In the R_AARCH64_ADR_GOT_PAGE case on AARCH64, we may encounter an ADR
instead of an ADRP when the toolchain is working around Cortex-A53
erratum #843419. If that's the case, be sure to calculate the offset
appropriately.
This resolves an issue experienced when building a StandaloneMm image
with
Ard, Pedro,
It's all connecting for me:
* I was able to recreate this problem with a crosstool-ng built toolchain.
That confirms it's not specific to Ubuntu and is indeed related to
"--enable-fix-cortex-a53-843419".
* I also verified that this issue is specific to our
Fantastic!
When we hit this in GCC 12.3 toolchain, the ADR was indeed at a 0xffc page
offset. So, that connects.
This also explains why the issue seemed to be specific to stack protection:
Because it's comparing values right away. If we hit this with other loads, we
might not notice until
Totally understand and agree, Ard.
In the meantime, I've now experienced the issue with Ubuntu22's GCC 12.3.
Originally, we didn't see the issue on this toolchain, but a developer ran into
when preparing a change. Even more concerning, when I instrumented that
change, it went away. So, it
Thanks, Pedro and Ard,
An update on this issue:
* It seems to be very specific to Ubuntu20's 10.5 build of GCC.
* I could not reproduce it using a crosstool-ng build of 10.5, even after
trying to configure it identically to Ubuntu20's. It might be something in
Ubuntu's patchset, but
Any further comments on this change?
I'd like to get it merged.
Thanks,
Jake
From: Jake Garver
Sent: Wednesday, October 18, 2023 10:45 AM
To: gaoliming ; devel@edk2.groups.io
Cc: michael.d.kin...@intel.com ;
zhiguang@intel.com
Subject: Re: [PATCH v2]
t; On Fri, Oct 27, 2023 at 3:13 PM Ard Biesheuvel wrote:
> >
> > On Fri, 27 Oct 2023 at 16:09, Jake Garver via groups.io
> > wrote:
> > >
> > > Hi Ard:
> > >
> > > > Can you double check the object file? I suspect this is a linker
> >
ange opcode when
converting ADR to ADRP
External email: Use caution opening links or attachments
On Fri, 27 Oct 2023 at 16:26, Pedro Falcato wrote:
>
> On Fri, Oct 27, 2023 at 3:13 PM Ard Biesheuvel wrote:
> >
> > On Fri, 27 Oct 2023 at 16:09, Jake Garver via groups.io
> >
ternal email: Use caution opening links or attachments
Hello all,
Apologies for the late response.
On Fri, 27 Oct 2023 at 14:44, Jake Garver via groups.io
wrote:
>
> Thanks for your response, Pedro.
>
> I chased this as a toolchain bug originally, but concluded that the ADR
>
bob.c.f...@intel.com ;
yuwei.c...@intel.com ; ardb+tianoc...@kernel.org
Subject: Re: [edk2-devel] [PATCH] BaseTools/GenFw: Change opcode when
converting ADR to ADRP
External email: Use caution opening links or attachments
On Thu, Oct 26, 2023 at 4:32 PM Jake Garver via groups.io
wrote:
>
In the R_AARCH64_ADR_GOT_PAGE case on AARCH64, be sure to change the
opcode to ADRP. Prior to this change, we updated the address, but not
the opcode.
This resolves an issue experienced when building a StandaloneMm image
with stack protection enabled on GCC 10.5. This scenario generates an
ADR
Thanks for the review, Liming Gao.
Any further comments on this change?
Thanks,
Jake
From: gaoliming
Sent: Saturday, October 7, 2023 1:05 AM
To: Jake Garver ; devel@edk2.groups.io
Cc: michael.d.kin...@intel.com ;
zhiguang@intel.com
Subject: 回复: [PATCH v2]
__builtin_return_address returns a pointer, not a string. Fix the STACK
FAULT message in BaseStackCheckLib appropriately.
Signed-off-by: Jake Garver
---
MdePkg/Library/BaseStackCheckLib/BaseStackCheckGcc.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
Thanks. I'll submit a v2 with that change.
From: Kinney, Michael D
Sent: Tuesday, October 3, 2023 12:00 PM
To: Name j...@nvidia.com ; devel@edk2.groups.io
Cc: Gao, Liming ; Liu, Zhiguang
; Jake Garver ; Kinney, Michael D
Subject: RE: [PATCH]
Ignore old style declaration warnings in oniguruma/src/st.c. This was
already ignored for MSFT, but newer versions of clang complain as well.
Signed-off-by: Jake Garver
---
.../Universal/RegularExpressionDxe/RegularExpressionDxe.inf | 1 +
1 file changed, 1 insertion(+)
diff --git
These sections were removed mistakenly. SmmCryptLib.inf supports these
architectures.
Signed-off-by: Jake Garver
---
CryptoPkg/Library/BaseCryptLib/SmmCryptLib.inf | 6 ++
1 file changed, 6 insertions(+)
diff --git a/CryptoPkg/Library/BaseCryptLib/SmmCryptLib.inf
When checking the version in DevicePath's Makefile, use BUILD_CC instead
of assuming "gcc". BUILD_CC is set in header.makefile and is the
compiler that will actually be used to build DevicePath. It defaults to
"gcc", but may be overridden.
Signed-off-by: Jake Garver
---
Resend with BaseTools maintainers CC'd.
Hi, folks. Is there interest in accepting this change? Any comments?
Thanks,
Jake
From: Jake Garver
Sent: Monday, December 19, 2022 11:36 AM
To: devel@edk2.groups.io ; Jeff Brasen
; Ashish Singhal
Subject: Re: [PATCH]
Hi, folks. Is there interest in accepting this change? Any comments?
Thanks,
Jake
From: Jake Garver
Sent: Thursday, December 8, 2022 11:22 AM
To: devel@edk2.groups.io ; Jeff Brasen
; Ashish Singhal
Cc: Jake Garver
Subject: [PATCH] BaseTools: Generate deps
Thanks, Konstantin,
I approve this version.
Thanks,
Jake
From: Konstantin Aladyshev
Sent: Tuesday, December 13, 2022 11:22 AM
To: devel@edk2.groups.io
Cc: bob.c.f...@intel.com ; gaolim...@byosoft.com.cn
; yuwei.c...@intel.com ; Jake
Garver ; Konstantin
Hi, Konstantin,
Thanks for working with me offline to reproduce this.
Short answer: We can live without the "$(DEBUG_DIR)(+)$(MODULE_NAME).efi"
output and I think removing that will resolve your issue. Rather than revert
the original change, can you submit a new patch that removes just that
Hi, Konstantin,
Do you have a fix for the cyclic redundancy issue when building OptionROMs? If
not, can you help me reproduce it?
I'd hate to revert d372ab as it fixed dependency issues we frequently ran into
during parallel builds.
Thanks,
Jake
From:
Prior to this change, deps were not generated for Arm and AARCH64
libraries when MODULE_TYPE was BASE, SEC, PEI_CORE, or PIEM. That
resulted in bad incremental builds.
Signed-off-by: Jake Garver
Reviewed-by: Jeff Brasen
---
BaseTools/Conf/build_rule.template | 2 +-
1 file changed, 1
Hello maintainers,
Any interest in accepting this and the "BaseTools/Conf: Fix
Dynamic-Library-File template" patch?
Thanks,
Jake
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#89268): https://edk2.groups.io/g/devel/message/89268
Mute
In the Dynamic-Library-File template, add missing output file
declarations. These files are generated by the template and other rules
explicitly depend on them.
This change resolves missing dependency issues we encountered while
running a recursive make with job control.
Signed-off-by: Jake
The generated Makefile was missing a dependency. This resulted in a
build-time race condition if the recursive make is multi-threaded and
shares job control.
Signed-off-by: Jake Garver
---
BaseTools/Source/Python/Workspace/DscBuildData.py | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
Hi folks,
Any interest in accepting this simple fix?
Thanks,
Jake
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#86654): https://edk2.groups.io/g/devel/message/86654
Mute This Topic: https://groups.io/mt/88573504/21656
Group Owner:
This change modularizes the Ext4 DSC/FDF files following the model used
in edk2/NetworkPkg.
A platform DSC can include Ext4 using "!include Features/Ext4.dsc.inc".
Ext.dsc.inc includes all the required information to enable Ext4
features. Similarly, "!include Features/Ext4.fdf.inc" would be used
Fix a Edk2Logger.warn() message format to match the arguments.
We ran into this after a failure in PcdValueInit. The failure was
masked by a new exception, "TypeError: not all arguments converted
during string formatting".
Signed-off-by: Jake Garver
---
29 matches
Mail list logo
