From: Philippe Gerum <rpm@xenomai.org>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: Xenomai <xenomai@lists.linux.dev>
Subject: Re: Missing v5.10.195-dovetail2-rebase?
Date: Tue, 17 Oct 2023 19:06:03 +0200 [thread overview]
Message-ID: <87bkcx2mc9.fsf@xenomai.org> (raw)
In-Reply-To: <87fs292xxq.fsf@xenomai.org>
Philippe Gerum <rpm@xenomai.org> writes:
> Jan Kiszka <jan.kiszka@siemens.com> writes:
>
>> Hi Philippe,
>>
>> I've seen you tagged "genirq: irq_pipeline: do not request IPI on
>> uniprocessor" via v5.15.129-dovetail2-rebase and
>> v6.1.54-dovetail2-rebase, but there is no tag on the 5.10 side -
>> intentionally?
>>
>
> Yes, I noticed an unrelated DMA issue with CONFIG_IMX_SDMA_OOB. Working
> on it.
The tag has landed. However, the recent fix to support stacked irqchips
[1] uncovered a nasty issue which dates back to the dark ages, with a
(fortunately) not-so-common practice of relaying _hardware_ irq events
from oob to inband using irq_post_inband(), so that the same IRQ handler
can implement a two-step handling. This does not affect synthetic irqs,
only real/device-originated ones, observed with fasteoi so
far. Introducing the explicit deferral status (IRQS_DEFERRED) in irq
descriptors revealed the problem. IOW, this is no regression, rather an
ancient bug who crawled out from under a rock.
Fortunately, I have a real use case at hand (imx-dma). Working on
providing a sane replacement for the current open-coded way.
[1] b9132dbc8c31c genirq: irq_pipeline: fix irq flow handling with stacked irqchips
--
Philippe.
prev parent reply other threads:[~2023-10-17 17:24 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-17 12:36 Missing v5.10.195-dovetail2-rebase? Jan Kiszka
2023-10-17 13:12 ` Philippe Gerum
2023-10-17 17:06 ` Philippe Gerum [this message]
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=87bkcx2mc9.fsf@xenomai.org \
--to=rpm@xenomai.org \
--cc=jan.kiszka@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 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 read-only IMAP folder(s) and NNTP newsgroup(s).