On 04/04/2013 05:44 PM, Tim Lewis wrote:
> My point was: wouldn't it be easier to ship with an ACPI table that is not
> AML. For example, like IBST for iSCSI, where a table has a name and the data
> portion is just the start and size. As a point of reference, please see
> Appendix O of the UEFI
My point was: wouldn't it be easier to ship with an ACPI table that is not AML.
For example, like IBST for iSCSI, where a table has a name and the data portion
is just the start and size. As a point of reference, please see Appendix O of
the UEFI specification, where a GUID can be used to creat
On 04/04/2013 05:03 PM, Andrew Fish wrote:
>
> I think the tricky part is telling the OS that this memory range has special
> properties. You want to treat it like EFIACPIReclaimMemory, as the OS could
> convert the range to pageable memory. If you make the memory
> EfiRuntimeServicesData then
On 04/04/2013 04:54 PM, Tim Lewis wrote:
> Or an EFI System Configuration table (which uses a GUID to identify it) Not
> sure about SSDT, since that would require an AML interpreter.
>
> Tim
The OS has an AML interpreter already. The SSDT can be pre-generated
with only the address and length
On Apr 4, 2013, at 4:54 PM, Tim Lewis wrote:
> Or an EFI System Configuration table (which uses a GUID to identify it) Not
> sure about SSDT, since that would require an AML interpreter.
>
I think the tricky part is telling the OS that this memory range has special
properties. You want to
Or an EFI System Configuration table (which uses a GUID to identify it) Not
sure about SSDT, since that would require an AML interpreter.
Tim
-Original Message-
From: H. Peter Anvin [mailto:h...@zytor.com]
Sent: Thursday, April 04, 2013 4:50 PM
To: edk2-devel@lists.sourceforge.net
C
Thanks, Jordan! I will look at your code tomorrow morning; it will
probably help me out a lot.
On Thu, Apr 4, 2013 at 6:32 PM, Jordan Justen wrote:
> On Thu, Apr 4, 2013 at 4:15 PM, Andrew Fish wrote:
> > On Apr 4, 2013, at 3:50 PM, Thomas Rognon wrote:
> >
> > I found this on the internet,
Thanks! I understand how it works now. This is the first time I've needed
to use the BLOCK_IO protocol.
I'll tackle the ram disk driver after I finish the code to get the ISO
mounted and booted. I have no clue where to begin, but I always figure
things out eventually!
On Thu, Apr 4, 2013 at 6
On 04/04/2013 12:27 PM, Tim Lewis wrote:
> Produce an instance of BLOCK_IO protocol which returns pieces of the
> in-memory image. Then perform a ConnectController() on that handle, which
> should cause a DISK_IO and SIMPLE_FILE_SYSTEM to be mounted automatically.
>
> Then, you use the file syst
On Thu, Apr 4, 2013 at 4:15 PM, Andrew Fish wrote:
> On Apr 4, 2013, at 3:50 PM, Thomas Rognon wrote:
>
> I found this on the internet,
> The UEFI adaptation of ElTorito results in mapping image offset 14800-A27FF
> as a FAT file system, and executing file EFI/BOOT/BOOTX64.efi in the 64-bit
> UEF
On Apr 4, 2013, at 3:50 PM, Thomas Rognon wrote:
> I found this on the internet,
> The UEFI adaptation of ElTorito results in mapping image offset 14800-A27FF
> as a FAT file system, and executing file EFI/BOOT/BOOTX64.efi in the 64-bit
> UEFI environment.
> (link: http://www.seabios.org/pipe
I found this on the internet,
The UEFI adaptation of ElTorito results in mapping image offset 14800-A27FF
as a FAT file system, and executing file EFI/BOOT/BOOTX64.efi in the 64-bit
UEFI environment.
(link: http://www.seabios.org/pipermail/seabios/2012-June/003937.html)
So here are the steps, if I
Duane,
There were some updates to the OVMF ACPI based TimerLib back in September and
October 2012 to address some hang conditions in SEC phase.
Are you using the latest qemu emulator and OVMF sources? You might try
verifying using the qemu monitor that the PMBA and PMREGMISC registers have
be
Thanks!
So it appears that my OVMF boot in qemu is hanging in a loop that
includes InternalAcpiDelay(). Is there a way to build OVMF that changes
the way ACPI timers are handled?
--
Minimize network downtime and maximize
So remember El Torito != ISO-9660, so the image will look different in the OS
vs. the EFI. The file system that EFI sees will show up as a file in the OS,
and EFI will not see all the ISO-9660 files you can see in the OS.
Also remember that the OS Installer is going to call ExitBootServices() a
Oh ok, that's good news regarding El Torito. I'll start wrapping my head
around the code you mentioned. The ISOs I will be booting are OS
installers. Specifically, I'm trying to boot a legit Windows 7 installer
from memory right now.
On Thu, Apr 4, 2013 at 3:41 PM, Tim Lewis wrote:
> The pa
The partition driver already exists (see
MdeModulePkg/Universal/Disk/PartitionDxe). This already recognizes El Torito
format, so, as long as the partition you boot from is FAT32 you should have no
trouble (see the FatBinPkg).
If you are trying to boot an OS, then you have to figure out how to
Thanks! That gives me a great start. I'm new to a lot of this stuff,
though, so the steps you outlined will be tough for me without guidance.
Is there any UEFI optical drive code floating around the internet
somewhere? I looked but couldn't find any. I'm essential doing the same
thing as booti
Produce an instance of BLOCK_IO protocol which returns pieces of the in-memory
image. Then perform a ConnectController() on that handle, which should cause a
DISK_IO and SIMPLE_FILE_SYSTEM to be mounted automatically.
Then, you use the file system to load the boot loader (LoadImage()), signal th
Let's say I'm running a UEFI_APPLICATION called IsoBootTest.efi and I have
an ISO image in volatile memory and I have the start address and size of
the ISO image. Can I boot the ISO image? If so, can someone give me a
hint on how I would do it? (If it makes any difference, I'm using UDK2010.)
T
No, at the moment I don't have a patch for this issue.
Thanks,
Stefan
-Ursprüngliche Nachricht-
Von: Carsey, Jaben [mailto:jaben.car...@intel.com]
Gesendet: Donnerstag, 4. April 2013 17:14
An: edk2-devel@lists.sourceforge.net
Betreff: Re: [edk2] ShellPrintEx and ShellPrintHiiEx formating
This is the first report of this. I do not know what the priority of this will
be given.
Do you want to submit a patch file to fix the error?
Thanks,
-Jaben
-Original Message-
From: Mutschler Stefan [mailto:s.mutsch...@ads-tec.de]
Sent: Thursday, April 04, 2013 8:10 AM
To: edk2-devel@
Are there any plans to fix this issue in future?
Thanks,
Stefan
-Ursprüngliche Nachricht-
Von: Carsey, Jaben [mailto:jaben.car...@intel.com]
Gesendet: Donnerstag, 4. April 2013 16:32
An: edk2-devel@lists.sourceforge.net
Betreff: Re: [edk2] ShellPrintEx and ShellPrintHiiEx formating flags
That is a side effect.
-Jaben
-Original Message-
From: Mutschler Stefan [mailto:s.mutsch...@ads-tec.de]
Sent: Thursday, April 04, 2013 6:54 AM
To: edk2-devel@lists.sourceforge.net
Subject: [edk2] ShellPrintEx and ShellPrintHiiEx formating flags
Hi,
if I type something like the followin
Hi,
if I type something like the following into the shell:
set BLUE something
echo %BLUE%
it's defining a variable and prints out the variable. If I hit the
up-key to see the command history the second string gets formatted. It
is blue after the blank and only the string "echo LUE%" is visible
On 03/07/13 11:07, Gary Ching-Pang Lin wrote:
> On Thu, Mar 07, 2013 at 09:17:13AM +, Fu, Siyuan wrote:
>> Hi, Lin
>>
>> R14141 is the only patch for multiple signature support. R14165
>> fixes a bug when computing the signature count of an
>> EFI_SIGANTURE_LIST in db/dbx. Please notice that t
26 matches
Mail list logo
