From: Philippe Gerum <rpm@xenomai.org>
To: Florian Bezdeka <florian.bezdeka@siemens.com>
Cc: xenomai@lists.linux.dev, Jan Kiszka <jan.kiszka@siemens.com>,
Tobias Schaffner <tobias.schaffner@siemens.com>
Subject: Re: [PATCH Dovetail v3 5/6] dovetail: Decouple kernel/irq/pipeline.c from irqstate bit definitions
Date: Fri, 26 Jun 2026 08:16:00 +0200 [thread overview]
Message-ID: <877bnlrfmn.fsf@xenomai.org> (raw)
In-Reply-To: <20260622-wip-flo-v7-1-arm-pipelining-fixes-v3-5-230f03227abb@siemens.com> (Florian Bezdeka's message of "Mon, 22 Jun 2026 10:05:29 +0200")
Florian Bezdeka <florian.bezdeka@siemens.com> writes:
> This only affects architectures where CONFIG_GENERIC_ENTRY is not set.
>
> Moving the stall bit definitions into the architecture specific
> implementations allows re-use of those definition within the arch
> specifics.
>
> Moving the defines to include/linux/irq_pipeline.h turned out to
> trigger a lot of trouble as re-use will live in asm/ where including
> a "general" header is quite hard.
>
> The platform independent implementation will now need the following
> arch specific hooks, all operating on the platform specific
> struct pt_regs:
> - arch_kentry_irqs_enabled()
I would go for arch_kentry_{test/is_}stalled() since this would clearly
hint at the stage stall bit.
> - arch_kentry_lockdep_set()
Name could be confusing since you also have
arch_kentry_set_lockdep(). arch_kentry_test_hardirq() instead?
> - arch_kentry_clear_irq_state()
> - arch_kentry_set_irqs_stalled()
arch_kentry_{set_}stall{ed}()
> - arch_kentry_set_lockdep()
>
I would pick arch_kentry_set_hardirq() to better match the general
naming scheme you have been using for these helpers, fixing the original
misnomer in the same move (the idea conveyed by the original name was
too vague/broadly scoped already).
> diff --git a/arch/arm/include/asm/irq_pipeline.h b/arch/arm/include/asm/irq_pipeline.h
> diff --git a/arch/arm64/include/asm/irq_pipeline.h b/arch/arm64/include/asm/irq_pipeline.h
> diff --git a/kernel/irq/pipeline.c b/kernel/irq/pipeline.c
Please split this into separate commits (arm: irq_pipeline, arm64:
irq_pipeline and genirq: irq_pipeline).
--
Philippe.
next prev parent reply other threads:[~2026-06-26 6:16 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-22 8:05 [PATCH Dovetail v3 0/6] Dovetail: arm: Finalize review results Florian Bezdeka
2026-06-22 8:05 ` [PATCH Dovetail v3 1/6] arm: irq_pipeline: Rename and move fault_entry_{enter,exit} Florian Bezdeka
2026-06-22 8:05 ` [PATCH Dovetail v3 2/6] arm: irq_pipeline: Fix dovetail_fault_{entry,exit} bypass in do_kernel_address_page_fault Florian Bezdeka
2026-06-22 8:05 ` [PATCH Dovetail v3 3/6] arm: irq_pipeline: Fix dovetail_fault_{entry,exit} bypass in hw_breakpoint_pending Florian Bezdeka
2026-06-22 13:44 ` Florian Bezdeka
2026-06-22 13:51 ` Philippe Gerum
2026-06-26 5:51 ` Philippe Gerum
2026-06-26 13:37 ` Florian Bezdeka
2026-06-22 8:05 ` [PATCH Dovetail v3 4/6] arm: irq_pipeline: Fix ordering problem in alignment trap handling Florian Bezdeka
2026-06-22 8:05 ` [PATCH Dovetail v3 5/6] dovetail: Decouple kernel/irq/pipeline.c from irqstate bit definitions Florian Bezdeka
2026-06-26 6:16 ` Philippe Gerum [this message]
2026-06-22 8:05 ` [PATCH Dovetail v3 6/6] arm: irq_pipeline: Fix IRQ state checking in trap entry paths Florian Bezdeka
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=877bnlrfmn.fsf@xenomai.org \
--to=rpm@xenomai.org \
--cc=florian.bezdeka@siemens.com \
--cc=jan.kiszka@siemens.com \
--cc=tobias.schaffner@siemens.com \
--cc=xenomai@lists.linux.dev \
/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.