You are right, we can’t say exception specific non-zero value here. Also, same is true for virtual instruction trap as well. Let me quickly re-phrase this. Regards, Anup From:

Reviewed-by: Atish Patra <atish.patra@...> -- Regards, Atish

Reviewed-by: Atish Patra <atish.patra@...> -- Regards, Atish

Reviewed-by: Atish Patra <atish.patra@...> -- Regards, Atish

Reviewed-by: Atish Patra <atish.patra@...> -- Regards, Atish

+** When an illegal instruction trap is taken into S-mode, exception-specific + non-zere value must be written to the `stval` CSR. nit: When the illegal instruction is c.unimp (encoded as 0x0000)

Instead of describe the machine-level timer for the M platform, we should simply point to the RISC-V ACLINT MTIMER defined by the RISC-V ACLINT specification. Signed-off-by: Anup Patel

The image files (i.e. PNG files) are committed in the repo so we should not remove it when doing "make clean". Signed-off-by: Anup Patel <anup.patel@...> --- Makefile | 1 - 1 file changed, 1

We should follow profile naming as-per latest RISC-V profiles specification. Also, we should avoid explicit mentions of "RV32xxx" and "RV64xxx" ISA strings. Signed-off-by: Anup Patel

We don't need a table for required SBI extensions and UEFI interfaces implemented by UEFI firmware. Signed-off-by: Anup Patel <anup.patel@...> Reviewed-by: Alistair Francis

We re-write the interrupts and timer section to align the interrupts and timer table. We also add more fine grained requirements for AIA. Signed-off-by: Anup Patel <anup.patel@...> --- Makefile

The terminology table should be at start and specification table should be at the end. Also, specification table should be replaced with bibliography reference list. Signed-off-by: Anup Patel

This patch updates terminology and specification table in following ways: 1) Provide complete expansion of terms for APLIC and ACLINT 2) Add links to unprivileged and privileged specifications 3) Add

To have a meaningful H-extension support, both OS/A-base and OS/A-server platforms must comply with additional requirements for H-extension. Also, the little-endian requirement in OS/A-base should

On 10/08/21, 12:11 AM, "Atish Patra" <Atish.Patra@...> wrote: Here are more details on profile naming convention https://github.com/riscv/riscv-profiles/blob/master/profiles.adoc

On 09/08/21, 5:06 AM, "Alistair Francis" <Alistair.Francis@...> wrote: Why do these have 64 at the end? The "64" at the end means RV64. Regards, Anup Alistair

I think we should move the requirement to OS-A base and clearly state that non-zero exception specific information should be written to mtinst or htinst. Regards, Anup From:

Yes, both the ACLINT and PLIC specs need to go through the process ASAP of becoming approved standards. Obviously that can overlap with the process for the platform specs, but we should get both

The above adoc is very tentative and hasn't been reviewed at all. Even Krste would acknowledge that the naming details will change. So whatever we use in the platform specs is just a placeholder that

Reviewed-by: Atish Patra <atish.patra@...> -- Regards, Atish