Skip to content

Preserve shortreal hierarchy width - #10

Open
muzafferkal wants to merge 1 commit into
gtkwave:mainfrom
muzafferkal:fix-shortreal-metadata
Open

muzafferkal wants to merge 1 commit into
gtkwave:mainfrom
muzafferkal:fix-shortreal-metadata

Conversation

@muzafferkal

Copy link
Copy Markdown

Keep 32-bit hierarchy metadata while using eight-byte real storage for promoted shortreal values.

Keep 32-bit hierarchy metadata while using eight-byte real storage for promoted shortreal values.
@muzafferkal

Copy link
Copy Markdown
Author

@rfuest how do I get this reviewed pls?

@rfuest

rfuest commented Aug 13, 2026

Copy link
Copy Markdown
Contributor

I think that both GTKWave branches ignore the bit width for real and shortreal variables, so this change shouldn't break anything.

Why do you require this change for shortreal when real still outputs 0 as the bitwidth? I would prefer to have one consistent behavior for all floating point types.

@muzafferkal

Copy link
Copy Markdown
Author

I think that both GTKWave branches ignore the bit width for real and shortreal variables, so this change shouldn't break anything.

Why do you require this change for shortreal when real still outputs 0 as the bitwidth? I would prefer to have one consistent behavior for all floating point types.

I'm trying to add shortreal support to verilator and Wilson asked me to make this change here first so that verilator can inherit.

@muzafferkal

Copy link
Copy Markdown
Author

@rfuest could you please give me some guidance on how to proceed here?

@rfuest

rfuest commented Aug 31, 2026

Copy link
Copy Markdown
Contributor

Sorry for the late reply.

I still don't understand why this change is required, if the bit width is ignored for all real values. The 32 bit width is implied in the SV_SHORTREAL vartype. Could you point me to where something breaks without this change applied?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants