dyninst.git
13 years agoAdding BPatch_dependenceGraphNode and -Edge to represent graphs used in slicing
tugrul [Mon, 17 Sep 2007 15:25:00 +0000 (15:25 +0000)]
Adding BPatch_dependenceGraphNode and -Edge to represent graphs used in slicing

13 years agoMoving Annotation Framework into common library and making BPatch_function annotatable
tugrul [Mon, 17 Sep 2007 15:23:13 +0000 (15:23 +0000)]
Moving Annotation Framework into common library and making BPatch_function annotatable

13 years agoMoving Annotation Framework into common library
tugrul [Mon, 17 Sep 2007 15:13:57 +0000 (15:13 +0000)]
Moving Annotation Framework into common library

13 years agoFix off-by-one memory allocation error
cooksey [Fri, 14 Sep 2007 17:06:00 +0000 (17:06 +0000)]
Fix off-by-one memory allocation error

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:05 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Change in call to decodeStartupSysCalls, which is now in class process

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:04 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added some private helper functions, grouped them together

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:03 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Just re-ordering of process construction initializations to avoid a compiler warning

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:02 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added extra assert(0); function stubs and fixed a compile time bug, assert(0) functions require return address in windows

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:01 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added copy of link structure to help in parsing of malformed binaries for 64 bit linux

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:55:00 +0000 (16:55 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added 64 bit functionality to the backup find main code

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:54:57 +0000 (16:54 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Made bugfix in analyzeImage

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:54:56 +0000 (16:54 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added extra assert(0); function stubs

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:54:55 +0000 (16:54 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Bugfix to the code I added last time, fixes function offsets in new regions

13 years agoExtended backup method for finding main to linux x86 64 bit and fixed some bugs that...
roundy [Fri, 14 Sep 2007 16:54:53 +0000 (16:54 +0000)]
Extended backup method for finding main to linux x86 64 bit and fixed some bugs that I had introduced.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoDisable save-the-world tests
legendre [Fri, 14 Sep 2007 15:37:41 +0000 (15:37 +0000)]
Disable save-the-world tests

13 years agoFixed test race condition that was causing Linux/x86 test13 failures
legendre [Thu, 13 Sep 2007 20:13:07 +0000 (20:13 +0000)]
Fixed test race condition that was causing Linux/x86 test13 failures

13 years agoChange asserts in dwarf code to proper error returns
legendre [Thu, 13 Sep 2007 20:13:06 +0000 (20:13 +0000)]
Change asserts in dwarf code to proper error returns

13 years agoWorkaround for issue where system() would lead to mutatee crashes
legendre [Thu, 13 Sep 2007 20:13:04 +0000 (20:13 +0000)]
Workaround for issue where system() would lead to mutatee crashes

13 years agoFix issue where some Linux/x86-64 systems could not load RT library
legendre [Thu, 13 Sep 2007 20:13:03 +0000 (20:13 +0000)]
Fix issue where some Linux/x86-64 systems could not load RT library

13 years agoTurn an warnings from fprintf(stderr, ... to a dyn_uwn_printf
legendre [Thu, 13 Sep 2007 20:13:01 +0000 (20:13 +0000)]
Turn an warnings from fprintf(stderr, ... to a dyn_uwn_printf

13 years agoTest if module has been unloaded before allowing some operations
legendre [Thu, 13 Sep 2007 20:13:00 +0000 (20:13 +0000)]
Test if module has been unloaded before allowing some operations

13 years agoMark BPatch_threads as cleaned after exec
legendre [Thu, 13 Sep 2007 20:12:57 +0000 (20:12 +0000)]
Mark BPatch_threads as cleaned after exec

13 years agoAdded dyninst version #defines as per request of O|SS
legendre [Thu, 13 Sep 2007 20:12:56 +0000 (20:12 +0000)]
Added dyninst version #defines as per request of O|SS

13 years agoAdd symtabAPI and common to link lines
legendre [Thu, 13 Sep 2007 20:12:54 +0000 (20:12 +0000)]
Add symtabAPI and common to link lines

13 years agoDon't abort build if 'gcc -v' errors
legendre [Thu, 13 Sep 2007 20:12:52 +0000 (20:12 +0000)]
Don't abort build if 'gcc -v' errors

13 years agoMinor updates
bernat [Wed, 12 Sep 2007 22:15:02 +0000 (22:15 +0000)]
Minor updates

13 years agoMore updates from previous commit
bernat [Wed, 12 Sep 2007 20:59:34 +0000 (20:59 +0000)]
More updates from previous commit

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:18 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Propagate s/process/AddressSpace/g

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:16 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Pull out non-dynamic code from process class

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:15 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Centralize inferior heap code

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:11 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Define and implement AddressSpace refactoring of process class

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:08 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Add AddressSpace as an InstructionSource

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:06 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Use addressSpace instead of process

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:05 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Remove obsolete code

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:04 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
SaveTheWorld work

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:02 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Refactor process parameter into AddressSpace parameter

13 years agoRefactoring of AddressSpace from process class - - - - - - - - - - - - -
bernat [Wed, 12 Sep 2007 20:57:00 +0000 (20:57 +0000)]
Refactoring of AddressSpace from process class  - - - - - - - - - - - - -
Add addressSpace.[Ch] and infHeap.C

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:18:36 +0000 (20:18 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added API function addMemModule

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:15:04 +0000 (20:15 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Replaced sanity checks with warnings

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:15:03 +0000 (20:15 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
made change to handleProcessCreate so that we don't fail if main has not been located

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:15:02 +0000 (20:15 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Made changes to the fileDescriptor class, added a new constructor and fields: startAddr, endAddr

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:15:01 +0000 (20:15 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Changed findmain so that it doesn't automatically fail when it can't find a text section

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:59 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
added handling for two new events evtLibcLoaded and evtLibcTrap

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:58 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
added SignalGeneratorCommon::decodeStartupSysCalls() function

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:57 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
decodeIfProcessStartup now calls decodeStartupSysCalls if we're tracing startup events

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:56 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added a few functions for system call tracing so we can tell when and where libc is loaded

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:55 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I added a new bootstrap state, 'libcLoaded_bs', that comes before initialized_bs.  Made changes to process::loadDYNINSTlib(), we now wait for the 'loadedLibc' bootstrap state instead of 'initialized' before starting

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:53 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
If our findMain heuristics fail it is detected here and I activate the libc instrumentation method for finding main

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:52 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Replaced sanity check with warning and added special case code

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:51 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Made change to continueLWP_ so that it issues PTRACE_SYSCALL instead of PTRACE_CONTINUE under certain conditions

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:50 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added code for ...

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:47 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
changed image::analyzedImage(), so it searches for functions from the beginning of the text segment to its end instead of just in between functions.

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:46 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
fixed bug in int_function::addPrettyName

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:45 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added assert(0) function shells

13 years agoAdd backup method for finding main and added API function BPatch_module *BPatch_image...
roundy [Thu, 6 Sep 2007 20:14:43 +0000 (20:14 +0000)]
Add backup method for finding main and added API function BPatch_module *BPatch_image::addMemModule, which creates a new memory module and parses its contents.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Added two new events evtLibcLoaded and evtLibcTrap

13 years agoBugfix: we weren't checking the beginning of a jump table candidate for validity...
bill [Fri, 31 Aug 2007 21:40:00 +0000 (21:40 +0000)]
Bugfix: we weren't checking the beginning of a jump table candidate for validity correctly.  This caused assertions on the Maryland machines.

13 years agoAdded Passed Test Message to test16
rutar [Thu, 23 Aug 2007 18:27:28 +0000 (18:27 +0000)]
Added Passed Test Message to test16

13 years agoBugfix: change to setCurrentAddress() semantics in InstrucIter broke stack walk test...
bill [Mon, 20 Aug 2007 22:58:22 +0000 (22:58 +0000)]
Bugfix: change to setCurrentAddress() semantics in InstrucIter broke stack walk test pm x86-64.  setcurrentAddress() again behaves properly, this time without causing an infinite loop.

M dyninstAPI/src/InstrucIter-x86.C
M dyninstAPI/src/InstrucIter-xplat.C
M dyninstAPI/src/InstrucIter.h
M dyninstAPI/src/image-flowGraph.C

13 years agoMoved getBPInstruction to InstrucIter-xplat.C
bill [Fri, 17 Aug 2007 20:07:50 +0000 (20:07 +0000)]
Moved getBPInstruction to InstrucIter-xplat.C

M dyninstAPI/src/InstrucIter-ia64.C
M dyninstAPI/src/InstrucIter-power.C
M dyninstAPI/src/InstrucIter-sparc.C
M dyninstAPI/src/InstrucIter-xplat.C
M dyninstAPI/src/InstrucIter.h

13 years agoFixing x86 InstrucIter semantics to be consistent with other platforms. Also, this...
bill [Fri, 17 Aug 2007 16:59:37 +0000 (16:59 +0000)]
Fixing x86 InstrucIter semantics to be consistent with other platforms.  Also, this removes the infinite loop that produced some of our sea of red tests this morning.

M InstrucIter-x86.C

13 years agoInstrucIter refactoring:
bill [Thu, 16 Aug 2007 20:43:46 +0000 (20:43 +0000)]
InstrucIter refactoring:

* new InstructionSource abstraction, replacing dual proc/image ownership; processes and images now are InstructionSources
* getNextInstruction/getPrevInstruction removed, as side effects were inconsistent
* increment/decrement made to work cross-platform
* IA32 previous instruction stack made more compact

Bill

13 years agoBrought .cvsignores up to date - - - - - - - - - - - - - - - - - - - - -
bill [Thu, 16 Aug 2007 20:32:54 +0000 (20:32 +0000)]
Brought .cvsignores up to date  - - - - - - - - - - - - - - - - - - - - -
cvs -nq update from core will no longer give ?s on legitimate build products

13 years agoDebug output to figure out nightly build problem
rutar [Thu, 16 Aug 2007 02:34:21 +0000 (02:34 +0000)]
Debug output to figure out nightly build problem

13 years agoFixes a memory bug that affects the Fortran tests - - - - - - - - - - -
cooksey [Tue, 14 Aug 2007 16:46:38 +0000 (16:46 +0000)]
Fixes a memory bug that affects the Fortran tests  - - - - - - - - - - -
Fix off-by-one memory allocation bug and buffer overflow in findVariable()

13 years agoComply with 64-bit ELF PowerPC ABI by adding function descriptors for DYNINSTthreadIn...
ssuen [Fri, 10 Aug 2007 15:14:13 +0000 (15:14 +0000)]
Comply with 64-bit ELF PowerPC ABI by adding function descriptors for DYNINSTthreadIndexFAST() and atomic_set().
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoinstruction::generateMemAccess64() expects a signed 16-bit integer as an immediate...
ssuen [Thu, 9 Aug 2007 22:24:01 +0000 (22:24 +0000)]
instruction::generateMemAccess64() expects a signed 16-bit integer as an immediate load offset argument.  Following an instruction::loadPartialImmIntoReg(), the lowest 16 bits of a load address needs to be explicitly converted to a signed 16-bit integer before being passed to instruction::generateMemAccess64().  We now use an explict (int16_t) cast to perform this needed conversion.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoFix a problem caused by branching to a displacement rather than
bernat [Thu, 9 Aug 2007 22:13:06 +0000 (22:13 +0000)]
Fix a problem caused by branching to a displacement rather than
absolute address.

13 years agoNot having OpenMP compiler doesn't break build
rutar [Thu, 9 Aug 2007 21:17:56 +0000 (21:17 +0000)]
Not having OpenMP compiler doesn't break build

13 years agoincremental commit; ppc64_linux passes about one-third of tests. - - - -
ssuen [Thu, 9 Aug 2007 18:22:23 +0000 (18:22 +0000)]
incremental commit; ppc64_linux passes about one-third of tests.  - - - -
Handle 64-bit ELF functions that have aliases.

13 years agoincremental commit; ppc64_linux passes about one-third of tests. - - - -
ssuen [Thu, 9 Aug 2007 18:22:22 +0000 (18:22 +0000)]
incremental commit; ppc64_linux passes about one-third of tests.  - - - -
Use region_hi_64() and region_lo_64() for 64-bit mutatees.

13 years agoincremental commit; ppc64_linux passes about one-third of tests. - - - -
ssuen [Thu, 9 Aug 2007 18:22:20 +0000 (18:22 +0000)]
incremental commit; ppc64_linux passes about one-third of tests.  - - - -
Move region_hi() and region_lo() out of linux-power.h.

13 years agoBugfix: use mkstemp() to create tempfiles, rather than dying when we run out of the...
bill [Thu, 9 Aug 2007 16:51:45 +0000 (16:51 +0000)]
Bugfix: use mkstemp() to create tempfiles, rather than dying when we run out of the 100 we knew how to generate.

13 years agoPush POWER branch overflow handling deeper into POWER specific code.
rchen [Thu, 9 Aug 2007 10:15:36 +0000 (10:15 +0000)]
Push POWER branch overflow handling deeper into POWER specific code.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Remove unneeded branch range check.

13 years agoPush POWER branch overflow handling deeper into POWER specific code.
rchen [Thu, 9 Aug 2007 10:15:33 +0000 (10:15 +0000)]
Push POWER branch overflow handling deeper into POWER specific code.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Add trap-inst to generateBranch().  Fix generateInterFunctionBranch() bug.

13 years agoPush POWER branch overflow handling deeper into POWER specific code.
rchen [Thu, 9 Aug 2007 10:15:22 +0000 (10:15 +0000)]
Push POWER branch overflow handling deeper into POWER specific code.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Remove unneeded definition of instruction::offsetWithinRangeOfBranchInsn().

13 years agoFix for 1_29 failure
rutar [Wed, 8 Aug 2007 20:09:19 +0000 (20:09 +0000)]
Fix for 1_29 failure

13 years agoPOWER fix. Check return branch displacement when generating in trampEnd.
rchen [Wed, 8 Aug 2007 15:26:01 +0000 (15:26 +0000)]
POWER fix.  Check return branch displacement when generating in trampEnd.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Check branch displacement before generating return branch.

13 years agoPOWER fix. Check return branch displacement when generating in trampEnd.
rchen [Wed, 8 Aug 2007 15:25:51 +0000 (15:25 +0000)]
POWER fix.  Check return branch displacement when generating in trampEnd.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Modify instruction::offsetWithinRangeOfBranchInsn() to use long parameter.

13 years agoPOWER fix. Check return branch displacement when generating in trampEnd.
rchen [Wed, 8 Aug 2007 15:25:24 +0000 (15:25 +0000)]
POWER fix.  Check return branch displacement when generating in trampEnd.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Add instruction::offsetWithinRangeOfBranchInsn() method.

13 years agoMakes compilation work whether OpenMP compiler is present or not
rutar [Wed, 8 Aug 2007 03:04:01 +0000 (03:04 +0000)]
Makes compilation work whether OpenMP compiler is present or not

13 years agoFix some C++-isms that was giving the AIX native compiler fits.
bernat [Tue, 7 Aug 2007 23:01:25 +0000 (23:01 +0000)]
Fix some C++-isms that was giving the AIX native compiler fits.

13 years agoprocess::loadDYNINSTlib_hidden() works for 64-bit ELF. - - - - - - - - -
ssuen [Tue, 7 Aug 2007 22:17:25 +0000 (22:17 +0000)]
process::loadDYNINSTlib_hidden() works for 64-bit ELF.  - - - - - - - - -

13 years agoOpenMP support in form of test16
rutar [Tue, 7 Aug 2007 18:35:39 +0000 (18:35 +0000)]
OpenMP support in form of test16

13 years agoOpenMP test added to testsuite as test16
rutar [Tue, 7 Aug 2007 16:31:30 +0000 (16:31 +0000)]
OpenMP test added to testsuite as test16

13 years agotest_driver.C: added quiet mode for interactive use.
bill [Tue, 7 Aug 2007 15:52:44 +0000 (15:52 +0000)]
test_driver.C: added quiet mode for interactive use.
--Bill

13 years agoppc64_linux (64-bit ELF) passes first test: test1_41. - - - - - - - - -
ssuen [Tue, 7 Aug 2007 15:16:06 +0000 (15:16 +0000)]
ppc64_linux (64-bit ELF) passes first test:  test1_41.  - - - - - - - - -
Implement calculation of the TOC (table of contents) offset for a given module.  The TOC is a section within the data segment that contains global symbols.

13 years agoppc64_linux (64-bit ELF) passes first test: test1_41. - - - - - - - - -
ssuen [Tue, 7 Aug 2007 15:16:04 +0000 (15:16 +0000)]
ppc64_linux (64-bit ELF) passes first test:  test1_41.  - - - - - - - - -
Put in changes to handle both 32-bit and 64-bit mutatees.  Read/write 32 or 64 bits when appropriate.  Use different offsets to read registers and stack.  getAddressWidth() is used a lot.

13 years agofix memory leakage
mjbrim [Mon, 6 Aug 2007 21:18:40 +0000 (21:18 +0000)]
fix memory leakage

13 years agofix memory leakage of SubTest objects
mjbrim [Mon, 6 Aug 2007 21:17:52 +0000 (21:17 +0000)]
fix memory leakage of SubTest objects

13 years agocheckPPC64DescriptorSymbols fix - check for all pretty&mangled names
giri [Mon, 6 Aug 2007 21:15:13 +0000 (21:15 +0000)]
checkPPC64DescriptorSymbols fix - check for all pretty&mangled names

13 years agoChange Address to OFFSET
giri [Fri, 3 Aug 2007 16:19:23 +0000 (16:19 +0000)]
Change Address to OFFSET

13 years agoFix a special case for ppc64 ELF binary symbols
giri [Fri, 3 Aug 2007 16:19:19 +0000 (16:19 +0000)]
Fix a special case for ppc64 ELF binary symbols

13 years agoAdd new ppc32_linux platform. - - - - - - - - - - - - - - - - - - - - -
ssuen [Wed, 1 Aug 2007 18:47:26 +0000 (18:47 +0000)]
Add new ppc32_linux platform.  - - - - - - - - - - - - - - - - - - - - -

13 years agoHandle 64-bit ELF PowerPC Linux stack frame convention. Changes in stack walking...
ssuen [Wed, 1 Aug 2007 14:57:51 +0000 (14:57 +0000)]
Handle 64-bit ELF PowerPC Linux stack frame convention.  Changes in stack walking:  getCallerFrame().  Changes in function parameter access:  emitR(), case getParamOp.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years ago 64-bit ELF PowerPC Linux (and AIX) uses r2 as TOC base register; changes in emitFunc...
ssuen [Tue, 31 Jul 2007 18:17:23 +0000 (18:17 +0000)]
 64-bit ELF PowerPC Linux (and AIX) uses r2 as TOC base register; changes in emitFuncCall().  64-bit ELF PowerPC Linux (and AIX) also has a three-part function descriptor; changes in writeFunctionPtr().
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoUpdate stack frame pointer handling to satisfy both 32-bit and 64-bit PowerPC ELF...
ssuen [Tue, 31 Jul 2007 15:42:02 +0000 (15:42 +0000)]
Update stack frame pointer handling to satisfy both 32-bit and 64-bit PowerPC ELF ABI documents.  Change STACKSKIP from 220 bytes to 288 bytes.  16-byte (quadword) align all stack frame pointer updates.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoCorrected data type of two register-save variables from 'unsigned' to 'unsigned long'.
ssuen [Mon, 30 Jul 2007 16:37:52 +0000 (16:37 +0000)]
Corrected data type of two register-save variables from 'unsigned' to 'unsigned long'.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agogeneratedBlock().proc() returns NULL. Use (process *) parameter instead.
ssuen [Fri, 27 Jul 2007 18:04:24 +0000 (18:04 +0000)]
generatedBlock().proc() returns NULL.  Use (process *) parameter instead.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agorecommit: Linux and AIX save registers into different places on the stack.
ssuen [Fri, 27 Jul 2007 16:58:49 +0000 (16:58 +0000)]
recommit:  Linux and AIX save registers into different places on the stack.
  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

13 years agoAdding implementation of sparc specific functions that will be used by slicing
tugrul [Fri, 27 Jul 2007 16:09:35 +0000 (16:09 +0000)]
Adding implementation of sparc specific functions that will be used by slicing