10 years agosnapshot: don't export page-aligning functions
Brian Norris [Thu, 14 Jun 2012 06:53:44 +0000 (23:53 -0700)]
snapshot: don't export page-aligning functions

Declare these functions static and move them upward in the file.

10 years agorename struct Snapshot_t => struct Snapshot
Brian Norris [Thu, 14 Jun 2012 06:51:07 +0000 (23:51 -0700)]
rename struct Snapshot_t => struct Snapshot

10 years agosnapshot: don't declare sTheRecord in both #if and #else
Brian Norris [Thu, 14 Jun 2012 06:39:59 +0000 (23:39 -0700)]
snapshot: don't declare sTheRecord in both #if and #else

10 years agosnapshot: remove 'extern "C"' block
Brian Norris [Thu, 14 Jun 2012 06:28:10 +0000 (23:28 -0700)]
snapshot: remove 'extern "C"' block

We do not need to wrap this function in 'extern "C"' for two reasons:

1) It's not actually exported and used in C code
2) You only need this sort of declaration in the header file that may be
   included in both C++ and C compilation. But snapshot.cc should only be
   compiled by the C++ compiler, and snapshot.h is not included in any C

10 years agosnapshot-interface: don't export SnapshotGlobalSegments()
Brian Norris [Thu, 14 Jun 2012 06:20:43 +0000 (23:20 -0700)]
snapshot-interface: don't export SnapshotGlobalSegments()

SnapshotGlobalSegments() is only used within snapshot-interface.cc, so declare
it static and remove it from the header.

10 years agosnapshot: replace DumpIntoLog() with lighter-weight debugging
Brian Norris [Thu, 14 Jun 2012 06:13:29 +0000 (23:13 -0700)]
snapshot: replace DumpIntoLog() with lighter-weight debugging

This file I/O "debug" messaging gets in the way. Make the debugging code simple
and succinct, or don't put it in at all. Now, you can enabled SnapShot
debugging by adding this near the top of snapshot.cc:


10 years agosnapshot: move comments to the right place
Brian Norris [Thu, 14 Jun 2012 05:08:57 +0000 (22:08 -0700)]
snapshot: move comments to the right place

10 years agoRemoving comment asking to fix fork(), has been fixed now
Subramanian Ganapathy [Wed, 13 Jun 2012 20:07:22 +0000 (13:07 -0700)]
Removing comment asking to fix fork(), has been fixed now

10 years agofixing calloc(), fix Makefile
Subramanian Ganapathy [Wed, 13 Jun 2012 19:58:54 +0000 (12:58 -0700)]
fixing calloc(), fix Makefile

Adding sbrk() calls and hooked into free also to avoid freeing sbrked memory...
checked both fork and paged implementation. seems to work Also everytime we
switch between fork and paged impl, we need to rebuild so adding snapshot.h as
dependency to mymemory.cc in makefile

10 years agolibatomic: perform write before context switch
Brian Norris [Thu, 7 Jun 2012 00:34:47 +0000 (17:34 -0700)]
libatomic: perform write before context switch

Atomic actions block a thread at the "switch_to_master" function call, so under
the current structure (which isn't quite fit for relaxed modeling yet...) we
should perform the memory write before calling "switch_to_master".

If not, we can observe sequences like the following, where x is an atomic
variable. All actions are seq_cst:

 Initially, x = 0

 Thread    Action
 ------    ------
 1         r1 = x;       // r1 = 0
 1          x = r1 + 1;  //  x = 1, not stored yet?
 2         r2 = x;       // r2 = 0
 2          x = r2       //  x = 1, not stored yet?

Then, depending on scheduling, Thread 1 or Thread 2 might complete first, with
its write being performed *after* it receives control again.

10 years ago.gitignore: ignore docs directory
Brian Norris [Wed, 6 Jun 2012 19:22:07 +0000 (12:22 -0700)]
.gitignore: ignore docs directory

10 years agoaction: fix some comments
Brian Norris [Wed, 6 Jun 2012 16:48:54 +0000 (09:48 -0700)]
action: fix some comments

10 years agoMakefile: have `mrclean' depend on `clean'
Brian Norris [Wed, 6 Jun 2012 16:46:44 +0000 (09:46 -0700)]
Makefile: have `mrclean' depend on `clean'

10 years agomore comments
Brian Demsky [Wed, 6 Jun 2012 09:16:04 +0000 (02:16 -0700)]
more comments

10 years agomore doc changes
Brian Demsky [Wed, 6 Jun 2012 09:01:38 +0000 (02:01 -0700)]
more doc changes

10 years agomore documentation
Brian Demsky [Wed, 6 Jun 2012 08:53:30 +0000 (01:53 -0700)]
more documentation

10 years ago1) Add more comments.
Brian Demsky [Wed, 6 Jun 2012 08:46:49 +0000 (01:46 -0700)]
1) Add more comments.
2) Change is_dependent to is_synchronizing and make it capture actual synchronizing behavior...
3) Add more support for RMW operations.

10 years agomore documentation
Brian Demsky [Wed, 6 Jun 2012 07:45:15 +0000 (00:45 -0700)]
more documentation

10 years agomore docs
Brian Demsky [Wed, 6 Jun 2012 07:13:23 +0000 (00:13 -0700)]
more docs

10 years agoadd support for docs
Brian Demsky [Wed, 6 Jun 2012 06:51:02 +0000 (23:51 -0700)]
add support for docs

10 years agosnapshot: zero out entire siginfo_t
Brian Norris [Tue, 5 Jun 2012 19:54:52 +0000 (12:54 -0700)]
snapshot: zero out entire siginfo_t

We should initialize this variable to all zeros, just to be safe.

10 years agoChanges needed to run on OS X... Example runs on my laptop now. No need to push...
Brian Demsky [Mon, 4 Jun 2012 05:35:41 +0000 (22:35 -0700)]
Changes needed to run on OS X...  Example runs on my laptop now.  No need to push code back and forth to test...

Out of time for fun hack, have to get back to real work now.  :(

10 years agohack some stuff towards running on mac... unrelated bugs still..
Brian Demsky [Mon, 4 Jun 2012 01:05:39 +0000 (18:05 -0700)]
hack some stuff towards running on mac...  unrelated bugs still..

10 years agosnapshot: remove duplicate definition of finalize()
Brian Norris [Thu, 31 May 2012 18:40:25 +0000 (11:40 -0700)]
snapshot: remove duplicate definition of finalize()

finalize() was moved to snapshot-interface.h but wasn't removed from snapshot.h

10 years agouse EXIT_SUCCESS and EXIT_FAILURE
Brian Norris [Thu, 31 May 2012 02:49:03 +0000 (19:49 -0700)]

We were wildly inconsistent with our exit status. We might as well follow the
C standard.

10 years agouserprog: use atomics allocated on "heap"
Brian Norris [Thu, 31 May 2012 02:42:07 +0000 (19:42 -0700)]
userprog: use atomics allocated on "heap"

Just for kicks, since snapshotting is working properly. We might as well test

10 years agoAdding fixes for the fork based implementation, also removed some redundant code...
Subramanian Ganapathy [Wed, 30 May 2012 23:52:38 +0000 (16:52 -0700)]
Adding fixes for the fork based implementation, also removed some redundant code pointed to by Prof. Demsky

10 years agocheck in message
Brian Demsky [Wed, 30 May 2012 07:23:38 +0000 (00:23 -0700)]
check in message

10 years agosnapshot-interface: replace Subramanian's /proc/*/maps checking
Brian Norris [Tue, 29 May 2012 20:37:33 +0000 (13:37 -0700)]
snapshot-interface: replace Subramanian's /proc/*/maps checking

Subramanian's code needed rewriting. It didn't actual snapshot any globals, and
it was also written so that it would never snapshot libmodel.so. I rewrite this
code and fix these problems, after Brian Demsky has fixed the problem with
dynamic linking of the HandlePF() code.

10 years agofix bug...this is another evil one...
Brian Demsky [Wed, 30 May 2012 07:18:44 +0000 (00:18 -0700)]
fix bug...this is another evil one...

10 years agonode: do not use static member variable
Brian Norris [Wed, 30 May 2012 15:06:02 +0000 (08:06 -0700)]
node: do not use static member variable

Static member variables of a class are going to be snapshotted. Move them so
they aren't static in order to prevent this.

(This can create interesting behavior, where the "total number of nodes
created" is decreasing at times!)

10 years agosnapshot: move declarations to the right interface header
Brian Norris [Wed, 30 May 2012 06:35:15 +0000 (23:35 -0700)]
snapshot: move declarations to the right interface header

We should have a minimal set of code/declarations in the snapshot-interface.h
header. And it should not include other headers unless necessary. With this
change, the model checker code only needs to use snapshot-interface.h.

10 years agosnapshot: rename MyFuncPtr to VoidFuncPtr
Brian Norris [Wed, 30 May 2012 06:29:06 +0000 (23:29 -0700)]
snapshot: rename MyFuncPtr to VoidFuncPtr

Better descriptive name.

10 years agosnapshot: rename USE_CHECKPOINTING to USE_MPROTECT_SNAPSHOT
Brian Norris [Wed, 30 May 2012 06:20:37 +0000 (23:20 -0700)]

"Checkpointing" is a generic name for either method. USE_MPROTECT_SNAPSHOT is a
more descriptive name.

10 years agosnapshot-interface: don't redefine PAGESIZE
Brian Norris [Tue, 29 May 2012 20:22:09 +0000 (13:22 -0700)]
snapshot-interface: don't redefine PAGESIZE

10 years agosnapshot-interface: bugfix - fixup array indeces
Brian Norris [Tue, 29 May 2012 19:30:35 +0000 (12:30 -0700)]
snapshot-interface: bugfix - fixup array indeces

When removing code, I missed a few magic numbers in the code.

10 years agonodestack: push 'create_cv' functionality responsibility back to ModelChecker
Brian Norris [Tue, 29 May 2012 17:46:42 +0000 (10:46 -0700)]
nodestack: push 'create_cv' functionality responsibility back to ModelChecker

It doesn't really make sense to have NodeStack call the 'create_cv' function
for a ModelAction. This commit separates the functionality of 'explore_action'
from 'create_cv', calling each separately from the higher-level
'check_current_action' model-checking function.

10 years agomodel: thread creation establishes synchronization
Brian Norris [Tue, 29 May 2012 17:29:59 +0000 (10:29 -0700)]
model: thread creation establishes synchronization

First, I add a get_parent_action() wrapper which finds either the last action
in the current thread (get_last_action()) or if NULL, finds the parent
thread-creation action.

Then, using this functionality, I provide the correct 'parent' to the
explore_action() function, which performs the clock-vector creation.

10 years agothreads: add 'creation' information
Brian Norris [Tue, 29 May 2012 17:21:06 +0000 (10:21 -0700)]
threads: add 'creation' information

The 'creation' field will hold the ModelAction that created the thread. For the
main user-thread, this is NULL.

10 years agolibthreads: pass 'class Thread' object as ModelAction 'location'
Brian Norris [Tue, 29 May 2012 17:17:23 +0000 (10:17 -0700)]
libthreads: pass 'class Thread' object as ModelAction 'location'

Currently, the 'location' parameter is unused for THREAD_CREATE. I need to
provide a reference from a Thread object to the ModelAction which created it,
so I will use this parameter as an entry point into the ModelChecker, which can
build the necessary reference later.

10 years agomodel: move bookkeeping to add_action_to_lists()
Brian Norris [Mon, 28 May 2012 19:58:56 +0000 (12:58 -0700)]
model: move bookkeeping to add_action_to_lists()

I keep a lot of bookkeeping info for efficiently locating ModelActions (e.g.,
in lists, maps, vectors). Let's just perform the dirty calculations in a single
function to help separate logical functions a little better.

10 years agomodel: set 'last action in thread' as an action's parent
Brian Norris [Mon, 28 May 2012 19:45:54 +0000 (12:45 -0700)]
model: set 'last action in thread' as an action's parent

This is a step toward synchronizing clock vectors properly: on exploration of
an action, we use only the last action in the current thread as the 'parent'
(or NULL).

Note that this still does not include the parent thread in determining each
ModelAction's parent.

10 years agomodel: log the last action in each thread
Brian Norris [Mon, 28 May 2012 19:42:54 +0000 (12:42 -0700)]
model: log the last action in each thread

The last action in each thread is needed for some model-checking computations.
This could be expanded to a full-fledged per-thread list if needed...

10 years agonodestack: compute parent ModelAction externally
Brian Norris [Mon, 28 May 2012 19:20:01 +0000 (12:20 -0700)]
nodestack: compute parent ModelAction externally

For a particular ModelAction, the 'parent' may be the last action in the
current thread, or otherwise, the action in the parent thread that created the
current thread. This calculation is not suitable for within the NodeStack (and
the current implementation is wrong, taking the last action performed by *any*
thread as the parent). Thus, I set up the method calls to pass the 'parent' to
explore_action() and leave the details to further work.

10 years agoclockvector: fixup initialization
Brian Norris [Mon, 28 May 2012 19:12:04 +0000 (12:12 -0700)]
clockvector: fixup initialization

Two related fixes:

Even if parent is NULL, pass the current ModelAction to ClockVector

When no parent is present, do not make special case for initializing
clock[0] = 1. This is only correct for a single case (the root Action)
and is already handled in other ways.

10 years agonodestack: construct Node with 'number of threads', not 'parent node'
Brian Norris [Mon, 28 May 2012 17:16:34 +0000 (10:16 -0700)]
nodestack: construct Node with 'number of threads', not 'parent node'

I don't really want to base the number of threads off of a 'parent'; I can
just record the global 'number of threads' as a Node parameter.

10 years agomodel: add obj_thrd_map
Brian Norris [Sat, 26 May 2012 02:17:51 +0000 (19:17 -0700)]
model: add obj_thrd_map

Provides a mapping of ModelActions so that we can efficiently sort through the
actions on a particular object (memory location) in a particular thread.

10 years agoMakefile: add 'make debug'
Brian Norris [Sat, 26 May 2012 01:35:27 +0000 (18:35 -0700)]
Makefile: add 'make debug'

This is sort of hacky, but it works! Now (after cleaning), you can recompile
with 'make debug' to enable my debugging features (not Subramanian's).

10 years agoaction: enhance print() message
Brian Norris [Sat, 26 May 2012 02:19:15 +0000 (19:19 -0700)]
action: enhance print() message

Include clock vector, realign

10 years agonodestack: create ModelAction clock vectors
Brian Norris [Sat, 26 May 2012 02:16:06 +0000 (19:16 -0700)]
nodestack: create ModelAction clock vectors

Only create the clock vector if the ModelAction will be retained in our
nodestack, and hence will be used for synchronization, etc.

10 years agoaction: add create_cv() and read_from()
Brian Norris [Sat, 26 May 2012 02:13:36 +0000 (19:13 -0700)]
action: add create_cv() and read_from()

These functions provide basic wrapper functionality for creating and updating
the clock vector for a particular ModelAction.

10 years agoclockvector: bugfix - fixup vector initialization, size
Brian Norris [Sat, 26 May 2012 03:28:58 +0000 (20:28 -0700)]
clockvector: bugfix - fixup vector initialization, size

This fixes a few bugs in ClockVector:

- The parent-less vector is the initial-thread creation, so set its vector
  clock[0] = 1
- The clock vector's data should be initialized to 0
- A clock vector cannot determine its size simply by the size of its parent (+1
  if 'act' is THREAD_CREATE); other threads could have been created, giving an
  inconsistent result here. Instead of trying to be smart about calculating
  num_threads from the parent, then, I just ask for it globally.

10 years agomodel: add get_num_threads()
Brian Norris [Sat, 26 May 2012 03:28:30 +0000 (20:28 -0700)]
model: add get_num_threads()

10 years agoclockvector: fixup print message
Brian Norris [Sat, 26 May 2012 02:36:06 +0000 (19:36 -0700)]
clockvector: fixup print message

10 years agoclockvector: add print() method
Brian Norris [Wed, 16 May 2012 19:45:03 +0000 (12:45 -0700)]
clockvector: add print() method

10 years agoaction: add clock vector field, destructor
Brian Norris [Wed, 16 May 2012 18:45:04 +0000 (11:45 -0700)]
action: add clock vector field, destructor

10 years agoaction: use proper location comparison
Brian Norris [Sat, 26 May 2012 02:00:37 +0000 (19:00 -0700)]
action: use proper location comparison

Now that snapshotting is in place for the initial state, stack allocation is
deterministic across executions, so we can compare stack-allocated addresses

10 years agoclockvector: add snapshotting new/delete operators
Brian Norris [Sat, 26 May 2012 01:44:23 +0000 (18:44 -0700)]
clockvector: add snapshotting new/delete operators

10 years agorename snapshotStack -> SnapshotStack
Brian Norris [Sat, 26 May 2012 01:28:49 +0000 (18:28 -0700)]
rename snapshotStack -> SnapshotStack

10 years agosnapshot: remove #include, use snapshot_id typedef
Brian Norris [Sat, 26 May 2012 01:23:44 +0000 (18:23 -0700)]
snapshot: remove #include, use snapshot_id typedef

10 years agosnapshot-interface: cleanup interface header
Brian Norris [Sat, 26 May 2012 01:10:58 +0000 (18:10 -0700)]
snapshot-interface: cleanup interface header

The MyString type does not belong in an interface header. It's only used in one
file, so move its typedef locally.

Remove/move a bunch of misplaced #includes.

There's an unnecessary forward declaration of class snapshotStack.

10 years agomodel: remove scheduler comment
Brian Norris [Sat, 26 May 2012 01:00:10 +0000 (18:00 -0700)]
model: remove scheduler comment

It's being snapshotted now.

10 years agomodel: don't clear thread_map (snapshot)
Brian Norris [Sat, 26 May 2012 00:59:13 +0000 (17:59 -0700)]
model: don't clear thread_map (snapshot)

The thread_map structure is being snapshotted, so no need to clear it.

10 years agomodel: don't reset action_trace manually
Brian Norris [Sat, 26 May 2012 00:11:09 +0000 (17:11 -0700)]
model: don't reset action_trace manually

Snapshotting takes care of this one

10 years agomain: take/revert snapshots
Brian Norris [Thu, 24 May 2012 18:33:36 +0000 (11:33 -0700)]
main: take/revert snapshots

Use basic snapshotting for 'reset_to_initial_state()'.

10 years agomodel: make thread_map heap allocated
Brian Norris [Sat, 26 May 2012 00:53:56 +0000 (17:53 -0700)]
model: make thread_map heap allocated

I want to snapshot this structure, so make it heap-allocated.

10 years agoaction: add all action_lists to snapshot...
Brian Norris [Fri, 25 May 2012 23:50:55 +0000 (16:50 -0700)]
action: add all action_lists to snapshot...

10 years agocommon: remove unneeded macros/includes
Brian Norris [Sat, 26 May 2012 00:14:56 +0000 (17:14 -0700)]
common: remove unneeded macros/includes

10 years agothreads: fixup stack allocation (snapshotting...)
Brian Norris [Sat, 26 May 2012 00:14:18 +0000 (17:14 -0700)]
threads: fixup stack allocation (snapshotting...)

10 years agosnapshot: remove time information
Brian Norris [Thu, 24 May 2012 18:31:39 +0000 (11:31 -0700)]
snapshot: remove time information

Why do we need timestamps? This is unneeded extra code and might cause

10 years agosnapshot: fix indentation
Brian Norris [Fri, 25 May 2012 23:45:37 +0000 (16:45 -0700)]
snapshot: fix indentation

10 years agosnapshot: use %p for printing pointers
Brian Norris [Fri, 25 May 2012 00:14:30 +0000 (17:14 -0700)]
snapshot: use %p for printing pointers

10 years agomore allocation fixes (use snapshotting)
Brian Norris [Fri, 25 May 2012 23:13:02 +0000 (16:13 -0700)]
more allocation fixes (use snapshotting)

10 years agonodestack: bugfix - fixup allocators for vectors
Brian Norris [Fri, 25 May 2012 23:05:23 +0000 (16:05 -0700)]
nodestack: bugfix - fixup allocators for vectors

10 years agoadd more classes to snapshotting region
Brian Norris [Thu, 24 May 2012 18:32:55 +0000 (11:32 -0700)]
add more classes to snapshotting region

Classes that don't use MyAlloc or MEMALLOC are snapshotted by default.

10 years agomymemory: use the right *&^% allocator!!
Brian Norris [Fri, 25 May 2012 00:23:38 +0000 (17:23 -0700)]
mymemory: use the right *&^% allocator!!

We're using MYMALLOC in the global new/delete instead of defaulting to
snapshottable malloc, as planned. What a stupid bug.

Also extend new/delete operators to be sure.

10 years agosnapshot: fix EOL spaces
Brian Norris [Thu, 24 May 2012 23:39:08 +0000 (16:39 -0700)]
snapshot: fix EOL spaces

10 years agowow, this is a nasty bug...
Brian Demsky [Thu, 24 May 2012 20:32:55 +0000 (13:32 -0700)]
wow, this is a nasty bug...

the last part of the snapshot bug is the following:
we snapshot the user threads stack...  when we get a seg fault,
the signal handler is using the same write protected stack...
obviously this is going to cause problems.  luckily there is support
for a special stack for the signal handler.  this checkin switches
the signal handler to run on a different stack than the program stack.

10 years agofix one segfault bug...something is still strange, but has to do with stack
Brian Demsky [Thu, 24 May 2012 19:09:39 +0000 (12:09 -0700)]
fix one segfault bug...something is still strange, but has to do with stack

10 years agofactor page alignment into function call...place near existing call
Brian Demsky [Thu, 24 May 2012 18:50:19 +0000 (11:50 -0700)]
factor page alignment into function call...place near existing call

10 years agofix page alignment issue...
Brian Demsky [Thu, 24 May 2012 18:40:14 +0000 (11:40 -0700)]
fix page alignment issue...

10 years agosnapshot: use perror() on failed library calls
Brian Norris [Thu, 24 May 2012 17:57:20 +0000 (10:57 -0700)]
snapshot: use perror() on failed library calls

This will print more informative messages when we fail.

10 years agomymemory: define the "opposite" of MEMALLOC
Brian Norris [Thu, 24 May 2012 17:32:19 +0000 (10:32 -0700)]
mymemory: define the "opposite" of MEMALLOC

We may want an identifier to show that a particular class is intended to be

10 years agomymemory: include some spacing
Brian Norris [Thu, 24 May 2012 17:31:50 +0000 (10:31 -0700)]
mymemory: include some spacing

10 years agofixup EOL whitespace
Brian Norris [Mon, 21 May 2012 19:33:12 +0000 (12:33 -0700)]
fixup EOL whitespace

10 years agosnapshot: use 'SSDEBUG', not 'DEBUG'
Brian Norris [Mon, 21 May 2012 19:16:30 +0000 (12:16 -0700)]
snapshot: use 'SSDEBUG', not 'DEBUG'

10 years agomodel: remove incorrect #undef
Brian Norris [Mon, 21 May 2012 19:13:33 +0000 (12:13 -0700)]
model: remove incorrect #undef

Don't 'undef' something; rename it. Anyway, this is unnecessary now, since
someone renamed the new DEBUG.

10 years agouserprog: use modulo
Brian Norris [Mon, 21 May 2012 18:59:02 +0000 (11:59 -0700)]
userprog: use modulo

Increasing the number of loops is useless unless we have modulo here... I
don't know why this was changed by Subramanian.

10 years agolibthreads: fix thrd_create() to use typedef'd start_routine
Brian Norris [Mon, 21 May 2012 18:58:26 +0000 (11:58 -0700)]
libthreads: fix thrd_create() to use typedef'd start_routine

10 years agofix indentation + spacing, esp. for MEMALLOC macro
Brian Norris [Mon, 21 May 2012 18:51:07 +0000 (11:51 -0700)]
fix indentation + spacing, esp. for MEMALLOC macro

10 years agomymemory: fix indentation, spacing
Brian Norris [Mon, 21 May 2012 18:32:56 +0000 (11:32 -0700)]
mymemory: fix indentation, spacing

10 years agonodestack: allocate from "mymemory" region
Brian Norris [Mon, 21 May 2012 18:24:43 +0000 (11:24 -0700)]
nodestack: allocate from "mymemory" region

10 years agoMerge commit: branch 'work'
Brian Norris [Mon, 21 May 2012 18:07:44 +0000 (11:07 -0700)]
Merge commit: branch 'work'

tree.{cc,h} were replaced with NodeStack. (Will need to update them)

Other trivial conflicts in Makefile and model.h


10 years agosnapshot: remove references to libmymemory.so
Brian Norris [Mon, 21 May 2012 17:57:16 +0000 (10:57 -0700)]
snapshot: remove references to libmymemory.so

This library doesn't exist any more.

10 years agosnapshot: move local defines after #includes
Brian Norris [Mon, 21 May 2012 17:55:28 +0000 (10:55 -0700)]
snapshot: move local defines after #includes

Prevents accidental dependencies, where a header *might* use PAGESIZE, e.g.

10 years agomain: clear out redundant comments
Brian Norris [Mon, 21 May 2012 17:52:46 +0000 (10:52 -0700)]
main: clear out redundant comments

We don't really need these. I can read the source :)

10 years agocommon: disable debugging
Brian Norris [Mon, 21 May 2012 17:32:34 +0000 (10:32 -0700)]
common: disable debugging

Note to others: do not enable debugging in the checked-in code. It is for
testint purposes only.

10 years agorun.sh: provide gdb option
Brian Norris [Mon, 21 May 2012 17:32:05 +0000 (10:32 -0700)]
run.sh: provide gdb option

10 years agoMakefile: only build a single shared library
Brian Norris [Mon, 21 May 2012 17:05:47 +0000 (10:05 -0700)]
Makefile: only build a single shared library

Now that everything is uniform, we can easily just build a single library.

10 years agoMakefile: clean up Makefile
Brian Norris [Mon, 21 May 2012 16:47:05 +0000 (09:47 -0700)]
Makefile: clean up Makefile

There are a number of inconsistencies and superfluous arguments:

- For libmymemory, we don't need (or want) to hardcode a shared library
  relative path
- For libmodel, we don't really need the 'soname' parameter
- Linking C++ probably should be done with the C++ compiler, not the C compiler
- We don't need CPPFLAGS for linking-only stages
- We want to use LDFLAGS for an early-stage linking rather than the late-stage
  linking of the user program

Overall, simpler is better