Mohamed Gaber

### Description Some OpenROAD steps are observed to behave at least a bit differently with multiple corners loaded. ### Proposal All steps should be able to read multiple corners, similar...

### Description The config.tcl files for new PDKs are laborious to create and quite confusing. ### Proposal A lot of the data can be extracted using a number of [odb](https://github.com/The-OpenROAD-Project/OpenROAD)...

### Description The RC values are not a universal fit for all designs (because of differences in congestion, layer utilization, and so on), so there may be a large discrepancy...

Sweep and understand the effect of OpenROAD placement flags

1

### Description OpenROAD has approximately 9 million parameters for placement. It is unclear what any of them do without conducting a literature review. ### Proposal Study the effect of the...

### Description `run_designs` was an unwieldy piece of crap. However, what it did have is a pretty neat sweeping feature. ### Proposal We should migrate this script to OpenLane 2...

### Description OpenROAD has a feature that supports antenna repair post-detailed routing. This is the most accurate information for antenna repair possible, and may result in better metrics. ### Proposal...

### Description Currently, a user may specify one macro for all corners. This is not granular enough and will likely result in bad chips because of not being accounted for...

### Description Currently, OpenLane has its clock information strewn all over the place in variables. The problem is while OpenLane steps in general assume the source of this information is...

set_debug_level DRT snapshot save directory

3

### Description When you `set_debug_level DRT snapshot 1`, the snapshots are saved to the current working directory. An option to set where they go would be pretty cool. ### Suggested...

## Is your feature request related to a problem? What I assume is a fairly common use case for setting up caches is to only copy parts of the closure...