qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v2 1/1] target-ppc: Fix booke206 tlbwe TLB instr


From: David Gibson
Subject: Re: [Qemu-devel] [PATCH v2 1/1] target-ppc: Fix booke206 tlbwe TLB instruction
Date: Tue, 16 Jan 2018 15:58:50 +1100
User-agent: Mutt/1.9.1 (2017-09-22)

On Mon, Jan 15, 2018 at 10:32:20AM +0100, Luc MICHEL wrote:
> When overwritting a valid TLB entry with a new one, the previous page
> were not flushed in QEMU TLB, leading to incoherent mapping. This commit
> fixes this.
> 
> Signed-off-by: Luc MICHEL <address@hidden>

Applied to ppc-for-2.12.

> ---
>  target/ppc/mmu_helper.c | 32 +++++++++++++++++++++++++++-----
>  1 file changed, 27 insertions(+), 5 deletions(-)
> 
> diff --git a/target/ppc/mmu_helper.c b/target/ppc/mmu_helper.c
> index 2a1f9902c9..298c15e961 100644
> --- a/target/ppc/mmu_helper.c
> +++ b/target/ppc/mmu_helper.c
> @@ -2570,6 +2570,17 @@ void helper_booke_setpid(CPUPPCState *env, uint32_t 
> pidn, target_ulong pid)
>      tlb_flush(CPU(cpu));
>  }
>  
> +static inline void flush_page(CPUPPCState *env, ppcmas_tlb_t *tlb)
> +{
> +    PowerPCCPU *cpu = ppc_env_get_cpu(env);
> +
> +    if (booke206_tlb_to_page_size(env, tlb) == TARGET_PAGE_SIZE) {
> +        tlb_flush_page(CPU(cpu), tlb->mas2 & MAS2_EPN_MASK);
> +    } else {
> +        tlb_flush(CPU(cpu));
> +    }
> +}
> +
>  void helper_booke206_tlbwe(CPUPPCState *env)
>  {
>      PowerPCCPU *cpu = ppc_env_get_cpu(env);
> @@ -2628,6 +2639,21 @@ void helper_booke206_tlbwe(CPUPPCState *env)
>      if (msr_gs) {
>          cpu_abort(CPU(cpu), "missing HV implementation\n");
>      }
> +
> +    if (tlb->mas1 & MAS1_VALID) {
> +        /* Invalidate the page in QEMU TLB if it was a valid entry.
> +         *
> +         * In "PowerPC e500 Core Family Reference Manual, Rev. 1",
> +         * Section "12.4.2 TLB Write Entry (tlbwe) Instruction":
> +         * (https://www.nxp.com/docs/en/reference-manual/E500CORERM.pdf)
> +         *
> +         * "Note that when an L2 TLB entry is written, it may be displacing 
> an
> +         * already valid entry in the same L2 TLB location (a victim). If a
> +         * valid L1 TLB entry corresponds to the L2 MMU victim entry, that L1
> +         * TLB entry is automatically invalidated." */
> +        flush_page(env, tlb);
> +    }
> +
>      tlb->mas7_3 = ((uint64_t)env->spr[SPR_BOOKE_MAS7] << 32) |
>          env->spr[SPR_BOOKE_MAS3];
>      tlb->mas1 = env->spr[SPR_BOOKE_MAS1];
> @@ -2663,11 +2689,7 @@ void helper_booke206_tlbwe(CPUPPCState *env)
>          tlb->mas1 &= ~MAS1_IPROT;
>      }
>  
> -    if (booke206_tlb_to_page_size(env, tlb) == TARGET_PAGE_SIZE) {
> -        tlb_flush_page(CPU(cpu), tlb->mas2 & MAS2_EPN_MASK);
> -    } else {
> -        tlb_flush(CPU(cpu));
> -    }
> +    flush_page(env, tlb);
>  }
>  
>  static inline void booke206_tlb_to_mas(CPUPPCState *env, ppcmas_tlb_t *tlb)

-- 
David Gibson                    | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
                                | _way_ _around_!
http://www.ozlabs.org/~dgibson

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]