From: Richard Henderson <richard.henderson@linaro.org>
To: "Peter Maydell" <peter.maydell@linaro.org>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>
Cc: qemu-devel@nongnu.org, qemu-arm@nongnu.org,
qemu-s390x@nongnu.org, qemu-riscv@nongnu.org,
qemu-ppc@nongnu.org, Paolo Bonzini <pbonzini@redhat.com>,
Pierrick Bouvier <pierrick.bouvier@linaro.org>,
Peter Xu <peterx@redhat.com>
Subject: Re: [PATCH 00/22] hw/core/cpu: Remove @CPUState::as field
Date: Wed, 1 Oct 2025 09:35:40 -0700 [thread overview]
Message-ID: <b5280074-e87c-4838-b816-eb9e42dd090c@linaro.org> (raw)
In-Reply-To: <CAFEAcA-daogqBPxt72qTM6BuMuD1wuw+vXZwxLUQftwY2FZk=w@mail.gmail.com>
On 10/1/25 09:08, Peter Maydell wrote:
> On Wed, 1 Oct 2025 at 16:05, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
>>
>> Instead of using the convenient @CPUState::as shortcut, use
>> cpu_get_address_space(asidx=0) to get the vCPU first address
>> space.
>>
>> The goal is to reduce the risk of AS mis-use for targets
>> that uses multiple ASes per vCPU.
>
> I'm not sure I agree with this. The cpu->as is handy for
> the large number of targets which only use one AddressSpace:
> which is everything except arm and x86 right now.
>
> Targets that use multiple address spaces need to be careful
> about which AS they use by definition. Unless we have a
> track record of code in those targets carelessly using
> cpu->as when it should do something else, I'm not sure
> that there's a benefit that makes it worth making the
> "use the one and only AS this CPU will ever have" more awkward.
I agree.
Moreover, '0' as a default is just as bad as cpu->as, because you're giving no thought to
the selection. Similarly using MEMATTRS_UNSPECIFIED. When such things matter, we need
the real attributes and/or as piped down from the proper memory reference.
r~
next prev parent reply other threads:[~2025-10-01 16:38 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-01 15:05 [PATCH 00/22] hw/core/cpu: Remove @CPUState::as field Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 01/22] system/qtest: Use &address_space_memory for first vCPU address space Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 02/22] disas/disas-mon: Get cpu first addr space with cpu_get_address_space() Philippe Mathieu-Daudé
2025-10-01 15:34 ` BALATON Zoltan
2025-10-01 15:05 ` [PATCH 03/22] monitor/hmp-cmds: " Philippe Mathieu-Daudé
2025-10-01 15:35 ` BALATON Zoltan
2025-10-01 16:23 ` Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 04/22] hw/core/loader: " Philippe Mathieu-Daudé
2025-10-01 15:08 ` Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 05/22] hw/ppc: " Philippe Mathieu-Daudé
2025-10-01 15:23 ` Miles Glenn
2025-10-01 15:05 ` [PATCH 06/22] hw/m86k: " Philippe Mathieu-Daudé
2025-10-18 5:52 ` Thomas Huth
2025-10-01 15:05 ` [PATCH 07/22] target/xtensa: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 08/22] target/riscv: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 09/22] semihosting: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 10/22] target/alpha: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 11/22] target/arm: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 12/22] target/hppa: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 13/22] target/i386: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 14/22] target/loongarch: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 15/22] target/m68k: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 16/22] target/microblaze: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 17/22] target/ppc: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 18/22] target/s390x: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 19/22] target/sparc: " Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 20/22] hw/core/cpu: Remove @CPUState::as field Philippe Mathieu-Daudé
2025-10-01 15:05 ` [PATCH 21/22] exec/cpu: Declare cpu_memory_rw_debug() in 'hw/core/cpu.h' and document Philippe Mathieu-Daudé
2025-10-06 19:08 ` Philippe Mathieu-Daudé
2025-10-08 16:25 ` Zhao Liu
2025-10-01 15:05 ` [PATCH 22/22] target/sparc: Reduce inclusions of 'exec/cpu-common.h' Philippe Mathieu-Daudé
2025-10-01 15:39 ` [PATCH 00/22] hw/core/cpu: Remove @CPUState::as field BALATON Zoltan
2025-10-01 16:08 ` Peter Maydell
2025-10-01 16:35 ` Richard Henderson [this message]
2025-10-01 16:42 ` Philippe Mathieu-Daudé
2025-10-01 18:38 ` Philippe Mathieu-Daudé
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=b5280074-e87c-4838-b816-eb9e42dd090c@linaro.org \
--to=richard.henderson@linaro.org \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=peterx@redhat.com \
--cc=philmd@linaro.org \
--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 \
/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).