a9e985f", "c60ac31542e93499b58dcfc1e3f6550ba5b5728e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:06:09 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 12:06:09 2013 -0800" }, "message": "Merge tag \u0027metag-v3.9-rc1-v4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag\n\nPull new ImgTec Meta architecture from James Hogan:\n \"This adds core architecture support for Imagination\u0027s Meta processor\n cores, followed by some later miscellaneous arch/metag cleanups and\n fixes which I kept separate to ease review:\n\n - Support for basic Meta 1 (ATP) and Meta 2 (HTP) core architecture\n - A few fixes all over, particularly for symbol prefixes\n - A few privilege protection fixes\n - Several cleanups (setup.c includes, split out a lot of\n metag_ksyms.c)\n - Fix some missing exports\n - Convert hugetlb to use vm_unmapped_area()\n - Copy device tree to non-init memory\n - Provide dma_get_sgtable()\"\n\n* tag \u0027metag-v3.9-rc1-v4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag: (61 commits)\n metag: Provide dma_get_sgtable()\n metag: prom.h: remove declaration of metag_dt_memblock_reserve()\n metag: copy devicetree to non-init memory\n metag: cleanup metag_ksyms.c includes\n metag: move mm/init.c exports out of metag_ksyms.c\n metag: move usercopy.c exports out of metag_ksyms.c\n metag: move setup.c exports out of metag_ksyms.c\n metag: move kick.c exports out of metag_ksyms.c\n metag: move traps.c exports out of metag_ksyms.c\n metag: move irq enable out of irqflags.h on SMP\n genksyms: fix metag symbol prefix on crc symbols\n metag: hugetlb: convert to vm_unmapped_area()\n metag: export clear_page and copy_page\n metag: export metag_code_cache_flush_all\n metag: protect more non-MMU memory regions\n metag: make TXPRIVEXT bits explicit\n metag: kernel/setup.c: sort includes\n perf: Enable building perf tools for Meta\n metag: add boot time LNKGET/LNKSET check\n metag: add __init to metag_cache_probe()\n ...\n" }, { "commit": "529e5fbcd8d3cb48cf824ac8fde91cc80a9e985f", "tree": "5b6792ea3fe35ba3a2900978b1679d5d0a8bbee3", "parents": [ "686c09407d24123755ee05aae5be680be886a233", "16af43fef87512f7324205783526f543ddcf09cf" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 11:54:39 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 11:54:39 2013 -0800" }, "message": "Merge branch \u0027for-linus\u0027 of git://git.linaro.org/people/rmk/linux-arm\n\nPull late ARM updates from Russell King:\n \"Here is the late set of ARM updates for this merge window; in here is:\n\n - The ARM parts of the broadcast timer support, core parts merged\n through tglx\u0027s tree. This was left over from the previous merge to\n allow the dependency on tglx\u0027s tree to be resolved.\n\n - A fix to the VFP code which shows up on Raspberry Pi\u0027s, as well as\n fixing the fallout from a previous commit in this area.\n\n - A number of smaller fixes scattered throughout the ARM tree\"\n\n* \u0027for-linus\u0027 of git://git.linaro.org/people/rmk/linux-arm:\n ARM: Fix broken commit 0cc41e4a21d43 corrupting kernel messages\n ARM: fix scheduling while atomic warning in alignment handling code\n ARM: VFP: fix emulation of second VFP instruction\n ARM: 7656/1: uImage: Error out on build of multiplatform without LOADADDR\n ARM: 7640/1: memory: tegra_ahb_enable_smmu() depends on TEGRA_IOMMU_SMMU\n ARM: 7654/1: Preserve L_PTE_VALID in pte_modify()\n ARM: 7653/2: do not scale loops_per_jiffy when using a constant delay clock\n ARM: 7651/1: remove unused smp_timer_broadcast #define\n" }, { "commit": "68b86a25225d03f134f306caffc46df80906c3f8", "tree": "d5f6556e07e3fb376199c2ba70153c17d47b7607", "parents": [ "527c680f7c36ff17d49efc99632232dba3549c51", "41e9f3f71bc7a5d41a2b925cfdc0dc22a77f7d8c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 10:23:29 2013 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Sun Mar 03 10:23:29 2013 -0800" }, "message": "Merge git://www.linux-watchdog.org/linux-watchdog\n\nPull watchdog updates from Wim Van Sebroeck:\n \"This contains:\n - fixes and improvements\n - devicetree bindings\n - conversion to watchdog generic framework of the following drivers:\n - booke_wdt\n - bcm47xx_wdt.c\n - at91sam9_wdt\n - Removal of old STMP3xxx driver\n - Addition of following new drivers:\n - new driver for STMP3xxx and i.MX23/28\n - Retu watchdog driver\"\n\n* git://www.linux-watchdog.org/linux-watchdog: (30 commits)\n watchdog: sp805_wdt depends on ARM\n watchdog: davinci_wdt: update to devm_* API\n watchdog: davinci_wdt: use devm managed clk get\n watchdog: at91rm9200: add DT support\n watchdog: add timeout-sec property binding\n watchdog: at91sam9_wdt: Convert to use the watchdog framework\n watchdog: omap_wdt: Add option nowayout\n watchdog: core: dt: add support for the timeout-sec dt property\n watchdog: bcm47xx_wdt.c: add hard timer\n watchdog: bcm47xx_wdt.c: rename wdt_time to timeout\n watchdog: bcm47xx_wdt.c: rename ops methods\n watchdog: bcm47xx_wdt.c: use platform device\n watchdog: bcm47xx_wdt.c: convert to watchdog core api\n watchdog: Convert BookE watchdog driver to watchdog infrastructure\n watchdog: s3c2410_wdt: Use devm_* functions\n watchdog: remove old STMP3xxx driver\n watchdog: add new driver for STMP3xxx and i.MX23/28\n rtc: stmp3xxx: add wdt-accessor function\n watchdog: introduce retu_wdt driver\n watchdog: intel_scu_watchdog: fix Kconfig dependency\n ...\n" }, { "commit": "16af43fef87512f7324205783526f543ddcf09cf", "tree": "28995abd8604d61ba84d97d28448ccccb91467c9", "parents": [ "9664ffe6a16676fa4d6a238ad3d9bb6cc24825a1", "ded3ef0fa71696d3cf631fa4b8ac65313f370b4c", "653a761e4ba70bf12174dedcd5fbbf54f431abb3" ], "author": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 00:32:50 2013 +0000" }, "committer": { "name": "Russell King", "email": "rmk+kernel@arm.linux.org.uk", "time": "Sun Mar 03 00:32:50 2013 +0000" }, "message": "Merge branches \u0027devel-stable\u0027, \u0027fixes\u0027 and \u0027mmci\u0027 into for-linus\n" }, { "commit": "c60ac31542e93499b58dcfc1e3f6550ba5b5728e", "tree": "9546eac961af7a343eefc302f81f0e70d1298cf9", "parents": [ "2742c52655cff867fbf55e2e9cb993b8f965b7d7" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Feb 19 13:25:46 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:17 2013 +0000" }, "message": "metag: Provide dma_get_sgtable()\n\nmetag/allmodconfig:\n\ndrivers/media/v4l2-core/videobuf2-dma-contig.c: In function \u0027vb2_dc_get_base_sgt\u0027:\ndrivers/media/v4l2-core/videobuf2-dma-contig.c:387: error: implicit declaration of function \u0027dma_get_sgtable\u0027\n\nFor architectures using dma_map_ops, dma_get_sgtable() is provided in\n\u003casm-generic/dma-mapping-common.h\u003e.\n\nMetag does not use dma_map_ops yet, hence it should implement it as an\ninline stub using dma_common_get_sgtable().\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Geert Uytterhoeven \u003cgeert@linux-m68k.org\u003e\nAcked-by: Marek Szyprowski \u003cm.szyprowski@samsung.com\u003e\n" }, { "commit": "2742c52655cff867fbf55e2e9cb993b8f965b7d7", "tree": "e0e13e15d11f5e211fe3f43d0e6ecfc9f0a45a63", "parents": [ "2270e6d30bb6601ffd42e9d972442df0499ad547" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 20 14:19:54 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:17 2013 +0000" }, "message": "metag: prom.h: remove declaration of metag_dt_memblock_reserve()\n\nMetag doesn\u0027t have a metag_dt_memblock_reserve() function so remove the\ndeclaration from asm/prom.h.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "2270e6d30bb6601ffd42e9d972442df0499ad547", "tree": "c72e32a15d041cfe38560dca277f335da3f6e7cf", "parents": [ "d7900504480ea9d49a6b5f8811f2678ad08d3255" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 20 13:59:13 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:17 2013 +0000" }, "message": "metag: copy devicetree to non-init memory\n\nMake a copy of the device tree blob in non-init memory. It is required\nwhen using built-in device tree files that the platform code copies the\nblob to non-init memory prior to calling unflatten_device_tree(),\notherwise the strings that the device tree refer to will get poisoned\nand potentially reused, breaking later reading of the device tree\npost-init (such as compatible matching in modules, debugfs, and the\nprocfs interface).\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nReviewed-by: Vineet Gupta \u003cvgupta@synopsys.com\u003e\n" }, { "commit": "d7900504480ea9d49a6b5f8811f2678ad08d3255", "tree": "f2759833b469d4a3aa25984da9d17b4ec8ec15e6", "parents": [ "44c2451080ab3896688c852253814f5eabb2ecce" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 13:30:15 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:16 2013 +0000" }, "message": "metag: cleanup metag_ksyms.c includes\n\nMinimise metag_ksyms.c includes to directly include the \u003casm/*.h\u003e files\nthat declare a particular symbol, and not include any unnecessary ones.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "44c2451080ab3896688c852253814f5eabb2ecce", "tree": "2b7063e255da7a127cd8264ca0f392cccd01e232", "parents": [ "9da3ee9aa8a9c9fcf94188d4c0ae280afbeb63c1" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 13:19:15 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:16 2013 +0000" }, "message": "metag: move mm/init.c exports out of metag_ksyms.c\n\nIt\u0027s less error prone to have function symbols exported immediately\nafter the function rather than in metag_ksyms.c. Move each EXPORT_SYMBOL\nin metag_ksyms.c for symbols defined in mm/init.c into mm/init.c.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "9da3ee9aa8a9c9fcf94188d4c0ae280afbeb63c1", "tree": "b6c3890534b2b3ee49fda2bc5a342030378e4079", "parents": [ "7293dbed9d8be1916034dbfcf2f203e96bd8fae1" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 12:57:10 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:15 2013 +0000" }, "message": "metag: move usercopy.c exports out of metag_ksyms.c\n\nIt\u0027s less error prone to have function symbols exported immediately\nafter the function rather than in metag_ksyms.c. Move each EXPORT_SYMBOL\nin metag_ksyms.c for symbols defined in usercopy.c into usercopy.c\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "7293dbed9d8be1916034dbfcf2f203e96bd8fae1", "tree": "c8f691277091762579a264d976a72c5f9acd45f9", "parents": [ "aa29ec5f79f61ad880771d53d37a341cc7f5dfa6" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 12:55:51 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:15 2013 +0000" }, "message": "metag: move setup.c exports out of metag_ksyms.c\n\nIt\u0027s less error prone to have function symbols exported immediately\nafter the function rather than in metag_ksyms.c. Move each EXPORT_SYMBOL\nin metag_ksyms.c for symbols defined in setup.c into setup.c\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "aa29ec5f79f61ad880771d53d37a341cc7f5dfa6", "tree": "7e2aa91b2acd4a0466b1b7b38285219350ffe7dc", "parents": [ "9fb4aa8723f6b332ee3089dc59c9ff7dcf2c0d47" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 13:01:55 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:15 2013 +0000" }, "message": "metag: move kick.c exports out of metag_ksyms.c\n\nIt\u0027s less error prone to have function symbols exported immediately\nafter the function rather than in metag_ksyms.c. Move each EXPORT_SYMBOL\nin metag_ksyms.c for symbols defined in kick.c into kick.c\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "9fb4aa8723f6b332ee3089dc59c9ff7dcf2c0d47", "tree": "2c69f2ed034ca57ab8f172f75fe6c8659f9cdf62", "parents": [ "fa771d029af8f8a23089c97b6ab6a5745e98ad7e" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Wed Feb 13 12:19:03 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:14 2013 +0000" }, "message": "metag: move traps.c exports out of metag_ksyms.c\n\nIt\u0027s less error prone to have function symbols exported immediately\nafter the function rather than in metag_ksyms.c. Move each EXPORT_SYMBOL\nin metag_ksyms.c for symbols defined in traps.c into traps.c\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "fa771d029af8f8a23089c97b6ab6a5745e98ad7e", "tree": "949fbbe9278e5da6711d2da055888e82772c2a62", "parents": [ "97c3ec63089fdcd2abf66619b913900909841dc0" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Feb 12 16:04:53 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:14 2013 +0000" }, "message": "metag: move irq enable out of irqflags.h on SMP\n\nThe SMP version of arch_local_irq_enable() uses preempt_disable(), but\n\u003casm/irqflags.h\u003e doesn\u0027t include \u003clinux/preempt.h\u003e causing the following\nerrors on SMP when pstore/ftrace is enabled (caught by buildbot smp\nallyesconfig):\n\nIn file included from include/linux/irqflags.h:15,\n from fs/pstore/ftrace.c:16:\narch/metag/include/asm/irqflags.h: In function \u0027arch_local_irq_enable\u0027:\narch/metag/include/asm/irqflags.h:84: error: implicit declaration of function \u0027preempt_disable\u0027\narch/metag/include/asm/irqflags.h:86: error: implicit declaration of function \u0027preempt_enable_no_resched\u0027\n\nHowever \u003clinux/preempt.h\u003e cannot be easily included from\n\u003casm/irqflags.h\u003e as it can cause circular include dependencies in the\n!SMP case, and potentially in the SMP case in the future. Therefore move\nthe SMP implementation of arch_local_irq_enable() into traps.c and use\nan inline version of get_trigger_mask() which is also defined in traps.c\nfor SMP.\n\nThis adds an extra layer of function call / stack push when\npreempt_disable needs to call other functions, however in the\nnon-preemptive SMP case it should be about as fast, as it was already\ncalling the get_trigger_mask() function which is now used inline.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "f75c28d896f4dd0064e60bba1e82a4c98908d239", "tree": "18097d57e812bdcd86634d86b6089fc07605946a", "parents": [ "c838e72a35e49ea51c39c2c634ece028fa49c565" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Mon Feb 11 17:28:10 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:13 2013 +0000" }, "message": "metag: hugetlb: convert to vm_unmapped_area()\n\nConvert hugetlb_get_unmapped_area_new_pmd() to use vm_unmapped_area()\nrather than searching the virtual address space itself. This fixes the\nfollowing errors in linux-next due to the specified members being\nremoved after other architectures have already been converted:\n\narch/metag/mm/hugetlbpage.c: In function \u0027hugetlb_get_unmapped_area_new_pmd\u0027:\narch/metag/mm/hugetlbpage.c:199: error: \u0027struct mm_struct\u0027 has no member named \u0027cached_hole_size\u0027\narch/metag/mm/hugetlbpage.c:200: error: \u0027struct mm_struct\u0027 has no member named \u0027free_area_cache\u0027\narch/metag/mm/hugetlbpage.c:215: error: \u0027struct mm_struct\u0027 has no member named \u0027cached_hole_size\u0027\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nAcked-by: Michel Lespinasse \u003cwalken@google.com\u003e\n" }, { "commit": "c838e72a35e49ea51c39c2c634ece028fa49c565", "tree": "1fdd87314187b1e2e1f429d803154f0ecc0f75f0", "parents": [ "f626dc704e761761b37efbb6320308461f159375" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Mon Feb 11 11:47:02 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:13 2013 +0000" }, "message": "metag: export clear_page and copy_page\n\nVarious file systems use clear_page() and copy_page(), so when they\u0027re\nbuilt as modules we get build errors like the following:\n\nERROR: \"clear_page\" [fs/ntfs/ntfs.ko] undefined!\nERROR: \"copy_page\" [fs/nilfs2/nilfs2.ko] undefined!\n\nTherefore export these functions to modules from metag_ksyms.c to fix\nthe errors. This was hit by a randconfig build.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "f626dc704e761761b37efbb6320308461f159375", "tree": "c9fe9bf98a74c359a6f4790a871ce3647a7eb5b4", "parents": [ "3d6b7bb0a2c518d24bc3036f9bbb6f3fb35462c3" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Mon Feb 11 11:43:20 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:11:12 2013 +0000" }, "message": "metag: export metag_code_cache_flush_all\n\nVarious file systems indirectly use metag_code_cache_flush_all(), so\nwhen they\u0027re built as modules we get build errors like the following:\n\nERROR: \"metag_code_cache_flush_all\" [fs/xfs/xfs.ko] undefined!\n\nTherefore export this function to modules to fix the errors. This was\nhit by a randconfig build.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "3d6b7bb0a2c518d24bc3036f9bbb6f3fb35462c3", "tree": "d9f8427f6be3a3ec2882ac4c3e9ee2e0dbe9256d", "parents": [ "c787c2d62fe0c482f5fb3e5b869cd262fe69b244" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Thu Jan 31 13:42:03 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:58 2013 +0000" }, "message": "metag: protect more non-MMU memory regions\n\nRename setup_txprivext() to setup_priv() and add initialisation of some\nmore per-thread privilege protection registers:\n\n - TxPRIVSYSR: 0x04400000-0x047fffff\n 0x05000000-0x07ffffff\n 0x84000000-0x87ffffff\n - TxPIOREG: 0x02000000-0x02ffffff\n 0x04800000-0x048fffff\n - TxSYREG: 0x04000000-0x04000fff (except write fetch system event)\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "c787c2d62fe0c482f5fb3e5b869cd262fe69b244", "tree": "ad3a653e8ce81536a9d2d57976f09a2211de4d80", "parents": [ "82f0167aa4c4bbc06b5a2e1e83d252792f7c5754" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Thu Jan 31 13:27:35 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:58 2013 +0000" }, "message": "metag: make TXPRIVEXT bits explicit\n\nDefine PRIV_BITS using explicit constants from \u003casm/metag_regs.h\u003e rather\nthan with a hard coded value. This also adds a couple of missing\ndefinitions for the TXPRIVEXT priv bits for protecting writes to TXTIMER\nand the trace registers.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "82f0167aa4c4bbc06b5a2e1e83d252792f7c5754", "tree": "6d945c917963c7ea32e19fb7f3db6ba9d4120dfc", "parents": [ "1bea5b8188e98611e4d6961647809f87b023e14c" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Thu Jan 31 13:38:48 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:57 2013 +0000" }, "message": "metag: kernel/setup.c: sort includes\n\nSort includes in kernel/setup.c.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "883a635591ef25bc6a325f8a43af17ec8ecfe011", "tree": "2bc087451e14343e35cc6159cb91bfe0b5dd48f5", "parents": [ "0a38a8adc5d479c1e59a82d1fcee3bdf59b85ef3" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Thu Jan 31 11:06:03 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:56 2013 +0000" }, "message": "metag: add boot time LNKGET/LNKSET check\n\nAdd boot time check for whether LNKGET/LNKSET go through or around the\ncache. Depending on the configuration an info message (no harm), warning\n(technically wrong but no harm), or big WARN (expect failure in either\nkernel or userland) may be emitted if the behaviour is not as expected:\n\nConfiguration Hardware Response\n------------------------------------------ -------- --------\nAROUND_CACHE through pr_info\n!AROUND_CACHE \u0026\u0026 ATOMICITY_LNKGET around WARN (kernel)\n \" \u0026\u0026 !ATOMICITY_LNKGET \u0026\u0026 SMP around WARN (user)\n \" \" \u0026\u0026 !SMP around pr_warn\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "0a38a8adc5d479c1e59a82d1fcee3bdf59b85ef3", "tree": "1284cb30541ef39b7c4a5a97891a482e9ad160c1", "parents": [ "ae85ac71b7433fa974759109c4380c620258f07f" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Thu Jan 31 11:04:49 2013 +0000" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:56 2013 +0000" }, "message": "metag: add __init to metag_cache_probe()\n\nmetag_cache_probe() is only called from setup_arch(), so add the __init\nattribute to it.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "ae85ac71b7433fa974759109c4380c620258f07f", "tree": "066915c6c1f0f830c35ff12f4a84e233b6fd15da", "parents": [ "690998b629a554d8004d3129a42176afafce9fae" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Sep 21 17:38:15 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:56 2013 +0000" }, "message": "metag: Add JTAG Debug Adapter (DA) support\n\nAdd basic JTAG Debug Adapter (DA) support so that drivers which\ncommunicate with the DA can detect whether one is actually present\n(otherwise the target will halt indefinitely).\n\nThis allows the metag_da TTY driver and imgdafs filesystem driver to be\nbuilt, updates defconfigs, and sets up the metag_da console early if\nit\u0027s configured in.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "00512bdd4573674d10af1c1d60328b4b0f9dcafd", "tree": "31f0db0a3ea2887bc6b6bb17dd1e09c508d2f803", "parents": [ "903b20ad6810e05bc5f7cc038257e80463e71001" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:27:31 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:55 2013 +0000" }, "message": "metag: ftrace support\n\nAdd ftrace support for metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Frederic Weisbecker \u003cfweisbec@gmail.com\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nReviewed-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "903b20ad6810e05bc5f7cc038257e80463e71001", "tree": "592087306a334c55c00aba52ba1246da68f5f636", "parents": [ "5633004cc2498ff50a5b88d415d3746ff0c301f2" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:54:55 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:54 2013 +0000" }, "message": "metag: Perf\n\nAdd Perf support for metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Peter Zijlstra \u003ca.p.zijlstra@chello.nl\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Ingo Molnar \u003cmingo@redhat.com\u003e\nCc: Arnaldo Carvalho de Melo \u003cacme@ghostprotocols.net\u003e\n" }, { "commit": "5633004cc2498ff50a5b88d415d3746ff0c301f2", "tree": "b60265906c89e68022b8013e2aa2a19127fe3a24", "parents": [ "79f83c02944ccb3dffbc5852e63e966ea34ed63e" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:32 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:54 2013 +0000" }, "message": "metag: Build infrastructure\n\nAdd metag build infrastructure.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "1e57372eaccdd5959b8ae00bf99aa10d330f207d", "tree": "aec9b42d31ba4026b7a6e24772f5877af984e973", "parents": [ "e8de3486a4b06389d4f996eb2dda678a39f20115" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:22:14 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:52 2013 +0000" }, "message": "metag: Various other headers\n\nAdd the remaining metag header files:\n - byteorder.h, swab.h (byte order and swapping)\n - barrier.h, cpu.h. hwthread.h, processor.h (hardware thread related)\n - bug.h, elf.h, gpio.h, linkage.h, resource.h (other)\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "e8de3486a4b06389d4f996eb2dda678a39f20115", "tree": "2a72c3bf3394335fb638cb27ae187919387c94f1", "parents": [ "086e9dc0e2ca925b1b58caefd04ed2757d14790b" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 17:01:38 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:52 2013 +0000" }, "message": "metag: Stack unwinding\n\nAdd stack unwinding support for metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "086e9dc0e2ca925b1b58caefd04ed2757d14790b", "tree": "cfe182f1d07d40d09d2cae09e337423462250cac", "parents": [ "f507758ccbed5c354cc1ce3b8f53ea072d7bc222" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 17:02:09 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:52 2013 +0000" }, "message": "metag: Optimised library functions\n\nAdd optimised library functions for metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "f507758ccbed5c354cc1ce3b8f53ea072d7bc222", "tree": "dd474b63b194039b5c6c97790016f55a02a93643", "parents": [ "42682c6c42a5765b2c7cccfca170368fef6191ef" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:27:03 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:51 2013 +0000" }, "message": "metag: DMA\n\nAdd DMA mapping code.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "42682c6c42a5765b2c7cccfca170368fef6191ef", "tree": "adcfa5e96cd98527ee75fc541efc279357bbe6a2", "parents": [ "fdabf525b4b7aab3945c19eac39d3a65b68d0c4f" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:56:56 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:51 2013 +0000" }, "message": "metag: SMP support\n\nAdd SMP support for metag. This allows Linux to take control of multiple\nhardware threads on a single Meta core, treating them as separate Linux\nCPUs.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "6006c0d8ce9441dd1363bf14f18a8e28d3588460", "tree": "786183053c89e11b3058b8a16f7953744b819340", "parents": [ "9b802d1f43978869fcd98e92b854fd8785cefee7" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 11:00:24 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:50 2013 +0000" }, "message": "metag: Atomics, locks and bitops\n\nAdd header files to implement Meta hardware thread locks (used by some\nother atomic operations), atomics, spinlocks, and bitops.\n\nThere are 2 main types of atomic primitives for metag (in addition to\nIRQs off on UP):\n - LOCK instructions provide locking between hardware threads.\n - LNKGET/LNKSET instructions provide load-linked/store-conditional\n operations allowing for lighter weight atomics on Meta2\n\nLOCK instructions allow for hardware threads to acquire voluntary or\nexclusive hardware thread locks:\n - LOCK0 releases exclusive and voluntary lock from the running hardware\n thread.\n - LOCK1 acquires the voluntary hardware lock, blocking until it becomes\n available.\n - LOCK2 implies LOCK1, and additionally acquires the exclusive hardware\n lock, blocking all other hardware threads from executing.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "9b802d1f43978869fcd98e92b854fd8785cefee7", "tree": "67bc3e7d033cab101dd1cdaa00500b2787572376", "parents": [ "44dea393cf98a09b4b9f00dc3dd7e2c211f4b0e8" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:55:00 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:49 2013 +0000" }, "message": "metag: Module support\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "44dea393cf98a09b4b9f00dc3dd7e2c211f4b0e8", "tree": "a2d5aed0af6fcf05b9ea882a4915ce446cffb6a4", "parents": [ "26025bbfbba33a9425be1b89eccb4664ea4c17b6" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:54 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:49 2013 +0000" }, "message": "metag: Scheduling/Process management\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "26025bbfbba33a9425be1b89eccb4664ea4c17b6", "tree": "f8a7f754525812c4d758b638d9bcade297af67da", "parents": [ "5698c50d9da4ab2f57d98c64ea97675dcaf2a608" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:51 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:49 2013 +0000" }, "message": "metag: System Calls\n\nAdd metag system call and gateway page interfaces. The metag\narchitecture port uses the generic system call numbers from\nasm-generic/unistd.h, as well as a user gateway page mapped at\n0x6ffff000 which contains fast atomic primitives (depending on SMP) and\na fast method of accessing TLS data.\n\nSystem calls use the SWITCH instruction with the immediate 0x440001 to\nsignal a system call.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "5698c50d9da4ab2f57d98c64ea97675dcaf2a608", "tree": "5b82a24daa8bbb85703489340cbfdcb00a2f20a4", "parents": [ "63047ea36070d11f902ab7d09a5a18aea037c0f7" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:47 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:48 2013 +0000" }, "message": "metag: Internal and external irqchips\n\nMeta core internal interrupts (from HWSTATMETA and friends) are vectored\nonto the TR1 core trigger for the current thread. This is demultiplexed\nin irq-metag.c to individual Linux IRQs for each internal interrupt.\n\nExternal SoC interrupts (from HWSTATEXT and friends) are vectored onto\nthe TR2 core trigger for the current thread. This is demultiplexed in\nirq-metag-ext.c to individual Linux IRQs for each external SoC interrupt.\nThe external irqchip has devicetree bindings for configuring the number\nof irq banks and the type of masking available.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Grant Likely \u003cgrant.likely@secretlab.ca\u003e\nCc: Rob Herring \u003crob.herring@calxeda.com\u003e\nCc: Rob Landley \u003crob@landley.net\u003e\nCc: Dom Cobley \u003cpopcornmix@gmail.com\u003e\nCc: Simon Arlott \u003csimon@fire.lp0.eu\u003e\nCc: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nCc: Maxime Ripard \u003cmaxime.ripard@free-electrons.com\u003e\nCc: devicetree-discuss@lists.ozlabs.org\nCc: linux-doc@vger.kernel.org\n" }, { "commit": "63047ea36070d11f902ab7d09a5a18aea037c0f7", "tree": "f82e359be810c8b747ed390fc942e8a3d91226f3", "parents": [ "ac919f0883e53d7785745566692c8a0620abd7ea" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:47 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:48 2013 +0000" }, "message": "metag: IRQ handling\n\nAdd core IRQ handling for metag. The code in irq.c exposes the TBX\nsignal numbers as Linux IRQs.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "ac919f0883e53d7785745566692c8a0620abd7ea", "tree": "521a0b0ddcab5176a1998d0b7b9faefde6e3f0ae", "parents": [ "a2c5d4ed92bbc02ff4a37efc2adffe7d145abe4f" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:43 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:45 2013 +0000" }, "message": "metag: Traps\n\nAdd trap code for metag. At the lowest level Meta traps (and return from\ninterrupt instruction - RTI) simply swap the PC and PCX registers and\noptionally toggle the interrupt status bit (ISTAT). Low level TBX code\nin tbipcx.S handles the core context save, determine the TBX signal\nnumber based on the core trigger that fired (using the TXSTATI status\nregister), and call TBX signal handlers (mostly in traps.c) via a vector\ntable.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "a2c5d4ed92bbc02ff4a37efc2adffe7d145abe4f", "tree": "21fc65e4f0b04928025565f208a410a7a64ab523", "parents": [ "bc3966bf1583a6c22b76397535174445c43952de" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:39 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:22 2013 +0000" }, "message": "metag: Time keeping\n\nAdd time keeping code for metag. Meta hardware threads have 2 timers.\nThe background timer (TXTIMER) is used as a free-running time base, and\nthe interrupt timer (TXTIMERI) is used for the timer interrupt. Both\ncounters traditionally count at approximately 1MHz.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: John Stultz \u003cjohnstul@us.ibm.com\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\n" }, { "commit": "bc3966bf1583a6c22b76397535174445c43952de", "tree": "8854b33e95c672078bb32fe878f39b9d716accb5", "parents": [ "29dd78cf0b526d24063364a8c634b3e92514c1a2" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:36 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:22 2013 +0000" }, "message": "metag: ptrace\n\nThe ptrace interface for metag provides access to some core register\nsets using the PTRACE_GETREGSET and PTRACE_SETREGSET operations. The\ndetails of the internal context structures is abstracted into user API\nstructures to both ease use and allow flexibility to change the internal\ncontext layouts. Copyin and copyout functions for these register sets\nare exposed to allow signal handling code to use them to copy to and\nfrom the signal context.\n\nstruct user_gp_regs (NT_PRSTATUS) provides access to the core general\npurpose register context.\n\nstruct user_cb_regs (NT_METAG_CBUF) provides access to the TXCATCH*\nregisters which contains information abuot a memory fault, unaligned\naccess error or watchpoint. This can be modified to alter the way the\nfault is replayed on resume (\"catch replay\"), or to prevent the replay\ntaking place.\n\nstruct user_rp_state (NT_METAG_RPIPE) provides access to the state of\nthe Meta read pipeline which can be used to hide memory latencies in\nhand optimised data loops.\n\nExtended DSP register state, DSP RAM, and hardware breakpoint registers\naren\u0027t yet exposed through ptrace.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nCc: Denys Vlasenko \u003cvda.linux@googlemail.com\u003e\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Tony Lindgren \u003ctony@atomide.com\u003e\nCc: \"Paul E. McKenney\" \u003cpaulmck@linux.vnet.ibm.com\u003e\n" }, { "commit": "29dd78cf0b526d24063364a8c634b3e92514c1a2", "tree": "4dfa291da2d101242f5536be0146cba25e398838", "parents": [ "262d96b0deb44ed58823447825d6efa5dddb4108" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Fri Oct 05 16:02:12 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:22 2013 +0000" }, "message": "metag: Device tree\n\nAdd device tree files to arch/metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nReviewed-by: Vineet Gupta \u003cvgupta@synopsys.com\u003e\n" }, { "commit": "262d96b0deb44ed58823447825d6efa5dddb4108", "tree": "03712555691c86256a409ceccf529f5bb7ebb37c", "parents": [ "c438b58e65462cfff172b396d03d6bc45c971fca" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:27 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:21 2013 +0000" }, "message": "metag: Signal handling\n\nAdd signal handling code for metag.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\nCc: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n" }, { "commit": "c438b58e65462cfff172b396d03d6bc45c971fca", "tree": "3d8ea3cfbe75f4e6b41f5f41f36c889df5b0f7d8", "parents": [ "bbc17704d5d3a8e4325dae74c2a2e77218c26057" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:17 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:21 2013 +0000" }, "message": "metag: TCM support\n\nAdd some TCM support\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "bbc17704d5d3a8e4325dae74c2a2e77218c26057", "tree": "fe13881c1f85fa781e3ac413216954dd946fafae", "parents": [ "e624e95bd88f94fc70bbe612789bcac44c6f0923" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:17 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:20 2013 +0000" }, "message": "metag: Highmem support\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "e624e95bd88f94fc70bbe612789bcac44c6f0923", "tree": "04fd5485d400ca9fa3aedcbbf61a81e9a434dc72", "parents": [ "373cd784d0fc83f076c899ca7da50ecca7286e42" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:17 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:20 2013 +0000" }, "message": "metag: Huge TLB\n\nAdd huge TLB support to the metag architecture.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "373cd784d0fc83f076c899ca7da50ecca7286e42", "tree": "46165054cae3e5444a02ddd071e1979aa029fdf6", "parents": [ "f5df8e268f749987c32c7eee001f7623fd7be69c" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:17 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:19 2013 +0000" }, "message": "metag: Memory handling\n\nMeta has instructions for accessing:\n - bytes - GETB (1 byte)\n - words - GETW (2 bytes)\n - doublewords - GETD (4 bytes)\n - longwords - GETL (8 bytes)\n\nAll accesses must be aligned. Unaligned accesses can be detected and\nmade to fault on Meta2, however it isn\u0027t possible to fix up unaligned\nwrites so we don\u0027t bother fixing up reads either.\n\nThis patch adds metag memory handling code including:\n - I/O memory (io.h, ioremap.c): Actually any virtual memory can be\n accessed with these helpers. A part of the non-MMUable address space\n is used for memory mapped I/O. The ioremap() function is implemented\n one to one for non-MMUable addresses.\n - User memory (uaccess.h, usercopy.c): User memory is directly\n accessible from privileged code.\n - Kernel memory (maccess.c): probe_kernel_write() needs to be\n overwridden to use the I/O functions when doing a simple aligned\n write to non-writecombined memory, otherwise the write may be split\n by the generic version.\n\nNote that due to the fact that a portion of the virtual address space is\nnon-MMUable, and therefore always maps directly to the physical address\nspace, metag specific I/O functions are made available (metag_in32,\nmetag_out32 etc). These cast the address argument to a pointer so that\nthey can be used with raw physical addresses. These accessors are only\nto be used for accessing fixed core Meta architecture registers in the\nnon-MMU region, and not for any SoC/peripheral registers.\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" }, { "commit": "f5df8e268f749987c32c7eee001f7623fd7be69c", "tree": "eb727e5917b4f4e4e0de0ec085fd36a11aa8d898", "parents": [ "99ef7c2ac1e3b01f532bfdebbe92e9960e95bebc" ], "author": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Tue Oct 09 10:54:17 2012 +0100" }, "committer": { "name": "James Hogan", "email": "james.hogan@imgtec.com", "time": "Sat Mar 02 20:09:19 2013 +0000" }, "message": "metag: Memory management\n\nAdd memory management files for metag.\n\nMeta\u0027s 32bit virtual address space is split into two halves:\n - local (0x08000000-0x7fffffff): traditionally local to a hardware\n thread and incoherent between hardware threads. Each hardware thread\n has it\u0027s own local MMU table. On Meta2 the local space can be\n globally coherent (GCOn) if the cache partitions coincide.\n - global (0x88000000-0xffff0000): coherent and traditionally global\n between hardware threads. On Meta2, each hardware thread has it\u0027s own\n global MMU table.\n\nThe low 128MiB of each half is non-MMUable and maps directly to the\nphysical address space:\n - 0x00010000-0x07ffffff: contains Meta core registers and maps SoC bus\n - 0x80000000-0x87ffffff: contains low latency global core memories\n\nLinux usually further splits the local virtual address space like this:\n - 0x08000000-0x3fffffff: user mappings\n - 0x40000000-0x7fffffff: kernel mappings\n\nSigned-off-by: James Hogan \u003cjames.hogan@imgtec.com\u003e\n" } ], "next": "99ef7c2ac1e3b01f532bfdebbe92e9960e95bebc" }