From: Alistair Francis <alistair23@gmail.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 for-8.2 2/2] target/riscv/cpu.c: add zihpm extension flag
Date: Mon, 24 Jul 2023 12:00:43 +1000 [thread overview]
Message-ID: <CAKmqyKN-sK1czTCwBmjb5BArBeWVBuAX+GS7z814+PB4KwKufA@mail.gmail.com> (raw)
In-Reply-To: <20230717215419.124258-3-dbarboza@ventanamicro.com>
On Tue, Jul 18, 2023 at 7:55 AM Daniel Henrique Barboza
<dbarboza@ventanamicro.com> wrote:
>
> zihpm is the Hardware Performance Counters extension described in
> chapter 12 of the unprivileged spec. It describes support for 29
> unprivileged performance counters, hpmcounter3-hpmcounter21.
>
> As with zicntr, QEMU already implements zihpm before it was even an
> extension. zihpm is also part of the RVA22 profile, so add it to QEMU
> to complement the future future profile implementation.
>
> Default it to 'true' since it was always present in the code. Change the
> realize() time validation to disable it in case 'icsr' isn't present and
> if there's no hardware counters (cpu->cfg.pmu_num is zero).
>
> There's a small tweak needed in riscv_cpu_realize_tcg() made:
> riscv_cpu_validate_set_extensions() must be executed after the block
> that executes riscv_pmu_init(). The reason is that riscv_pmu_init() will
> do "cpu->cfg.pmu_num = 0" if PMU support cannot be enabled. We want to
> get the latest, definite value of cfg.pmu_num during the validation() to
> ensure we do the right thing.
>
> Signed-off-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Alistair
> ---
> target/riscv/cpu.c | 20 +++++++++++++-------
> target/riscv/cpu_cfg.h | 1 +
> 2 files changed, 14 insertions(+), 7 deletions(-)
>
> diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c
> index 7ec88659be..5836640d5c 100644
> --- a/target/riscv/cpu.c
> +++ b/target/riscv/cpu.c
> @@ -89,6 +89,7 @@ static const struct isa_ext_data isa_edata_arr[] = {
> ISA_EXT_DATA_ENTRY(zicsr, PRIV_VERSION_1_10_0, ext_icsr),
> ISA_EXT_DATA_ENTRY(zifencei, PRIV_VERSION_1_10_0, ext_ifencei),
> ISA_EXT_DATA_ENTRY(zihintpause, PRIV_VERSION_1_10_0, ext_zihintpause),
> + ISA_EXT_DATA_ENTRY(zihpm, PRIV_VERSION_1_12_0, ext_ihpm),
> ISA_EXT_DATA_ENTRY(zawrs, PRIV_VERSION_1_12_0, ext_zawrs),
> ISA_EXT_DATA_ENTRY(zfa, PRIV_VERSION_1_12_0, ext_zfa),
> ISA_EXT_DATA_ENTRY(zfbfmin, PRIV_VERSION_1_12_0, ext_zfbfmin),
> @@ -1296,6 +1297,10 @@ void riscv_cpu_validate_set_extensions(RISCVCPU *cpu, Error **errp)
> cpu->cfg.ext_icntr = false;
> }
>
> + if (cpu->cfg.ext_ihpm && (!cpu->cfg.ext_icsr || cpu->cfg.pmu_num == 0)) {
> + cpu->cfg.ext_ihpm = false;
> + }
> +
> /*
> * Disable isa extensions based on priv spec after we
> * validated and set everything we need.
> @@ -1426,12 +1431,6 @@ static void riscv_cpu_realize_tcg(DeviceState *dev, Error **errp)
> return;
> }
>
> - riscv_cpu_validate_set_extensions(cpu, &local_err);
> - if (local_err != NULL) {
> - error_propagate(errp, local_err);
> - return;
> - }
> -
> #ifndef CONFIG_USER_ONLY
> CPU(dev)->tcg_cflags |= CF_PCREL;
>
> @@ -1446,6 +1445,12 @@ static void riscv_cpu_realize_tcg(DeviceState *dev, Error **errp)
> }
> }
> #endif
> +
> + riscv_cpu_validate_set_extensions(cpu, &local_err);
> + if (local_err != NULL) {
> + error_propagate(errp, local_err);
> + return;
> + }
> }
>
> static void riscv_cpu_realize(DeviceState *dev, Error **errp)
> @@ -1784,10 +1789,11 @@ static Property riscv_cpu_extensions[] = {
> DEFINE_PROP_BOOL("svpbmt", RISCVCPU, cfg.ext_svpbmt, false),
>
> /*
> - * Always default true - we'll disable it during
> + * Always default true - we'll disable them during
> * realize() if needed.
> */
> DEFINE_PROP_BOOL("zicntr", RISCVCPU, cfg.ext_icntr, true),
> + DEFINE_PROP_BOOL("zihpm", RISCVCPU, cfg.ext_ihpm, true),
>
> DEFINE_PROP_BOOL("zba", RISCVCPU, cfg.ext_zba, true),
> DEFINE_PROP_BOOL("zbb", RISCVCPU, cfg.ext_zbb, true),
> diff --git a/target/riscv/cpu_cfg.h b/target/riscv/cpu_cfg.h
> index d36dc12b92..85c7a71853 100644
> --- a/target/riscv/cpu_cfg.h
> +++ b/target/riscv/cpu_cfg.h
> @@ -66,6 +66,7 @@ struct RISCVCPUConfig {
> bool ext_icsr;
> bool ext_icbom;
> bool ext_icboz;
> + bool ext_ihpm;
> bool ext_zicond;
> bool ext_zihintpause;
> bool ext_smstateen;
> --
> 2.41.0
>
>
next prev parent reply other threads:[~2023-07-24 2:01 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-17 21:54 [PATCH for-8.2 0/2] target/riscv: add zicntr and zihpm flags Daniel Henrique Barboza
2023-07-17 21:54 ` [PATCH for-8.2 1/2] target/riscv/cpu.c: add zicntr extension flag Daniel Henrique Barboza
2023-07-24 1:56 ` Alistair Francis
2023-07-17 21:54 ` [PATCH for-8.2 2/2] target/riscv/cpu.c: add zihpm " Daniel Henrique Barboza
2023-07-24 2:00 ` Alistair Francis [this message]
2023-07-17 22:33 ` [PATCH for-8.2 0/2] target/riscv: add zicntr and zihpm flags Conor Dooley
2023-07-17 23:11 ` Daniel Henrique Barboza
2023-07-18 8:23 ` Conor Dooley
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=CAKmqyKN-sK1czTCwBmjb5BArBeWVBuAX+GS7z814+PB4KwKufA@mail.gmail.com \
--to=alistair23@gmail.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).