From: Eric Auger <eric.auger@linaro.org>
To: Paolo Bonzini <pbonzini@redhat.com>,
eric.auger@st.com, qemu-devel@nongnu.org,
alex.williamson@redhat.com, peter.maydell@linaro.org,
agraf@suse.de
Cc: kvmarm@lists.cs.columbia.edu, christoffer.dall@linaro.org,
patches@linaro.org
Subject: Re: [Qemu-devel] [PATCH 0/2] irq: add get_gsi callback
Date: Thu, 23 Apr 2015 11:40:41 +0200 [thread overview]
Message-ID: <5538BE19.6030009@linaro.org> (raw)
In-Reply-To: <5538BBD3.9040404@redhat.com>
Hi Paolo,
On 04/23/2015 11:30 AM, Paolo Bonzini wrote:
>
>
> On 23/04/2015 10:49, Eric Auger wrote:
>> This series introduces a new callback function in IRQState, named
>> get_gsi_cb. It is supposed to be populated by the interrupt controller
>> and its role is to convert the interrupt controller pin number into
>> the global system interrupt (gsi) number. The gsi is used when setting
>> irqfd up.
>>
>> With PCI there is a PCIINTxRoute bus lookup mechanism that enables to
>> retrieve the gsi from the PCI host controller/bridge pin. The conversion
>> is implemented by the PCI host controller. With platform devices, this
>> conversion function is implemented by the interrupt controller.
>>
>> Besides the callback member, a setter is introduced. First user is
>> arm_gic_kvm. A public function wraps the callback, qemu_irq_get_gsi.
>>
>> The first user of qemu_irq_get_gsi might be the VFIO platform device.
>> This will come in the "KVM platform device passthrough" series.
>
> I'm sorry, I cannot understand the point of this without seeing a user.
no problem, I intend to send the KVM platform device passthrough v13
today which will illustrate the usage.
>
> Why can't you just use a GHashTable?
You mean implementing this hash table in the interrupt controller?
The problem is my VFIO device currently has no link to the interrupt
controller object. Besides it has a platform bus in between. Previously
I devised a reset notifier called in the machine file to pass this link
but Alex was not keen about this method. So the idea here is when the
VFIO sysbus device qemuirq is getting connected, a notifier is called,
VFIO device can retrieve the gsi of this qemu_irq and call KVM_IRQFD
with that gsi.
Best Regards
Eric
This mustn't be a hot path, since
> VFIO does all the signalling in the kernel via irqfds.
>
> Paolo
>
next prev parent reply other threads:[~2015-04-23 9:43 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-23 8:49 [Qemu-devel] [PATCH 0/2] irq: add get_gsi callback Eric Auger
2015-04-23 8:49 ` [Qemu-devel] [PATCH 1/2] " Eric Auger
2015-04-23 8:49 ` [Qemu-devel] [PATCH 2/2] intc: arm_gic_kvm: set the " Eric Auger
2015-04-23 9:30 ` [Qemu-devel] [PATCH 0/2] irq: add " Paolo Bonzini
2015-04-23 9:40 ` Eric Auger [this message]
2015-04-23 9:58 ` Paolo Bonzini
2015-04-23 11:25 ` Eric Auger
2015-04-24 9:01 ` Eric Auger
2015-04-24 9:11 ` Paolo Bonzini
2015-04-24 9:18 ` Eric Auger
2015-04-24 9:29 ` Paolo Bonzini
2015-04-24 9:48 ` Eric Auger
2015-04-24 10:02 ` Paolo Bonzini
2015-04-24 11:59 ` Eric Auger
2015-04-24 12:04 ` Paolo Bonzini
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=5538BE19.6030009@linaro.org \
--to=eric.auger@linaro.org \
--cc=agraf@suse.de \
--cc=alex.williamson@redhat.com \
--cc=christoffer.dall@linaro.org \
--cc=eric.auger@st.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=patches@linaro.org \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@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).