src/mte_tag: HLV/HSV/MPRV support for opt-out based on PTE.MTAG for c…

[deepak0414]

…odepage

Guest might have set PTE.MTAG for code page and thus wouldn't be expecting tag checks even if data pointer had PTE.MTAG set. On an exit to hypervisor, hypervisor emulating this load/store must do emulation similarly and thus HLVX* is updated to collect to MTAG bit in hstatus. MTAG_I and use that during HLV/HSV.

Similar mechanisms when in M-mode and MPRV=1.

[deepak0414]

@radimkrcmar @SiFiveHolland

I re-worked the treatment of MTAG from higher priv to lower priv. Please take a look.
Following are changes from earlier

• MTAG_I is only reported on traps when *tinst is non-zero
• MTAG_I is not collected/harvested
• If MTAG_I is clear and first-stage paging rules say "check tags", then tags will be checked.

If guest code page had PTE.MTAG=1, that's fine. HLV/HSV or ld/st with MPRV=1 should still do tagchecks
PTE.MTAG=1 is just an optimization for guest to avoid tag checks. In this case, its already trapped to
higher priv for emulation and it is expected that if PTE.MTAG=1 for data page for guest, its tag must have
been set.

Let me know if this looks ok.

[SiFiveHolland]

Looks reasonable to me