- 422c790 In fbench, compute sin/cos/sqrt from "first principles", so that the by sewardj · 18 years ago
- f62d496 Update. by sewardj · 18 years ago
- 3ae3dd3 VG_(getgroups) fix (Shinichi Noda) by sewardj · 18 years ago
- f843ad2 Check that the two sigprocmasks haven't failed. by sewardj · 18 years ago
- 3387dda More dispatcher tuning for ppc32/64. Makes a big difference for perf/tinycc. by sewardj · 18 years ago
- 95c2329 code layout wibbles only by njn · 18 years ago
- b1e1a7a Attempt to make tinycc work on ppc32, by (1) getting rid of various by sewardj · 18 years ago
- c0ec8e9 by njn · 18 years ago
- 694202c Update for recent change to OSet interface. by njn · 18 years ago
- 02280ba add extra note about tinycc by njn · 18 years ago
- 7d414c1 update by njn · 18 years ago
- 2bf9ba6 A minor performance improvement -- make swizzle() faster. by njn · 18 years ago
- 755e126 Fix read-after-free in VG_(HT_destruct). This fixes by sewardj · 19 years ago
- 5267526 Tidy up ppc32 jm-insns test some more. by cerion · 19 years ago
- afa1226 by njn · 19 years ago
- c281233 ppc64 ELF allows the 288 bytes below the stack pointer to be accessed. by sewardj · 19 years ago
- db2ac81 Make ppc64-linux stack unwinding work. by sewardj · 19 years ago
- fbfbd4f Change default --reps value from 3 to 1. by njn · 19 years ago
- 5fed8c0 ppc64-linux fixes by sewardj · 19 years ago
- 5bb8682 ppc64-linux fixes. by sewardj · 19 years ago
- b00944a Handle ppc64-related primops. by sewardj · 19 years ago
- 0fe1441 Tracking VEX r1507 adjusted references to renamed VEX dirs guest-ppc/, host-ppc/ by cerion · 19 years ago
- 53ee1fc Deal with function pointer vs function entry crazyness on ppc64-linux. by sewardj · 19 years ago
- 8a43f81 Tidied up jm-insns.c test a little. by cerion · 19 years ago
- 1f0d814 Track changes to PPC naming convention - VEX r1504 by cerion · 19 years ago
- 1bc8210 Handle CmpORD64{U,S} as ppc64 requires those. by sewardj · 19 years ago
- 26c80bd Use rt_sigprocmask, and check for errors correctly. (Not yet done: amd64-linux). by sewardj · 19 years ago
- 3d6320c Add enough syscalls to run bzip2 on ppc64-linux. by sewardj · 19 years ago
- 39e8160 Properly return error codes resulting from sigprocmask failures. by sewardj · 19 years ago
- 331dc98 Comment-only changes by sewardj · 19 years ago
- cda391f Add comment about log2(). by njn · 19 years ago
- 2537616 Make async-style syscalls work on ppc64, by using rt_sigprocmask by sewardj · 19 years ago
- f7fe0ee More ppc64-linux syscalls. by sewardj · 19 years ago
- 297c88f fixed up ppc64 assembly with .opd sections by cerion · 19 years ago
- 55c9fa9 When switching threads on ppc64, clear the reservation pseudo-reg, as on ppc32. by sewardj · 19 years ago
- e004d72 Add a destructor function to OSet_Destroy() which can be called for each node. by njn · 19 years ago
- 78cce3f Add the name of the input file to cg_annotate's output. by njn · 19 years ago
- 860b8dd Make a start adding syscalls. by sewardj · 19 years ago
- 0d8ef0b Save %CIA correctly (caused ppc64-linux to loop at the first syscall, by sewardj · 19 years ago
- 93fe3b2 Renamed coregrind.h to pub_core_clreq.h and introduced pub_tool_clreq.h to by njn · 19 years ago
- 7654d5e update by njn · 19 years ago
- 920050d Fixed cg_annotate bug -- when using the --sort option the primary threshold by njn · 19 years ago
- 99c8558 Add a new performance benchmark: tinycc. by njn · 19 years ago
- 1ade697 Rewrite ppc64 dispatch loop to avoid profiling overhead, as per ppc32 rewrite (r5352). by cerion · 19 years ago
- bcd75fc Remove all remaining profiling gunk. by njn · 19 years ago
- 13a4926 Remove remaining profiling gunk from Memcheck. by njn · 19 years ago
- fe5115a Remove the old profiling stuff from Memcheck. by njn · 19 years ago
- da033f5 Fix for bug #117096. by njn · 19 years ago
- baecb10 Bugfix update by njn · 19 years ago
- f7ba703 Fix a minor --gen-suppressions output bug. by njn · 19 years ago
- 5666ee6 update by njn · 19 years ago
- 6b23958 Clarify leak checker output. by njn · 19 years ago
- 288d0df update by njn · 19 years ago
- 81e60b2 Add comment from commit log. by njn · 19 years ago
- 8609d39 Check that noinst_PROGRAMS and noinst_LIBRARIES are not empty strings by tom · 19 years ago
- d55121e Fixed to check whether __builtin_expect is supported before using it. by tom · 19 years ago
- a879a47 Make sure EM_X86_64 is defined as elf.h doesn't define it on some older systems. by tom · 19 years ago
- 5741d40 Duh, remove debug printing. Duh. by sewardj · 19 years ago
- 8d1fa88 Update. by sewardj · 19 years ago
- 7289274 Increase buffer size, so that long debugger command lines don't crash by sewardj · 19 years ago
- 75d8eb9 Nasty hack to stop the leak checker dying on custom-allocator example by sewardj · 19 years ago
- 7e9ed71 When using a custom allocator that allocates with no intervening by sewardj · 19 years ago
- 49bdd7a findSb: gradually rearrange the superblock list to bring frequently by sewardj · 19 years ago
- bc6a06c Update known-bugs summary. by sewardj · 19 years ago
- 444e4b4 Add a malloc/free stress test. by njn · 19 years ago
- a527a49 Add info about overhead in heap blocks and OSet nodes. by njn · 19 years ago
- 5ee4948 Hold the event count in r29 rather than the count register, since the by sewardj · 19 years ago
- 1f430d3 Add missing cases in debug printing. by sewardj · 19 years ago
- 5d3a1c9 Rewrite ppc32 dispatch loop to avoid profiling overhead, as per by sewardj · 19 years ago
- 265bf44 add note about recent performance improvement by njn · 19 years ago
- 30e23ac by njn · 19 years ago
- 0d3a1a8 Make this work on platforms where r != x. by sewardj · 19 years ago
- 0d9efdd Rewrite amd64 dispatch loop to add performance enhancements as per x86 by sewardj · 19 years ago
- 274807d - Track vex r1494 (x86/amd64 change of conventions for getting by sewardj · 19 years ago
- 586fb03 Added fp regtest - needed some hackery to get around VEX's loss of accuracy. by cerion · 19 years ago
- 57c3e5d Changed jm_insns.c usage to use one of flags 'i|f|a' to run int|fp|av insns respectively. by cerion · 19 years ago
- 9dca92c svn:ignores by cerion · 19 years ago
- f131b3a Added a stress test which measures the cost of translation. by njn · 19 years ago
- 79076e3 Ensure that f() doesn't get inlined. by njn · 19 years ago
- 5778186 Add --vg option to vg_perf, which lets you specify one or more Valgrinds by njn · 19 years ago
- b2b4f46 whoops by njn · 19 years ago
- fd91dc9 Added some more svn:ignores. by cerion · 19 years ago
- ee45f1d Added a --reps option to control how many times each program is run. by njn · 19 years ago
- 866a9b4 remove bogus whitespace by njn · 19 years ago
- c9582a2 Die if a performance benchmark is missing. by njn · 19 years ago
- 02383bc Use user time instead of wall-clock time. by njn · 19 years ago
- 5096a39 Improve vg_SP_update_pass() to catch more constant offset cases. Improves by njn · 19 years ago
- fa4d68c File recording info about each performance benchmark. by njn · 19 years ago
- 57ea0c2 Remove nanosleep and adjust iteration count and array size accordingly. by sewardj · 19 years ago
- ecb4744 Abort the performance timings if any of the programs fail, by njn · 19 years ago
- c870d18 Make sarp return zero. by njn · 19 years ago
- 86c2349 Add two new programs to the performance suite. by njn · 19 years ago
- 038ee9e Make it clearer that internal errors are Valgrind's fault. by njn · 19 years ago
- 5a19e7b Update ignores. by rjwalsh · 19 years ago
- ec0c27a First attempt at some performance tracking tools. Includes a script vg_perf by njn · 19 years ago
- 932d5ed Remove dead line. by njn · 19 years ago
- ef1c8a8 Add missing declaration. by njn · 19 years ago
- 4311fe6 Fix minor Cachegrind bug that was occasionally causing misattributions of by njn · 19 years ago
- 52783ca Fail gracefully if 'make check' hasn't been run. by njn · 19 years ago
- c726652 More V setup for ppc64 - client stack setup - init_thread1state: set TOC ptr by cerion · 19 years ago