> On Oct 29, 2014, at 10:08 PM, Tian, Feng wrote:
>
> Thanks for your great contribution, Paulo.
>
> For the device path issue, my personal idea is vendor device path may not be
> a good idea and I prefer to introduce a new UDF device path node. But let's
> us wait other's comments at first.
Thanks for your great contribution, Paulo.
For the device path issue, my personal idea is vendor device path may not be a
good idea and I prefer to introduce a new UDF device path node. But let's us
wait other's comments at first.
Thanks
Feng
-Original Message-
From: Paulo Alcantara [m
IA32 Peim module cannot provide x64bit C interrupt handler. That’s why
DxeIplPeim does not support this dump feature.
However, DXE Core will link CPU Exception Handler Lib(in UefiCpuPkg) to support
dump registers from beginning of long mode.
Please search InitializeCpuExceptionHandlers() in DxeM
Hi, Jeff:
MdeModulePkg\Core\DxeIplPeim\Ia32\DxeLoadFunc.c will create IDT before handoff
to Dxe.
The first 32 IDT entries’ interrupt handler is very simple(ref to
Ia32\IdtVectorAsm.asm):
cli
jmp $
So, why not hook a dump register function for these 32 reserved IDT entries?
Such as :
The original IDT entries from 0 to OldIdtSize still work to capture all
exceptions.
And, it also does not impact CpuArch->CpuRegisterInterrupt() to register new
Interrupt Handlers from vector 0 to vector 0xff.
From: tiger...@via-alliance.com [mailto:tiger...@via-alliance.com]
Sent: Thursday, Oc
Hi, Andrew:
Thanks for your reply!
So, I made a conclusion:
If OldIdtSize > = 32, even below code not run, it’s aslo ok?
//
for (Index = OldIdtSize; Index < 32; Index++) {
Status = CpuRegisterInterruptHandler (&gCpu, Index, CommonExceptionHandler);
ASSERT_EFI_ERROR (Status);
Correct. The old CpuDxe(one year ago) has such code to protect IDT entries used
by soft debugger.
Now the UefiCpuPkg/CpuDxe has already been updated to support Method to Reserve
Interrupt and Exception Vectors (Introduced from PI spec 1.2.1). And removed
such code logic.
Thanks!
Jeff
From: And
This patch adds support to finding UDF/ECMA-167 volumes on disks by looking at
the UDF Standard Identifiers (BEA01, NSR02/NSR03, TEA01) and the Anchor Volume
Descriptor Pointer (AVDP) at LBAs 256, N - 256 or N.
A new PcdsFeatureFlags (PcdUdfFileSystemSupport) has been added to
enable/disable such
> On Oct 29, 2014, at 7:19 PM, tiger...@via-alliance.com wrote:
>
> Hi, experts:
> I am confused by some codes in InitInterruptDescriptorTable() function (in
> UefiCpuPkg/CpuDxe/CpuDxe.c).
> ……
> if ((OldIdtPtr.Base != 0) && ((OldIdtPtr.Limit & 7) == 7)) {
> // Based on my understanding, becau
Hi, experts:
I am confused by some codes in InitInterruptDescriptorTable() function
(in UefiCpuPkg/CpuDxe/CpuDxe.c).
..
if ((OldIdtPtr.Base != 0) && ((OldIdtPtr.Limit & 7) == 7)) {
// Based on my understanding, because DxeIpl has created IDT when
switching from P-Mode to Long-Mode.
// So
Hi, Euqune
The original intention of r16104 is to use logic bar index from PciIo protocol
instead of physical offset in PCI configuration header, but it incorrectly add
a new restriction about the existence of the memory bar and IO bar. Here is a
patch to remove this check, please help to check
Eugene
Thanks for raising this issue. We are working on this and will get back to you
if we have progress.
Thanks
Feng
-Original Message-
From: Cohen, Eugene [mailto:eug...@hp.com]
Sent: Wednesday, October 29, 2014 22:25
To: edk2-devel@lists.sourceforge.net
Subject: [edk2] SnpDxe BAR c
Olivier,
I have some questions about your patch submission. The majority are inserted,
inline, into your quoted message below.
Any patch marked ' Accepted.' has been committed as Reviewed by: Daryl
McDaniel .
The remainder require further discussion. All of my comments are prefixed with
.
Bruce Cran [mailto:bruce.c...@gmail.com] wrote:
]On Tue, Oct 28, 2014 at 10:59 PM, Scott Duplichan wrote:
]> Optimization is not enabled for x64 builds using gcc 4.4-4.9. For IA32
]> builds, -Os (optimize for small code size) is used. Why is this? Apparently
]> it is because variable argument lis
Huh!
So how do I implement non-volatile variables ?
Is there a variable service that uses flash device to store non-volatile
variables ?
Thanx.
On Tue, Oct 28, 2014 at 7:25 PM, Gao, Liming wrote:
> DxeCore will install FV protocol for PI format FV FFS image. Variable
> storage is not FFS format
SCT repository is a UEFI member only repository (any UEFI member can request to
get access - you only need to send an email to eric@intel.com).
Work on this repository cannot be made public (current UEFI Forum restriction).
So patches cannot be advertised on public mailing-list and I could no
Tim Lewis [mailto:tim.le...@insyde.com] wrote:
]Scott --
]
]For historical perspective, the EDK2 build flags have focused on space over
speed because of the code size constraints placed on ]flash-resident code. Not
being as familiar with gcc as I am with VS20xx, I don't know whether these can
b
> On Oct 29, 2014, at 11:45 AM, Olivier Martin wrote:
>
> I have just noticed I have replied one of your question.
>
> To answer your question on "how to restrict to use only first memory range".
> I know one way to do it, but I am not sure it is the recommended way...
>
> If you do not want t
That's how it's done on NUMA platforms I've seen. Only low memory is
marked as EFI_RESOURCE_ATTRIBUTE_TESTED at first. Then in the BDS
phase, after all DXE modules have loaded, a memory test is run which
transitions the remaining memory to "EFI_RESOURCE_ATTRIBUTE_PRESENT |
EFI_RESOURCE_ATTRI
I have just noticed I have replied one of your question.
To answer your question on "how to restrict to use only first memory range".
I know one way to do it, but I am not sure it is the recommended way...
If you do not want to run UEFI into the upper memory space. You can declare
this memory ran
On 10/29/2014 10:18 AM, Tim Lewis wrote:
> For historical perspective, the EDK2 build flags have focused on space over
> speed because of the code size constraints placed on flash-resident code. Not
> being as familiar with gcc as I am with VS20xx, I don't know whether these
> can be set togeth
Scott --
For historical perspective, the EDK2 build flags have focused on space over
speed because of the code size constraints placed on flash-resident code. Not
being as familiar with gcc as I am with VS20xx, I don't know whether these can
be set together.
Tim
-Original Message-
Fro
Rafael –
Create a device path using the port/multiplier port information (see
BuildDevicePath), then use LocateDevicePath.
From: Rafael Machado [mailto:rafaelrodrigues.mach...@gmail.com]
Sent: Wednesday, October 29, 2014 3:32 AM
To: edk2-devel
Subject: Re: [edk2] Get BlockIo protocol from Ata De
On Tue, Oct 28, 2014 at 10:59 PM, Scott Duplichan wrote:
> Optimization is not enabled for x64 builds using gcc 4.4-4.9. For IA32
> builds, -Os (optimize for small code size) is used. Why is this? Apparently
> it is because variable argument list handling fails when gcc X64 optimization
> is enabl
I forgot to add to the commit message but some of the fix should prevent
some errors too. Example:
Status = pSocketProtocol->pfnListen ( pSocketProtocol,
backlog,
&errno );
+if (EFI_ERROR (Status)) {
+
Olivier Martin [mailto:olivier.mar...@arm.com] wrote:
]Some of the variables are unused that trigger a GCC warning/error.
]
]Contributed-under: TianoCore Contribution Agreement 1.0
]Signed-off-by: Olivier Martin
Thanks Oliver. I ran this through some Visual Studio build
testing and it is OK th
Dear MdeModulePkg maintainer,
On September 14 SVN rev 16104 there was a change to SnpDxe to improve the
detection of BARs to replace the hardcoded MemoryBarIndex=0 and IoBarIndex=1
with some intelligent scanning using PCI_IO GetBarAttributes instead.
This change had the side effect of breaking
Hi Feng
Thanks for the answer.
Just to explain better my scenario.
What I need to do is execute some ata commands, like read verify sector,
this is done sending ata commands to some Port / Multiplier port, but in
case this device has a Block Io protocol, I need to use this protocol
instead of sen
Some of the variables are unused that trigger a GCC warning/error.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Olivier Martin
---
StdLib/BsdSocketLib/bind.c| 3 +++
StdLib/BsdSocketLib/getnameinfo.c | 8 ++--
StdLib/BsdSocketLib/ge
Hi, experts:
Based on PI Spec 1.3, there is EFI_HEALTH_FLAGS struct definition.
And it is info about BPS's BIST info.
So, my question is:
1. Based on IA-32 Architectures software developer manual, when CPU
is powered on, it will run BIST automatically.
So, how to retrieve BIST result i
30 matches
Mail list logo
