PR for the spec fix, in case anyone is interested. Found a couple of other
references to Smcsrind that I also removed.
https://github.com/riscv/riscv-control-transfer-records/pull/29
On Tue, Jun 4, 2024 at 8:41 PM Beeman Strong wrote:
> Ah, good catch. We removed that dependency late. I
on for VU-mode.
>
> If there is any misunderstanding, please let me know. Also Sstateen0 does
> not involve in CTR. Am I correct?
>
That all looks correct. sctrdepth gets that special treatment in VS-mode
(bold and underlined above) because it is expected that a hypervisor will
wis
hapter 1 states that:
> Smctr depends on the Smcsrind extension, while Ssctr depends on the
> Sscsrind extension. Further, both Smctr and Ssctr depend upon
> implementation of S-mode.
> Beeman Strong 於 2024/6/5 上午 06:32 寫道:
>
> There is no dependency on Smcsrind, only Sscsrind.
>
There is no dependency on Smcsrind, only Sscsrind.
On Tue, Jun 4, 2024 at 12:29 AM Jason Chien wrote:
> Smctr depends on the Smcsrind extension, Ssctr depends on the Sscsrind
> extension, and both Smctr and Ssctr depend upon implementation of S-mode.
> There should be a dependency check in riscv
On Tue, Jun 4, 2024 at 10:19 AM Jason Chien wrote:
>
> Rajnesh Kanwal 於 2024/5/30 上午 12:09 寫道:
> > CTR extension adds a new instruction sctrclr to quickly
> > clear the recorded entries buffer.
> >
> > Signed-off-by: Rajnesh Kanwal
> > ---
> > target/riscv/cpu.h |