16-bit imp_id field seems too small

[MarkHillCodasip]

The allocation of just 16-bits to vendor_n_imp_id.imp_id to cover all of a companies portfolio of IPs and versions seems quite small to me, especially if a vendor wants to break this broken down into subfields for classes of IP and major/minor/patch versions etc. Specifically in the case of RISCV processor implementations it seems reasonable to support a one-one mapping between RERI registers and mvendorid, marchid and mimpid.

[ved-rivos]

Its reasonable to widen the imp_id to 32-bits by expanding into the 16 reserved bits adjacent to it. I will post a PR shortly with a change.

[MarkHillCodasip]

Thanks!

…

On Mon, 18 Mar 2024 at 14:01, Ved Shanbhogue ***@***.***> wrote: Its reasonable to widen the imp_id to 32-bits by expanding into the 16 reserved bits adjacent to it. I will post a PR shortly with a change. — Reply to this email directly, view it on GitHub <#46 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BEROVD3WHG3KRUTFGXUWYT3YY3XTJAVCNFSM6AAAAABEWA4MUKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMBUGAYDEMZSGI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[ved-rivos]

Update in PR #47