Here is a trace snippet that I'm wondering about:
999996000: system.cpu.execution_unit: [tid:0] 
Executing [sn:1745612] [PC:
(0x1200527e4=>0x1200527e8)] bne.

999996000: system.cpu.execution_unit: [tid:0]: 
[sn:1745612] Squashing from stage 2. 
Redirecting  fetch to (0x120052848=>0x12005284c).

999996000: system.cpu.execution_unit: [tid:0] [sn:1745612] 
bne        r8,0x120052848 ...
PC (0x1200527e4=>0x120052848) ... 
Mispredicts! (Prediction: Taken)

It seems as though the predictor has predicted taken 
and begun fetching starting at 
0x120052848 (the branch target). 
Then the execution unit causes a squash and 
redirection to 0x120052848 (the branch target). 
It seems like this should not be a misprediction. 
Am I reading this trace incorrectly?

Thanks!
Henry


_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to