)]}' { "log": [ { "commit": "88314ee73fd75eb32abdcb3119cd303c116d4500", "tree": "913c23ad89ec778d62eb517487dc6ab545d869c5", "parents": [ "7c963ad1d113790a8c723a178988b675868f3abe" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 31 19:13:52 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 31 19:13:52 2005 -0700" }, "message": "[SPARC64]: Refine PCI strbuf ctx-based flush.\n\nThe initial peek read PIO of the match register is just a waste.\nJust do the flush writes first, as that is more efficient.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7c963ad1d113790a8c723a178988b675868f3abe", "tree": "2e1cc54795aeca06a11801636737901ba71a2ed8", "parents": [ "2e3e80c2b75e3815a0160cbd23d4fdb767d66b35" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 31 16:57:59 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 31 16:57:59 2005 -0700" }, "message": "[SPARC64]: Fix streaming buffer flushing on PCI and SBUS.\n\nFirstly, if the direction is TODEVICE, then dirty data in the\nstreaming cache is impossible so we can elide the flush-flag\nsynchronization in that case.\n\nNext, the context allocator is broken. It is highly likely\nthat contexts get used multiple times for different dma\nmappings, which confuses the strbuf flushing code and makes\nit run inefficiently.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "816242da3735957bee99aeba40aa60c4f120a101", "tree": "c5a1092a52911143b0c4f9bc4257c1dbafbb8bdd", "parents": [ "ab3fc403633c38aef5ef48844f8e5dbfee7c34f8" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon May 23 15:52:08 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon May 23 15:52:08 2005 -0700" }, "message": "[SPARC64]: Add boot option to force UltraSPARC-III P-Cache on.\n\nOlder UltraSPARC-III chips have a P-Cache bug that makes us disable it\nby default at boot time.\n\nHowever, this does hurt performance substantially, particularly with\nmemcpy(), and the bug is _incredibly_ obscure. I have never seen it\ntriggered in practice, ever.\n\nSo provide a \"-P\" boot option that forces the P-Cache on. It taints\nthe kernel, so if it does trigger and cause some data corruption or\nOOPS, we will find out in the logs that this option was on when it\nhappened.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "a228dfd5dc4b92288ea22d427b2bfc48ba5bb8b0", "tree": "bd3c9a5892a5e3823427c9d614fd6c9aee96c1fb", "parents": [ "37e0915b701281182cea9fc90e894d10addf134a" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 20 11:40:32 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Fri May 20 11:40:32 2005 -0700" }, "message": "[SPARC64]: Fix bad performance side effect of strbuf timeout changes.\n\nThe recent change to add a timeout to strbuf flushing had\na negative performance impact. The udelay()\u0027s are too long,\nand they were done in the wrong order wrt. the register read\nchecks. Fix both, and things are happy again.\n\nThere are more possible improvements in this area. In fact,\nPCI streaming buffer flushing seems to be part of the bottleneck\nin network receive performance on my SunBlade1000 box.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "4dbc30fb27ac4e647e6efadb382ff7d38c3d368e", "tree": "e2fb7329e60be37e17f133a28ab6a5e77007db32", "parents": [ "e4fdee8e3b41239242a8f421a28736ef8e08ca55" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 11 11:37:00 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Wed May 11 11:37:00 2005 -0700" }, "message": "[SPARC64]: Add timeouts to streaming buffer synchronization.\n\nIf some hardware error occurs and the flush flag never updates,\nwe will hang forever in these routines. Add a timeout, and\nprint out a diagnostic if it is reached.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7cc1712b8a778c8077048969848857895d242009", "tree": "99407d822cde93c61c8dd9e871cd1a607d7fc277", "parents": [ "8edf72ebce06d52e855438ec18fe20dea7a4cc04" ], "author": { "name": "Coywolf Qi Hunt", "email": "coywolf@lovecn.org", "time": "Thu May 05 14:53:01 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 05 14:53:01 2005 -0700" }, "message": "[SPARC]: Remove legacy stuff from cpu_idle().\n\nCurrently sparc and sparc64\u0027s UP cpu_idle() checks current pid. This\nis old time legacy. Now it\u0027s paranoia.\n\nSigned-off-by: Coywolf Qi Hunt \u003ccoywolf@lovecn.org\u003e\nAcked-by: William Irwin \u003cwli@holomorphy.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "8edf72ebce06d52e855438ec18fe20dea7a4cc04", "tree": "a75409d5ae04a31209d52b11983158d8d4d4a199", "parents": [ "bfd4bda097f8758d28e632ff2035e25577f6b060" ], "author": { "name": "Christoph Hellwig", "email": "hch@lst.de", "time": "Thu May 05 14:27:56 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Thu May 05 14:27:56 2005 -0700" }, "message": "[SPARC64]: Kill useless __pte_alloc_one_kernel indirection\n\nwarning: untested, but it there\u0027s not too much chance for screwups\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "41832a08feca695158e15a6e58c26b224a7bfae2", "tree": "62bb245b25a5eeebe6202f84987613d66a4eaa7c", "parents": [ "cee2824f85414c98fff4004e335a6bc4072c8809" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 03 22:05:43 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 03 22:05:43 2005 -0700" }, "message": "[SPARC64]: Disable IRQ forwarding.\n\nThere is some race whereby IRQs get stuck, the IRQ status\nis pending but no processor actually handles the IRQ vector\nand thus the interrupt.\n \nThis is a temporary workaround.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "cee2824f85414c98fff4004e335a6bc4072c8809", "tree": "374fff3b0ad0bb6f28aa0d5a0a0c90a09ce78d35", "parents": [ "8800cea62025a5209d110c5fa5990429239d6eee" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 03 22:04:36 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Tue May 03 22:04:36 2005 -0700" }, "message": "[SPARC64]: Fix goal_cpu tracking in retarget_one_irq().\n\nWe would never advance the goal_cpu counter like we\nshould, so all IRQs would go to a single processor.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "7ed20e1ad521b5f5df61bf6559ae60738e393741", "tree": "90931724e45eaedb3445314e8b94e78253642395", "parents": [ "e5bdd883a189243541e7a132385580703b049102" ], "author": { "name": "Jesper Juhl", "email": "juhl-lkml@dif.dk", "time": "Sun May 01 08:59:14 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun May 01 08:59:14 2005 -0700" }, "message": "[PATCH] convert that currently tests _NSIG directly to use valid_signal()\n\nConvert most of the current code that uses _NSIG directly to instead use\nvalid_signal(). This avoids gcc -W warnings and off-by-one errors.\n\nSigned-off-by: Jesper Juhl \u003cjuhl-lkml@dif.dk\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "ef0299bf8e868c09bff13e6f3c4fc9b835f60471", "tree": "0b734b9d95cf5fe38a9a15a48e4900d28447502e", "parents": [ "fdca124a1bcc7e624f6b887c6f26153f40ee43ee" ], "author": { "name": "Al Viro", "email": "viro@parcelfarce.linux.theplanet.co.uk", "time": "Sun Apr 24 12:28:36 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun Apr 24 12:28:36 2005 -0700" }, "message": "[PATCH] mostek bogus sparse annotations fixed\n\nvoid * __iomem foo is not a pointer to iomem - it\u0027s an iomem variable\ncontaining void *. A pile of such guys in arch/sparc64/kernel/time.c,\ndrivers/sbus/char/rtc.c and include/asm-sparc64/mostek.h turned into\nintended void __iomem *. \n\nSigned-off-by: Al Viro \u003cviro@parcelfarce.linux.theplanet.co.uk\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "b4bca26c0160f48b4eb04f21d31a229832732013", "tree": "b4a6d736fcb8664753066df3e6a93ed24c0f409b", "parents": [ "d7be828e03969ea7f922f299acb8daa0d8ce7006" ], "author": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Thu Apr 21 21:42:34 2005 -0700" }, "committer": { "name": "David S. Miller", "email": "davem@sunset.davemloft.net", "time": "Thu Apr 21 21:42:34 2005 -0700" }, "message": "[SPARC]: Provide generic ioctls in Sparc RTC driver.\n\nProvide support for drivers/char/rtc.c ioctls in the\nMostek rtc driver as well as the Sparc specific RTCGET\nand RTCSET.\n\nThis allows userspace to be much less messy. Currently\nutil-linux and other spots jump through hoops trying\nvarious ioctl variants until it hits the right one whatever\ndriver actually being used supports.\n\nEventually all of this should move over to the genrtc.c\ndriver, but not today...\n\nWhile we are here, fix up the register types for sparse.\n\nThanks to Frans Pop for helping point out this issue.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "0ba4da03cc84f54b2574e1abcdaa18ef8456a9fb", "tree": "6730c36a12833eaf05ddddd7b26fc4de1284ff5b", "parents": [ "0b2cad2f30d0353f2576b1a2207c0792ba713157" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Apr 18 15:13:15 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Mon Apr 18 15:13:15 2005 -0700" }, "message": "[PATCH] sparc64: Fix stat\n\nLike Alpha, sparc64\u0027s struct stat was defined before we had the\nnanosecond et al. fields added. So like Alpha I have to cons up a\nstruct stat64 to get this stuff. I\u0027ll work on the glibc bits soon. \n\nAlso, we were forgetting to fill in the nanosecond fields in the sparc\ncompat stat64 syscalls. \n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n\n" }, { "commit": "9c7d3b3a6b6aaeded9d9e5c5111dbcc65b0b0f91", "tree": "19b2c9e85dcab6df9250ba38df885d951c96e0a6", "parents": [ "dadeafdfc8da8c27e5a68e0706b9856eaac89391" ], "author": { "name": "Jurij Smakov", "email": "jurij@wooyd.org", "time": "Sun Apr 17 18:03:12 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun Apr 17 18:03:12 2005 -0700" }, "message": "[PATCH] sparc64: Fix copy_sigingo_to_user32()\n\nThe compat routine to copy over this data structure was not\nhandling SI_POLL correctly, breaking various fcntl() variants\nin compat tasks.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "dadeafdfc8da8c27e5a68e0706b9856eaac89391", "tree": "17993d26e93e598a2f449063fe213afad2a45814", "parents": [ "fb65b9619b756793d824df7501c895a2c2871f40" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 17 18:03:11 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun Apr 17 18:03:11 2005 -0700" }, "message": "[PATCH] sparc64: Reduce ptrace cache flushing\n\nWe were flushing the D-cache excessively for ptrace() processing\nand this makes debugging threads so slow as to be totally unusable.\n\nAll process page accesses via ptrace() go via access_process_vm().\nThis routine, for each process page, uses get_user_pages(). That\nin turn does a flush_dcache_page() on the child pages before we\ncopy in/out the ptrace request data.\n\nTherefore, all we need to do after the data movement is:\n\n1) Flush the D-cache pages if the kernel maps the page to a different\n color than userspace does.\n2) If we wrote to the page, we need to flush the I-cache on older cpus.\n\nPreviously we just flushed the entire cache at the end of a ptrace()\nrequest, and that was beyond stupid.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "fb65b9619b756793d824df7501c895a2c2871f40", "tree": "8a594fd4be652d14e3795d046eeda2dc33103c33", "parents": [ "961f8bc9fc91c39a3fe169f3b6d3df1b096769d3" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 17 18:03:11 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun Apr 17 18:03:11 2005 -0700" }, "message": "[PATCH] sparc: Fix PTRACE_CONT bogosity\n\nSunOS aparently had this weird PTRACE_CONT semantic which\nwe copied. If the addr argument is something other than\n1, it sets the process program counter to whatever that\nvalue is.\n\nThis is different from every other Linux architecture, which\ndon\u0027t do anything with the addr and data args.\n\nThis difference in particular breaks the Linux native GDB support\nfor fork and vfork tracing on sparc and sparc64.\n\nThere is no interest in running SunOS binaries using this weird\nPTRACE_CONT behavior, so just delete it so we behave like other\nplatforms do.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "961f8bc9fc91c39a3fe169f3b6d3df1b096769d3", "tree": "2d7af71ae798fda8be5e1979fe61bfca3f2d3b20", "parents": [ "a9546f59e90d6ec1b65952c547302e0678e3a5ac" ], "author": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Sun Apr 17 18:03:10 2005 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@ppc970.osdl.org", "time": "Sun Apr 17 18:03:10 2005 -0700" }, "message": "[PATCH] sparc64: use message queue compat syscalls\n\nA couple message queue system call entries for compat tasks\nwere not using the necessary compat_sys_*() functions, causing\nsome glibc test cases to fail.\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\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" } ] }