Hi,
Is this the right place to submit bug reports?
When I restore with the detailed cpu after booting X86 Linux
(2.6.28.4) and checkpointing, gem5 is aborting. I have reproduced this
issue using the gem5 tip, and with the pre-packaged
x86_64-vmlinux-2.6.22.9.
I was able to checkpoint and restore in detailed mode with ARM using
vmlinux.arm.smp.fb.2.6.38.8 and linux-arm-ael.img.
$ build/X86/gem5.debug --debug-flags=Drain configs/example/fs.py -r 1
--restore-with-cpu=detailed >debug.txt
warn: add_child('terminal'): child 'terminal' already has parent
Listening for com_1 connection on port 3456
warn: Reading current count from inactive timer.
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
hack: be nice to actually delete the event here
...
gem5.debug: build/X86/cpu/o3/fetch_impl.hh:432: void
DefaultFetch<Impl>::drainSanityCheck() const [with Impl = O3CPUImpl]:
Assertion `!memReq[i]' failed.
Program aborted at cycle 17350227428500
The output from Drain flags:
Switched CPUS @ tick 17346781476000
switching cpus
17346781476000: system.physmem: DRAM controller not drained, write: 0,
read: 0, resp: 1
17346781476000: system.cpu: Draining...
17346781476000: system.cpu: Fetch not drained.
17346781476000: system.cpu: CPU not drained
info: Entering event queue @ 17346781476000. Starting simulation...
17346781476500: system.cpu: Fetch not drained.
17346781477000: system.cpu: Fetch not drained.
17346781477500: system.cpu: Fetch not drained.
17346781478000: system.cpu: Fetch not drained.
17346781478500: system.cpu: Fetch not drained.
.... 5.9 million lines later
17350227427000: system.cpu: Main CPU structures not drained.
17350227427000: system.cpu: Fetch not drained.
17350227427000: system.cpu: Commit not drained.
17350227427500: system.cpu: Main CPU structures not drained.
17350227427500: system.cpu: Fetch not drained.
17350227427500: system.cpu: Commit not drained.
17350227428000: system.cpu.commit: Draining:
0:(0xffffffff8020bd20=>0xffffffff8020bd28).(0=>1)
17350227428000: system.cpu.fetch: 0: Thread drained.
17350227428000: system.cpu: Fetch not drained.
17350227428500: system.cpu: CPU done draining, processing drain event
17350227428500: system.membus.reqLayer: Bus not drained
17350227428500: system.membus.respLayer: Bus not drained
17350227428500: system.physmem: DRAM controller not drained, write: 5,
read: 0, resp: 2
17350227428500: system.cpu: Draining...
17350227428500: system.cpu: CPU is already drained
That is where gem5 aborts. Any advice or workaround would be appreciated.
-Gedare
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
