I had used gem5-stable for some weeks, and the gem5-stable didn't cause
seg fault with ruby model. However I NEVER successfully restored from a
checkpoint when I use gem5-stable with ruby_fs.py.
On 03/25/2012 04:44 AM, Xiao Guo wrote:
I also have the same problem. Did you create the checkpoint with fs.py or
ruby_fs.py?
On Mar 23, 2012, at 5:11 PM, Michael Moeng wrote:
I looked at src/mem/bus.cc -- there are two places it tries to access
peers which will segfault if there are none. I was able to get past
that particular segfault by guarding those statements with
if((*intIter)->isConnected()).
This appears to let the timing model run; but when I tried to run the
PARSEC benchmark I got another error with an atomic request being sent
to a ruby port:
command line: ./build/ALPHA/gem5.opt --remote-gdb-port=0
./configs/example/ruby_fs.py -n 16
--checkpoint-dir=checkpoints/n16/blackscholes/large --cpu-type=timing
-r 2 -m 10000000000
warning: overwriting port<orphan LinuxAlphaSystem>.tsunami.ide.dma
value<orphan LinuxAlphaSystem>.piobus.slave[0] with<orphan
LinuxAlphaSystem>.dma_cntrl0.dma_sequencer.slave[0]
warning: overwriting port<orphan LinuxAlphaSystem>.tsunami.ide.dma
value<orphan LinuxAlphaSystem>.dma_cntrl0.dma_sequencer.slave[0] with
<orphan LinuxAlphaSystem>.dma_cntrl0.dma_sequencer.slave[1]
warning: overwriting port<orphan
LinuxAlphaSystem>.tsunami.ethernet.dma value<orphan
LinuxAlphaSystem>.piobus.slave[1] with<orphan
LinuxAlphaSystem>.dma_cntrl1.dma_sequencer.slave[0]
warning: overwriting port<orphan
LinuxAlphaSystem>.tsunami.ethernet.dma value<orphan
LinuxAlphaSystem>.dma_cntrl1.dma_sequencer.slave[0] with<orphan
LinuxAlphaSystem>.dma_cntrl1.dma_sequencer.slave[1]
Global frequency set at 1000000000000 ticks per second
info: kernel located at:
/home/moeng/work/m5/full_system/binaries/vmlinux_2.6.27-gcc_4.3.4
Listening for system connection on port 3456
0: system.tsunami.io.rtc: Real-time clock set to Thu Jan 1 00:00:00 2009
Switch at curTick count:10000
info: Entering event queue @ 2747178166000. Starting simulation...
panic: RubyPort::M5Port::recvAtomic() not implemented!
@ cycle 2747178166500
[recvAtomic:build/ALPHA/mem/ruby/system/RubyPort.cc, line 125]
Memory Usage: 1496548 KBytes
Program aborted at cycle 2747178166500
Aborted (core dumped)
On Thu, Mar 22, 2012 at 8:50 AM, Hamid Reza Khaleghzadeh
<[email protected]> wrote:
Hi,
I have similar problem. please inform me, If you can find a solution.
Thanks
On 3/20/12, Chuanlei<[email protected]> wrote:
Hi,
I am trying to run blackscholes benchmark on ruby memory model, but this
cause a segmentation fault.
Then I try to use gdb to debug this fault, and following is the output.
=================gdb output===========
(gdb) r
Starting program: /home/zheng/gem5/build/ALPHA_MOESI_hammer/gem5.opt
configs/example/ruby_fs.py --cpu-type=timing
[Thread debugging using libthread_db enabled]
gem5 Simulator System. http://gem5.org
gem5 is copyrighted software; use the --copyright option for details.
gem5 compiled Mar 20 2012 18:48:26
gem5 started Mar 20 2012 19:44:20
gem5 executing on zheng-desktop
command line: /home/zheng/gem5/build/ALPHA_MOESI_hammer/gem5.opt
configs/example/ruby_fs.py --cpu-type=timing
warning: overwriting port<orphan LinuxAlphaSystem>.tsunami.ide.dma
value<orphan LinuxAlphaSystem>.piobus.slave[0] with<orphan
LinuxAlphaSystem>.dma_cntrl0.dma_sequencer.slave[0]
warning: overwriting port<orphan LinuxAlphaSystem>.tsunami.ethernet.dma
value<orphan LinuxAlphaSystem>.piobus.slave[1] with<orphan
LinuxAlphaSystem>.dma_cntrl1.dma_sequencer.slave[0]
Global frequency set at 1000000000000 ticks per second
info: kernel located at: /home/zheng/dist/binaries/vmlinux
Listening for system connection on port 3456
0: system.tsunami.io.rtc: Real-time clock set to Thu Jan 1
00:00:00 2009
Program received signal SIGSEGV, Segmentation fault.
Bus::init (this=0x95fbe48) at build/ALPHA_MOESI_hammer/mem/bus.cc:132
132 if ((*intIter)->getPeer()->isSnooping()) {
(gdb)
==========================================
Any help really appreciated!
Thanks.
----------------
Chuanlei
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
--
Hamid Reza Khaleghzadeh
hkhaleghzadeh.webs.com
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
