Qemu v1.7.0-rc0 features an ACPI linker/loader interface, available over
fw_cfg, written by Michael Tsirkin.
Qemu composes all ACPI tables on the host side, according to the target
hardware configuration, and makes the tables available to any guest
firmware over fw_cfg.
The feature moves the burd
QemuFwCfgLib is a BASE type library module. It should be available to all
UEFI phases. We made it depend on UefiBootServicesTableLib, even though we
don't use boot services in the code.
This dependency prevents its usage in PEIM module types, the build system
catches it. Remove the dependency so t
v1->v2:
- addressed Jordan's comments in v2 4/4 (patch #1 in v1)
- fixed the 32-bit PCI hole discrepancy in PEI after realizing that qemu
does export the window for us in easily consumable form
Retested with Fedora and Windows 2012 R2 guests, RAM sizes 1G/2.5G/4G.
Laszlo Ersek (4):
OvmfPkg/Qe
Introduce this dynamic PCD to communicate presence *and validity* of the
"etc/pci-info" fw_cfg file, from PEI to DXE. We can always check the
presence of any fw_cfg file, but re-checking validity later is wasteful to
run and to code up -- we'd have to share code between PEI and DXE, which
is overki
Upstream qemu commits
commit f8c457b88d72a48989f190bc3d7b79f4f3b7d11c
Author: Michael S. Tsirkin
Date: Mon May 13 20:00:23 2013 +0300
pc: pass PCI hole ranges to Guests
Guest currently has to jump through lots of hoops to guess the PCI
hole ranges. It's fragile, and m
On 11/26/13 07:44, Gary Ching-Pang Lin wrote:
> On Tue, Nov 26, 2013 at 06:21:21AM +0100, Laszlo Ersek wrote:
>> Hi,
>>
>> as soon as Timer.efi is installed and the first timer interrupt is
>> delivered, OVMF hangs with the VCPU spinning at 100%.
>>
>> I can see that SVN r14887 is the commit that d
On Tue, Nov 26, 2013 at 06:21:21AM +0100, Laszlo Ersek wrote:
> Hi,
>
> as soon as Timer.efi is installed and the first timer interrupt is
> delivered, OVMF hangs with the VCPU spinning at 100%.
>
> I can see that SVN r14887 is the commit that directly switches the
> libraries that OvmfPkg uses.
Hi Ray,
On 11/26/13 06:48, Ni, Ruiyu wrote:
> Laszlo,
> Which OVMF image did you use? I successfully booted the one built from
> OvmfPkgIa32X64.dsc in Windows OS using QEMU 0.13.0.
I'm using a recent qemu git master (after v1.7.0-rc0), X64 build of
OVMF, and fresh edk2 master. Host Linux kernel
Laszlo,
Which OVMF image did you use? I successfully booted the one built from
OvmfPkgIa32X64.dsc in Windows OS using QEMU 0.13.0.
Thanks,
Ray
-Original Message-
From: Laszlo Ersek [mailto:ler...@redhat.com]
Sent: Tuesday, November 26, 2013 1:21 PM
To: edk2-devel@lists.sourceforge.net;
Hi,
as soon as Timer.efi is installed and the first timer interrupt is
delivered, OVMF hangs with the VCPU spinning at 100%.
I can see that SVN r14887 is the commit that directly switches the
libraries that OvmfPkg uses.
However, at that point "MdeModulePkg/Library/CpuExceptionHandlerLibNull"
(o
Wei,
The PCD declaration in the DEC file declares the PCD type the PCD is allowed to
have. For this specific PCD, you may want to allow it to be FixedAtBuild or
PatahcbleInModule or Dynamic or DynamicEx. This provides the maximum
flexibility.
For future DUET use cases, the DSC file could set
Mike,
Is this something to consider for MdePkg/Include/IndustryStandard?
-Jordan
On Tue, Nov 19, 2013 at 12:38 PM, Wei Liu wrote:
> E820 definitions copied from IntelFrameworkModulePkg/Csm/
> LegacyBiosDxe/LegacyBiosInterface.h.
>
> Signed-off-by: Wei Liu
> ---
> OvmfPkg/Include/IndustryStand
Regarding patches 5-7, it seems like the mem-map code flow could be
more shared. It is a bit challenging to unravel things though.
I guess the only specific thing I can really point out is that
PcdPciAllowFullEnumeration should be initialized in a different patch,
and not within the mem-map init p
On 11/25/13 16:32, Markus Armbruster wrote:
> Laszlo Ersek writes:
>
>> This patch allows the user to usefully specify
>>
>> -drive file=img_1,if=pflash,format=raw,readonly \
>> -drive file=img_2,if=pflash,format=raw
>>
>> on the command line. The flash images will be mapped under 4G in their
On Mon, Nov 25, 2013 at 3:41 AM, Wei Liu wrote:
> On Sun, Nov 24, 2013 at 05:46:42PM -0800, Jordan Justen wrote:
>> On Tue, Nov 19, 2013 at 12:38 PM, Wei Liu wrote:
>> > + *
>> > + * Each element is a 32 bit value contianing the physical address
>> > + * of a BIOS table.
>> > + */
>> > +
On 11/25/13 16:22, Markus Armbruster wrote:
> Laszlo Ersek writes:
>
>> On 11/22/13 13:21, Markus Armbruster wrote:
>>> Laszlo Ersek writes:
>>>
This patch allows the user to usefully specify
-drive file=img_1,if=pflash,format=raw,readonly \
-drive file=img_2,if=pflash,fo
UNSUBSCRIBE
-Original Message-
From: Wei Liu [mailto:wei.l...@citrix.com]
Sent: Monday, November 25, 2013 02:56
To: Kinney, Michael D
Subject: Re: [edk2] [PATCH RFC v2 1/7] MdeModulePkg: introduce
PcdPciAllowFullEnumeration ... [and one more message]
On Mon, Nov 25, 2013 at 03:12:04AM +
Following the UEFI spec 2.4, SMBIOS data should be set as Runtime
Services data type(previously Reserved type), or dmidecode on arm64
will not work properly.
Signed-off-by: Yi Li
Contributed-under: TianoCore Contribution Agreement 1.0
---
In Charter 2.3.2 spec 2.4, it descripts SMBIOS tables sho
On Sun, Nov 24, 2013 at 06:01:03PM -0800, Jordan Justen wrote:
[...]
> >mXenInfo.HyperPages = AllocatePages (TransferPages);
> > @@ -72,6 +78,31 @@ XenConnect (
> >/* TBD: Locate hvm_info and reserve it away. */
> >mXenInfo.HvmInfo = NULL;
> >
> > + if (!AsciiStrCmp ((CHAR8 *) Info->Si
On Sun, Nov 24, 2013 at 05:46:42PM -0800, Jordan Justen wrote:
> On Tue, Nov 19, 2013 at 12:38 PM, Wei Liu wrote:
> > EFI_XEN_OVMF_INFO is defined to accept configurations from hvmloader. It
> > must match the definition on Xen side.
> >
> > XenInfo is extended to include those bits as well. Curre
On Sun, Nov 24, 2013 at 02:05:01PM -0800, Jordan Justen wrote:
> Feng, what do you think of this change to MdeModulePkg?
>
> Wei, How about PcdPciDisableBusEnumeration instead?
>
I have no objection to the new name.
Wei.
> -Jordan
--
On Mon, Nov 25, 2013 at 03:12:04AM +, Kinney, Michael D wrote:
> Jordan,
>
> I agree that removing code duplication is a good idea.
>
> I believe we can make the one in the MdeModulePkg functional
> everywhere. Size will be the only remaining difference. This can be
> addressed longer term.
22 matches
Mail list logo
