This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: K6 and decoding bottleneck


> So I am thinking about better implementation. I want to ask scheduling
> gurus for ideas to try.  I really don't know how to make scheduler handle
> this optimally so hope someone will get the idea.

I suggest to add one or more function units for the decoder, and add
appropriate usages of this unit / these units.  If there is a common
minimum latency between decoding and execution, you can just ignore this
common latency.  It only makes a difference when the pipelines need to
be refilled, i.e. when taking mispredicted branches, and the scheduler
doesn't take these into account anyways.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]