Dear all,

I'm using the latest ALPHA_FS/gem5.opt simulator.
I confront a problem when memsize in Benchmarks.py is set to 2048MB.
If doing so, I see simulator stops at the following step (in m5's terminal)
for a few minutes,

Inode-cache hash table entries: 2097152 (order: 11, 16777216 bytes)

then, exits with following assertion (in host's terminal):

m5.opt: build/ALPHA_FS/cpu/simple/atomic.cc:349: Fault
AtomicSimpleCPU::readMem(Addr, uint8_t*, unsigned int, unsigned int):
Assertion `!pkt.isError()' failed.
Program aborted at cycle 468935876000

If I set memsize to 2047MB then everything goes well.
What I notice is when I use 2048MB configuration, memsize seems overflow,
i.e., FFFFFFFF80000000 in m5's terminal as shown as following.

==== m5 slave terminal: Terminal 0 ====
M5 console: m5AlphaAccess @ 0xFFFFFD0200000000
Got Configuration 623
memsize FFFFFFFF80000000 pages FFFFFFFFFFFC0000
...

I use 64-bit machines. I'm wondering if I do any thing wrong.
I'd very appreciate it if you could give me any hint.
Thanks in advance.

Sincerely,
RS

Following lists detailed system information

uname -a
2.6.32-5-amd64 #1 SMP Tue Jun 14 09:42:28 UTC 2011 x86_64 GNU/Linux

gcc -v
gcc version 4.4.5 (Debian 4.4.5-8)

simulation command line:
gem5/build/ALPHA_FS/m5.opt \
gem5/configs/example/fs.py \
--prog-interval=50000000 \
--caches \
--l1d_size=16kB \
--l1i_size=16kB \
--l1d_assoc=4 \
--l1i_assoc=4 \
--cacheline_size=64 \
--l2cache \
--l2_size=1MB \
--l2_assoc=16 \
--cacheline_size=64 \
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to