qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Jones <ajones@ventanamicro.com>
To: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Cc: qemu-devel@nongnu.org, qemu-riscv@nongnu.org,
	alistair.francis@wdc.com,  bmeng@tinylab.org,
	liweiwei@iscas.ac.cn, zhiwei_liu@linux.alibaba.com,
	 palmer@rivosinc.com
Subject: Re: [PATCH 14/20] target/riscv/kvm: do not use riscv_cpu_add_misa_properties()
Date: Thu, 31 Aug 2023 13:50:27 +0200	[thread overview]
Message-ID: <20230831-f967ffa982698458cdb384f1@orel> (raw)
In-Reply-To: <20230825130853.511782-15-dbarboza@ventanamicro.com>

On Fri, Aug 25, 2023 at 10:08:47AM -0300, Daniel Henrique Barboza wrote:
> riscv_cpu_add_misa_properties() is being used to fill the missing KVM
> MISA properties but it is a TCG helper that was adapted to do so. We'll
> move it to tcg-cpu.c in the next patches, meaning that KVM needs to fill
> the remaining MISA properties on its own.
> 
> Do not use riscv_cpu_add_misa_properties(). Let's create a new array
> with all available MISA bits we support that can be read by KVM. Then,
> inside kvm_riscv_add_cpu_user_properties(), we'll create all KVM MISA
> properties as usual and then use this array to add any missing MISA
> properties with the riscv_cpu_add_kvm_unavail_prop() helper.
> 
> Note that we're creating misa_bits[], and not using the existing
> 'riscv_single_letter_exts[]', because the latter is tuned for riscv,isa
> related functions and it doesn't have all MISA bits we support. Commit
> 0e2c377023 ("target/riscv: misa to ISA string conversion fix") has the
> full context.
> 
> While we're at it, move both satp and the multi-letter extension
> properties to kvm_riscv_add_cpu_user_properties() as well.
> 
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
> ---
>  target/riscv/cpu.c         |  2 ++
>  target/riscv/cpu.h         |  3 ++-
>  target/riscv/kvm/kvm-cpu.c | 17 +++++++++++------
>  3 files changed, 15 insertions(+), 7 deletions(-)
> 
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index bf6c8519b1..f9aea6a80a 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -38,6 +38,8 @@
>  
>  /* RISC-V CPU definitions */
>  static const char riscv_single_letter_exts[] = "IEMAFDQCPVH";
> +const uint32_t misa_bits[] = {RVI, RVE, RVM, RVA, RVF, RVD, RVV,
> +                              RVC, RVS, RVU, RVH, RVJ, RVG};
>  
>  struct isa_ext_data {
>      const char *name;
> diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
> index 950c2301f2..9ec3b98bd2 100644
> --- a/target/riscv/cpu.h
> +++ b/target/riscv/cpu.h
> @@ -43,7 +43,7 @@
>  #define RV(x) ((target_ulong)1 << (x - 'A'))
>  
>  /*
> - * Consider updating misa_ext_info_arr[] and misa_ext_cfgs[]
> + * Update misa_bits[], misa_ext_info_arr[] and misa_ext_cfgs[]
>   * when adding new MISA bits here.
>   */
>  #define RVI RV('I')
> @@ -60,6 +60,7 @@
>  #define RVJ RV('J')
>  #define RVG RV('G')
>  
> +extern const uint32_t misa_bits[13];
                                   ^ maintaining this will be a PITA

I suggest misa_bits either have a terminating zero (zero is an invalid bit
number) or that a const nr_misa_bits is also exported from riscv/cpu.c
which be set to ARRAY_SIZE(misa_bits) and would be used for the loop
condition.

>  const char *riscv_get_misa_ext_name(uint32_t bit);
>  const char *riscv_get_misa_ext_description(uint32_t bit);
>  
> diff --git a/target/riscv/kvm/kvm-cpu.c b/target/riscv/kvm/kvm-cpu.c
> index 85e8b0a927..501384924b 100644
> --- a/target/riscv/kvm/kvm-cpu.c
> +++ b/target/riscv/kvm/kvm-cpu.c
> @@ -387,6 +387,8 @@ static void kvm_riscv_add_cpu_user_properties(Object *cpu_obj)
>  {
>      int i;
>  
> +    riscv_add_satp_mode_properties(cpu_obj);
> +
>      for (i = 0; i < ARRAY_SIZE(kvm_misa_ext_cfgs); i++) {
>          KVMCPUConfig *misa_cfg = &kvm_misa_ext_cfgs[i];
>          int bit = misa_cfg->offset;
> @@ -402,6 +404,11 @@ static void kvm_riscv_add_cpu_user_properties(Object *cpu_obj)
>                                          misa_cfg->description);
>      }
>  
> +    for (i = 0; i < ARRAY_SIZE(misa_bits); i++) {
> +        const char *ext_name = riscv_get_misa_ext_name(misa_bits[i]);
> +        riscv_cpu_add_kvm_unavail_prop(cpu_obj, ext_name);
> +    }
> +
>      for (i = 0; i < ARRAY_SIZE(kvm_multi_ext_cfgs); i++) {
>          KVMCPUConfig *multi_cfg = &kvm_multi_ext_cfgs[i];
>  
> @@ -418,6 +425,10 @@ static void kvm_riscv_add_cpu_user_properties(Object *cpu_obj)
>      object_property_add(cpu_obj, "cboz_blocksize", "uint16",
>                          NULL, kvm_cpu_set_cbomz_blksize,
>                          NULL, &kvm_cboz_blocksize);
> +
> +    riscv_cpu_add_kvm_unavail_prop_array(cpu_obj, riscv_cpu_extensions);
> +    riscv_cpu_add_kvm_unavail_prop_array(cpu_obj, riscv_cpu_vendor_exts);
> +    riscv_cpu_add_kvm_unavail_prop_array(cpu_obj, riscv_cpu_experimental_exts);
>  }
>  
>  static int kvm_riscv_get_regs_core(CPUState *cs)
> @@ -1301,12 +1312,6 @@ static void kvm_cpu_instance_init(CPUState *cs)
>  
>      if (rcc->user_extension_properties) {
>          kvm_riscv_add_cpu_user_properties(obj);
> -        riscv_add_satp_mode_properties(obj);
> -        riscv_cpu_add_misa_properties(obj);
> -
> -        riscv_cpu_add_kvm_unavail_prop_array(obj, riscv_cpu_extensions);
> -        riscv_cpu_add_kvm_unavail_prop_array(obj, riscv_cpu_vendor_exts);
> -        riscv_cpu_add_kvm_unavail_prop_array(obj, riscv_cpu_experimental_exts);
>      }
>  
>      for (Property *prop = riscv_cpu_options; prop && prop->name; prop++) {
> -- 
> 2.41.0
> 
> 

Otherwise,

Reviewed-by: Andrew Jones <ajones@ventanamicro.com>


  reply	other threads:[~2023-08-31 11:51 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-25 13:08 [PATCH 00/20] riscv: split TCG/KVM accelerators from cpu.c Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 01/20] target/riscv: introduce TCG AccelCPUClass Daniel Henrique Barboza
2023-08-31 10:17   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 02/20] target/riscv: move riscv_cpu_realize_tcg() to TCG::cpu_realizefn() Daniel Henrique Barboza
2023-08-31 10:21   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 03/20] target/riscv: move riscv_cpu_validate_set_extensions() to tcg-cpu.c Daniel Henrique Barboza
2023-08-31 10:31   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 04/20] target/riscv: move riscv_tcg_ops " Daniel Henrique Barboza
2023-08-28 16:30   ` Philippe Mathieu-Daudé
2023-08-31 10:38   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 05/20] target/riscv/cpu.c: add 'user_extension_properties' class prop Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 06/20] target/riscv: add 'max_features' CPU flag Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 07/20] target/riscv/cpu.c: add .instance_post_init() Daniel Henrique Barboza
2023-08-31 11:00   ` Andrew Jones
2023-09-01 20:08     ` Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 08/20] target/riscv: move 'host' CPU declaration to kvm.c Daniel Henrique Barboza
2023-08-28 16:35   ` Philippe Mathieu-Daudé
2023-08-31 11:04   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 09/20] target/riscv/cpu.c: mark extensions arrays as 'const' Daniel Henrique Barboza
2023-08-31 11:10   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 10/20] target/riscv: move riscv_cpu_add_kvm_properties() to kvm.c Daniel Henrique Barboza
2023-08-31 11:22   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 11/20] target/riscv: introduce KVM AccelCPUClass Daniel Henrique Barboza
2023-08-28 16:38   ` Philippe Mathieu-Daudé
2023-08-29 13:16     ` Daniel Henrique Barboza
2023-08-31 11:26   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 12/20] target/riscv: move KVM only files to kvm subdir Daniel Henrique Barboza
2023-08-28 16:47   ` Philippe Mathieu-Daudé
2023-08-30 18:21     ` Daniel Henrique Barboza
2023-08-30 20:54       ` Philippe Mathieu-Daudé
2023-08-31 11:30   ` Andrew Jones
2023-09-01 17:19     ` Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 13/20] target/riscv/kvm: refactor kvm_riscv_init_user_properties() Daniel Henrique Barboza
2023-08-31 11:34   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 14/20] target/riscv/kvm: do not use riscv_cpu_add_misa_properties() Daniel Henrique Barboza
2023-08-31 11:50   ` Andrew Jones [this message]
2023-08-25 13:08 ` [PATCH 15/20] target/riscv/tcg: introduce tcg_cpu_instance_init() Daniel Henrique Barboza
2023-08-31 11:56   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 16/20] target/riscv/tcg: move riscv_cpu_add_misa_properties() to tcg-cpu.c Daniel Henrique Barboza
2023-08-31 12:01   ` Andrew Jones
2023-09-04 14:21     ` Daniel Henrique Barboza
2023-08-25 13:08 ` [PATCH 17/20] target/riscv/cpu.c: export isa_edata_arr[] Daniel Henrique Barboza
2023-08-31 12:06   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 18/20] target/riscv/cpu: move priv spec functions to tcg-cpu.c Daniel Henrique Barboza
2023-08-31 12:07   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 19/20] target/riscv: add 'tcg_supported' class property Daniel Henrique Barboza
2023-08-31 12:25   ` Andrew Jones
2023-08-25 13:08 ` [PATCH 20/20] target/riscv: add 'kvm_supported' " Daniel Henrique Barboza
2023-08-31 12:47   ` Andrew Jones
2023-09-01 20:57     ` Daniel Henrique Barboza
2023-09-04  9:05       ` Andrew Jones

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230831-f967ffa982698458cdb384f1@orel \
    --to=ajones@ventanamicro.com \
    --cc=alistair.francis@wdc.com \
    --cc=bmeng@tinylab.org \
    --cc=dbarboza@ventanamicro.com \
    --cc=liweiwei@iscas.ac.cn \
    --cc=palmer@rivosinc.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-riscv@nongnu.org \
    --cc=zhiwei_liu@linux.alibaba.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).