Hi James, That is indeed a bug that shouldn't have made it into the release. The problem is that DPRINTF's are not included when compiling m5.fast, and in that code the variable tid is only used in a DPRINTF. If you comment out the line that declares the int tid, and the two DPRINTFs that use tid, that should fix that error.
Kevin -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steve Reinhardt Sent: Friday, September 22, 2006 7:55 PM To: M5 users mailing list Subject: Re: [m5-users] Example detailed mode configuration bug Hi James, This looks to be a problem with m5.fast... I get the same error when I try to compile that, but m5.opt compiles just fine. m5.fast isn't always a lot faster than m5.opt anyway :-). Kevin, I added this bug to flyspray for you. If it's really Korey's, feel free to reassign it to him. Steve James Anon wrote: > Hi Kevin, > > Thanks, that was indeed the problem. Unfortunately when I add the O3CPU > to the scons command, m5 no longer properly compiles. I used the command: > > % scons build/ALPHA_FS/m5.fast > CPU_MODELS=AtomicSimpleCPU,TimingSimpleCPU,O3CPU > > and scons terminates with: > > /usr/bin/g++-3.4 -o build/ALPHA_FS/cpu/o3/base_dyn_inst.fo -c -pipe > -fno-strict-aliasing -Wall -Wno-sign-compare -Werror -Wundef -O3 > -DTHE_IS A=ALPHA_ISA -DNDEBUG -Iext/dnet -I/usr/include/python2.4 > -Ibuild/libelf/include -Ibuild/ALPHA_FS > build/ALPHA_FS/cpu/o3/base_dyn_inst.cc > /usr/bin/g++-3.4 -o build/ALPHA_FS/cpu/o3/bpred_unit.fo -c -pipe > -fno-strict-aliasing -Wall -Wno-sign-compare -Werror -Wundef -O3 > -DTHE_ISA=A LPHA_ISA -DNDEBUG -Iext/dnet -I/usr/include/python2.4 > -Ibuild/libelf/include -Ibuild/ALPHA_FS build/ALPHA_FS/cpu/o3/bpred_unit.cc > /usr/bin/g++-3.4 -o build/ALPHA_FS/cpu/o3/commit.fo -c -pipe > -fno-strict-aliasing -Wall -Wno-sign-compare -Werror -Wundef -O3 > -DTHE_ISA=ALPHA _ISA -DNDEBUG -Iext/dnet -I/usr/include/python2.4 > -Ibuild/libelf/include -Ibuild/ALPHA_FS build/ALPHA_FS/cpu/o3/commit.cc > build/ALPHA_FS/cpu/o3/commit_impl.hh: In member function `void > DefaultCommit<Impl>::skidInsert() [with Impl = O3CPUImpl]': > build/ALPHA_FS/cpu/o3/commit.cc:34: instantiated from here > build/ALPHA_FS/cpu/o3/commit_impl.hh:1204: warning: unused variable 'tid' > scons: *** [build/ALPHA_FS/cpu/o3/commit.fo] Error 1 > scons: building terminated because of errors. > > > I haven't had any compilation problems with any of the other > configurations. This occurs both when trying to compile the full system > and syscall emulation modes. Have you experienced this? > > Thanks again, > > James > > > On 9/13/06, *Kevin Lim* <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> wrote: > > Hi James, > > > > The problem is that the detailed model is not included in the models > being compiled by default. Normally we only compile the > AtomicSimpleCPU and TimingSimpleCPU models. When compiling M5, try > adding "CPU_MODELS=AtomicSimpleCPU,TimingSimpleCPU,O3CPU' to the > scons command. > > > > For example: > > > % scons build/ALPHA_SE/m5.debug > CPU_MODELS=AtomicSimpleCPU,TimingSimpleCPU,O3CPU > > > > That tells scons which CPU models to compile. Sorry about the > confusion, and I'll add this to the FAQ. > > > > Kevin > > > > ------------------------------------------------------------------------ > > *From:* [EMAIL PROTECTED] > <mailto:[EMAIL PROTECTED]> [mailto: > [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>] *On > Behalf Of *James Anon > *Sent:* Tuesday, September 05, 2006 2:30 PM > *To:* [email protected] <mailto:[email protected]> > *Subject:* [m5-users] Example detailed mode configuration bug > > > > Hi all, > > After downloading, compiling, and setting up m5 to run in the full > system mode, I tried to boot the simulator in the detailed mode with: > > ./m5.fast -d output/ configs/example/fs.py -d > > and I get the following error: > > M5 compiled Sep 1 2006 17:35:27 > M5 started Tue Sep 5 14:08:08 2006 > M5 executing on idealjames > command line: ./m5.fast -d output/ configs/example/fs.py -d > panic: Simulator object type 'DerivO3CPU' not found. > @ cycle 0 > [createObject:build/ALPHA_FS/sim/builder.cc, line 121] > Program aborted at cycle 0 > Aborted > > > Note that the system boots perfectly with the -t (for timing mode) > or with no argument (atomic mode). I found the Deriv03CPU class > within the src/python/m5/objects directory and attempted to setup > symlinks to see if this was simply a path error of some sort, but I > have never used Python before and didn't have much luck. I have also > installed the latest patch for m5 beta 2. > > I also posted this as a bug in the "Flyspray" bug tracker linked off > of the website (I noticed I was the first, is this up and running yet?). > > Thanks > > > _______________________________________________ > m5-users mailing list > [email protected] <mailto:[email protected]> > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > > ------------------------------------------------------------------------ > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
