From: Marc Zyngier <maz@kernel.org>
To: "Cédric Le Goater" <clg@kaod.org>
Cc: <linux-kernel@vger.kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Michael Ellerman <mpe@ellerman.id.au>,
PowerPC <linuxppc-dev@lists.ozlabs.org>,
Greg Kurz <groug@kaod.org>
Subject: Re: [PATCH 16/39] irqdomain: Make normal and nomap irqdomains exclusive
Date: Tue, 16 Nov 2021 08:33:42 +0000 [thread overview]
Message-ID: <87a6i48pp5.wl-maz@kernel.org> (raw)
In-Reply-To: <1fe9d629-0f5f-4807-b97c-77b3b3c7de72@kaod.org>
On Mon, 15 Nov 2021 19:05:17 +0000,
Cédric Le Goater <clg@kaod.org> wrote:
>
> Hello Mark,
s/k/c/
>
> On 5/20/21 18:37, Marc Zyngier wrote:
> > Direct mappings are completely exclusive of normal mappings, meaning
> > that we can refactor the code slightly so that we can get rid of
> > the revmap_direct_max_irq field and use the revmap_size field
> > instead, reducing the size of the irqdomain structure.
> >
> > Signed-off-by: Marc Zyngier <maz@kernel.org>
>
>
> This patch is breaking the POWER9/POWER10 XIVE driver (these are not
> old PPC systems :) on machines sharing the same LSI HW IRQ. For instance,
> a linux KVM guest with a virtio-rng and a virtio-balloon device. In that
> case, Linux creates two distinct IRQ mappings which can lead to some
> unexpected behavior.
Either the irq domain translates, or it doesn't. If the driver creates
a nomap domain, and yet expects some sort of translation to happen,
then the driver is fundamentally broken. And even without that: how do
you end-up with a single HW interrupt having two mappings?
> A fix to go forward would be to change the XIVE IRQ domain to use a
> 'Tree' domain for reverse mapping and not the 'No Map' domain mapping.
> I will keep you updated for XIVE.
I bet there is a bit more to it. From what you are saying above,
something rather ungodly is happening in the XIVE code.
M.
--
Without deviation from the norm, progress is not possible.
next prev parent reply other threads:[~2021-11-16 8:33 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-20 16:37 [PATCH 00/39] irqdomain: Simplify interrupt handling Marc Zyngier
2021-05-20 16:37 ` [PATCH 01/39] nios2: Do not include linux/irqdomain.h from asm/irq.h Marc Zyngier
2021-05-20 16:37 ` [PATCH 02/39] staging: octeon-hcd: Directly include linux/of.h Marc Zyngier
2021-05-20 16:37 ` [PATCH 03/39] mfd: ioc3: Directly include linux/irqdomain.h Marc Zyngier
2021-05-21 7:07 ` Lee Jones
2021-05-20 16:37 ` [PATCH 04/39] watchdog/octeon-wdt: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 05/39] irqchip/mips-gic: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 06/39] MIPS: lantiq: Directly include linux/of.h in xway/dma.c Marc Zyngier
2021-05-20 16:37 ` [PATCH 07/39] MIPS: Add missing linux/irqdomain.h includes Marc Zyngier
2021-05-20 16:37 ` [PATCH 08/39] MIPS: Do not include linux/irqdomain.h from asm/irq.h Marc Zyngier
2021-05-20 16:37 ` [PATCH 09/39] powerpc: Add missing linux/{of.h,irqdomain.h} include directives Marc Zyngier
2021-05-20 16:37 ` [PATCH 10/39] scsi/ibmvscsi: Directly include linux/{of.h,irqdomain.h} Marc Zyngier
2021-05-20 16:37 ` [PATCH 11/39] powerpc: Convert irq_domain_add_legacy_isa use to irq_domain_add_legacy Marc Zyngier
2021-05-20 16:37 ` [PATCH 12/39] powerpc: Drop dependency between asm/irq.h and linux/irqdomain.h Marc Zyngier
2021-05-20 16:37 ` [PATCH 13/39] irqdomain: Kill irq_domain_add_legacy_isa Marc Zyngier
2021-05-20 16:37 ` [PATCH 14/39] irqdomain: Reimplement irq_linear_revmap() with irq_find_mapping() Marc Zyngier
2021-05-20 16:37 ` [PATCH 15/39] powerpc: Move the use of irq_domain_add_nomap() behind a config option Marc Zyngier
2021-05-20 16:37 ` [PATCH 16/39] irqdomain: Make normal and nomap irqdomains exclusive Marc Zyngier
2021-11-15 19:05 ` Cédric Le Goater
2021-11-16 8:33 ` Marc Zyngier [this message]
2021-11-16 10:27 ` Cédric Le Goater
2021-05-20 16:37 ` [PATCH 17/39] irqdomain: Use struct_size() helper when allocating irqdomain Marc Zyngier
2021-05-20 16:37 ` [PATCH 18/39] irqdomain: Cache irq_data instead of a virq number in the revmap Marc Zyngier
2021-05-20 16:37 ` [PATCH 19/39] irqdomain: Implement irq_domain_clear_mapping() with irq_domain_set_mapping() Marc Zyngier
2021-05-20 16:37 ` [PATCH 20/39] irqdomain: Protect the linear revmap with RCU Marc Zyngier
2021-06-19 11:37 ` Thomas Gleixner
2021-05-20 16:37 ` [PATCH 21/39] irqdomain: Introduce irq_resolve_mapping() Marc Zyngier
2021-06-10 2:22 ` Qian Cai
2021-06-10 7:24 ` Marc Zyngier
2021-05-20 16:37 ` [PATCH 22/39] genirq: Use irq_resolve_mapping() to implement __handle_domain_irq() and co Marc Zyngier
2021-05-20 16:37 ` [PATCH 23/39] irqdesc: Fix __handle_domain_irq() comment Marc Zyngier
2021-05-20 16:37 ` [PATCH 24/39] irqchip/nvic: Convert from handle_IRQ() to handle_domain_irq() Marc Zyngier
2021-05-20 16:37 ` [PATCH 25/39] genirq: Add generic_handle_domain_irq() helper Marc Zyngier
2021-05-20 16:37 ` [PATCH 26/39] genirq: Move non-irqdomain handle_domain_irq() handling into ARM's handle_IRQ() Marc Zyngier
2021-05-20 16:37 ` [PATCH 27/39] irqchip: Bulk conversion to generic_handle_domain_irq() Marc Zyngier
2021-05-21 8:54 ` Geert Uytterhoeven
2021-05-21 21:10 ` Krzysztof Kozlowski
2021-05-20 16:37 ` [PATCH 28/39] gpio: " Marc Zyngier
2021-05-21 8:59 ` Geert Uytterhoeven
2021-05-20 16:37 ` [PATCH 29/39] pinctrl: " Marc Zyngier
2021-05-21 21:16 ` Krzysztof Kozlowski
2021-05-20 16:37 ` [PATCH 30/39] PCI: " Marc Zyngier
2021-05-20 17:47 ` Rob Herring
2021-05-24 8:38 ` Marc Zyngier
2021-05-24 13:28 ` Rob Herring
2021-05-24 15:06 ` Marc Zyngier
2021-05-21 8:57 ` Geert Uytterhoeven
2021-05-20 16:37 ` [PATCH 31/39] mfd: " Marc Zyngier
2021-05-21 7:06 ` Lee Jones
2021-05-20 16:37 ` [PATCH 32/39] gpu: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 33/39] SH: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 34/39] ARM: " Marc Zyngier
2021-05-20 18:04 ` Rob Herring
2021-05-24 8:32 ` Marc Zyngier
2021-05-21 21:17 ` Krzysztof Kozlowski
2021-05-20 16:37 ` [PATCH 35/39] mips: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 36/39] arc: " Marc Zyngier
2021-06-01 18:00 ` Vineet Gupta
2021-05-20 16:37 ` [PATCH 37/39] xtensa: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 38/39] nios2: " Marc Zyngier
2021-05-20 16:37 ` [PATCH 39/39] powerpc: " Marc Zyngier
2021-05-20 23:33 ` [PATCH 00/39] irqdomain: Simplify interrupt handling Linus Walleij
2021-05-24 8:53 ` Marc Zyngier
2021-05-24 9:39 ` Linus Walleij
2021-06-01 14:29 ` Marc Zyngier
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=87a6i48pp5.wl-maz@kernel.org \
--to=maz@kernel.org \
--cc=clg@kaod.org \
--cc=groug@kaod.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=tglx@linutronix.de \
/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).