)]}' { "log": [ { "commit": "859deea949c382d9ccb6397fe33df3703ecef45d", "tree": "b0fe2d7a814143f3ff61a73a0727522a1a3dd6e4", "parents": [ "21c4ff80cba5e24932f3ef79c8482c0491630b2b" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Fri Oct 20 14:37:05 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Oct 25 11:54:18 2006 +1000" }, "message": "[POWERPC] Cell timebase bug workaround\n\nThe Cell CPU timebase has an erratum. When reading the entire 64 bits\nof the timebase with one mftb instruction, there is a handful of cycles\nwindow during which one might read a value with the low order 32 bits\nalready reset to 0x00000000 but the high order bits not yet incremeted\nby one. This fixes it by reading the timebase again until the low order\n32 bits is no longer 0. That might introduce occasional latencies if\nhitting mftb just at the wrong time, but no more than 70ns on a cell\nblade, and that was considered acceptable.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nAcked-by: Olof Johansson \u003colof@lixom.net\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" }, { "commit": "0909c8c2d547e45ca50e2492b08ec93a37b35237", "tree": "23e66e1dc9a5bd674ba1375b5fccd2cb0d5787a8", "parents": [ "7aeb732428fc8e2ecae6d432873770c12f04a979" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Fri Oct 20 11:47:18 2006 +1000" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Wed Oct 25 11:54:07 2006 +1000" }, "message": "[POWERPC] Support feature fixups in vdso\u0027s\n\nThis patch reworks the feature fixup mecanism so vdso\u0027s can be fixed up.\nThe main issue was that the construct:\n\n .long label (or .llong on 64 bits)\n\nwill not work in the case of a shared library like the vdso. It will\ngenerate an empty placeholder in the fixup table along with a reloc,\nwhich is not something we can deal with in the vdso.\n\nThe idea here (thanks Alan Modra !) is to instead use something like:\n\n1:\n .long label - 1b\n\nThat is, the feature fixup tables no longer contain addresses of bits of\ncode to patch, but offsets of such code from the fixup table entry\nitself. That is properly resolved by ld when building the .so\u0027s. I\u0027ve\nmodified the fixup mecanism generically to use that method for the rest\nof the kernel as well.\n\nAnother trick is that the 32 bits vDSO included in the 64 bits kernel\nneed to have a table in the 64 bits format. However, gas does not\nsupport 32 bits code with a statement of the form:\n\n .llong label - 1b (Or even just .llong label)\n\nThat is, it cannot emit the right fixup/relocation for the linker to use\nto assign a 32 bits address to an .llong field. Thus, in the specific\ncase of the 32 bits vdso built as part of the 64 bits kernel, we are\nusing a modified macro that generates:\n\n .long 0xffffffff\n .llong label - 1b\n\nNote that is assumes that the value is negative which is enforced by\nthe .lds (those offsets are always negative as the .text is always\nbefore the fixup table and gas doesn\u0027t support emiting the reloc the\nother way around).\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\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": "cbd27b8ced4b1888c93f69b4dd108a69ac4d733f", "tree": "c3bf062e9473261cbd4774996c96cfc0f3759239", "parents": [ "9d624ea474a3ddf3a0702d9b47e428ce1f8488a6" ], "author": { "name": "Benjamin Herrenschmidt", "email": "benh@kernel.crashing.org", "time": "Wed Oct 12 11:39:33 2005 +1000" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@g5.osdl.org", "time": "Wed Oct 12 08:24:47 2005 -0700" }, "message": "[PATCH] ppc32: Fix timekeeping\n\nInterestingly enough, ppc32 had broken timekeeping for ages... It\nworked, but probably drifted a bit more than could be explained by the\nactual bad precision of the timebase calibration. We discovered that\nrecently when somebody figured out that the common code was using\nCLOCK_TICK_RATE to correct the timekeeing, and ppc32 had a completely\nbogus value for it.\n\nThis patch turns it into something saner. Probably not as good as doing\nsomething based on the actual timebase frequency precision but I\u0027ll\nleave that sort of math to others. This at least makes it better for\nthe common HZ values.\n\nSigned-off-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n" }, { "commit": "dd56fdf23dfa0127d512b73d4238dbd2b5a7c1eb", "tree": "59f952b87d4bec5f11c578a384c3aaa2ddf5860d", "parents": [ "64807081e38703617cf9a5d71db14ea2b3e1cb04" ], "author": { "name": "jdl@freescale.com", "email": "jdl@freescale.com", "time": "Wed Sep 07 15:59:48 2005 -0500" }, "committer": { "name": "Paul Mackerras", "email": "paulus@samba.org", "time": "Fri Sep 09 22:11:35 2005 +1000" }, "message": "[PATCH] powerpc: Merge a few more include files\n\nMerge a few asm-ppc and asm-ppc64 header files.\nNote: the merge of setup.h intentionally does not carry\nforward the m68k cruft. That means this patch continues\nto break the already broken amiga on the ppc32.\n\nSigned-off-by: Jon Loeliger \u003cjdl@freescale.com\u003e\nSigned-off-by: Kumar Gala \u003ckumar.gala@freescale.com\u003e\nSigned-off-by: Paul Mackerras \u003cpaulus@samba.org\u003e\n" } ] }