I tried setting the flag but another sort of the same error came back:
scons STATS_BINNING=no ALPHA_FS/test/opt/quick
scons: Reading SConscript files ...
Checking for C header file fenv.h... yes
KeyError: 'USE_MYSQL':
File "SConstruct", line 348:
sticky_opts.Update(env)
File "/m/imageua/helou/lib/scons/SCons/Options/__init__.py", line 176:
option.validator(option.key, env.subst('${%s}'%option.key), env)
File "/m/imageua/helou/lib/scons/SCons/Options/BoolOption.py", line 75:
if not env[key] in (True, False):
File "/m/imageua/helou/lib/scons/SCons/Environment.py", line 290:
return self._dict[key]
It has to do with the self_dict[key] again.
Steve Reinhardt wrote:
That's very odd... it sure looks like a scons bug (all that code you
see in the stack trace is scons code), but we use 0.96.1 on several of
our machines too and it works fine there.
Try explicitly setting the flag:
scons STATS_BINNING=no ALPHA_FS/test/opt/quick
I'm actually stumped but that's the best idea I can come up with right
now.
Also, please cc [email protected] on your questions.
Steve
Jirar Nicolas Helou wrote:
I am using scons-0.96.1
Steve Reinhardt wrote:
I've never seen this error before... what version of scons are you
using?
Jirar Nicolas Helou wrote:
Dr Reinhardt,
Thank you very much for the quick reply. I looked at the readme
file and went through the instructions and when I tried to compile
the full system by teh command:
scons ALPHA_FS/test/opt/quick
I get the following error:
scons: Reading SConscript files ...
Checking for C header file fenv.h... yes
KeyError: 'STATS_BINNING':
File "SConstruct", line 348:
sticky_opts.Update(env)
File "/m/imageua/helou/lib/scons/SCons/Options/__init__.py", line
176:
option.validator(option.key, env.subst('${%s}'%option.key), env)
File "/m/imageua/helou/lib/scons/SCons/Options/BoolOption.py",
line 75:
if not env[key] in (True, False):
File "/m/imageua/helou/lib/scons/SCons/Environment.py", line 290:
return self._dict[key]
Now I am not sure whether I need to install the whole system or not
but the goal is to be able to run any benchmark on M5 and observe
what kind of results it gives me. I think that you guys provide
some Benchmarks that I might be able to run but do you think I need
to compile and build the whole system and if so how do I fix this
problem. Thank you for your time.
Steve Reinhardt wrote:
It's our fault... that document is out of date. As of the 1.1
release, the regression tests are built into scons. The
up-to-date instructions are in the README file included in the
release.
Steve
Jirar Nicolas Helou wrote:
Hi,
My name is Jirar Helou and I am trying to compile and run some
benchmarks on the M5 simulator as a part of my project for my
parallel computing class at the University of delaware. I am
having difficutlies in getting the M5 to compile on a linux
machine. I can run the scons and I get some of the results in:
http://m5.eecs.umich.edu/docs/compile.html
but when i try to go to m5-test and run ./do-tests ..... I can't
find any do-test.py file anywhere?
Am I missing something or doing something wrong?
Any help will be very appreciated especially with the project due
next week and having this simulator running will be very essential.
Thanks!
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
m5sim-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/m5sim-users
