Permalink
Switch branches/tags
Nothing to show
Commits on Dec 14, 2011
  1. Noted reported compatibility bug with boost

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 14, 2011
Commits on Jun 20, 2011
  1. Mac defines malloc_size to be const void *ptr, not void *ptr

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jun 20, 2011
  2. Ignoring .DS_Store

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jun 20, 2011
Commits on May 31, 2011
  1. scons now tests for C++0x support before turning it on and tries mult…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed May 31, 2011
    …iple libraries for clock_gettime() rather than assuming it lives in librt. This ought to fix miscompilation on Mac OS X. Thanks to Robert D. Blanchet Jr. for reporting this.
  2. Fixed stupid typo when compiling in C++

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed May 31, 2011
Commits on May 19, 2011
  1. Marked off the v2 malloc API as deprecated in preparation for beta re…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed May 19, 2011
    …lease. Updated CHM documentation.
Commits on Apr 23, 2011
  1. Added override of _GNU_SOURCE for when HAVE_MREMAP is auto-detected. …

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Apr 23, 2011
    …Thanks to Maxim Zakharov for reporting this issue.
Commits on Dec 9, 2010
  1. Fixed a problem where the binary patcher was early outing too soon an…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 9, 2010
    …d therefore failing to patch all the binaries properly. It would seem that the Microsoft linker doesn't sort the import table like I had thought it did - I would guess it sorts per DLL location, otherwise is unsorted. Thanks to Roman Tatkin for reporting this bug.
  2. Very embarrassingly, nedmalloc has been severely but unpredictably br…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 9, 2010
    …oken on POSIX for over a year now when built with DEBUG defined. This was turning on DEFAULT_GRANULARITY_ALIGNED whose POSIX implementation was causing random segfaults so mysterious that neither gdb nor valgrind could pick them up - in other words, the very worst kind of memory corruption: undetectable, untraceable and undebuggable. I only found them myself due to a recent bug report for TnFOX on POSIX where due to luck, very recent Linux kernels just happened by pure accident to cause this bug to manifest itself as preventing process init right at the very start - so early that no debugger could attach. After over a week of trial & error I narrowed it down to being somewhere in nedmalloc, then having something to do with DEBUG being defined or not, then two hours ago the eureka moment arrived and I quite literally did a jig around the room in joy. Problem is now fixed thank the heavens!!!
Commits on Dec 6, 2010
  1. Disabled stack backtrace finding of MSVCRT when there is just one MSV…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 6, 2010
    …CRT loaded in the process. Also moved nedmalloc_loader.c into unsupported.
  2. Turned off patching msvcrt.dll. Breaks OpenGL :(

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 6, 2010
Commits on Dec 5, 2010
  1. As the release of v1.10 RC1 approaches, fixed a long standing problem…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 5, 2010
    … with the binary patcher where multiple MSVCRT versions in the process weren't handled - everything was sent to one MSVCRT only, and needless to say that sorta worked sometimes and sometimes not. Now when nedmalloc passes a foreign block to the system allocator, it runs a stack backtrace to figure out what MSVCRT in the process it ought to pass it to. It's slow, but fixes a very common segfault on process exit on VS2010.
Commits on Dec 3, 2010
  1. Added note about large pages hanging the computer sometimes and anoth…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 3, 2010
    …er note about the binary patcher only patching the release MSVCRT
Commits on Dec 1, 2010
  1. Fixed a slew of compilation bugs on POSIX as well as a new static lib…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 1, 2010
    …rary build option
  2. Was compiled umpa related code when umpa was disabled

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Dec 1, 2010
Commits on Nov 29, 2010
  1. Added new facility to have the umpa preload its free page cache. Also…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 29, 2010
    … added VirtualAlloc and VirtualFree emulations based on the umpa to the patcher - this can quite literally completely replace the Windows kernel page allocator implementation!
  2. Added a little bit of debugging for large pages. Also disabled user m…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 29, 2010
    …ode page allocator in normal nedmalloc.dll (use nedumpa.dll if you want the user mode page allocator)
  3. Added new build option --adminuac

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 29, 2010
Commits on Nov 28, 2010
  1. Finally got round to retiring the MSVC project files as they were sou…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 28, 2010
    …rces of never ending hassle due to being out of sync with the SConstruct config. Rebuilt scons build system to be fully compatible with MSVC instead (long overdue!)
Commits on Nov 26, 2010
  1. Have DEFAULT_GRANULARITY only be observed when it is larger than the …

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 26, 2010
    …page size .
  2. Fixed missing ) in ENABLE_LARGE_PAGES path. Thanks to Roman Tatkin fo…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 26, 2010
    …r reporting this.
Commits on Nov 25, 2010
  1. Fixed nedflushlogs() not actually flushing data and/or causing a segf…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 25, 2010
    …ault. Thanks to Roman Tatkin for reporting this.
  2. Fixed a potential segfault if no logging ops had been made before cal…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 25, 2010
    …ling nedflushlogs()
Commits on Nov 1, 2010
  1. Fixed excessive lock destruction

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Nov 1, 2010
Commits on Oct 31, 2010
  1. Fixed leak of mutexes when using pthread or win32 mutexs as locks. Th…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Oct 31, 2010
    …anks to Gavin Lambert for reporting this.
Commits on Aug 6, 2010
  1. Merge branch 'master' of ssh://mediacentre/Repositories/Git/nedmalloc…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Aug 6, 2010
    … into master.kate
    
    Conflicts:
    	SConscript
    	SConstruct
    	nedmalloc.c
    	scalingtest.cpp
    	test.c
    	usermodepageallocator.c
    	winpatcher.c
  2. Made user mode page allocator API export for use by winpatcher

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Aug 6, 2010
    Added patches for VirtualAlloc, VirtualFree and VirtualQuery
  3. Fixed up CRLF

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Aug 6, 2010
Commits on Aug 4, 2010
  1. Added traverse and notraverse to scalingtest

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Aug 4, 2010
Commits on Aug 1, 2010
  1. Fixed a series of compile errors under GCC and POSIX in general

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Aug 1, 2010
Commits on Jul 29, 2010
  1. Updated scaling test results to use CPU cycles instead of picoseconds…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jul 29, 2010
    …, and with a new "nothing" test to discount loop overheads
Commits on Jul 14, 2010
  1. Added nedmalloc vs system

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jul 14, 2010
  2. Added allocator scaling test

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jul 14, 2010
Commits on Jul 13, 2010
  1. Removed MSVCRT dependencies (prevents a lot of DLL injection failure …

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jul 13, 2010
    …scenarios).
    
    Added REALLOC_ZERO_BYTES_FREES support to nedrealloc functions.
Commits on Jul 12, 2010
  1. Fixed small bug which was preventing the windows patcher from correct…

    Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com) committed Jul 12, 2010
    …ly finding the proper MSVCRT.