qemu-riscv
[Top][All Lists]
Advanced

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

Re: [PATCH v5 01/10] target/riscv/tcg: add 'zic64b' support


From: Andrew Jones
Subject: Re: [PATCH v5 01/10] target/riscv/tcg: add 'zic64b' support
Date: Thu, 26 Oct 2023 14:04:53 +0200

On Wed, Oct 25, 2023 at 08:44:50PM -0300, Daniel Henrique Barboza wrote:
> zic64b is defined in the RVA22U64 profile [1] as a named feature for
> "Cache blocks must be 64 bytes in size, naturally aligned in the address
> space".  It's a fantasy name for 64 bytes cache blocks. RVA22U64
> mandates this feature, meaning that applications using it expects 64
> bytes cache blocks.
> 
> In theory we're already compliant to it since we're using 64 bytes cache
> block sizes by default, but nothing is stopping users from enabling a
> profile and changing the cache block size at the same time.
> 
> We'll add zic64b as a 'named feature', not a regular extension, in a
> sense that we won't write it in riscv,isa. It'll be used solely to track
> whether the user changed cache sizes and if we should warn about it.
> 
> zic64b is default to 'true' since we're already using 64 bytes blocks.
> If any cache block size (cbom_blocksize or cboz_blocksize) is changed to
> something different than 64, zic64b is set to 'false' and, if zic64b was
> set to 'true' in the command line, also throw an user warning.
> 
> Our profile implementation set mandatory extensions as if users enabled
> them in the command line, so this logic will extend to the future RVA22U64
> implementation as well.
> 
> [1] 
> https://github.com/riscv/riscv-profiles/releases/download/v1.0/profiles.pdf
> 
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
> ---
>  target/riscv/cpu.c         | 12 ++++++++++--
>  target/riscv/cpu.h         |  3 +++
>  target/riscv/cpu_cfg.h     |  1 +
>  target/riscv/tcg/tcg-cpu.c | 26 ++++++++++++++++++++++++++
>  4 files changed, 40 insertions(+), 2 deletions(-)
> 
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index f40da4c661..5095f093ba 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -1394,6 +1394,12 @@ const RISCVCPUMultiExtConfig 
> riscv_cpu_experimental_exts[] = {
>      DEFINE_PROP_END_OF_LIST(),
>  };
>  
> +const RISCVCPUMultiExtConfig riscv_cpu_named_features[] = {
> +    MULTI_EXT_CFG_BOOL("zic64b", zic64b, true),
> +
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
>  /* Deprecated entries marked for future removal */
>  const RISCVCPUMultiExtConfig riscv_cpu_deprecated_exts[] = {
>      MULTI_EXT_CFG_BOOL("Zifencei", ext_zifencei, true),
> @@ -1423,8 +1429,10 @@ Property riscv_cpu_options[] = {
>      DEFINE_PROP_UINT16("vlen", RISCVCPU, cfg.vlen, 128),
>      DEFINE_PROP_UINT16("elen", RISCVCPU, cfg.elen, 64),
>  
> -    DEFINE_PROP_UINT16("cbom_blocksize", RISCVCPU, cfg.cbom_blocksize, 64),
> -    DEFINE_PROP_UINT16("cboz_blocksize", RISCVCPU, cfg.cboz_blocksize, 64),
> +    DEFINE_PROP_UINT16("cbom_blocksize", RISCVCPU,
> +                       cfg.cbom_blocksize, CB_DEF_VALUE),
> +    DEFINE_PROP_UINT16("cboz_blocksize", RISCVCPU,
> +                       cfg.cboz_blocksize, CB_DEF_VALUE),
>  
>      DEFINE_PROP_END_OF_LIST(),
>  };
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index 8efc4d83ec..ee9abe61d6 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -64,6 +64,8 @@ extern const uint32_t misa_bits[];
>  const char *riscv_get_misa_ext_name(uint32_t bit);
>  const char *riscv_get_misa_ext_description(uint32_t bit);
>  
> +#define CB_DEF_VALUE 64
> +
>  #define CPU_CFG_OFFSET(_prop) offsetof(struct RISCVCPUConfig, _prop)
>  
>  /* Privileged specification version */
> @@ -745,6 +747,7 @@ typedef struct RISCVCPUMultiExtConfig {
>  extern const RISCVCPUMultiExtConfig riscv_cpu_extensions[];
>  extern const RISCVCPUMultiExtConfig riscv_cpu_vendor_exts[];
>  extern const RISCVCPUMultiExtConfig riscv_cpu_experimental_exts[];
> +extern const RISCVCPUMultiExtConfig riscv_cpu_named_features[];
>  extern const RISCVCPUMultiExtConfig riscv_cpu_deprecated_exts[];
>  extern Property riscv_cpu_options[];
>  
> diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
> index 6eef4a51ea..b6693320d3 100644
> --- a/target/riscv/cpu_cfg.h
> +++ b/target/riscv/cpu_cfg.h
> @@ -107,6 +107,7 @@ struct RISCVCPUConfig {
>      bool ext_smepmp;
>      bool rvv_ta_all_1s;
>      bool rvv_ma_all_1s;
> +    bool zic64b;
>  
>      uint32_t mvendorid;
>      uint64_t marchid;
> diff --git a/target/riscv/tcg/tcg-cpu.c b/target/riscv/tcg/tcg-cpu.c
> index 093bda2e75..ac5f65a757 100644
> --- a/target/riscv/tcg/tcg-cpu.c
> +++ b/target/riscv/tcg/tcg-cpu.c
> @@ -264,6 +264,27 @@ static void 
> riscv_cpu_disable_priv_spec_isa_exts(RISCVCPU *cpu)
>      }
>  }
>  
> +static void riscv_cpu_validate_zic64b(RISCVCPU *cpu)
> +{
> +    const char *warn = "zic64b set to 'true' but %s is set to %u. "
> +                       "zic64b changed to 'false'";
> +    bool send_warn = cpu_cfg_ext_is_user_set(CPU_CFG_OFFSET(zic64b));
> +
> +    if (cpu->cfg.cbom_blocksize != CB_DEF_VALUE) {
> +        cpu->cfg.zic64b = false;
> +        if (send_warn) {
> +            warn_report(warn, "cbom_blocksize", cpu->cfg.cbom_blocksize);
> +        }
> +    }
> +
> +    if (cpu->cfg.cboz_blocksize != CB_DEF_VALUE) {
> +        cpu->cfg.zic64b = false;
> +        if (send_warn) {
> +            warn_report(warn, "cboz_blocksize", cpu->cfg.cboz_blocksize);
> +        }

I think if a user does X=1,X=2 and expects X to be the superposition of 1
and 2 they should get an error instead of a warning, and that's what
zic64b=on,cbom_blocksize=128 would effectively be asking to do, so let's
error out here instead of warn.

> +    }
> +}
> +
>  /*
>   * Check consistency between chosen extensions while setting
>   * cpu->cfg accordingly.
> @@ -394,6 +415,10 @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, 
> Error **errp)
>          return;
>      }
>  
> +    if (cpu->cfg.zic64b) {
> +        riscv_cpu_validate_zic64b(cpu);
> +    }

We set this on by default, so if a user just gives cbom_blocksize=128,
then we'll get the warning (or failure). They'd have to do
zic64b=off,cbom_blocksize=128 instead to avoid it. I think zic64b needs
to be off by default and then automatically turn on during cpu finalize
if all blocksize properties are 64.

> +
>      if (cpu->cfg.ext_zvfh) {
>          cpu_cfg_ext_auto_update(cpu, CPU_CFG_OFFSET(ext_zvfhmin), true);
>      }
> @@ -889,6 +914,7 @@ static void riscv_cpu_add_user_properties(Object *obj)
>      riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_extensions);
>      riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_vendor_exts);
>      riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_experimental_exts);
> +    riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_named_features);
>  
>      riscv_cpu_add_multiext_prop_array(obj, riscv_cpu_deprecated_exts);
>  
> -- 
> 2.41.0
>

Thanks,
drew



reply via email to

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