Re: [libvirt] [PATCH] Add AHCI support to qemu driver
On Tue, Oct 18, 2011 at 07:49:11AM -0500, Adam Litke wrote: > On Mon, Oct 17, 2011 at 02:08:42PM -0600, Jim Fehlig wrote: > > Hi Adam, > > > > I've been traveling, sorry for the delay... > > > > > With the following config, I end up booting from the cdrom. If I take > > > away the > > > cdrom definition completely, I am unable to boot at all. > > > > > > > I can boot fine with your config, either from sda () or > > cdrom (). > > > > Have you verified your seabios supports booting from ahci as Bruce > > mentioned? > > No, I have not been able to verify it. Seabios refuses to build on my system > with the version of binutils I have installed. I am fairly certain that I am > using too old of a version of seabios (as you suggested). I am happy to > withdraw my problem report. And I just now was able to verify it. With an updated seabios, everything seems to be working fine for me. I can boot from the disks without problems. -- Adam Litke IBM Linux Technology Center -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
On Mon, Oct 17, 2011 at 02:08:42PM -0600, Jim Fehlig wrote: > Hi Adam, > > I've been traveling, sorry for the delay... > > > With the following config, I end up booting from the cdrom. If I take away > > the > > cdrom definition completely, I am unable to boot at all. > > > > I can boot fine with your config, either from sda () or > cdrom (). > > Have you verified your seabios supports booting from ahci as Bruce > mentioned? No, I have not been able to verify it. Seabios refuses to build on my system with the version of binutils I have installed. I am fairly certain that I am using too old of a version of seabios (as you suggested). I am happy to withdraw my problem report. -- Adam Litke IBM Linux Technology Center -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
Eric Blake wrote: > On 09/28/2011 04:43 PM, Jim Fehlig wrote: >> Tested with multiple AHCI controllers and multiple disks attached >> to a controller. E.g., >> >> +++ b/src/qemu/qemu_capabilities.c >> @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST, >> "no-shutdown", >> >> "cache-unsafe", /* 75 */ >> + "ich9-ahci", > > You'll have to rebase here, but nothing too difficult. Yep, simple rebase > > ACK. and pushed. Thanks! Jim -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
On 09/28/2011 04:43 PM, Jim Fehlig wrote: Tested with multiple AHCI controllers and multiple disks attached to a controller. E.g., +++ b/src/qemu/qemu_capabilities.c @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST, "no-shutdown", "cache-unsafe", /* 75 */ + "ich9-ahci", You'll have to rebase here, but nothing too difficult. ACK. -- Eric Blake ebl...@redhat.com+1-801-349-2682 Libvirt virtualization library http://libvirt.org -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
Adam Litke wrote: > On Mon, Oct 03, 2011 at 10:46:18PM -0600, Jim Fehlig wrote: > >> Adam Litke wrote: >> >>> Hi Jim. I was testing this and found that I could not boot from the sata >>> disks >>> I defined. When I switch them back to ide, they can be booted just fine. >>> Perhaps something is missing from the boot order logic? >>> >>> >> Hmm, I didn't notice this in my testing. sda in the below config >> contained /boot. Can you provide the domain config? >> > > Hi Adam, I've been traveling, sorry for the delay... > With the following config, I end up booting from the cdrom. If I take away > the > cdrom definition completely, I am unable to boot at all. > I can boot fine with your config, either from sda () or cdrom (). Have you verified your seabios supports booting from ahci as Bruce mentioned? Thanks, Jim > > blockPull-test > 262144 > 262144 > 1 > > hvm > > > > > > > > > destroy > restart > restart > > > /home/aglitke/src/qemu/x86_64-softmmu/qemu-system-x86_64 > > > > > > > > > > > > > > > > > > > > > > > -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
Adam,
Are you using a seabios which supports booting from ahci? It needs to be
somewhat recent, not more than around 6 weeks old, as there have been
bug fixes in that area.
Bruce
>>> Adam Litke 10/03/11 4:43 PM >>>
Hi Jim. I was testing this and found that I could not boot from the sata disks
I defined. When I switch them back to ide, they can be booted just fine.
Perhaps something is missing from the boot order logic?
On Wed, Sep 28, 2011 at 04:43:25PM -0600, Jim Fehlig wrote:
> Tested with multiple AHCI controllers and multiple disks attached
> to a controller. E.g.,
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> ---
> docs/formatdomain.html.in |9 +++--
> docs/schemas/domaincommon.rng |1 +
> src/conf/domain_conf.c | 14 +
> src/qemu/qemu_capabilities.c |3 ++
> src/qemu/qemu_capabilities.h |1 +
> src/qemu/qemu_command.c| 30
>
> .../qemuxml2argv-disk-sata-device.args |6
> .../qemuxml2argv-disk-sata-device.xml | 25
> tests/qemuxml2argvtest.c |3 ++
> 9 files changed, 82 insertions(+), 10 deletions(-)
>
> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in
> index 3087d01..b6a0c66 100644
> --- a/docs/formatdomain.html.in
> +++ b/docs/formatdomain.html.in
> @@ -972,11 +972,12 @@
> as a device ordering hint. The optional bus
> attribute specifies the type of disk device to emulate;
> possible values are driver specific, with typical values being
> -"ide", "scsi", "virtio", "xen" or "usb". If omitted, the bus type is
> -inferred from the style of the device name. eg, a device named 'sda'
> -will typically be exported using a SCSI bus.
> +"ide", "scsi", "virtio", "xen", "usb" or "sata". If omitted, the bus
> +type is inferred from the style of the device name. eg, a device
> named
> +'sda' will typically be exported using a SCSI bus.
> Since 0.0.3; bus attribute since 0.4.3;
> -"usb" attribute value since after 0.4.4
> +"usb" attribute value since after 0.4.4; "sata" attribute value since
> +0.9.7
>driver
>
> The optional driver element allows specifying further details
> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng
> index be98be0..675d55d 100644
> --- a/docs/schemas/domaincommon.rng
> +++ b/docs/schemas/domaincommon.rng
> @@ -790,6 +790,7 @@
> xen
> usb
> uml
> +sata
>
>
>
> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
> index a918679..6a7f296 100644
> --- a/src/conf/domain_conf.c
> +++ b/src/conf/domain_conf.c
> @@ -2157,6 +2157,15 @@ virDomainDiskDefAssignAddress(virCapsPtr caps,
> virDomainDiskDefPtr def)
> def->info.addr.drive.unit = (idx % 2);
> break;
>
> +case VIR_DOMAIN_DISK_BUS_SATA:
> +/* For SATA we define the default mapping to be 6 units
> + * per bus, 1 bus per controller, many controllers */
> +def->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_DRIVE;
> +def->info.addr.drive.controller = idx / 6;
> +def->info.addr.drive.bus = 0;
> +def->info.addr.drive.unit = idx % 6;
> +break;
> +
> case VIR_DOMAIN_DISK_BUS_FDC:
> /* For FDC we define the default mapping to be 2 units
> * per bus, 1 bus per controller, many controllers */
> @@ -8675,6 +8684,11 @@ int virDomainDefAddImplicitControllers(virDomainDefPtr
> def)
>VIR_DOMAIN_DISK_BUS_IDE) < 0)
> return -1;
>
> +if (virDomainDefAddDiskControllersForType(def,
> +
> VIR_DOMAIN_CONTROLLER_TYPE_SATA,
> + VIR_DOMAIN_DISK_BUS_SATA) < 0)
> +return -1;
> +
> if (virDomainDefMaybeAddVirtioSerialController(def) < 0)
> return -1;
>
> diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
> index 8e20e3f..7122756 100644
> --- a/src/qemu/qemu_capabilities.c
> +++ b/src/qemu/qemu_capabilities.c
> @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
>"no-shutdown",
>
>"cache-unsafe", /* 75 */
> + "ich9-ahci",
> );
>
> struct qemu_feature_flags {
> @@ -1241,6 +1242,8 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr
> flags)
> qemuCapsSet(flags, QEMU_CAPS_USB_REDIR);
> if (strstr(str, "name \"usb-hub\""))
> qemuCapsSet(flags, QEMU_CAPS_USB_HUB);
> +if (strstr(str, "name \"
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
On Mon, Oct 03, 2011 at 10:46:18PM -0600, Jim Fehlig wrote: > Adam Litke wrote: > > Hi Jim. I was testing this and found that I could not boot from the sata > > disks > > I defined. When I switch them back to ide, they can be booted just fine. > > Perhaps something is missing from the boot order logic? > > > > Hmm, I didn't notice this in my testing. sda in the below config > contained /boot. Can you provide the domain config? With the following config, I end up booting from the cdrom. If I take away the cdrom definition completely, I am unable to boot at all. blockPull-test 262144 262144 1 hvm destroy restart restart /home/aglitke/src/qemu/x86_64-softmmu/qemu-system-x86_64 -- Adam Litke IBM Linux Technology Center -- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
Adam Litke wrote:
> Hi Jim. I was testing this and found that I could not boot from the sata
> disks
> I defined. When I switch them back to ide, they can be booted just fine.
> Perhaps something is missing from the boot order logic?
>
Hmm, I didn't notice this in my testing. sda in the below config
contained /boot. Can you provide the domain config?
Thanks for taking a look!
Jim
> On Wed, Sep 28, 2011 at 04:43:25PM -0600, Jim Fehlig wrote:
>
>> Tested with multiple AHCI controllers and multiple disks attached
>> to a controller. E.g.,
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> > function='0x0'/>
>>
>>
>> > function='0x0'/>
>>
>> ---
>> docs/formatdomain.html.in |9 +++--
>> docs/schemas/domaincommon.rng |1 +
>> src/conf/domain_conf.c | 14 +
>> src/qemu/qemu_capabilities.c |3 ++
>> src/qemu/qemu_capabilities.h |1 +
>> src/qemu/qemu_command.c| 30
>>
>> .../qemuxml2argv-disk-sata-device.args |6
>> .../qemuxml2argv-disk-sata-device.xml | 25
>> tests/qemuxml2argvtest.c |3 ++
>> 9 files changed, 82 insertions(+), 10 deletions(-)
>>
>> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in
>> index 3087d01..b6a0c66 100644
>> --- a/docs/formatdomain.html.in
>> +++ b/docs/formatdomain.html.in
>> @@ -972,11 +972,12 @@
>> as a device ordering hint. The optional bus
>> attribute specifies the type of disk device to emulate;
>> possible values are driver specific, with typical values being
>> -"ide", "scsi", "virtio", "xen" or "usb". If omitted, the bus type is
>> -inferred from the style of the device name. eg, a device named 'sda'
>> -will typically be exported using a SCSI bus.
>> +"ide", "scsi", "virtio", "xen", "usb" or "sata". If omitted, the bus
>> +type is inferred from the style of the device name. eg, a device
>> named
>> +'sda' will typically be exported using a SCSI bus.
>> Since 0.0.3; bus attribute since
>> 0.4.3;
>> -"usb" attribute value since after 0.4.4
>> +"usb" attribute value since after 0.4.4; "sata" attribute value
>> since
>> +0.9.7
>>driver
>>
>> The optional driver element allows specifying further details
>> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng
>> index be98be0..675d55d 100644
>> --- a/docs/schemas/domaincommon.rng
>> +++ b/docs/schemas/domaincommon.rng
>> @@ -790,6 +790,7 @@
>> xen
>> usb
>> uml
>> +sata
>>
>>
>>
>> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
>> index a918679..6a7f296 100644
>> --- a/src/conf/domain_conf.c
>> +++ b/src/conf/domain_conf.c
>> @@ -2157,6 +2157,15 @@ virDomainDiskDefAssignAddress(virCapsPtr caps,
>> virDomainDiskDefPtr def)
>> def->info.addr.drive.unit = (idx % 2);
>> break;
>>
>> +case VIR_DOMAIN_DISK_BUS_SATA:
>> +/* For SATA we define the default mapping to be 6 units
>> + * per bus, 1 bus per controller, many controllers */
>> +def->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_DRIVE;
>> +def->info.addr.drive.controller = idx / 6;
>> +def->info.addr.drive.bus = 0;
>> +def->info.addr.drive.unit = idx % 6;
>> +break;
>> +
>> case VIR_DOMAIN_DISK_BUS_FDC:
>> /* For FDC we define the default mapping to be 2 units
>> * per bus, 1 bus per controller, many controllers */
>> @@ -8675,6 +8684,11 @@ int
>> virDomainDefAddImplicitControllers(virDomainDefPtr def)
>>VIR_DOMAIN_DISK_BUS_IDE) < 0)
>> return -1;
>>
>> +if (virDomainDefAddDiskControllersForType(def,
>> +
>> VIR_DOMAIN_CONTROLLER_TYPE_SATA,
>> + VIR_DOMAIN_DISK_BUS_SATA) < 0)
>> +return -1;
>> +
>> if (virDomainDefMaybeAddVirtioSerialController(def) < 0)
>> return -1;
>>
>> diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
>> index 8e20e3f..7122756 100644
>> --- a/src/qemu/qemu_capabilities.c
>> +++ b/src/qemu/qemu_capabilities.c
>> @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
>>"no-shutdown",
>>
>>"cache-unsafe", /* 75 */
>> + "ich9-ahci",
>> );
>>
>> struct qemu_feature_flags {
>> @@ -1241,6 +1242,8 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr
>> flags)
>> qemuCapsSet(flags, QEMU_CAPS_USB_R
Re: [libvirt] [PATCH] Add AHCI support to qemu driver
Hi Jim. I was testing this and found that I could not boot from the sata disks
I defined. When I switch them back to ide, they can be booted just fine.
Perhaps something is missing from the boot order logic?
On Wed, Sep 28, 2011 at 04:43:25PM -0600, Jim Fehlig wrote:
> Tested with multiple AHCI controllers and multiple disks attached
> to a controller. E.g.,
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>function='0x0'/>
>
>
>function='0x0'/>
>
> ---
> docs/formatdomain.html.in |9 +++--
> docs/schemas/domaincommon.rng |1 +
> src/conf/domain_conf.c | 14 +
> src/qemu/qemu_capabilities.c |3 ++
> src/qemu/qemu_capabilities.h |1 +
> src/qemu/qemu_command.c| 30
>
> .../qemuxml2argv-disk-sata-device.args |6
> .../qemuxml2argv-disk-sata-device.xml | 25
> tests/qemuxml2argvtest.c |3 ++
> 9 files changed, 82 insertions(+), 10 deletions(-)
>
> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in
> index 3087d01..b6a0c66 100644
> --- a/docs/formatdomain.html.in
> +++ b/docs/formatdomain.html.in
> @@ -972,11 +972,12 @@
> as a device ordering hint. The optional bus
> attribute specifies the type of disk device to emulate;
> possible values are driver specific, with typical values being
> -"ide", "scsi", "virtio", "xen" or "usb". If omitted, the bus type is
> -inferred from the style of the device name. eg, a device named 'sda'
> -will typically be exported using a SCSI bus.
> +"ide", "scsi", "virtio", "xen", "usb" or "sata". If omitted, the bus
> +type is inferred from the style of the device name. eg, a device
> named
> +'sda' will typically be exported using a SCSI bus.
> Since 0.0.3; bus attribute since
> 0.4.3;
> -"usb" attribute value since after 0.4.4
> +"usb" attribute value since after 0.4.4; "sata" attribute value since
> +0.9.7
>driver
>
> The optional driver element allows specifying further details
> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng
> index be98be0..675d55d 100644
> --- a/docs/schemas/domaincommon.rng
> +++ b/docs/schemas/domaincommon.rng
> @@ -790,6 +790,7 @@
> xen
> usb
> uml
> +sata
>
>
>
> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
> index a918679..6a7f296 100644
> --- a/src/conf/domain_conf.c
> +++ b/src/conf/domain_conf.c
> @@ -2157,6 +2157,15 @@ virDomainDiskDefAssignAddress(virCapsPtr caps,
> virDomainDiskDefPtr def)
> def->info.addr.drive.unit = (idx % 2);
> break;
>
> +case VIR_DOMAIN_DISK_BUS_SATA:
> +/* For SATA we define the default mapping to be 6 units
> + * per bus, 1 bus per controller, many controllers */
> +def->info.type = VIR_DOMAIN_DEVICE_ADDRESS_TYPE_DRIVE;
> +def->info.addr.drive.controller = idx / 6;
> +def->info.addr.drive.bus = 0;
> +def->info.addr.drive.unit = idx % 6;
> +break;
> +
> case VIR_DOMAIN_DISK_BUS_FDC:
> /* For FDC we define the default mapping to be 2 units
> * per bus, 1 bus per controller, many controllers */
> @@ -8675,6 +8684,11 @@ int virDomainDefAddImplicitControllers(virDomainDefPtr
> def)
>VIR_DOMAIN_DISK_BUS_IDE) < 0)
> return -1;
>
> +if (virDomainDefAddDiskControllersForType(def,
> +
> VIR_DOMAIN_CONTROLLER_TYPE_SATA,
> + VIR_DOMAIN_DISK_BUS_SATA) < 0)
> +return -1;
> +
> if (virDomainDefMaybeAddVirtioSerialController(def) < 0)
> return -1;
>
> diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
> index 8e20e3f..7122756 100644
> --- a/src/qemu/qemu_capabilities.c
> +++ b/src/qemu/qemu_capabilities.c
> @@ -139,6 +139,7 @@ VIR_ENUM_IMPL(qemuCaps, QEMU_CAPS_LAST,
>"no-shutdown",
>
>"cache-unsafe", /* 75 */
> + "ich9-ahci",
> );
>
> struct qemu_feature_flags {
> @@ -1241,6 +1242,8 @@ qemuCapsParseDeviceStr(const char *str, virBitmapPtr
> flags)
> qemuCapsSet(flags, QEMU_CAPS_USB_REDIR);
> if (strstr(str, "name \"usb-hub\""))
> qemuCapsSet(flags, QEMU_CAPS_USB_HUB);
> +if (strstr(str, "name \"ich9-ahci\""))
> +qemuCapsSet(flags, QEMU_CAPS_ICH9_AHCI);
>
> /* Prefer -chardev spicevmc (detected earlier) over -device spicevmc */
> if (!qemuCapsGet(flags, QE
