Optimizing Index Computations in SWG for Improved Timing#1509
Optimizing Index Computations in SWG for Improved Timing#1509
Conversation
STFleming
left a comment
There was a problem hiding this comment.
Looks good to me and tests are passing on my end.
fpjentzsch
left a comment
There was a problem hiding this comment.
LGTM, all tests passing on our end.
I did a small spot check on 16 exemplary SWG configurations and the average Fmax actually decreased from 194 MHz to 192 MHz. Maybe I didn't cover any cases where these changes would have a positive impact on timing.
Where there any specific edge cases that caused you trouble before?
|
@preusser Just out of curiosity I did another experiment with 224 different SWG configurations (using OOC synth with 200 MHz target clock):
Respective configurations with min/max Fmax: [Before MIN] fmax=141.18 MHz |

No description provided.