[PATCH] Adds Cortex-A53 and Cortex-A57 subtargets.

Dave Estes cestes at codeaurora.org
Mon Mar 3 09:44:49 PST 2014

> If this is already working for you, that's cool. But I should point out that typical in-order machines would set BufferSize=0 on the resources. That would prevent two instructions that need to be issued on the same pipe from being scheduled back-to-back in the same "cycle".
> This comment just looks stale:
> +// A53UnitFPMDS is the only exception, wheras is has BufferSize = 1 since it is
> +// not pipelined for most of its instructions.
> Otherwise looks good.

Thanks, Andy. I did set BufferSize=0 for the ProcResources and it 
introduced a regression. I haven't tracked it down but I intend to when 
I model the forwarding logic. As it stands today, the latencies are far 
from accurate and I want to get that shored up first.

I'll fix the comment now.


Employee of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
hosted by The Linux Foundation

More information about the llvm-commits mailing list