From: "Andreas Färber" <afaerber@suse.de>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Peter Crosthwaite <peter.crosthwaite@xilinx.com>,
Igor Mitsyanko <i.mitsyanko@samsung.com>,
Evgeny Voevodin <e.voevodin@samsung.com>,
patches@linaro.org, Hu Tao <hutao@cn.fujitsu.com>,
qemu-devel@nongnu.org,
Jean-Christophe DUBOIS <jcd@tribudubois.net>,
Peter Chubb <peter.chubb@nicta.com.au>,
Dmitry Solodkiy <d.solodkiy@samsung.com>,
Anthony Liguori <anthony@codemonkey.ws>,
Paolo Bonzini <pbonzini@redhat.com>,
Maksim Kozlov <m.kozlov@samsung.com>,
Gerd Hoffmann <kraxel@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] Avoid use of QOM type name macros in VMStateDescriptions
Date: Sun, 30 Jun 2013 09:20:26 +0200 [thread overview]
Message-ID: <51CFDC3A.5090509@suse.de> (raw)
In-Reply-To: <1372331024-3783-1-git-send-email-peter.maydell@linaro.org>
Am 27.06.2013 13:03, schrieb Peter Maydell:
> The name field in a VMStateDescription is part of the migration state
> versioning, so changing it will break migration. It's therefore a
> bad idea to use a QOM typename macro to initialize it, because in
> general we're free to rename QOM types as part of code refactoring
> and cleanup. For the handful of devices that were doing this by
> mistake, replace the QOM typenames with the corresponding literal
> strings.
>
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> ---
> As per recent discussion. There are also a few devices which use
> the typename in memory_region_init_io(). Since that is suboptimal
> but not a problem in the way that possible migration breaks would
> be, I haven't fixed those since they'd just clash with Paolo's
> memory-region-owner patches.
>
> The one I didn't touch was hw/usb/host-linux.c, since that changes
> the QOM typename and the VMStateDescription name depending on
> whether QEMU was built with CONFIG_USB_LIBUSB defined or not.
> That seems a bit fishy to me but I've left it alone.
>
> hw/i2c/exynos4210_i2c.c | 2 +-
> hw/scsi/vmw_pvscsi.c | 2 +-
> hw/timer/imx_epit.c | 2 +-
> hw/timer/imx_gpt.c | 2 +-
> hw/usb/ccid-card-passthru.c | 2 +-
> hw/usb/dev-smartcard-reader.c | 2 +-
> 6 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/hw/i2c/exynos4210_i2c.c b/hw/i2c/exynos4210_i2c.c
> index 196f889..a75abef 100644
> --- a/hw/i2c/exynos4210_i2c.c
> +++ b/hw/i2c/exynos4210_i2c.c
> @@ -271,7 +271,7 @@ static const MemoryRegionOps exynos4210_i2c_ops = {
> };
>
> static const VMStateDescription exynos4210_i2c_vmstate = {
> - .name = TYPE_EXYNOS4_I2C,
> + .name = "exynos4210.i2c",
> .version_id = 1,
> .minimum_version_id = 1,
> .fields = (VMStateField[]) {
> diff --git a/hw/scsi/vmw_pvscsi.c b/hw/scsi/vmw_pvscsi.c
> index 7cf4044..f2f0c00 100644
> --- a/hw/scsi/vmw_pvscsi.c
> +++ b/hw/scsi/vmw_pvscsi.c
> @@ -1136,7 +1136,7 @@ pvscsi_post_load(void *opaque, int version_id)
> }
>
> static const VMStateDescription vmstate_pvscsi = {
> - .name = TYPE_PVSCSI,
> + .name = "pvscsi",
> .version_id = 0,
> .minimum_version_id = 0,
> .minimum_version_id_old = 0,
Apparently someone confused TypeInfo and VMStateDescription here, the
TypeInfo .name is by contrast not using the constant - fixing up.
Rest is verified to match constants.
Applied to my new qom-next staging tree:
https://github.com/afaerber/qemu-cpu/commits/qom-next
Andreas
> diff --git a/hw/timer/imx_epit.c b/hw/timer/imx_epit.c
> index 7cdb006..8cefd74a 100644
> --- a/hw/timer/imx_epit.c
> +++ b/hw/timer/imx_epit.c
> @@ -370,7 +370,7 @@ static const MemoryRegionOps imx_epit_ops = {
> };
>
> static const VMStateDescription vmstate_imx_timer_epit = {
> - .name = TYPE_IMX_EPIT,
> + .name = "imx.epit",
> .version_id = 2,
> .minimum_version_id = 2,
> .minimum_version_id_old = 2,
> diff --git a/hw/timer/imx_gpt.c b/hw/timer/imx_gpt.c
> index de53b13..eebd2b7 100644
> --- a/hw/timer/imx_gpt.c
> +++ b/hw/timer/imx_gpt.c
> @@ -142,7 +142,7 @@ typedef struct {
> } IMXGPTState;
>
> static const VMStateDescription vmstate_imx_timer_gpt = {
> - .name = TYPE_IMX_GPT,
> + .name = "imx.gpt",
> .version_id = 3,
> .minimum_version_id = 3,
> .minimum_version_id_old = 3,
> diff --git a/hw/usb/ccid-card-passthru.c b/hw/usb/ccid-card-passthru.c
> index 01c7e6f..5f01ff1 100644
> --- a/hw/usb/ccid-card-passthru.c
> +++ b/hw/usb/ccid-card-passthru.c
> @@ -364,7 +364,7 @@ static int passthru_exitfn(CCIDCardState *base)
> }
>
> static VMStateDescription passthru_vmstate = {
> - .name = PASSTHRU_DEV_NAME,
> + .name = "ccid-card-passthru",
> .version_id = 1,
> .minimum_version_id = 1,
> .fields = (VMStateField[]) {
> diff --git a/hw/usb/dev-smartcard-reader.c b/hw/usb/dev-smartcard-reader.c
> index 125cc2c..b33eb25 100644
> --- a/hw/usb/dev-smartcard-reader.c
> +++ b/hw/usb/dev-smartcard-reader.c
> @@ -1397,7 +1397,7 @@ static VMStateDescription usb_device_vmstate = {
> };
>
> static VMStateDescription ccid_vmstate = {
> - .name = CCID_DEV_NAME,
> + .name = "usb-ccid",
> .version_id = 1,
> .minimum_version_id = 1,
> .post_load = ccid_post_load,
>
--
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
prev parent reply other threads:[~2013-06-30 7:20 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-27 11:03 [Qemu-devel] [PATCH] Avoid use of QOM type name macros in VMStateDescriptions Peter Maydell
2013-06-27 11:41 ` Gerd Hoffmann
2013-06-30 7:20 ` Andreas Färber [this message]
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=51CFDC3A.5090509@suse.de \
--to=afaerber@suse.de \
--cc=anthony@codemonkey.ws \
--cc=d.solodkiy@samsung.com \
--cc=e.voevodin@samsung.com \
--cc=hutao@cn.fujitsu.com \
--cc=i.mitsyanko@samsung.com \
--cc=jcd@tribudubois.net \
--cc=kraxel@redhat.com \
--cc=m.kozlov@samsung.com \
--cc=patches@linaro.org \
--cc=pbonzini@redhat.com \
--cc=peter.chubb@nicta.com.au \
--cc=peter.crosthwaite@xilinx.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).