[libvirt] [PATCH 4/4] qemu: Add support for setting the TSEG size

[Martin Kletzander]

On Thu, Jun 07, 2018 at 09:58:04PM +0200, Ján Tomko wrote:
>On Thu, Jun 07, 2018 at 10:37:43AM +0200, Martin Kletzander wrote:
>>The default is stable per machine type so there should be no need to keep that.
>>
>>Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1469338
>>
>>Signed-off-by: Martin Kletzander <mkletzan at redhat.com>
>>---
>> src/conf/domain_conf.c                        |  3 +-
>> src/qemu/qemu_command.c                       | 18 ++++++++
>> src/qemu/qemu_domain.c                        | 35 ++++++++++++++
>> .../tseg-explicit-size.x86_64-latest.args     | 35 ++++++++++++++
>> tests/qemuxml2argvdata/tseg-explicit-size.xml | 23 ++++++++++
>> tests/qemuxml2argvdata/tseg-i440fx.xml        | 23 ++++++++++
>> tests/qemuxml2argvdata/tseg-invalid-size.xml  | 23 ++++++++++
>> tests/qemuxml2argvtest.c                      | 25 ++++++++++
>> .../qemuxml2xmloutdata/tseg-explicit-size.xml | 46 +++++++++++++++++++
>> .../tseg-old-machine-type.xml                 | 44 ++++++++++++++++++
>> tests/qemuxml2xmloutdata/tseg.xml             | 44 ++++++++++++++++++
>> tests/qemuxml2xmltest.c                       |  9 ++++
>> 12 files changed, 327 insertions(+), 1 deletion(-)
>> create mode 100644 tests/qemuxml2argvdata/tseg-explicit-size.x86_64-latest.args
>> create mode 100644 tests/qemuxml2argvdata/tseg-explicit-size.xml
>> create mode 100644 tests/qemuxml2argvdata/tseg-i440fx.xml
>> create mode 100644 tests/qemuxml2argvdata/tseg-invalid-size.xml
>> create mode 100644 tests/qemuxml2xmloutdata/tseg-explicit-size.xml
>> create mode 100644 tests/qemuxml2xmloutdata/tseg-old-machine-type.xml
>> create mode 100644 tests/qemuxml2xmloutdata/tseg.xml
>>
>>diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
>>index 62bf6bb803bb..e83487d6b0de 100644
>>--- a/src/conf/domain_conf.c
>>+++ b/src/conf/domain_conf.c
>>@@ -22043,7 +22043,8 @@ virDomainDefFeaturesCheckABIStability(virDomainDefPtr src,
>>             return false;
>>         }
>>
>>-        if (src->tseg_size != dst->tseg_size) {
>>+        if (src->tseg_specified &&
>
>Why this change?
>
>IIUC if they weren't specified on both sides, they should both be 0
>here.
>
>If you're sure it's needed, put it in the commit adding this check.
>

Oh, it is very easy to explain.  The first line (the one being removed) is from
the previous version of this series.  The second one (which is being added) is
from version two in which case I used fixup during interactive rebase on a wrong
commit.  Otherwise this would be part of the commit in which this was added.

And as said in the other reply to Laszlo's e-mail (which I had to separate due
to this par tof review being omitted there) it makes sense for the tseg size to
be 0 as that means the extended size is disabled.  I agree that it should be
part of the documentation, though.

@Laszlo: When thinking about it, even though it is a very stupid idea,
technically there isn't really a difference between 'extended-tseg-mbytes=0' and
'extended-tseg-mbytes=8' (of course the second option will still provide a
fw_cfg info about the size etc., but the guest will behave the same way), right?

That brings me to another question (maybe even more stupid), what size does OVMF
choose if I specify 'extended-tseg-mbytes=6'?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: Digital signature
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20180607/a2b6eb1d/attachment-0001.sig>