[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v2 11/14] spapr: nested: Use correct source for parttbl info
From: |
Nicholas Piggin |
Subject: |
Re: [PATCH v2 11/14] spapr: nested: Use correct source for parttbl info for nested PAPR API. |
Date: |
Wed, 29 Nov 2023 14:15:17 +1000 |
On Thu Oct 12, 2023 at 8:49 PM AEST, Harsh Prateek Bora wrote:
> For nested PAPR API, we use SpaprMachineStateNestedGuest struct to store
> partition table info, use the same in spapr_get_pate_nested() as well.
>
> Signed-off-by: Michael Neuling <mikey@neuling.org>
> Signed-off-by: Harsh Prateek Bora <harshpb@linux.ibm.com>
> ---
> hw/ppc/spapr.c | 5 +---
> hw/ppc/spapr_nested.c | 49 +++++++++++++++++++++++------------
> include/hw/ppc/spapr_nested.h | 3 +++
> 3 files changed, 36 insertions(+), 21 deletions(-)
>
> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> index 14196fdd11..e1f7b29842 100644
> --- a/hw/ppc/spapr.c
> +++ b/hw/ppc/spapr.c
> @@ -1356,10 +1356,7 @@ static bool spapr_get_pate(PPCVirtualHypervisor *vhyp,
> PowerPCCPU *cpu,
> return true;
> } else {
> assert(spapr->nested.api);
> - if (spapr->nested.api == NESTED_API_KVM_HV) {
> - return spapr_get_pate_nested(spapr, cpu, lpid, entry);
> - }
> - return false;
> + return spapr_get_pate_nested(spapr, cpu, lpid, entry);
If you don't permit the api to be changed to PAPR at the start, then you
don't need to juggle this back and forth.
> }
> }
>
> diff --git a/hw/ppc/spapr_nested.c b/hw/ppc/spapr_nested.c
> index f41b0c0944..1917d8bb13 100644
> --- a/hw/ppc/spapr_nested.c
> +++ b/hw/ppc/spapr_nested.c
> @@ -19,29 +19,45 @@ void spapr_nested_init(SpaprMachineState *spapr)
> bool spapr_get_pate_nested(SpaprMachineState *spapr, PowerPCCPU *cpu,
> target_ulong lpid, ppc_v3_pate_t *entry)
> {
> - uint64_t patb, pats;
> + if (spapr->nested.api == NESTED_API_KVM_HV) {
Maybe split out the implementations into two functions?
And actually instead of open coding the API test, I'm thinking
maybe you should add a helper function
spapr_cpu_nested_api(cpu) which returns NESTED_API_... and you
can just hard code it to return spapr->nested.api for now.
That should avoid a bunch of churn when we permit guests to
run with different APIs.
> + uint64_t patb, pats;
>
> - assert(lpid != 0);
> + assert(lpid != 0);
>
> - patb = spapr->nested.ptcr & PTCR_PATB;
> - pats = spapr->nested.ptcr & PTCR_PATS;
> + patb = spapr->nested.ptcr & PTCR_PATB;
> + pats = spapr->nested.ptcr & PTCR_PATS;
>
> - /* Check if partition table is properly aligned */
> - if (patb & MAKE_64BIT_MASK(0, pats + 12)) {
> - return false;
> - }
> + /* Check if partition table is properly aligned */
> + if (patb & MAKE_64BIT_MASK(0, pats + 12)) {
> + return false;
> + }
>
> - /* Calculate number of entries */
> - pats = 1ull << (pats + 12 - 4);
> - if (pats <= lpid) {
> - return false;
> + /* Calculate number of entries */
> + pats = 1ull << (pats + 12 - 4);
> + if (pats <= lpid) {
> + return false;
> + }
> +
> + /* Grab entry */
> + patb += 16 * lpid;
> + entry->dw0 = ldq_phys(CPU(cpu)->as, patb);
> + entry->dw1 = ldq_phys(CPU(cpu)->as, patb + 8);
> + return true;
> }
> +#ifdef CONFIG_TCG
Nested spapr is entirely TCG, no? If we're not properly guarding the
nested code with CONFIG_TCG, that should go as an initial prep patch.
Thanks,
Nick
> + /* Nested PAPR API */
> + SpaprMachineStateNestedGuest *guest;
> + assert(lpid != 0);
> + guest = spapr_get_nested_guest(spapr, lpid);
> + assert(guest != NULL);
> +
> + entry->dw0 = guest->parttbl[0];
> + entry->dw1 = guest->parttbl[1];
>
> - /* Grab entry */
> - patb += 16 * lpid;
> - entry->dw0 = ldq_phys(CPU(cpu)->as, patb);
> - entry->dw1 = ldq_phys(CPU(cpu)->as, patb + 8);
> return true;
> +#else
> + return false;
> +#endif
> }
>
> #ifdef CONFIG_TCG
> @@ -412,7 +428,6 @@ void spapr_exit_nested(PowerPCCPU *cpu, int excp)
> address_space_unmap(CPU(cpu)->as, regs, len, len, true);
> }
>
> -static
> SpaprMachineStateNestedGuest *spapr_get_nested_guest(SpaprMachineState
> *spapr,
> target_ulong guestid)
> {
> diff --git a/include/hw/ppc/spapr_nested.h b/include/hw/ppc/spapr_nested.h
> index d27688b1b8..e04202ecb4 100644
> --- a/include/hw/ppc/spapr_nested.h
> +++ b/include/hw/ppc/spapr_nested.h
> @@ -457,4 +457,7 @@ bool spapr_get_pate_nested(SpaprMachineState *spapr,
> PowerPCCPU *cpu,
> void spapr_register_nested_papr(void);
> void spapr_nested_init(SpaprMachineState *spapr);
> void spapr_nested_gsb_init(void);
> +SpaprMachineStateNestedGuest *spapr_get_nested_guest(SpaprMachineState
> *spapr,
> + target_ulong lpid);
> +
> #endif /* HW_SPAPR_NESTED_H */
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Re: [PATCH v2 11/14] spapr: nested: Use correct source for parttbl info for nested PAPR API.,
Nicholas Piggin <=