Closed zhangdujiao closed 3 months ago
This is a trade-off between encoding efficiency and area. Larger values will always produce more efficient encoding, but cost more in terms of gates. That said, even relatively small values (e.g. 2) will provide significant efficiency improvements.
Iain
From: zhangdujiao @.> Sent: 12 June 2024 04:05 To: riscv-non-isa/riscv-trace-spec @.> Cc: Subscribed @.***> Subject: [riscv-non-isa/riscv-trace-spec] Are there any recommended value of call_counter_size_p, return_stack_size_p, bpred_size_p and cache_size_p? (Issue #118)
-Implicit return mode call_counter_size_p return_stack_size_p
-Branch prediction mode bpred_size_p
-Jump target cache mode cache_size_p
- Reply to this email directly, view it on GitHubhttps://github.com/riscv-non-isa/riscv-trace-spec/issues/118, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALQOPSQK5XCNAHBE2QFCCVLZG63GTAVCNFSM6AAAAABJFO7VT6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGM2DONRZHA4DGNQ. You are receiving this because you are subscribed to this thread.Message ID: @.**@.>>
-Implicit return mode call_counter_size_p return_stack_size_p
-Branch prediction mode bpred_size_p
-Jump target cache mode cache_size_p