Re: [PATCH v2 20/28] target/ppc/mmu_common.c: Make get_physical_address_wtlb() static

["Nicholas Piggin" <npiggin@gmail.com>]

On Thu May 2, 2024 at 9:43 AM AEST, BALATON Zoltan wrote:
> This function is not used from any other files so make it static and
> fix the maybe used uninitialised warnings this has uncovered.
>
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
> ---
>  target/ppc/internal.h   | 5 +----
>  target/ppc/mmu_common.c | 5 ++++-
>  2 files changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/target/ppc/internal.h b/target/ppc/internal.h
> index 601c0b533f..7a99f08dc8 100644
> --- a/target/ppc/internal.h
> +++ b/target/ppc/internal.h
> @@ -261,10 +261,7 @@ typedef struct mmu_ctx_t mmu_ctx_t;
>  bool ppc_xlate(PowerPCCPU *cpu, vaddr eaddr, MMUAccessType access_type,
>                        hwaddr *raddrp, int *psizep, int *protp,
>                        int mmu_idx, bool guest_visible);
> -int get_physical_address_wtlb(CPUPPCState *env, mmu_ctx_t *ctx,
> -                                     target_ulong eaddr,
> -                                     MMUAccessType access_type, int type,
> -                                     int mmu_idx);
> +
>  /* Software driven TLB helpers */
>  int ppc6xx_tlb_getnum(CPUPPCState *env, target_ulong eaddr,
>                                      int way, int is_code);
> diff --git a/target/ppc/mmu_common.c b/target/ppc/mmu_common.c
> index 762b13805b..4852cb5571 100644
> --- a/target/ppc/mmu_common.c
> +++ b/target/ppc/mmu_common.c
> @@ -666,6 +666,7 @@ static int mmubooke_get_physical_address(CPUPPCState *env, mmu_ctx_t *ctx,
>      hwaddr raddr = (hwaddr)-1ULL;
>      int i, ret = -1;
>  
> +    ctx->prot = 0;
>      for (i = 0; i < env->nb_tlb; i++) {
>          tlb = &env->tlb.tlbe[i];
>          ret = mmubooke_check_tlb(env, tlb, &raddr, &ctx->prot, address,
> @@ -873,6 +874,7 @@ static int mmubooke206_get_physical_address(CPUPPCState *env, mmu_ctx_t *ctx,
>      hwaddr raddr = (hwaddr)-1ULL;
>      int i, j, ways, ret = -1;
>  
> +    ctx->prot = 0;
>      for (i = 0; i < BOOKE206_MAX_TLBN; i++) {
>          ways = booke206_tlb_ways(env, i);
>          for (j = 0; j < ways; j++) {

The prot warnings are valid AFAIKS, used uninit by qemu_log_mask call.

So, I see what the booke _check_tlb() functions are doing with
*prot now and that is to assign it iff return value is 0 or -2 or
-3, matching TLB address (and possibly mismatch prot).

Would it be better to fix it as:

    qemu_log_mask(CPU_LOG_MMU,
                  "%s: access %s " TARGET_FMT_lx " => " HWADDR_FMT_plx
                  " %d %d\n", __func__, ret < 0 ? "refused" : "granted",
                  address, raddr, ret == -1 ? 0 : ctx->prot, ret);

This way it's clearer that we're only printing it when a TLB was
found, and it won't silence other possible use-uninitialised?

> @@ -1144,7 +1146,7 @@ void dump_mmu(CPUPPCState *env)
>      }
>  }
>  
> -int get_physical_address_wtlb(CPUPPCState *env, mmu_ctx_t *ctx,
> +static int get_physical_address_wtlb(CPUPPCState *env, mmu_ctx_t *ctx,
>                                       target_ulong eaddr,
>                                       MMUAccessType access_type, int type,
>                                       int mmu_idx)
> @@ -1163,6 +1165,7 @@ int get_physical_address_wtlb(CPUPPCState *env, mmu_ctx_t *ctx,
>      if (real_mode && (env->mmu_model == POWERPC_MMU_SOFT_6xx ||
>                        env->mmu_model == POWERPC_MMU_SOFT_4xx ||
>                        env->mmu_model == POWERPC_MMU_REAL)) {
> +        memset(ctx, 0, sizeof(*ctx));
>          ctx->raddr = eaddr;
>          ctx->prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC;
>          return 0;

I wonder why the compiler doesn't see these, they are all in the return
not-zero cases that should be quite visible?

What if you leave the static change to the end of your series, do the
simplifications allow the compiler to work it out? I prefer not to
squash such compiler warnings if it can be avoided.

Thanks,
Nick