From: Igor Mammedov <imammedo@redhat.com>
To: Samuel Ortiz <sameo@linux.intel.com>
Cc: Yang Zhong <yang.zhong@intel.com>,
Peter Maydell <peter.maydell@linaro.org>,
Stefano Stabellini <sstabellini@kernel.org>,
Eduardo Habkost <ehabkost@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
qemu-devel@nongnu.org, Shannon Zhao <shannon.zhaosl@gmail.com>,
qemu-arm@nongnu.org, Paolo Bonzini <pbonzini@redhat.com>,
Anthony Perard <anthony.perard@citrix.com>,
xen-devel@lists.xenproject.org,
Richard Henderson <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH v5 12/24] hw: acpi: Export the MCFG getter
Date: Tue, 27 Nov 2018 14:54:15 +0100 [thread overview]
Message-ID: <20181127145415.17b49d64@redhat.com> (raw)
In-Reply-To: <20181121232106.GB4450@caravaggio>
On Thu, 22 Nov 2018 00:21:06 +0100
Samuel Ortiz <sameo@linux.intel.com> wrote:
> Hi Igor,
>
> On Thu, Nov 15, 2018 at 01:36:58PM +0100, Igor Mammedov wrote:
> > On Mon, 5 Nov 2018 02:40:35 +0100
> > Samuel Ortiz <sameo@linux.intel.com> wrote:
> >
> > > From: Yang Zhong <yang.zhong@intel.com>
> > >
> > > The ACPI MCFG getter is not x86 specific and could be called from
> > > anywhere within generic ACPI API, so let's export it.
> > So far it's x86 or more exactly q35 specific thing,
> It's property based, and it's using a generic PCIE property afaict.
> So it's up to each machine type to define those properties.
> I'm curious here: What's the idiomatic way to define a machine
> setting/attribute/property, let each instance define it or not, and
> make it available at run time?
> Would you be getting the PCI host pointer from the ACPI build state and
> getting that information back from there?
Cleaner way would be make arm/virt board set PCIE_HOST_MCFG_BASE/
PCIE_HOST_MCFG_SIZE properties and then use common build_mcfg()(in aml-build.c).
Something like this:
acpi_setup_reduced()
AcpiMcfgInfo mcfg_info = {
.base = object_property_get_uint(pcie, PCIE_HOST_MCFG_BASE, NULL),
.size = object_property_get_uint(pcie, PCIE_HOST_MCFG_SIZE, NULL)
};
acpi_build() {
build_mcfg("MCFG", &info);
}
}
and for legacy q35
acpi_build() {
if (pcie) {
AcpiMcfgInfo mcfg_info = {
.base = object_property_get_uint(pcie, PCIE_HOST_MCFG_BASE, NULL),
.size = object_property_get_uint(pcie, PCIE_HOST_MCFG_SIZE, NULL)
};
if (mcfg_info.base != PCIE_BASE_ADDR_UNMAPPED)
build_mcfg("MCFG", &info);
else
/* move comment here why we are doing it */
build_mcfg("QEMU", &info);
}
}
The thing I don't like about acpi_get_mcfg() is that it
does lookup acpi_get_i386_pci_host() each time it's called
and judges if it's PCI-E host by presence of properties.
I'd rather be explicit where PCI host be fetched once somewhere
in acpi_setup() or possibly passed down from the board as an argument
and board telling to i386/acpi_setup() if it's PCI or PCI-E host
so we don't have to guess it in acpi code.
> Cheers,
> Samuel.
next prev parent reply other threads:[~2018-11-27 13:54 UTC|newest]
Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-05 1:40 [Qemu-devel] [PATCH v5 00/24] ACPI reorganization for hardware-reduced API addition Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 01/24] hw: i386: Decouple the ACPI build from the PC machine type Samuel Ortiz
2018-11-09 14:23 ` Igor Mammedov
2018-11-21 14:42 ` Samuel Ortiz
2018-11-22 15:13 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 02/24] hw: acpi: Export ACPI build alignment API Samuel Ortiz
2018-11-09 14:27 ` Igor Mammedov
2018-11-21 14:42 ` Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 03/24] hw: acpi: The RSDP build API can return void Samuel Ortiz
2018-11-06 10:23 ` Paolo Bonzini
2018-11-06 10:43 ` Samuel Ortiz
2018-11-08 14:24 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 04/24] hw: acpi: Export the RSDP build API Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 05/24] hw: acpi: Implement XSDT support for RSDP Samuel Ortiz
2018-11-08 14:16 ` Igor Mammedov
2018-11-08 14:36 ` Samuel Ortiz
2018-11-08 14:53 ` Igor Mammedov
2018-11-19 18:27 ` Michael S. Tsirkin
2018-11-20 8:23 ` Igor Mammedov
2018-11-21 14:42 ` Samuel Ortiz
2018-11-22 16:26 ` Igor Mammedov
2018-11-23 9:36 ` Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 06/24] hw: acpi: Factorize the RSDP build API implementation Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 07/24] hw: acpi: Generalize AML build routines Samuel Ortiz
2018-11-09 13:37 ` Igor Mammedov
2018-11-21 15:00 ` Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 08/24] hw: acpi: Factorize _OSC AML across architectures Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 09/24] hw: i386: Move PCI host definitions to pci_host.h Samuel Ortiz
2018-11-09 14:30 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 10/24] hw: acpi: Export the PCI host and holes getters Samuel Ortiz
2018-11-13 15:59 ` Igor Mammedov
2018-11-21 15:43 ` Samuel Ortiz
2018-11-23 10:55 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 11/24] hw: acpi: Export and generalize the PCI host AML API Samuel Ortiz
2018-11-14 10:55 ` Igor Mammedov
2018-11-21 23:12 ` Samuel Ortiz
2018-11-23 11:04 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 12/24] hw: acpi: Export the MCFG getter Samuel Ortiz
2018-11-15 12:36 ` Igor Mammedov
2018-11-21 23:21 ` Samuel Ortiz
2018-11-27 13:54 ` Igor Mammedov [this message]
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 13/24] hw: acpi: Do not create hotplug method when handler is not defined Samuel Ortiz
2018-11-09 9:12 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 14/24] hw: i386: Make the hotpluggable memory size property more generic Samuel Ortiz
2018-11-15 12:49 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 15/24] hw: i386: Export the i386 ACPI SRAT build method Samuel Ortiz
2018-11-15 13:28 ` Igor Mammedov
2018-11-21 23:27 ` Samuel Ortiz
2018-11-26 15:47 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 16/24] hw: acpi: Fix memory hotplug AML generation error Samuel Ortiz
2018-11-08 14:23 ` Igor Mammedov
2019-01-14 18:35 ` Michael S. Tsirkin
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 17/24] hw: acpi: Export the PCI hotplug API Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 18/24] hw: i386: Export the MADT build method Samuel Ortiz
2018-11-16 9:27 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 19/24] hw: acpi: Retrieve the PCI bus from AcpiPciHpState Samuel Ortiz
2018-11-16 9:39 ` Igor Mammedov
2018-11-16 19:42 ` Boeuf, Sebastien
2018-11-19 15:37 ` Igor Mammedov
2018-11-19 18:02 ` Boeuf, Sebastien
2018-11-20 8:26 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 20/24] hw: acpi: Define ACPI tables builder interface Samuel Ortiz
2018-11-16 16:02 ` Igor Mammedov
2018-11-21 23:57 ` Samuel Ortiz
2018-11-27 14:08 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 21/24] hw: i386: Implement the ACPI builder interface for PC Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 22/24] hw: pci-host: piix: Return PCI host pointer instead of PCI bus Samuel Ortiz
2018-11-16 11:09 ` Igor Mammedov
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 23/24] hw: i386: Set ACPI configuration PCI host pointer Samuel Ortiz
2018-11-05 1:40 ` [Qemu-devel] [PATCH v5 24/24] hw: i386: Refactor PCI host getter Samuel Ortiz
2018-11-16 16:29 ` [Qemu-devel] [PATCH v5 00/24] ACPI reorganization for hardware-reduced API addition Igor Mammedov
2018-11-16 16:37 ` Paolo Bonzini
2018-11-19 15:31 ` Igor Mammedov
2018-11-19 17:14 ` Paolo Bonzini
2018-11-19 18:14 ` Michael S. Tsirkin
2018-11-20 21:35 ` Paolo Bonzini
2018-11-20 12:57 ` Igor Mammedov
2018-11-20 21:36 ` Paolo Bonzini
2018-11-21 12:35 ` Michael S. Tsirkin
2018-11-21 13:50 ` Samuel Ortiz
2018-11-21 13:57 ` Michael S. Tsirkin
2018-11-21 14:15 ` Igor Mammedov
2018-11-21 14:38 ` Samuel Ortiz
2018-11-22 10:39 ` Igor Mammedov
2018-11-22 0:17 ` Samuel Ortiz
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=20181127145415.17b49d64@redhat.com \
--to=imammedo@redhat.com \
--cc=anthony.perard@citrix.com \
--cc=ehabkost@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
--cc=sameo@linux.intel.com \
--cc=shannon.zhaosl@gmail.com \
--cc=sstabellini@kernel.org \
--cc=xen-devel@lists.xenproject.org \
--cc=yang.zhong@intel.com \
/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).