From: "Philippe Mathieu-Daudé" <philmd@oss.qualcomm.com>
To: qemu-devel@nongnu.org,
"Marc-André Lureau" <marcandre.lureau@redhat.com>,
"Pierrick Bouvier" <pierrick.bouvier@linaro.org>
Cc: "Paolo Bonzini" <pbonzini@redhat.com>,
qemu-arm@nongnu.org, qemu-s390x@nongnu.org,
"Richard Henderson" <richard.henderson@linaro.org>,
qemu-riscv@nongnu.org, qemu-ppc@nongnu.org,
"Philippe Mathieu-Daudé" <philmd@linaro.org>,
"Philippe Mathieu-Daudé" <philmd@mailo.com>,
"Zhao Liu" <zhao1.liu@intel.com>,
"Alex Bennée" <alex.bennee@linaro.org>,
"Eric Blake" <eblake@redhat.com>
Subject: Re: [PATCH v3 02/15] cpus: Add const-qualified CPU environment accessors
Date: Fri, 21 Aug 2026 17:51:40 +0200 [thread overview]
Message-ID: <22188776-edeb-44e6-b79d-0b877292221d@oss.qualcomm.com> (raw)
In-Reply-To: <3c9fe944-2b51-412c-bd30-26bf013cff46@oss.qualcomm.com>
On 2026-08-21 13:33, Philippe Mathieu-Daudé wrote:
> On 21/8/26 12:49, Philippe Mathieu-Daudé wrote:
>> From: Philippe Mathieu-Daudé <philmd@linaro.org>
>>
>> Use _Generic() controlling-expression to add the const-qualified
>> variants of cpu_env(), env_cpu() and env_archcpu(). This allows to
>> safely access CPU architecture state when it should not be modified.
>>
>> Alias env_cpu_const() which is still used.
>>
>> Suggested-by: Richard Henderson <richard.henderson@linaro.org>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
>> Signed-off-by: Philippe Mathieu-Daudé <philmd@oss.qualcomm.com>
>> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
>> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
>> ---
>> Following checkpatch.pl errors ignored:
>>
>> ERROR: spaces required around that ':' (ctx:VxE)
>> #46: FILE: include/exec/cpu-common.h:89:
>> + CPUArchState: \
>> ^
>> ERROR: spaces required around that ':' (ctx:VxE)
>> #62: FILE: include/exec/cpu-common.h:101:
>> + CPUArchState: \
>> ^
>> ERROR: spaces required around that ':' (ctx:VxE)
>> #90: FILE: include/hw/core/cpu.h:605:
>> + CPUState: \
>> ^
>> total: 3 errors, 0 warnings, 64 lines checked
>> ---
>> include/exec/cpu-common.h | 30 +++++++++++-------------------
>> include/hw/core/cpu.h | 17 ++++++++++++-----
>> 2 files changed, 23 insertions(+), 24 deletions(-)
>>
>> diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h
>> index 6594f7fa1be..bffef677607 100644
>> --- a/include/exec/cpu-common.h
>> +++ b/include/exec/cpu-common.h
>> @@ -85,21 +85,11 @@ static inline bool cpu_loop_exit_requested(const
>> CPUState *cpu)
>> *
>> * Return the ArchCPU associated with the environment.
>> */
>> -static inline ArchCPU *env_archcpu(CPUArchState *env)
>> -{
>> - return (void *)env - sizeof(CPUState);
>> -}
>> -
>> -/**
>> - * env_cpu_const(env)
>> - * @env: The architecture environment
>> - *
>> - * Return the CPUState associated with the environment.
>> - */
>> -static inline const CPUState *env_cpu_const(const CPUArchState *env)
>> -{
>> - return (void *)env - sizeof(CPUState);
>> -}
>> +#define env_archcpu(env) _Generic(*(env), \
>> + CPUArchState: \
>> + (ArchCPU *)((void *)env - sizeof(CPUState)), \
>> + const CPUArchState: \
>> + (const ArchCPU *)((const void *)env - sizeof(CPUState)))
> I was testing with a C17-ready compiler. Apparently C11 is ambiguous in
> how it treats qualifiers inside _Generic:
>
> ../../target/arm/internals.h:1777:25: warning: due to lvalue conversion
> of the controlling expression, association of type 'const CPUArchState'
> (aka 'const struct CPUArchState') will never be selected because it is
> qualified [-Wunreachable-code-generic-assoc]
> 1777 | const ARMCPU *cpu = env_archcpu(env);
> | ^
> include/exec/cpu-common.h:91:19: note: expanded from macro 'env_archcpu'
> 91 | const CPUArchState: \
> | ^
>
> Using typeof() makes it happier:
>
> -- >8 --
> diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h
> index d67d008236f..e3a5e40899f 100644
> --- a/include/exec/cpu-common.h
> +++ b/include/exec/cpu-common.h
> @@ -87,6 +87,6 @@ static inline bool cpu_loop_exit_requested(const
> CPUState *cpu)
> */
> -#define env_archcpu(env) _Generic(*(env), \
> - CPUArchState: \
> +#define env_archcpu(env) _Generic(typeof(*env), \
> + typeof(CPUArchState): \
> (ArchCPU *)((void *)env - sizeof(CPUState)), \
> - const CPUArchState: \
> + typeof(const CPUArchState): \
> (const ArchCPU *)((const void *)env - sizeof(CPUState)))
> @@ -99,6 +99,6 @@ static inline bool cpu_loop_exit_requested(const
> CPUState *cpu)
> */
> -#define env_cpu(env) _Generic(*(env), \
> - CPUArchState: \
> +#define env_cpu(env) _Generic(typeof(*env), \
> + typeof(CPUArchState): \
> (CPUState *)((void *)env - sizeof(CPUState)), \
> - const CPUArchState: \
> + typeof(const CPUArchState): \
> (const CPUState *)((const void *)env - sizeof(CPUState)))
> diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
> index 372485a2e54..ee522e1e705 100644
> --- a/include/hw/core/cpu.h
> +++ b/include/hw/core/cpu.h
> @@ -602,7 +602,7 @@ QEMU_BUILD_BUG_ON(offsetof(CPUState, neg) !=
> */
> -#define cpu_env(cpu) _Generic(*(cpu), \
> +#define cpu_env(cpu) _Generic(typeof(*cpu), \
> /* We validate that CPUArchState follows CPUState in cpu-
> target.c */ \
> - CPUState: \
> + typeof(CPUState): \
> (CPUArchState *)(cpu + 1), \
> - const CPUState: \
> + typeof(const CPUState): \
> (const CPUArchState *)(cpu + 1))
> ---
No, this now fails distinctly on Ubuntu 24.04:
In file included from ../accel/tcg/tcg-runtime.c:26:
../accel/tcg/tcg-runtime.c: In function ‘helper_exit_atomic’:
/home/runner/work/qemu/qemu/include/exec/cpu-common.h:100:31: error:
expected expression before ‘typeof’
100 | #define env_cpu(env) _Generic(typeof(*env), \
| ^~~~~~
../accel/tcg/tcg-runtime.c:135:26: note: in expansion of macro ‘env_cpu’
135 | cpu_loop_exit_atomic(env_cpu(env), GETPC());
| ^~~~~~~
I'm a bit clueless. In case you want to have a look, the
series is available here:
https://github.com/qualcomm/qemu/commits/refs/heads/philmd/cpu_has_work_const/
Thanks,
Phil.
next prev parent reply other threads:[~2026-08-21 15:52 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-21 10:49 [PATCH v3 00/15] cpus: Constify @cpu in SysemuCPUOps::has_work() handler Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 01/15] linux-user: Uncast void pointer argument as Object in target_cpu_free() Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 02/15] cpus: Add const-qualified CPU environment accessors Philippe Mathieu-Daudé
2026-08-21 11:33 ` Philippe Mathieu-Daudé
2026-08-21 15:51 ` Philippe Mathieu-Daudé [this message]
2026-08-21 17:07 ` Richard Henderson
2026-08-24 9:32 ` Philippe Mathieu-Daudé
2026-08-21 16:44 ` Richard Henderson
2026-08-21 10:49 ` [PATCH v3 03/15] linux-user: Replace env_cpu_const() by generic env_cpu() equivalent Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 04/15] system/cpus: Constify various CPUState arguments Philippe Mathieu-Daudé
2026-08-21 16:41 ` Richard Henderson
2026-08-21 10:49 ` [PATCH v3 05/15] target/avr: Constify CPUAVRState for some cpu_*() getters Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 06/15] target/hexagon: Constify CPUHexagonState in hexagon_thread_is_enabled() Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 07/15] target/i386: Constify CPU*State for cpu_*_interrupt() getters Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 08/15] target/loongarch: Constify CPULoongArchState for various cpu_*() getters Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 09/15] target/mips: Constify CPUMIPSState " Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 10/15] target/s390x: Constify S390CPU for cpu_has_*() getters Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 11/15] target/riscv: Constify @iprio argument in riscv_cpu_pending_to_irq() Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 12/15] target/riscv: Constify CPURISCVState for various cpu_*() getters Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 13/15] target/sparc: Constify CPUSPARCState " Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 14/15] target/tricore: Document architectural interrupts as not implemented Philippe Mathieu-Daudé
2026-08-21 10:49 ` [PATCH v3 15/15] cpus: Constify @cpu in SysemuCPUOps::has_work() handler Philippe Mathieu-Daudé
2026-08-21 16:42 ` Richard Henderson
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=22188776-edeb-44e6-b79d-0b877292221d@oss.qualcomm.com \
--to=philmd@oss.qualcomm.com \
--cc=alex.bennee@linaro.org \
--cc=eblake@redhat.com \
--cc=marcandre.lureau@redhat.com \
--cc=pbonzini@redhat.com \
--cc=philmd@linaro.org \
--cc=philmd@mailo.com \
--cc=pierrick.bouvier@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=qemu-riscv@nongnu.org \
--cc=qemu-s390x@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=zhao1.liu@intel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.