From: Jason Baron <jbaron@redhat.com>
To: Markus Armbruster <armbru@redhat.com>
Cc: aliguori@us.ibm.com, alex.williamson@redhat.com, mst@redhat.com,
jan.kiszka@siemens.com, qemu-devel@nongnu.org, agraf@suse.de,
blauwirbel@gmail.com, yamahata@valinux.co.jp, juzhang@redhat.com,
kevin@koconnor.net, avi@redhat.com, mkletzan@redhat.com,
pbonzini@redhat.com, lcapitulino@redhat.com, afaerber@suse.de,
kraxel@redhat.com
Subject: Re: [Qemu-devel] [PATCH v3 01/26] blockdev: Introduce a default machine blockdev interface field, QEMUMachine->mach_if
Date: Wed, 24 Oct 2012 15:41:53 -0400 [thread overview]
Message-ID: <20121024194153.GD7952@redhat.com> (raw)
In-Reply-To: <87liew10aj.fsf@blackfin.pond.sub.org>
On Wed, Oct 24, 2012 at 03:12:36PM +0200, Markus Armbruster wrote:
> Jason Baron <jbaron@redhat.com> writes:
>
> > From: Jason Baron <jbaron@redhat.com>
> >
> > The current QEMUMachine definition has a 'use_scsi' field to indicate if a
> > machine type should use scsi by default. However, Q35 wants to use ahci by
> > default. Thus, introdue a new field in the QEMUMachine defintion, mach_if.
>
> Even though q35's desire to default to IF_AHCI is driving this patch,
> generalizing the default interface type makes sense on its own. Even if
> we IF_AHCI should turn out to need more discussion.
>
> > This field should be initialized by the machine type to the default interface
> > type which it wants to use (IF_SCSI, IF_AHCI, etc.). If no mach_if is defined,
> > or it is set to 'IF_DEFAULT' or 'IF_NONE', we currently assume IF_IDE.
> >
> > Please use 'static inline int get_mach_if(int mach_if)', when accesssing the
> > new mach_if field.
> >
> > Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
> > Signed-off-by: Jason Baron <jbaron@redhat.com>
> > ---
> > blockdev.c | 4 ++--
> > blockdev.h | 19 +++++++++++++++++++
> > hw/boards.h | 2 +-
> > hw/device-hotplug.c | 2 +-
> > hw/highbank.c | 2 +-
> > hw/leon3.c | 2 +-
> > hw/mips_jazz.c | 4 ++--
> > hw/pc_sysfw.c | 2 +-
> > hw/puv3.c | 2 +-
> > hw/realview.c | 6 +++---
> > hw/spapr.c | 2 +-
> > hw/sun4m.c | 24 ++++++++++++------------
> > hw/versatilepb.c | 4 ++--
> > hw/vexpress.c | 4 ++--
> > hw/xilinx_zynq.c | 2 +-
> > vl.c | 20 +++++++++++---------
> > 16 files changed, 61 insertions(+), 40 deletions(-)
> >
> > diff --git a/blockdev.c b/blockdev.c
> > index 99828ad..c9a49c8 100644
> > --- a/blockdev.c
> > +++ b/blockdev.c
> > @@ -275,7 +275,7 @@ static bool do_check_io_limits(BlockIOLimit *io_limits)
> > return true;
> > }
> >
> > -DriveInfo *drive_init(QemuOpts *opts, int default_to_scsi)
> > +DriveInfo *drive_init(QemuOpts *opts, int mach_if)
>
> BlockInterfaceType mach_if
>
> Suggest to rename mach_if to something more descriptive. Kevin's
> default_drive_if works for me.
>
ok.
>
> > {
> > const char *buf;
> > const char *file = NULL;
> > @@ -325,7 +325,7 @@ DriveInfo *drive_init(QemuOpts *opts, int default_to_scsi)
> > return NULL;
> > }
> > } else {
> > - type = default_to_scsi ? IF_SCSI : IF_IDE;
> > + type = get_mach_if(mach_if);
> > }
> >
> > max_devs = if_max_devs[type];
> > diff --git a/blockdev.h b/blockdev.h
> > index 5f27b64..8b126ad 100644
> > --- a/blockdev.h
> > +++ b/blockdev.h
> > @@ -40,6 +40,22 @@ struct DriveInfo {
> > int refcount;
> > };
> >
> > +/*
> > + * Each qemu machine type defines a mach_if field for its default
> > + * interface type. If its unspecified, we set it to IF_IDE.
> > + */
> > +static inline int get_mach_if(int mach_if)
>
> BlockInterfaceType mach_if, and return type.
>
> > +{
> > + assert(mach_if < IF_COUNT);
> > + assert(mach_if >= IF_DEFAULT);
> > +
> > + if ((mach_if == IF_NONE) || (mach_if == IF_DEFAULT)) {
> > + return IF_IDE;
> > + }
> > +
> > + return mach_if;
> > +}
> > +
>
> I'm not sure we should map IF_NONE to IF_IDE.
>
> get_mach_if() should return an interface type the board supports. In
> theory, we could have a board that doesn't define any controllers, but
> still lets the user define some with -device (say because the board
> sports a PCI bus). Then IF_NONE would be the only interface type that
> makes any sense, and therefore the only sensible value of get_mach_if().
>
Ok, but no boards use IF_NONE in that sense currently. But planning for
the future is good :)
> If we drop the magic mapping of IF_NONE, only IF_DEFAULT is left, and
> that one's clearly marked "for use with drive_add() only". No real need
> for magic mapping then. Could drop get_mach_if() and use mach_if
> directly.
Meaning explicity set mach_if or default_drive_if to IF_IDE for all
machine types that are currently either IF_NONE, IF_DEFAULT, or not
explicitly defined?
Thanks,
-Jason
next prev parent reply other threads:[~2012-10-24 19:42 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-19 20:43 [Qemu-devel] [PATCH v3 00/26] q35 qemu support Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 01/26] blockdev: Introduce a default machine blockdev interface field, QEMUMachine->mach_if Jason Baron
2012-10-22 10:47 ` Michael S. Tsirkin
2012-10-22 11:26 ` Kevin Wolf
2012-10-22 18:02 ` Jason Baron
2012-10-24 13:12 ` Markus Armbruster
2012-10-24 19:41 ` Jason Baron [this message]
2012-10-26 10:28 ` Markus Armbruster
2012-10-26 9:53 ` Markus Armbruster
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 02/26] blockdev: Introduce IF_AHCI Jason Baron
2012-10-22 10:48 ` Michael S. Tsirkin
2012-10-22 11:40 ` Kevin Wolf
2012-10-22 18:11 ` Jason Baron
2012-10-24 15:50 ` Markus Armbruster
2012-10-24 19:36 ` Jason Baron
2012-10-26 12:56 ` Markus Armbruster
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 04/26] pci: introduce pci_swizzle_map_irq_fn() for standardized interrupt pin swizzle Jason Baron
2012-10-22 10:51 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 03/26] pci: pci capability must be in PCI space Jason Baron
2012-10-22 10:48 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 06/26] pc: Move ioapic_init() from pc_piix.c to pc.c Jason Baron
2012-10-22 13:28 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 05/26] pc, pc_piix: split out pc nic initialization Jason Baron
2012-10-22 13:27 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 07/26] pc/piix_pci: factor out smram/pam logic Jason Baron
2012-10-22 11:05 ` Michael S. Tsirkin
2012-10-29 16:21 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 08/26] pci_ids: add intel 82801BA pci-to-pci bridge id Jason Baron
2012-10-22 10:51 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 09/26] pci: Add class 0xc05 as 'SMBus' Jason Baron
2012-10-22 10:52 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 10/26] pcie: pass pcie window size to pcie_host_mmcfg_update() Jason Baron
2012-10-22 10:54 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 12/26] ich9: Add acpi support and definitions Jason Baron
2012-10-22 11:07 ` Michael S. Tsirkin
2012-10-22 11:22 ` Michael S. Tsirkin
2012-10-29 16:29 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 11/26] pcie: Convert PCIExpressHost to use the QOM Jason Baron
2012-10-22 10:55 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 13/26] ich9: Add the lpc chip Jason Baron
2012-10-22 11:12 ` Michael S. Tsirkin
2012-10-22 11:27 ` Michael S. Tsirkin
2012-10-23 4:22 ` Isaku Yamahata
2012-10-29 16:20 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 14/26] ich9: Add smbus Jason Baron
2012-10-22 11:13 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 15/26] q35: Introduce q35 pc based chipset emulator Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 16/26] ich9: Add i82801b11 dmi-to-pci bridge Jason Baron
2012-10-22 13:53 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 17/26] Add i21154 bridge chip Jason Baron
2012-10-22 13:26 ` Andreas Färber
2012-10-22 16:17 ` Michael S. Tsirkin
2012-10-22 18:18 ` Jason Baron
2012-10-22 18:53 ` Andreas Färber
2012-10-27 12:42 ` Blue Swirl
2012-10-22 14:03 ` Michael S. Tsirkin
2012-10-22 20:48 ` Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 18/26] q35: Suppress SMM BIOS initialization under KVM Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 20/26] q35: smbus: Remove PCI_STATUS_SIG_SYSTEM_ERROR and PCI_STATUS_DETECTED_PARITY from w1cmask Jason Baron
2012-10-21 12:26 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 19/26] q35: Fix non-PCI IRQ processing in ich9_lpc_update_apic Jason Baron
2012-10-22 14:04 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 21/26] q35: Add kvmclock support Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 22/26] Add a fallback bios file search, if -L fails Jason Baron
2012-10-21 7:26 ` Michael Tokarev
2012-10-21 9:52 ` Peter Maydell
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 23/26] q35: automatically load the q35 dsdt table Jason Baron
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 24/26] q35: add acpi-based pci hotplug Jason Baron
2012-10-22 14:09 ` Michael S. Tsirkin
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 25/26] q35: fill in usb pci slots with -usb Jason Baron
2012-10-22 5:54 ` Gerd Hoffmann
2012-10-24 17:10 ` Paolo Bonzini
2012-10-19 20:43 ` [Qemu-devel] [PATCH v3 26/26] ich9: add support pci assignment Jason Baron
2012-10-20 16:15 ` [Qemu-devel] [PATCH v3 00/26] q35 qemu support Michael Tokarev
2012-10-21 12:36 ` Michael S. Tsirkin
2012-10-21 12:43 ` Michael S. Tsirkin
2012-10-22 5:58 ` Gerd Hoffmann
2012-10-22 10:08 ` Michael S. Tsirkin
2012-10-22 10:37 ` Gerd Hoffmann
2012-10-22 13:16 ` Michael S. Tsirkin
2012-10-22 13:00 ` Eric Blake
2012-10-22 14:23 ` Michael S. Tsirkin
2012-10-22 14:03 ` Eric Blake
2012-10-22 14:39 ` Alexander Graf
2012-10-22 15:37 ` Anthony Liguori
2012-10-27 8:12 ` Michael Tokarev
2012-10-22 13:34 ` Michael S. Tsirkin
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=20121024194153.GD7952@redhat.com \
--to=jbaron@redhat.com \
--cc=afaerber@suse.de \
--cc=agraf@suse.de \
--cc=alex.williamson@redhat.com \
--cc=aliguori@us.ibm.com \
--cc=armbru@redhat.com \
--cc=avi@redhat.com \
--cc=blauwirbel@gmail.com \
--cc=jan.kiszka@siemens.com \
--cc=juzhang@redhat.com \
--cc=kevin@koconnor.net \
--cc=kraxel@redhat.com \
--cc=lcapitulino@redhat.com \
--cc=mkletzan@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=yamahata@valinux.co.jp \
/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).