)]}' { "log": [ { "commit": "90035fe378c7459ba19c43c63d5f878284224ce4", "tree": "e6daaaf5d9b0e2d077105e9263399f5f09f2c633", "parents": [ "9cb82f2f4692293a27c578c3038518ce4477de72" ], "author": { "name": "Tony Breeds", "email": "tony@bakeyournoodle.com", "time": "Thu Apr 24 13:43:49 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Apr 24 20:58:04 2008 +1000" }, "message": "[POWERPC] Raise the upper limit of NR_CPUS and move the pacas into the BSS\n\nThis adds the required functionality to fill in all pacas at runtime.\n\nWith NR_CPUS\u003d1024\ntext data bss dec hex filename\n 137 1704032 0 1704169 1a00e9 arch/powerpc/kernel/paca.o :Before\n 121 1179744 524288 1704153 1a00d9 arch/powerpc/kernel/paca.o :After\n\nAlso remove unneeded #includes from arch/powerpc/kernel/paca.c\n\nSigned-off-by: Tony Breeds \u003ctony@bakeyournoodle.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "30ff2e87ed55e83b4eb436f5f14a7e49ff81ad99", "tree": "9e1af946cb836c9af2059d4f8b41aa89dd3b780e", "parents": [ "3eb9cf076180ed2003db77bd2c33ac4ed0211089" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Thu Apr 10 16:43:47 2008 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Apr 15 21:21:25 2008 +1000" }, "message": "[POWERPC] iSeries: Make iseries_reg_save private to iSeries\n\nNow that we have the alpaca, the reg_save_ptr is no longer needed in the\npaca. Eradicate all global uses of it and make it static in the iSeries\nlpardata.c\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "06b8e878a9bc9301201cffe186eba99c4185f20a", "tree": "857434ed559cdb001177e81283be6f0b5693a781", "parents": [ "d9afa43532adf8a31b93c4c7601fda3f423d8972" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Wed Feb 06 01:36:12 2008 -0800" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Wed Feb 06 10:41:00 2008 -0800" }, "message": "taskstats scaled time cleanup\n\nThis moves the ability to scale cputime into generic code. This allows us\nto fix the issue in kernel/timer.c (noticed by Balbir) where we could only\nadd an unscaled value to the scaled utime/stime.\n\nThis adds a cputime_to_scaled function. As before, the POWERPC version\ndoes the scaling based on the last SPURR/PURR ratio calculated. The\ngeneric and s390 (only other arch to implement asm/cputime.h) versions are\nboth NOPs.\n\nAlso moves the SPURR and PURR snapshots closer.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCc: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "568e0c1119d7ff33fbcddc0dec9e2c799a53352d", "tree": "fea28b6bf1b3980f7a23d1bf721056f87636ac60", "parents": [ "9aa5e993fa0f1a8c2c900bdfb80697d77b6dc223" ], "author": { "name": "Julio M. Merino Vidal", "email": "jmerino@ac.upc.edu", "time": "Sat Oct 20 00:36:33 2007 +0200" }, "committer": { "name": "Adrian Bunk", "email": "bunk@kernel.org", "time": "Sat Oct 20 00:36:33 2007 +0200" }, "message": "powerpc: Remove obsolete prototype\n\nThe include/asm-powerpc/paca.h file has a prototype for a function that\ndoes not exist any more; its name is setup_boot_paca. This function was\nremoved in commit 4ba99b97dadd35b9ce1438b2bc7c992a4a14a8b1, so its\nprototype should have been removed at that time too.\n\nSigned-off-by: Julio M. Merino Vidal \u003cjmerino@ac.upc.edu\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@kernel.org\u003e\n" }, { "commit": "4603ac180a824197c2262747948d0179eb076e9c", "tree": "975f9a88b95f69e1099ad22c58717b182e6416fc", "parents": [ "f494f8fcb18e962839a82d542fbe6bed9cf5be07" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Thu Oct 18 03:06:37 2007 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@woody.linux-foundation.org", "time": "Thu Oct 18 14:37:28 2007 -0700" }, "message": "powerpc: add scaled time accounting\n\nThis adds POWERPC specific hooks for scaled time accounting.\n\nPOWER6 includes a SPURR register. The SPURR is based off the PURR register\nbut is scaled based on CPU frequency and issue rates. This gives a more\naccurate account of the instructions used per task. The PURR and timebase\nwill be constant relative to the wall clock, irrespective of the CPU\nfrequency.\n\nThis implementation reads the SPURR register in account_system_vtime which\nis only call called on context witch and hard and soft irq entry and exit.\nThe percentage of user and system time is then estimated using the ratio of\nthese accounted by the PURR. If the SPURR is not present, the PURR read.\n\nAn earlier implementation of this patch read the SPURR whenever the PURR\nwas read, which included the system call entry and exit path.\nUnfortunately this showed a performance regression on lmbench runs, so was\nre-implemented.\n\nI\u0027ve included the lmbench results here when run bare metal on POWER6. 1st\ncolumn is the unpatch results. 2nd column is the results using the below\npatch and the 3rd is the % diff of these results from the base. 4th and\n5th columns are the results and % differnce from the base using the older\npatch (SPURR read in syscall entry/exit path).\n\n Base Scaled-Acct SPURR-in-syscall\n Result Result % diff Result % diff\nSimple syscall: 0.3086 0.3086 0.0000 0.3452 11.8600\nSimple read: 0.4591 0.4671 1.7425 0.5044 9.86713\nSimple write: 0.4364 0.4366 0.0458 0.4731 8.40971\nSimple stat: 2.0055 2.0295 1.1967 2.0669 3.06158\nSimple fstat: 0.5962 0.5876 -1.442 0.6368 6.80979\nSimple open/close: 3.1283 3.1009 -0.875 3.2088 2.57328\nSelect on 10 fd\u0027s: 0.8554 0.8457 -1.133 0.8667 1.32101\nSelect on 100 fd\u0027s: 3.5292 3.6329 2.9383 3.6664 3.88756\nSelect on 250 fd\u0027s: 7.9097 8.1881 3.5197 8.2242 3.97613\nSelect on 500 fd\u0027s: 15.2659 15.836 3.7357 15.873 3.97814\nSelect on 10 tcp fd\u0027s: 0.9576 0.9416 -1.670 0.9752 1.83792\nSelect on 100 tcp fd\u0027s: 7.248 7.2254 -0.311 7.2685 0.28283\nSelect on 250 tcp fd\u0027s: 17.7742 17.707 -0.375 17.749 -0.1406\nSelect on 500 tcp fd\u0027s: 35.4258 35.25 -0.496 35.286 -0.3929\nSignal handler installation: 0.6131 0.6075 -0.913 0.647 5.52927\nSignal handler overhead: 2.0919 2.1078 0.7600 2.1831 4.35967\nProtection fault: 0.7345 0.7478 1.8107 0.8031 9.33968\nPipe latency: 33.006 16.398 -50.31 33.475 1.42368\nAF_UNIX sock stream latency: 14.5093 30.910 113.03 30.715 111.692\nProcess fork+exit: 219.8 222.8 1.3648 229.37 4.35623\nProcess fork+execve: 876.14 873.28 -0.32 868.66 -0.8533\nProcess fork+/bin/sh -c: 2830 2876.5 1.6431 2958 4.52296\nFile /var/tmp/XXX write bw: 1193497 1195536 0.1708 118657 -0.5799\nPagefaults on /var/tmp/XXX: 3.1272 3.2117 2.7020 3.2521 3.99398\n\nAlso, kernel compile times show no difference with this patch applied.\n\n[pbadari@us.ibm.com: Avoid unnecessary PURR reading]\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nCc: Balbir Singh \u003cbalbir@in.ibm.com\u003e\nCc: Jay Lan \u003cjlan@engr.sgi.com\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Badari Pulavarty \u003cpbadari@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "048c8bc90e53bf1f5feec020a7d482da94894e93", "tree": "6ca0891ac3320706873006104cde3d4e1039bcfc", "parents": [ "5669c3cf19fbadaa9120b59914beec8431277efe" ], "author": { "name": "Hugh Dickins", "email": "hugh@veritas.com", "time": "Wed Nov 01 05:44:54 2006 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Oct 03 11:48:44 2007 +1000" }, "message": "[POWERPC] ppc64: support CONFIG_DEBUG_PREEMPT\n\nAdd CONFIG_DEBUG_PREEMPT support to ppc64: it was useful for testing\nget_paca() preemption. Cheat a little, just use debug_smp_processor_id()\nin the debug version of get_paca(): it contains all the right checks and\nreporting, though get_paca() doesn\u0027t really use smp_processor_id().\n\nUse local_paca for what might have been called __raw_get_paca().\nSilence harmless warnings from io.h and lparcfg.c with local_paca -\nit is okay for iseries_lparcfg_data to be referencing shared_proc\nwith preemption enabled: all cpus should show the same value for\nshared_proc.\n\nWhy do other architectures need TRACE_IRQFLAGS_SUPPORT for DEBUG_PREEMPT?\nI don\u0027t know, ppc64 appears to get along fine without it.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "d0f13e3c20b6fb73ccb467bdca97fa7cf5a574cd", "tree": "a2de01a21dbb28449893102742e6b516a519c03e", "parents": [ "16f1c746755836aa823658000493cdab8ce7b098" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Tue May 08 16:27:27 2007 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed May 09 16:35:00 2007 +1000" }, "message": "[POWERPC] Introduce address space \"slices\"\n\nThe basic issue is to be able to do what hugetlbfs does but with\ndifferent page sizes for some other special filesystems; more\nspecifically, my need is:\n\n - Huge pages\n\n - SPE local store mappings using 64K pages on a 4K base page size\nkernel on Cell\n\n - Some special 4K segments in 64K-page kernels for mapping a dodgy\ntype of powerpc-specific infiniband hardware that requires 4K MMU\nmappings for various reasons I won\u0027t explain here.\n\nThe main issues are:\n\n - To maintain/keep track of the page size per \"segment\" (as we can\nonly have one page size per segment on powerpc, which are 256MB\ndivisions of the address space).\n\n - To make sure special mappings stay within their allotted\n\"segments\" (including MAP_FIXED crap)\n\n - To make sure everybody else doesn\u0027t mmap/brk/grow_stack into a\n\"segment\" that is used for a special mapping\n\nSome of the necessary mechanisms to handle that were present in the\nhugetlbfs code, but mostly in ways not suitable for anything else.\n\nThe patch relies on some changes to the generic get_unmapped_area()\nthat just got merged. It still hijacks hugetlb callbacks here or\nthere as the generic code hasn\u0027t been entirely cleaned up yet but\nthat shouldn\u0027t be a problem.\n\nSo what is a slice ? Well, I re-used the mechanism used formerly by our\nhugetlbfs implementation which divides the address space in\n\"meta-segments\" which I called \"slices\". The division is done using\n256MB slices below 4G, and 1T slices above. Thus the address space is\ndivided currently into 16 \"low\" slices and 16 \"high\" slices. (Special\ncase: high slice 0 is the area between 4G and 1T).\n\nDoing so simplifies significantly the tracking of segments and avoids\nhaving to keep track of all the 256MB segments in the address space.\n\nWhile I used the \"concepts\" of hugetlbfs, I mostly re-implemented\neverything in a more generic way and \"ported\" hugetlbfs to it.\n\nSlices can have an associated page size, which is encoded in the mmu\ncontext and used by the SLB miss handler to set the segment sizes. The\nhash code currently doesn\u0027t care, it has a specific check for hugepages,\nthough I might add a mechanism to provide per-slice hash mapping\nfunctions in the future.\n\nThe slice code provide a pair of \"generic\" get_unmapped_area() (bottomup\nand topdown) functions that should work with any slice size. There is\nsome trickiness here so I would appreciate people to have a look at the\nimplementation of these and let me know if I got something wrong.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "687304014f7ca8e2fbb3feaefef356b4a0da65ad", "tree": "2212a33688e74d8c8bef0affe67cabbf5a14e03c", "parents": [ "6cfef5b27e49e826125f12637ee0d7210a896044" ], "author": { "name": "Olof Johansson", "email": "olof@lixom.net", "time": "Tue Apr 24 01:11:55 2007 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Apr 24 22:06:59 2007 +1000" }, "message": "[POWERPC] Save trap number in bad_stack\n\nSave the trap number in the case of getting a bad stack in an exception\nhandler. It is sometimes useful to know what exception it was that caused\nthis to happen. Without this, no trap number is reported.\n\nSigned-off-by: Olof Johansson \u003colof@lixom.net\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "e91948fd84086020072e022d5463036033d449c1", "tree": "542886b186bad826c36664261c7a3427107e2eae", "parents": [ "0e6806734fd861c360ecbb4262d3d5678cea7faf" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Fri Mar 16 17:47:07 2007 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Mar 21 22:35:51 2007 +1100" }, "message": "[POWERPC] Minor paca optimisation\n\nMove the slb_shadow_ptr field into the first cache line since it is\n(like everything there) read-only after boot. It is in fact statically\ninitialised and thereafter only read.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nAcked-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "d04c56f73c30a5e593202ecfcf25ed43d42363a2", "tree": "bec1208293c904b73f12dd9179ebf8e88dbdb085", "parents": [ "284a940675a64df253e3dffc60b09bb4bbb149e4" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Oct 04 16:47:49 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Oct 16 16:31:36 2006 +1000" }, "message": "[POWERPC] Lazy interrupt disabling for 64-bit machines\n\nThis implements a lazy strategy for disabling interrupts. This means\nthat local_irq_disable() et al. just clear the \u0027interrupts are\nenabled\u0027 flag in the paca. If an interrupt comes along, the interrupt\nentry code notices that interrupts are supposed to be disabled, and\nclears the EE bit in SRR1, clears the \u0027interrupts are hard-enabled\u0027\nflag in the paca, and returns. This means that interrupts only\nactually get disabled in the processor when an interrupt comes along.\n\nWhen interrupts are enabled by local_irq_enable() et al., the code\nsets the interrupts-enabled flag in the paca, and then checks whether\ninterrupts got hard-disabled. If so, it also sets the EE bit in the\nMSR to hard-enable the interrupts.\n\nThis has the potential to improve performance, and also makes it\neasier to make a kernel that can boot on iSeries and on other 64-bit\nmachines, since this lazy-disable strategy is very similar to the\nsoft-disable strategy that iSeries already uses.\n\nThis version renames paca-\u003eproc_enabled to paca-\u003esoft_enabled, and\nchanges a couple of soft-disables in the kexec code to hard-disables,\nwhich should fix the crash that Michael Ellerman saw. This doesn\u0027t\nyet use a reserved CR field for the soft_enabled and hard_enabled\nflags. This applies on top of Stephen Rothwell\u0027s patches to make it\npossible to build a combined iSeries/other kernel.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c547fc28ab3e8716076fdaf4bd0260c5d63a18f7", "tree": "34af1fa64a63618660187ae58ad182665a1861ef", "parents": [ "3dd836a56de0d4f049438412959b905e1db4666e", "63b98080daa35f0d682db04f4fb7ada010888752" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Sep 14 07:07:18 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Sep 14 07:07:18 2006 +1000" }, "message": "Merge branch \u0027linux-2.6\u0027\n" }, { "commit": "f007cacffc8870702a1473d83ba5e4922d54e17c", "tree": "7faa1dbd7ccd2c4536f29852e0fedf7499d90508", "parents": [ "2e8e8dacc566cc91cd8707cb092e76c7bbfab178" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Sep 13 22:08:26 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Sep 13 22:08:26 2006 +1000" }, "message": "[POWERPC] Fix MMIO ops to provide expected barrier behaviour\n\nThis changes the writeX family of functions to have a sync instruction\nbefore the MMIO store rather than after, because the generally expected\nbehaviour is that the device receiving the MMIO store can be guaranteed\nto see the effects of any preceding writes to normal memory.\n\nTo preserve ordering between writeX and readX, and to preserve ordering\nbetween preceding stores and the readX, the readX family of functions\nhave had an sync added before the load.\n\nAlthough writeX followed by spin_unlock is not officially guaranteed\nto keep the writeX inside the spin-locked region unless an mmiowb()\nis used, there are currently drivers that depend on the previous\nbehaviour on powerpc, which was that the mmiowb wasn\u0027t actually required.\nTherefore we have a per-cpu flag that is set by writeX, cleared by\n__raw_spin_lock and mmiowb, and tested by __raw_spin_unlock. If it is\nset, __raw_spin_unlock does a sync and clears it.\n\nThis changes both 32-bit and 64-bit readX/writeX. 32-bit already has a\nsync in __raw_spin_unlock (since lwsync doesn\u0027t exist on 32-bit), and thus\ndoesn\u0027t need the per-cpu flag.\n\nTested on G5 (PPC970) and POWER5.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e" }, { "commit": "2f6093c84730b4bad65bcd0f2f904a5769b1dfc5", "tree": "ab4e64a0520e944062f418e91706ff968e23a6ea", "parents": [ "452b5e21216011f2f068e80443568f5f3f3f4d63" ], "author": { "name": "Michael Neuling", "email": "mikey@neuling.org", "time": "Mon Aug 07 16:19:19 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Tue Aug 08 17:08:56 2006 +1000" }, "message": "[POWERPC] Implement SLB shadow buffer\n\nThis adds a shadow buffer for the SLBs and regsiters it with PHYP.\nOnly the bolted SLB entries (top 3) are shadowed.\n\nThe SLB shadow buffer tells the hypervisor what the kernel needs to\nhave in the SLB for the kernel to be able to function. The hypervisor\ncan use this information to speed up partition context switches.\n\nSigned-off-by: Michael Neuling \u003cmikey@neuling.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "45c091bb2d453ce4a8b06cf19872ec7a77fc4799", "tree": "06fb2e05518ebfba163f8424e028e7faf5672d66", "parents": [ "d588fcbe5a7ba8bba2cebf7799ab2d573717a806", "2191fe3e39159e3375f4b7ec1420df149f154101" ], "author": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jun 22 22:11:30 2006 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Thu Jun 22 22:11:30 2006 -0700" }, "message": "Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc\n\n* git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: (139 commits)\n [POWERPC] re-enable OProfile for iSeries, using timer interrupt\n [POWERPC] support ibm,extended-*-frequency properties\n [POWERPC] Extra sanity check in EEH code\n [POWERPC] Dont look for class-code in pci children\n [POWERPC] Fix mdelay badness on shared processor partitions\n [POWERPC] disable floating point exceptions for init\n [POWERPC] Unify ppc syscall tables\n [POWERPC] mpic: add support for serial mode interrupts\n [POWERPC] pseries: Print PCI slot location code on failure\n [POWERPC] spufs: one more fix for 64k pages\n [POWERPC] spufs: fail spu_create with invalid flags\n [POWERPC] spufs: clear class2 interrupt status before wakeup\n [POWERPC] spufs: fix Makefile for \"make clean\"\n [POWERPC] spufs: remove stop_code from struct spu\n [POWERPC] spufs: fix spu irq affinity setting\n [POWERPC] spufs: further abstract priv1 register access\n [POWERPC] spufs: split the Cell BE support into generic and platform dependant parts\n [POWERPC] spufs: dont try to access SPE channel 1 count\n [POWERPC] spufs: use kzalloc in create_spu\n [POWERPC] spufs: fix initial state of wbox file\n ...\n\nManually resolved conflicts in:\n\tdrivers/net/phy/Makefile\n\tinclude/asm-powerpc/spu.h\n" }, { "commit": "bf72aeba2ffef599d1d386425c9e46b82be657cd", "tree": "ead8e5111dbcfa22e156999d1bb8a96e50f06fef", "parents": [ "31925323b1b51bb65db729e029472a8b1f635b7d" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jun 15 10:45:18 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jun 15 10:45:18 2006 +1000" }, "message": "powerpc: Use 64k pages without needing cache-inhibited large pages\n\nSome POWER5+ machines can do 64k hardware pages for normal memory but\nnot for cache-inhibited pages. This patch lets us use 64k hardware\npages for most user processes on such machines (assuming the kernel\nhas been configured with CONFIG_PPC_64K_PAGES\u003dy). User processes\nstart out using 64k pages and get switched to 4k pages if they use any\nnon-cacheable mappings.\n\nWith this, we use 64k pages for the vmalloc region and 4k pages for\nthe imalloc region. If anything creates a non-cacheable mapping in\nthe vmalloc region, the vmalloc region will get switched to 4k pages.\nI don\u0027t know of any driver other than the DRM that would do this,\nthough, and these machines don\u0027t have AGP.\n\nWhen a region gets switched from 64k pages to 4k pages, we do not have\nto clear out all the 64k HPTEs from the hash table immediately. We\nuse the _PAGE_COMBO bit in the Linux PTE to indicate whether the page\nwas hashed in as a 64k page or a set of 4k pages. If hash_page is\ntrying to insert a 4k page for a Linux PTE and it sees that it has\nalready been inserted as a 64k page, it first invalidates the 64k HPTE\nbefore inserting the 4k HPTE. The hash invalidation routines also use\nthe _PAGE_COMBO bit, to determine whether to look for a 64k HPTE or a\nset of 4k HPTEs to remove. With those two changes, we can tolerate a\nmix of 4k and 64k HPTEs in the hash table, and they will all get\nremoved when the address space is torn down.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "430644312810645a6e05855db50a978df9ba3ad3", "tree": "2363ddbb95af2b3df95c0071060d9c69044e28ba", "parents": [ "7a0c58d0513c246ac5438ef4a55ce8b93395ae0e" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 12 18:38:21 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jun 12 18:38:21 2006 +1000" }, "message": "powerpc: Remove unused paca-\u003epgdir field\n\nThe pgdir field in the paca was a leftover from the dynamic VSIDs\npatch, and is not used in the current kernel code. This removes it.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "62c4f0a2d5a188f73a94f2cb8ea0dba3e7cf0a7f", "tree": "e85ca2d0dd43f90dccf758338764c3caa55f333f", "parents": [ "089f26d5e31b7bf42a9a8fefec08b30cd27f4b0e" ], "author": { "name": "David Woodhouse", "email": "dwmw2@infradead.org", "time": "Wed Apr 26 12:56:16 2006 +0100" }, "committer": { "name": "David Woodhouse", "email": "dwmw2@infradead.org", "time": "Wed Apr 26 12:56:16 2006 +0100" }, "message": "Don\u0027t include linux/config.h from anywhere else in include/\n\nSigned-off-by: David Woodhouse \u003cdwmw2@infradead.org\u003e\n" }, { "commit": "4df20460a3ff0d60280738b094945c56cb5567a5", "tree": "39f831cf5f778b14a08b3453a8f798b2b8be8813", "parents": [ "b4f382a3e5e20ba867e7aa5b01189a3fd40eea2c" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Sat Mar 25 17:25:17 2006 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Mar 27 14:48:48 2006 +1100" }, "message": "[PATCH] powerpc: Allow non zero boot cpuids\n\nWe currently have a hack to flip the boot cpu and its secondary thread\nto logical cpuid 0 and 1. This means the logical - physical mapping will\ndiffer depending on which cpu is boot cpu. This is most apparent on\nkexec, where we might kexec on any cpu and therefore change the mapping\nfrom boot to boot.\n\nThe patch below does a first pass early on to work out the logical cpuid\nof the boot thread. We then fix up some paca structures to match.\n\nIve also removed the boot_cpuid_phys variable for ppc64, to be\nconsistent we use get_hard_smp_processor_id(boot_cpuid) everywhere.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "c6622f63db86fcbd41bf6fe05ddf2e00c1e51ced", "tree": "102f3ea0a891212603a3722fece337d6a74d450c", "parents": [ "a00428f5b149e36b8225b2a0812742a6dfb07b8c" ], "author": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri Feb 24 10:06:59 2006 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri Feb 24 14:05:56 2006 +1100" }, "message": "powerpc: Implement accurate task and CPU time accounting\n\nThis implements accurate task and cpu time accounting for 64-bit\npowerpc kernels. Instead of accounting a whole jiffy of time to a\ntask on a timer interrupt because that task happened to be running at\nthe time, we now account time in units of timebase ticks according to\nthe actual time spent by the task in user mode and kernel mode. We\nalso count the time spent processing hardware and software interrupts\naccurately. This is conditional on CONFIG_VIRT_CPU_ACCOUNTING. If\nthat is not set, we do tick-based approximate accounting as before.\n\nTo get this accurate information, we read either the PURR (processor\nutilization of resources register) on POWER5 machines, or the timebase\non other machines on\n\n* each entry to the kernel from usermode\n* each exit to usermode\n* transitions between process context, hard irq context and soft irq\n context in kernel mode\n* context switches.\n\nOn POWER5 systems with shared-processor logical partitioning we also\nread both the PURR and the timebase at each timer interrupt and\ncontext switch in order to determine how much time has been taken by\nthe hypervisor to run other partitions (\"steal\" time). Unfortunately,\nsince we need values of the PURR on both threads at the same time to\naccurately calculate the steal time, and since we can only calculate\nsteal time on a per-core basis, the apportioning of the steal time\nbetween idle time (time which we ceded to the hypervisor in the idle\nloop) and actual stolen time is somewhat approximate at the moment.\n\nThis is all based quite heavily on what s390 does, and it uses the\ngeneric interfaces that were added by the s390 developers,\ni.e. account_system_time(), account_user_time(), etc.\n\nThis patch doesn\u0027t add any new interfaces between the kernel and\nuserspace, and doesn\u0027t change the units in which time is reported to\nuserspace by things such as /proc/stat, /proc/\u003cpid\u003e/stat, getrusage(),\ntimes(), etc. Internally the various task and cpu times are stored in\ntimebase units, but they are converted to USER_HZ units (1/100th of a\nsecond) when reported to userspace. Some precision is therefore lost\nbut there should not be any accumulating error, since the internal\naccumulation is at full precision.\n\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "2ef9481e666b4654159ac9f847e6963809e3c470", "tree": "62abb35633702dcc585df1e2ee093aaf0dc6bb07", "parents": [ "75288c78c69020a574d93770c3a941b785f3d93d" ], "author": { "name": "Jon Mason", "email": "jdmason@us.ibm.com", "time": "Mon Jan 23 10:58:20 2006 -0600" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri Feb 10 16:53:51 2006 +1100" }, "message": "[PATCH] powerpc: trivial: modify comments to refer to new location of files\n\nThis patch removes all self references and fixes references to files\nin the now defunct arch/ppc64 tree. I think this accomplises\neverything wanted, though there might be a few references I missed.\n\nSigned-off-by: Jon Mason \u003cjdmason@us.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "3356bb9f7ba378a6e2709f9df95f4ea52111f4df", "tree": "84f370df6e58cec63132f9acce492d585226e671", "parents": [ "e58c3495e6007af59382540bb21ee941e470d88d" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Fri Jan 13 10:26:42 2006 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri Jan 13 21:17:39 2006 +1100" }, "message": "[PATCH] powerpc: Remove lppaca structure from the PACA\n\nAt present the lppaca - the structure shared with the iSeries\nhypervisor and phyp - is contained within the PACA, our own low-level\nper-cpu structure. This doesn\u0027t have to be so, the patch below\nremoves it, making a separate array of lppaca structures.\n\nThis saves approximately 500*NR_CPUS bytes of image size and kernel\nmemory, because we don\u0027t need aligning gap between the Linux and\nhypervisor portions of every PACA. On the other hand it means an\nextra level of dereference in many accesses to the lppaca.\n\nThe patch also gets rid of several places where we assign the paca\naddress to a local variable for no particular reason.\n\nSigned-off-by: David Gibson \u003cdwg@au1.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "7a0268fa1a3613f2c526a9b3058701b277f6abe1", "tree": "738bf94d9bb5d68d260805dbc1898ec40ebc20e5", "parents": [ "193cac99f6d8604aca71e5a966a8cd1dfb84819d" ], "author": { "name": "Anton Blanchard", "email": "anton@samba.org", "time": "Wed Jan 11 13:16:44 2006 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Jan 11 14:49:45 2006 +1100" }, "message": "[PATCH] powerpc/64: per cpu data optimisations\n\nThe current ppc64 per cpu data implementation is quite slow. eg:\n\n lhz 11,18(13) /* smp_processor_id() */\n ld 9,.LC63-.LCTOC1(30) /* per_cpu__variable_name */\n ld 8,.LC61-.LCTOC1(30) /* __per_cpu_offset */\n sldi 11,11,3 /* form index into __per_cpu_offset */\n mr 10,9\n ldx 9,11,8 /* __per_cpu_offset[smp_processor_id()] */\n ldx 0,10,9 /* load per cpu data */\n\n5 loads for something that is supposed to be fast, pretty awful. One\nreason for the large number of loads is that we have to synthesize 2\n64bit constants (per_cpu__variable_name and __per_cpu_offset).\n\nBy putting __per_cpu_offset into the paca we can avoid the 2 loads\nassociated with it:\n\n ld 11,56(13) /* paca-\u003edata_offset */\n ld 9,.LC59-.LCTOC1(30) /* per_cpu__variable_name */\n ldx 0,9,11 /* load per cpu data\n\nLonger term we can should be able to do even better than 3 loads.\nIf per_cpu__variable_name wasnt a 64bit constant and paca-\u003edata_offset\nwas in a register we could cut it down to one load. A suggestion from\nRusty is to use gcc\u0027s __thread extension here. In order to do this we\nwould need to free up r13 (the __thread register and where the paca\ncurrently is). So far Ive had a few unsuccessful attempts at doing that :)\n\nThe patch also allocates per cpu memory node local on NUMA machines.\nThis patch from Rusty has been sitting in my queue _forever_ but stalled\nwhen I hit the compiler bug. Sorry about that.\n\nFinally I also only allocate per cpu data for possible cpus, which comes\nstraight out of the x86-64 port. On a pseries kernel (with NR_CPUS \u003d\u003d 128)\nand 4 possible cpus we see some nice gains:\n\n total used free shared buffers cached\nMem: 4012228 212860 3799368 0 0 162424\n\n total used free shared buffers cached\nMem: 4016200 212984 3803216 0 0 162424\n\nA saving of 3.75MB. Quite nice for smaller machines. Note: we now have\nto be careful of per cpu users that touch data for !possible cpus.\n\nAt this stage it might be worth making the NUMA and possible cpu\noptimisations generic, but per cpu init is done so early we have to be\ncareful that all architectures have their possible map setup correctly.\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "88ced0314938814e1772b4d0d7ab20c52e4472b6", "tree": "3e06de882c0bf5706ee7a8994e06eb8c9ed3feca", "parents": [ "e1333803c3a8fb167ba67ffc5540dbb53fa7deb3" ], "author": { "name": "Arnd Bergmann", "email": "arnd@arndb.de", "time": "Fri Dec 16 22:43:46 2005 +0100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 15:13:08 2006 +1100" }, "message": "[PATCH] powerpc: sanitize header files for user space includes\n\ninclude/asm-ppc/ had #ifdef __KERNEL__ in all header files that\nare not meant for use by user space, include/asm-powerpc does\nnot have this yet.\n\nThis patch gets us a lot closer there. There are a few cases\nwhere I was not sure, so I left them out. I have verified\nthat no CONFIG_* symbols are used outside of __KERNEL__\nany more and that there are no obvious compile errors when\nincluding any of the headers in user space libraries.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "404849bbd2bfd62e05b36f4753f6e1af6050a824", "tree": "f233dd7d1ca0a0324fbfec7d33f83406c193cb27", "parents": [ "1888e7b51c0cb5db49911b59cb758ad2c7a530f2" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Thu Nov 24 16:51:31 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 14:50:35 2006 +1100" }, "message": "[PATCH] powerpc: Remove some unneeded fields from the paca\n\nThis patch removes several unnecessary fields from the paca:\n\n- next_jiffy_update_tb was simply unused. Remove trivially.\n\n- The exdsi exception save area was not used. There were plans to use\n it, but they never seem to have gone anywhere. If they ever do, we\n can put it back. Remove from the paca, and from asm-offsets.c\n\n- The default_decr field was used from asm, but was only ever assigned\n the value of tb_ticks_per_jiffy. Just access tb_ticks_per_jiffy from\n asm directly instead.\n\nBuilt and booted on POWER5 LPAR and iSeries RS64.\n\nSigned-off-by: David Gibson \u003cdwg@au1.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "1888e7b51c0cb5db49911b59cb758ad2c7a530f2", "tree": "dab7d7cbb895658845dce88f1b2bacc0f7238e50", "parents": [ "456752f7505ef8f580ffd157558e661da2767d99" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Thu Nov 24 16:34:45 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Mon Jan 09 14:50:32 2006 +1100" }, "message": "[PATCH] powerpc: Remove ItLpRegSave area from the paca\n\nOn iSeries, the paca contains, amongst other things an ItLpRegSave\nstructure used by the hypervisor to save registers. The hypervisor\nlocates this area through a pointer at the beginning of the paca, so\nthe structure itself can be located elsewhere. This patch moves the\nreg_save area out into its own array. This reduces the amount of\niSeries specific gunk which is visible to general powerpc code via\npaca.h\n\nBuilt and booted on POWER5 LPAR and iSeries RS64.\n\nSigned-off-by: David Gibson \u003cdwg@au1.ibm.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "8882a4da1c932c9f311c9f739e6719adea3e25d9", "tree": "1cf5e23780468d35305cde874f31c6223693908f", "parents": [ "584224e4095d8abcf2bef38efacc291be9a44c20" ], "author": { "name": "David Gibson", "email": "david@gibson.dropbear.id.au", "time": "Wed Nov 09 13:38:01 2005 +1100" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Nov 10 11:24:04 2005 +1100" }, "message": "[PATCH] powerpc: Move various ppc64 files with no ppc32 equivalent to powerpc\n\nThis patch moves a bunch of files from arch/ppc64 and\ninclude/asm-ppc64 which have no equivalents in ppc32 code into\narch/powerpc and include/asm-powerpc. The file affected are:\n\tabs_addr.h\n\tcompat.h\n\tlppaca.h\n\tpaca.h\n\ttce.h\n\tcpu_setup_power4.S\n\tioctl32.c\n\tfirmware.c\n\tpacaData.c\n\nThe only changes apart from the move and corresponding Makefile\nchanges are:\n\t- #ifndef/#define in includes updated to _ASM_POWERPC_ form\n\t- trailing whitespace removed\n\t- comments giving full paths removed\n\t- pacaData.c renamed paca.c to remove studlyCaps\n\t- Misplaced { moved in lppaca.h\n\nBuilt and booted on POWER5 LPAR (ARCH\u003dpowerpc and ARCH\u003dppc64), built\nfor 32-bit powermac (ARCH\u003dpowerpc).\n\nSigned-off-by: David Gibson \u003cdavid@gibson.dropbear.id.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "3c726f8dee6f55e96475574e9f645327e461884c", "tree": "f67c381e8f57959aa4a94bda4c68e24253cd8171", "parents": [ "f912696ab330bf539231d1f8032320f2a08b850f" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Mon Nov 07 11:06:55 2005 +1100" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Sun Nov 06 16:56:47 2005 -0800" }, "message": "[PATCH] ppc64: support 64k pages\n\nAdds a new CONFIG_PPC_64K_PAGES which, when enabled, changes the kernel\nbase page size to 64K. The resulting kernel still boots on any\nhardware. On current machines with 4K pages support only, the kernel\nwill maintain 16 \"subpages\" for each 64K page transparently.\n\nNote that while real 64K capable HW has been tested, the current patch\nwill not enable it yet as such hardware is not released yet, and I\u0027m\nstill verifying with the firmware architects the proper to get the\ninformation from the newer hypervisors.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "59ce20bb341282d5ecf17fd8b0ac093dd6350e5e", "tree": "fbd1db5c36167e4fbadca92ec6a143f80ef9d936", "parents": [ "8875ccfb7a6bd69d95a4e889ab36adda06c30d9e" ], "author": { "name": "Kelly Daly", "email": "kelly@au.ibm.com", "time": "Wed Nov 02 14:49:47 2005 +1100" }, "committer": { "name": "Kelly Daly", "email": "kelly@au.ibm.com", "time": "Wed Nov 02 14:49:47 2005 +1100" }, "message": "merge filename and modify references to iseries/it_lp_reg_save.h\n\nSigned-off-by: Kelly Daly \u003ckelly@au.ibm.com\u003e\n" }, { "commit": "ed094150bdeb6eca691f1b7a05bd34d4d1a95488", "tree": "6dc4428b3ce794c2e26db1583cab724f5bcefe19", "parents": [ "74889802a1585af4e1652f0cb853ac22a65816a4" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu Jun 30 15:16:09 2005 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jun 30 15:16:09 2005 +1000" }, "message": "[PATCH] ppc64: Simplify counting of lpevents, remove lpevent_count from paca\n\nCurrently there\u0027s a per-cpu count of lpevents processed, a per-queue (ie.\nglobal) total count, and a count by event type.\n\nReplace all that with a count by event for each cpu. We only need to add\nit up int the proc code.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "bea248fb30c3122ece8c34798527fac431c1d7b0", "tree": "9158d7a089312f92abcb6c8d5a8d942d543be24b", "parents": [ "b1bdfbd0a29d6da4dbe42736faac02c43a9afe76" ], "author": { "name": "Michael Ellerman", "email": "michael@ellerman.id.au", "time": "Thu Jun 30 15:07:09 2005 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Thu Jun 30 15:07:09 2005 +1000" }, "message": "[PATCH] ppc64: Remove lpqueue pointer from the paca on iSeries\n\nThe iSeries code keeps a pointer to the ItLpQueue in its paca struct. But\nall these pointers end up pointing to the one place, ie. xItLpQueue.\n\nSo remove the pointer from the paca struct and just refer to xItLpQueue\ndirectly where needed.\n\nThe only complication is that the spread_lpevents logic was implemented by\nhaving a NULL lpqueue pointer in the paca on CPUs that weren\u0027t supposed to\nprocess events. Instead we just compare the spread_lpevents value to the\nprocessor id to get the same behaviour.\n\nSigned-off-by: Michael Ellerman \u003cmichael@ellerman.id.au\u003e\nAcked-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "4a5304f5ba04eff979cd108309eae7a00f3fff77", "tree": "3dcf4bf1a32d1d456dd43b33bd7cac686de7dd1a", "parents": [ "c92877e0a079fe5a488cb244c7afb05137702b49" ], "author": { "name": "Stephen Rothwell", "email": "sfr@canb.auug.org.au", "time": "Tue Jun 21 17:15:40 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Tue Jun 21 18:46:28 2005 -0700" }, "message": "[PATCH] ppc64 iSeries: tidy up some includes and HvCall.h\n\nThis patch removes some unused bits from HvCall.h and some unneeded #includes\nfrom other files. Also includes ItLpQueue.h in paca.h in preference to a stub\ndeclaration of struct ItLpQueue.\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2", "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d", "parents": [], "author": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sat Apr 16 15:20:36 2005 -0700" }, "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n" } ] }