FastAlloc's reuse policies can mask allocation bugs, so
we typically want it disabled when debugging. Set
FORCE_FAST_ALLOC to enable even when debugging, and set
NO_FAST_ALLOC to disable even in non-debug builds.
This causes a lot of rebuilds that could have otherwise possibly been
avoided, and, more annoyingly, a lot of unnecessary rerunning of the
regressions. The benefits of having the revision in the output haven't
materialized, so this change removes it.
The global sticky options were being printed with a heading, and then the
global nonsticky options were being printed immediately after them without a
heading. Because the two lists ran together and the first had its own heading,
it looked like -all- those options where sticky even though some of them
aren't. This change adds a label to the second list so it's clear they're
different.
I've renamed the check_whitespace operation to check_style. You're going to
need to change your .hg/hgrc file. While you're at it, add a pre-qrefresh
hook please.
I like the brevity of Ali's recent change, but the ambiguity of
sometimes showing the source and sometimes the target is a little
confusing. This patch makes scons typically list all sources and
all targets for each action, with the common path prefix factored
out for brevity. It's a little more verbose now but also more
informative.
Somehow Ali talked me into adding colors too, which is a whole
'nother story.
clean up the code a little bit while we're at it.
I recommend that everyone adds the pre-qrefresh hook below since it
will make qref run the style hook and not just commit/qpush
[extensions]
style = <m5 path>/util/style.py
[hooks]
pretxncommit.style = python:style.check_whitespace
pre-qrefresh.style = python:style.check_whitespace
The arguments were added to the global_sticky_vars Variables object after the
basic help text was generated. As a result, the "actual:" value wouldn't
reflect the arguments to scons and wouldn't really be the "actual" value used
by the build. This change fixes that by updating global_sticky_vars slightly
earlier.
Get rid of misc.py and just stick misc things in __init__.py
Move utility functions out of SCons files and into m5.util
Move utility type stuff from m5/__init__.py to m5/util/__init__.py
Remove buildEnv from m5 and allow access only from m5.defines
Rename AddToPath to addToPath while we're moving it to m5.util
Rename read_command to readCommand while we're moving it
Rename compare_versions to compareVersions while we're moving it.
--HG--
rename : src/python/m5/convert.py => src/python/m5/util/convert.py
rename : src/python/m5/smartdict.py => src/python/m5/util/smartdict.py
Add the PROTOCOL sticky option sets the coherence protocol that slicc
will parse and therefore ruby will use. This whole process was made
difficult by the fact that the set of files that are output by slicc
are not easily known ahead of time. The easiest thing wound up being
to write a parser for slicc that would tell me. Incidentally this
means we now have a slicc grammar written in python.
Start by turning all of the *Source functions into classes
so we can do more calculations and more easily collect the data we need.
Add parameters to the new classes for indicating what sorts of flags the
objects should be compiled with so we can allow certain files to be compiled
without Werror for example.
env is used as a local variable all over the place and sometimes it is
easy to get confused as to whether the global env or local env is being
used. This will become especially important when I change the way we
support our variants.
1) -L is automatically added, so don't do it ourselves
2) prepend the paths for gzstream and libelf so they are certain to
come first. The problem is that python might add /usr/lib to the path
and the user might have a locally installed version of libelf installed.
Add some features to read_command so it works a little bit better
Clean up the mercurial checks.
Filter the user environment and only pick out the useful stuff.
This allows me to clean things up so we are up to date with respect to
deprecated features. There are many features scheduled for permanent failure
in scons 2.0 and 0.98.1 provides the most compatability for that. It
also paves the way for some nice new features that I will add soon
The gzstream package provides an ostream-interface for writing gzipped files.
The package comes from:
http://www.cs.unc.edu/Research/compgeom/gzstream/
And is distributed under the LGPL license. Both the license and version
information has been preservered, though all other files in the package have
been purged. Minor modifications to the code have been made. The output module
detects when a filename ends in .gz and constructs an ogzstream object instead
of an ofstream object. This works for both the create(...) and find(...)
commands. Additionally, since gzstream objects needs to be closed to ensure
proper file termination, I have the output deconstructor deleting all ostream's
that it manages on behalf of find(...). At the moment, the only output file
that I know this functionality works for is stats, i.e. by specifying
"--stats-file=m5stats.txt.gz" on the command line.
The current EXTRAS will fail if the top level directory pointed to by EXTRAS
has a SConscript file in it. We allow this by including the directory name
of the EXTRA in the build directory which prevents a clash between
src/SConscript and extra/SConscript. Maintain compatibility with older uses
of EXTRAS by adding a -I for each top level extra directory.
This should allow m5 to be more easily embedded into other simulators.
The m5 binary adds a simple main function which then calls into the m5
libarary to start the simulation. In order to make this work
correctly, it was necessary embed python code directly into the
library instead of the zipfile hack. This is because you can't just
append the zipfile to the end of a library the way you can a binary.
As a result, Python files that are part of the m5 simulator are now
compile, marshalled, compressed, and then inserted into the library's
data section with a certain symbol name. Additionally, a new Importer
was needed to allow python to get at the embedded python code.
Small additional changes include:
- Get rid of the PYTHONHOME stuff since I don't think anyone ever used
it, and it just confuses things. Easy enough to add back if I'm wrong.
- Create a few new functions that are key to initializing and running
the simulator: initSignals, initM5Python, m5Main.
The original code for creating libm5 was inspired by a patch Michael
Adler, though the code here was done by me.
