qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Markus Armbruster <armbru@redhat.com>
Cc: Alexander Graf <agraf@suse.de>,
	Anthony Liguori <aliguori@amazon.com>,
	QEMU Developers <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH v2] qdev: Keep global allocation counter per bus
Date: Sun, 22 Dec 2013 14:43:43 +0100	[thread overview]
Message-ID: <52B6EC8F.8080107@redhat.com> (raw)
In-Reply-To: <87lhzelc2k.fsf@blackfin.pond.sub.org>

Il 21/12/2013 11:42, Markus Armbruster ha scritto:
> Suggest to add: ", killing migration."

Not good.  But perhaps we can give a reason for this 2.0 thing.

It is certainly nice to schedule incompatible changes for obscure
machine types every 2 years.

> Which boards are affected?  Should be listed in the commit message!
> 
> I ran a quick test for all boards that actually make it to the monitor
> without special parameters or files, and survive "info qtree".  164
> boards can do that, 59 refuse to start, one crashes on start, 10 make it
> to the monitor but crash in info qtree.  Not nice.  If there's a way to
> start *any* board to the monitor, please let me know.

"-machine accel=qtest" probably helps with those that refuse to start.

Paolo

> Anyway, I found the following machines sporting non-unique bus names
> before your patch:
> 
>     target      machine         bus id              times
>     aarch64     n800            i2c-bus.0           2
>     aarch64     n810            i2c-bus.0           2
>     aarch64     nuri            i2c                 9
>     aarch64     smdkc210        i2c                 9
>     aarch64     vexpress-a15    virtio-mmio-bus.0   4
>     aarch64     vexpress-a9     virtio-mmio-bus.0   4
>     aarch64     virt            virtio-mmio-bus.0   32
>     aarch64     xilinx-zynq-a9  usb-bus.0           2
>     aarch64     xilinx-zynq-a9  spi0                3
>     arm         n800            i2c-bus.0           2
>     arm         n810            i2c-bus.0           2
>     arm         nuri            i2c                 9
>     arm         smdkc210        i2c                 9
>     arm         vexpress-a15    virtio-mmio-bus.0   4
>     arm         vexpress-a9     virtio-mmio-bus.0   4
>     arm         virt            virtio-mmio-bus.0   32
>     arm         xilinx-zynq-a9  usb-bus.0           2
>     arm         xilinx-zynq-a9  spi0                3
>     i386        isapc           ide.0               2
>     mips        mips            ide.0               2
>     mips64      mips            ide.0               2
>     mips64el    mips            ide.0               2
>     mipsel      mips            ide.0               2
>     ppc         g3beige         ide.0               2
>     ppc         mac99           ide.0               2
>     ppc         prep            ide.0               2
>     ppc64       g3beige         ide.0               2
>     ppc64       mac99           ide.0               2
>     ppc64       prep            ide.0               2
>     x86_64      isapc           ide.0               2
> 
> The isapc crash with v1 demonstrates that such machines need testing
> with your patch.
> 
> Not yet covered:
> 
>     target      machine         reason
>     aarch64     akita           info qtree crashes
>     aarch64     borzoi          info qtree crashes
>     aarch64     canon-a1100     refuses to start
>     aarch64     cheetah         refuses to start
>     aarch64     connex          refuses to start
>     aarch64     lm3s6965evb     refuses to start
>     aarch64     lm3s811evb      refuses to start
>     aarch64     mainstone       refuses to start
>     aarch64     spitz           info qtree crashes
>     aarch64     sx1             refuses to start
>     aarch64     sx1-v1          refuses to start
>     aarch64     terrier         info qtree crashes
>     aarch64     tosa            info qtree crashes
>     aarch64     verdex          refuses to start
>     aarch64     z2              refuses to start
>     arm         akita           info qtree crashes
>     arm         borzoi          info qtree crashes
>     arm         canon-a1100     refuses to start
>     arm         cheetah         refuses to start
>     arm         connex          refuses to start
>     arm         lm3s6965evb     refuses to start
>     arm         lm3s811evb      refuses to start
>     arm         mainstone       refuses to start
>     arm         spitz           info qtree crashes
>     arm         sx1             refuses to start
>     arm         sx1-v1          refuses to start
>     arm         terrier         info qtree crashes
>     arm         tosa            info qtree crashes
>     arm         verdex          refuses to start
>     arm         z2              refuses to start
>     cris        axis-dev88      refuses to start
>     i386        xenfv           refuses to start
>     i386        xenpv           refuses to start
>     lm32        milkymist       refuses to start
>     m68k        an5206          refuses to start
>     m68k        mcf5208evb      refuses to start
>     mips        magnum          refuses to start
>     mips        malta           refuses to start
>     mips        mipssim         refuses to start
>     mips        pica61          refuses to start
>     mips64      magnum          refuses to start
>     mips64      malta           refuses to start
>     mips64      mipssim         refuses to start
>     mips64      pica61          refuses to start
>     mips64el    fulong2e        refuses to start
>     mips64el    magnum          refuses to start
>     mips64el    malta           refuses to start
>     mips64el    mipssim         refuses to start
>     mips64el    pica61          refuses to start
>     mipsel      magnum          refuses to start
>     mipsel      malta           refuses to start
>     mipsel      mipssim         refuses to start
>     mipsel      pica61          refuses to start
>     ppc         ref405ep        refuses to start
>     ppc         taihu           refuses to start
>     ppc64       ref405ep        refuses to start
>     ppc64       taihu           refuses to start
>     ppcemb      g3beige         refuses to start
>     ppcemb      mac99           refuses to start
>     ppcemb      mpc8544ds       refuses to start
>     ppcemb      ppce500         refuses to start
>     ppcemb      prep            refuses to start
>     ppcemb      ref405ep        refuses to start
>     ppcemb      taihu           refuses to start
>     sh4         shix            refuses to start
>     sh4eb       shix            refuses to start
>     sparc       leon3_generic   refuses to start
>     unicore32   puv3            crashes on startup
>     x86_64      xenfv           refuses to start
>     x86_64      xenpv           refuses to start
> 
>>
>> This is intended and makes the bus enumeration work as expected.
>>
>> CC: Paolo Bonzini <pbonzini@redhat.com>
>> CC: Markus Armbruster <armbru@redhat.com>
>> CC: Anthony Liguori <aliguori@amazon.com>
>> Signed-off-by: Alexander Graf <agraf@suse.de>
>>
>> ---
>>
>> v1 -> v2:
>>
>>   - add fix for isapc which was searching for 2 buses called "ide.0"
> 
> Yes, this version no longer crashes on startup.  The secondary
> controller's bus gets renamed to "ide.1".
> 
>>   - explain the semantic change more in the commit message
> 
> Patch looks good to me, but I'd recommend more thorough testing, as
> outlined above.
> 

  reply	other threads:[~2013-12-22 13:43 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-20  1:41 [Qemu-devel] [PATCH v2] qdev: Keep global allocation counter per bus Alexander Graf
2013-12-21 10:42 ` Markus Armbruster
2013-12-22 13:43   ` Paolo Bonzini [this message]
2014-01-07 15:12     ` Markus Armbruster
2014-01-07 16:59       ` Paolo Bonzini
2014-01-07 17:34         ` Markus Armbruster
2014-01-08 14:04           ` Paolo Bonzini
2014-01-08 14:35             ` Markus Armbruster
2014-01-08 15:18               ` Paolo Bonzini
2014-01-08 16:52                 ` Markus Armbruster
2014-01-08  3:07         ` Peter Crosthwaite
2014-01-08  4:24           ` Andreas Färber
2014-01-08  8:00             ` Markus Armbruster
2014-01-08 10:11             ` Peter Maydell
2014-01-08  8:13           ` Markus Armbruster
2014-01-08  8:26             ` Peter Crosthwaite
2014-01-08 13:40             ` Andreas Färber
2014-01-08 13:47               ` Paolo Bonzini
2014-01-10  7:50                 ` Peter Crosthwaite
2014-01-10  8:48                   ` Markus Armbruster
2014-02-04  9:28                     ` Markus Armbruster
2014-02-05  5:19                       ` Peter Crosthwaite
2014-02-05  8:45                         ` Markus Armbruster
2014-01-08 11:02           ` Paolo Bonzini
2014-01-08 13:53             ` Andreas Färber
2014-01-08 14:07               ` 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=52B6EC8F.8080107@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=agraf@suse.de \
    --cc=aliguori@amazon.com \
    --cc=armbru@redhat.com \
    --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).