[OS-BUILD PATCH] [redhat] New configs in arch/s390
by GitLab Bridge on behalf of redhat-patchlab
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_DEBUG_USER_ASCE:
Check on exit to user space that address space control
elements are setup correctly.
If unsure, say N.
Symbol: DEBUG_USER_ASCE [=n]
Type : bool
Defined at arch/s390/Kconfig.debug:9
Prompt: Debug User ASCE
Location:
-> Kernel hacking
-> s390 Debugging
---
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_DEBUG_USER_ASCE | 1 +
.../generic/CONFIG_DEBUG_USER_ASCE | 18 ------------------
2 files changed, 1 insertion(+), 18 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_DEBUG_USER_ASCE
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_DEBUG_USER_ASCE
diff --git a/redhat/configs/common/generic/CONFIG_DEBUG_USER_ASCE b/redhat/configs/common/generic/CONFIG_DEBUG_USER_ASCE
new file mode 100644
index 000000000000..8a8327065991
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_DEBUG_USER_ASCE
@@ -0,0 +1 @@
+# CONFIG_DEBUG_USER_ASCE is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_DEBUG_USER_ASCE b/redhat/configs/pending-common/generic/CONFIG_DEBUG_USER_ASCE
deleted file mode 100644
index aeaa452caae0..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_DEBUG_USER_ASCE
+++ /dev/null
@@ -1,18 +0,0 @@
-# CONFIG_DEBUG_USER_ASCE:
-#
-# Check on exit to user space that address space control
-# elements are setup correctly.
-#
-# If unsure, say N.
-#
-# Symbol: DEBUG_USER_ASCE [=n]
-# Type : bool
-# Defined at arch/s390/Kconfig.debug:9
-# Prompt: Debug User ASCE
-# Location:
-# -> Kernel hacking
-# -> s390 Debugging
-#
-#
-#
-# CONFIG_DEBUG_USER_ASCE is not set
--
GitLab
3 years
[OS-BUILD PATCH] [redhat] New configs in fs/fuse
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_FUSE_DAX:
This allows bypassing guest page cache and allows mapping host page
cache directly in guest address space.
If you want to allow mounting a Virtio Filesystem with the "dax"
option, answer Y.
Symbol: FUSE_DAX [=y]
Type : bool
Defined at fs/fuse/Kconfig:42
Prompt: Virtio Filesystem Direct Host Memory Access support
Depends on: VIRTIO_FS [=m] && FS_DAX [=y] && DAX_DRIVER [=y]
Location:
-> File systems
-> FUSE (Filesystem in Userspace) support (FUSE_FS [=m])
-> Virtio Filesystem (VIRTIO_FS [=m])
Selects: INTERVAL_TREE [=y]
---
Cc: Miklos Szeredi <mszeredi(a)redhat.com>
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
redhat/configs/common/generic/CONFIG_FUSE_DAX | 1 +
.../pending-common/generic/CONFIG_FUSE_DAX | 22 -------------------
2 files changed, 1 insertion(+), 22 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_FUSE_DAX
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_FUSE_DAX
diff --git a/redhat/configs/common/generic/CONFIG_FUSE_DAX b/redhat/configs/common/generic/CONFIG_FUSE_DAX
new file mode 100644
index 000000000000..cc453d1c5bc7
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_FUSE_DAX
@@ -0,0 +1 @@
+CONFIG_FUSE_DAX=y
diff --git a/redhat/configs/pending-common/generic/CONFIG_FUSE_DAX b/redhat/configs/pending-common/generic/CONFIG_FUSE_DAX
deleted file mode 100644
index 69992890efd9..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_FUSE_DAX
+++ /dev/null
@@ -1,22 +0,0 @@
-# CONFIG_FUSE_DAX:
-#
-# This allows bypassing guest page cache and allows mapping host page
-# cache directly in guest address space.
-#
-# If you want to allow mounting a Virtio Filesystem with the "dax"
-# option, answer Y.
-#
-# Symbol: FUSE_DAX [=y]
-# Type : bool
-# Defined at fs/fuse/Kconfig:42
-# Prompt: Virtio Filesystem Direct Host Memory Access support
-# Depends on: VIRTIO_FS [=m] && FS_DAX [=y] && DAX_DRIVER [=y]
-# Location:
-# -> File systems
-# -> FUSE (Filesystem in Userspace) support (FUSE_FS [=m])
-# -> Virtio Filesystem (VIRTIO_FS [=m])
-# Selects: INTERVAL_TREE [=y]
-#
-#
-#
-CONFIG_FUSE_DAX=y
--
GitLab
3 years
[OS-BUILD PATCH] [redhat] New configs in drivers/thunderbolt
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_USB4_DEBUGFS_WRITE:
Enables writing to device configuration registers through
debugfs interface.
Only enable this if you know what you are doing! Never enable
this for production systems or distro kernels.
Symbol: USB4_DEBUGFS_WRITE [=n]
Type : bool
Defined at drivers/thunderbolt/Kconfig:21
Prompt: Enable write by debugfs to configuration spaces (DANGEROUS)
Depends on: USB4 [=y]
Location:
-> Device Drivers
-> Unified support for USB4 and Thunderbolt (USB4 [=y])
---
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_USB4_DEBUGFS_WRITE | 1 +
.../generic/CONFIG_USB4_DEBUGFS_WRITE | 20 -------------------
2 files changed, 1 insertion(+), 20 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_USB4_DEBUGFS_WRITE
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_USB4_DEBUGFS_WRITE
diff --git a/redhat/configs/common/generic/CONFIG_USB4_DEBUGFS_WRITE b/redhat/configs/common/generic/CONFIG_USB4_DEBUGFS_WRITE
new file mode 100644
index 000000000000..966dff9102ad
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_USB4_DEBUGFS_WRITE
@@ -0,0 +1 @@
+# CONFIG_USB4_DEBUGFS_WRITE is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_USB4_DEBUGFS_WRITE b/redhat/configs/pending-common/generic/CONFIG_USB4_DEBUGFS_WRITE
deleted file mode 100644
index 4aa0125ab425..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_USB4_DEBUGFS_WRITE
+++ /dev/null
@@ -1,20 +0,0 @@
-# CONFIG_USB4_DEBUGFS_WRITE:
-#
-# Enables writing to device configuration registers through
-# debugfs interface.
-#
-# Only enable this if you know what you are doing! Never enable
-# this for production systems or distro kernels.
-#
-# Symbol: USB4_DEBUGFS_WRITE [=n]
-# Type : bool
-# Defined at drivers/thunderbolt/Kconfig:21
-# Prompt: Enable write by debugfs to configuration spaces (DANGEROUS)
-# Depends on: USB4 [=y]
-# Location:
-# -> Device Drivers
-# -> Unified support for USB4 and Thunderbolt (USB4 [=y])
-#
-#
-#
-# CONFIG_USB4_DEBUGFS_WRITE is not set
--
GitLab
3 years
[OS-BUILD PATCH] redhat/Makefile.common: Fix 'fatal: Not a valid
object name
master' error
by GitLab Bridge on behalf of prarit
From: Prarit Bhargava <prarit(a)redhat.com>
Hey everyone,
I spent 5-10 mins looking at the 'fatal: Not a valid object name master' error
this morning. It comes from the redhat/Makefile.common:45
MERGE_BASE:=$(shell git merge-base HEAD master)
There is no master branch so 'git merge-base HEAD master' returns an error
with message 'fatal: Not a valid object name master'.
If I add a Makefile target to dump the MERGE_BASE, TAG, and SNAPSHOT
variables I get
MERGE_BASE= TAG=v5.8-rc7-1028-g305981a20220 SNAPSHOT=1
I thought about changing the MERGE_BASE value to
MERGE_BASE:=$(shell git merge-base HEAD master >& /dev/null && git merge-base HEAD master || git merge-base HEAD upstream/master)
but that seems really clunky. Anyone have a better idea? Maybe it should only
be
MERGE_BASE:$(git merge-base HEAD upstream/master) ?
In any case, after changing the MERGE_BASE line in Makefile.common, and
dumping the values the 'fatal' error no longer occurs and MERGE_BASE
is set correctly:
MERGE_BASE=d3590ebf6f91350192737dd1d1b219c05277f067 TAG=v5.8-rc7-39-gd3590ebf6f91 SNAPSHOT=1
P.
---8<----
Executing some redhat/Makefile targets results in a
'fatal: Not a valid object name master'
error. This occurs because MERGE_BASE is calculated against the master
branch which may not exist.
Check to see if master branch exists and if it doesn't use upstream/master
instead.
Signed-off-by: Prarit Bhargava <prarit(a)redhat.com>
---
redhat/Makefile.common | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/redhat/Makefile.common b/redhat/Makefile.common
index 6d048415ba6d..e67f2c9dec50 100644
--- a/redhat/Makefile.common
+++ b/redhat/Makefile.common
@@ -41,8 +41,7 @@ else
VERSION_ON_UPSTREAM:=0
endif
ifeq ($(VERSION_ON_UPSTREAM),1)
- # master is expected to track mainline.
- MERGE_BASE:=$(shell git merge-base HEAD master)
+ MERGE_BASE:=$(shell git merge-base HEAD master >& /dev/null && git merge-base HEAD master || git merge-base HEAD upstream/master)
TAG:=$(shell git describe $(MERGE_BASE))
# a snapshot off of a tagged git is of the form [tag]-[cnt]-g[hash]
SNAPSHOT:=$(shell echo $(TAG) | grep -c '\-g')
--
GitLab
3 years
[OS-BUILD PATCH] docs: Update docs to reflect newer workflow.
by GitLab Bridge on behalf of dzickusrh
From: Don Zickus <dzickus(a)redhat.com>
The workflow has recently changed such that all development is done
on the 'os-build' branch. Update the docs to show how easy it is
to make a change, commit it, generate the srpm and upload it to koji.
Also add it a build dep for making a srpm: patchutils (for filterdiff).
Cc: Bastien Nocera <bnocera(a)redhat.com>
Cc: Prarit Bhargava <prarit(a)redhat.com>
Cc: Justin Forbes <jforbes(a)redhat.com>
Signed-off-by: Don Zickus <dzickus(a)redhat.com>
---
redhat/docs/index.rst | 16 +++++-----------
redhat/docs/repository-layout.rst | 7 -------
redhat/docs/submitting-contributions.rst | 4 ++--
3 files changed, 7 insertions(+), 20 deletions(-)
diff --git a/redhat/docs/index.rst b/redhat/docs/index.rst
index 951845cc7be3..b451b9ba49ca 100644
--- a/redhat/docs/index.rst
+++ b/redhat/docs/index.rst
@@ -36,8 +36,8 @@ Once GitLab finishes forking the repository (this can take a while):
git remote add -f upstream git@gitlab.com:cki-project/kernel-ark.git
# Install build dependencies
- sudo dnf install -y make gcc flex bison bzip2 rpm-build
- git checkout upstream/ark-latest
+ sudo dnf install -y make gcc flex bison bzip2 rpm-build patchutils
+ git checkout upstream/os-build
# If you're on Fedora, you need to run:
# ln -s /usr/bin/python3 /usr/libexec/platform-python
make dist-srpm
@@ -48,18 +48,11 @@ Building an SRPM
----------------
The configuration and build scripts are in the ``os-build`` branch and
-are regularly updated to work with Linus's master branch. To build an
-SRPM, start by checking out the source tree you'd like to build. In this
-example, we'll assume that is Linus's master branch, but it could just
-as easily be Fedora's ``ark-patches`` branch (Linus's tree + Fedora
-patches) , a sub-system maintainer's tree, or your own creation.
+are regularly updated to work with Linus's master branch.
::
- git checkout linus/master
- git merge -m "Merge branch 'os-build'" os-build
- # Fedora carries a patch to alter this setting, so we need to change the configuration to build a vanilla tree.
- # If you're targeting RHEL and have brew/rhpkg installed, use "make DIST=.elrdy dist-srpm" instead
+ git checkout upstream/os-build
make dist-srpm
You can now build the SRPM however you like:
@@ -70,6 +63,7 @@ You can now build the SRPM however you like:
mock redhat/rpm/SRPMS/kernel*src.rpm
# Build the SRPM in Fedora's Koji
koji build --scratch rawhide redhat/rpm/SRPMS/kernel*src.rpm
+ koji build --scratch eln redhat/rpm/SRPMS/kernel*src.rpm
Want to add a patch? Just git-cherry-pick it or apply it with git-am and
re-run ``make dist-srpm``. Change configurations in ``redhat/configs/``
diff --git a/redhat/docs/repository-layout.rst b/redhat/docs/repository-layout.rst
index 5f6dcb08a1bd..851a2c5d715b 100644
--- a/redhat/docs/repository-layout.rst
+++ b/redhat/docs/repository-layout.rst
@@ -63,13 +63,6 @@ along with the configuration and build scripts. They can be checked out
and built into RPMs. The ``master`` branch points to the latest version
of these branches.
-rhpatches
-~~~~~~~~~
-
-This branch is no longer used. Previously, it held the Red Hat patches
-for the kernel as a quilt series and remains for historical reasons.
-Patch history up to v5.4 is available in this branch.
-
Tags
----
diff --git a/redhat/docs/submitting-contributions.rst b/redhat/docs/submitting-contributions.rst
index 07b25852ec66..65895a9ce49b 100644
--- a/redhat/docs/submitting-contributions.rst
+++ b/redhat/docs/submitting-contributions.rst
@@ -39,8 +39,8 @@ Patches
Quick start:
1. ``git fetch upstream``
-2. ``git checkout upstream/os-build && git checkout -b my-build-change``
-3. Make a change to a file or files in ``redhat/``.
+2. ``git checkout -b my-build-change upstream/os-build``
+3. Make a change to a file.
4. Add your changes with ``git add -A``.
5. Commit your changes and write a nice commit message that explains the
change: ``git commit -s``.
--
GitLab
3 years
[OS-BUILD PATCH] [redhat] New configs in drivers/gpio
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_GPIO_CDEV_V1:
Say Y here to support version 1 of the GPIO CDEV ABI.
This ABI version is deprecated.
Please use the latest ABI for new developments.
If unsure, say Y.
Symbol: GPIO_CDEV_V1 [=y]
Type : bool
Defined at drivers/gpio/Kconfig:85
Prompt: Support GPIO ABI Version 1
Depends on: GPIOLIB [=y] && GPIO_CDEV [=y]
Location:
-> Device Drivers
-> GPIO Support (GPIOLIB [=y])
-> Character device (/dev/gpiochipN) support (GPIO_CDEV [=y])
---
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_GPIO_CDEV_V1 | 1 +
.../generic/CONFIG_GPIO_CDEV_V1 | 22 -------------------
2 files changed, 1 insertion(+), 22 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_GPIO_CDEV_V1
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_GPIO_CDEV_V1
diff --git a/redhat/configs/common/generic/CONFIG_GPIO_CDEV_V1 b/redhat/configs/common/generic/CONFIG_GPIO_CDEV_V1
new file mode 100644
index 000000000000..2a4e60a576ae
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_GPIO_CDEV_V1
@@ -0,0 +1 @@
+CONFIG_GPIO_CDEV_V1=y
diff --git a/redhat/configs/pending-common/generic/CONFIG_GPIO_CDEV_V1 b/redhat/configs/pending-common/generic/CONFIG_GPIO_CDEV_V1
deleted file mode 100644
index 91a1e3450c38..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_GPIO_CDEV_V1
+++ /dev/null
@@ -1,22 +0,0 @@
-# CONFIG_GPIO_CDEV_V1:
-#
-# Say Y here to support version 1 of the GPIO CDEV ABI.
-#
-# This ABI version is deprecated.
-# Please use the latest ABI for new developments.
-#
-# If unsure, say Y.
-#
-# Symbol: GPIO_CDEV_V1 [=y]
-# Type : bool
-# Defined at drivers/gpio/Kconfig:85
-# Prompt: Support GPIO ABI Version 1
-# Depends on: GPIOLIB [=y] && GPIO_CDEV [=y]
-# Location:
-# -> Device Drivers
-# -> GPIO Support (GPIOLIB [=y])
-# -> Character device (/dev/gpiochipN) support (GPIO_CDEV [=y])
-#
-#
-#
-CONFIG_GPIO_CDEV_V1=y
--
GitLab
3 years
[OS-BUILD PATCH] [redhat] New configs in fs/Kconfig
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_TMPFS_INODE64:
tmpfs has historically used only inode numbers as wide as an unsigned
int. In some cases this can cause wraparound, potentially resulting
in multiple files with the same inode number on a single device. This
option makes tmpfs use the full width of ino_t by default, without
needing to specify the inode64 option when mounting.
But if a long-lived tmpfs is to be accessed by 32-bit applications so
ancient that opening a file larger than 2GiB fails with EINVAL, then
the INODE64 config option and inode64 mount option risk operations
failing with EOVERFLOW once 33-bit inode numbers are reached.
To override this configured default, use the inode32 or inode64
option when mounting.
If unsure, say N.
Symbol: TMPFS_INODE64 [=n]
Type : bool
Defined at fs/Kconfig:204
Prompt: Use 64-bit ino_t by default in tmpfs
Depends on: TMPFS [=y] && 64BIT [=y]
Location:
-> File systems
-> Pseudo filesystems
-> Tmpfs virtual memory file system support (former shm fs) (TMPFS [=y])
---
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_TMPFS_INODE64 | 1 +
.../generic/CONFIG_TMPFS_INODE64 | 31 -------------------
2 files changed, 1 insertion(+), 31 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_TMPFS_INODE64
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_TMPFS_INODE64
diff --git a/redhat/configs/common/generic/CONFIG_TMPFS_INODE64 b/redhat/configs/common/generic/CONFIG_TMPFS_INODE64
new file mode 100644
index 000000000000..cce8a2274ff4
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_TMPFS_INODE64
@@ -0,0 +1 @@
+# CONFIG_TMPFS_INODE64 is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_TMPFS_INODE64 b/redhat/configs/pending-common/generic/CONFIG_TMPFS_INODE64
deleted file mode 100644
index a1c19f28cc41..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_TMPFS_INODE64
+++ /dev/null
@@ -1,31 +0,0 @@
-# CONFIG_TMPFS_INODE64:
-#
-# tmpfs has historically used only inode numbers as wide as an unsigned
-# int. In some cases this can cause wraparound, potentially resulting
-# in multiple files with the same inode number on a single device. This
-# option makes tmpfs use the full width of ino_t by default, without
-# needing to specify the inode64 option when mounting.
-#
-# But if a long-lived tmpfs is to be accessed by 32-bit applications so
-# ancient that opening a file larger than 2GiB fails with EINVAL, then
-# the INODE64 config option and inode64 mount option risk operations
-# failing with EOVERFLOW once 33-bit inode numbers are reached.
-#
-# To override this configured default, use the inode32 or inode64
-# option when mounting.
-#
-# If unsure, say N.
-#
-# Symbol: TMPFS_INODE64 [=n]
-# Type : bool
-# Defined at fs/Kconfig:204
-# Prompt: Use 64-bit ino_t by default in tmpfs
-# Depends on: TMPFS [=y] && 64BIT [=y]
-# Location:
-# -> File systems
-# -> Pseudo filesystems
-# -> Tmpfs virtual memory file system support (former shm fs) (TMPFS [=y])
-#
-#
-#
-# CONFIG_TMPFS_INODE64 is not set
--
GitLab
3 years
[OS-BUILD PATCH] [redhat] New configs in arch/arm64
by GitLab Bridge on behalf of jeremycline
From: "CKI@GitLab" <cki-project(a)redhat.com>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_ARM64_BTI:
Branch Target Identification (part of the ARMv8.5 Extensions)
provides a mechanism to limit the set of locations to which computed
branch instructions such as BR or BLR can jump.
To make use of BTI on CPUs that support it, say Y.
BTI is intended to provide complementary protection to other control
flow integrity protection mechanisms, such as the Pointer
authentication mechanism provided as part of the ARMv8.3 Extensions.
For this reason, it does not make sense to enable this option without
also enabling support for pointer authentication. Thus, when
enabling this option you should also select ARM64_PTR_AUTH=y.
Userspace binaries must also be specifically compiled to make use of
this mechanism. If you say N here or the hardware does not support
BTI, such binaries can still run, but you get no additional
enforcement of branch destinations.
Symbol: ARM64_BTI [=y]
Type : bool
Defined at arch/arm64/Kconfig:1594
Prompt: Branch Target Identification support
Location:
-> Kernel Features
-> ARMv8.5 architectural features
---
Cc: Mark Salter <msalter(a)redhat.com>
Signed-off-by: CKI@GitLab <cki-project(a)redhat.com>
---
.../configs/common/generic/CONFIG_ARM64_BTI | 1 +
.../pending-common/generic/CONFIG_ARM64_BTI | 31 -------------------
2 files changed, 1 insertion(+), 31 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_ARM64_BTI
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_ARM64_BTI
diff --git a/redhat/configs/common/generic/CONFIG_ARM64_BTI b/redhat/configs/common/generic/CONFIG_ARM64_BTI
new file mode 100644
index 000000000000..fb0274de0d49
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_ARM64_BTI
@@ -0,0 +1 @@
+CONFIG_ARM64_BTI=y
diff --git a/redhat/configs/pending-common/generic/CONFIG_ARM64_BTI b/redhat/configs/pending-common/generic/CONFIG_ARM64_BTI
deleted file mode 100644
index 5af4d535b648..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_ARM64_BTI
+++ /dev/null
@@ -1,31 +0,0 @@
-# CONFIG_ARM64_BTI:
-#
-# Branch Target Identification (part of the ARMv8.5 Extensions)
-# provides a mechanism to limit the set of locations to which computed
-# branch instructions such as BR or BLR can jump.
-#
-# To make use of BTI on CPUs that support it, say Y.
-#
-# BTI is intended to provide complementary protection to other control
-# flow integrity protection mechanisms, such as the Pointer
-# authentication mechanism provided as part of the ARMv8.3 Extensions.
-# For this reason, it does not make sense to enable this option without
-# also enabling support for pointer authentication. Thus, when
-# enabling this option you should also select ARM64_PTR_AUTH=y.
-#
-# Userspace binaries must also be specifically compiled to make use of
-# this mechanism. If you say N here or the hardware does not support
-# BTI, such binaries can still run, but you get no additional
-# enforcement of branch destinations.
-#
-# Symbol: ARM64_BTI [=y]
-# Type : bool
-# Defined at arch/arm64/Kconfig:1594
-# Prompt: Branch Target Identification support
-# Location:
-# -> Kernel Features
-# -> ARMv8.5 architectural features
-#
-#
-#
-CONFIG_ARM64_BTI=y
--
2.26.2
3 years
[OS-BUILD PATCH] [redhat] New configs in drivers/watchdog
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_ARM_SMC_WATCHDOG:
Say Y here to include support for a watchdog timer
implemented by the EL3 Secure Monitor on ARM platforms.
Requires firmware support.
To compile this driver as a module, choose M here: the
module will be called arm_smc_wdt.
Symbol: ARM_SMC_WATCHDOG [=n]
Type : tristate
Defined at drivers/watchdog/Kconfig:871
Prompt: ARM Secure Monitor Call based watchdog support
Depends on: WATCHDOG [=y] && (ARM || ARM64 [=y]) && OF [=y] && HAVE_ARM_SMCCC [=y]
Location:
-> Device Drivers
-> Watchdog Timer Support (WATCHDOG [=y])
Selects: WATCHDOG_CORE [=y]
---
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_ARM_SMC_WATCHDOG | 1 +
.../generic/CONFIG_ARM_SMC_WATCHDOG | 21 -------------------
2 files changed, 1 insertion(+), 21 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_ARM_SMC_WATCHDOG
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_ARM_SMC_WATCHDOG
diff --git a/redhat/configs/common/generic/CONFIG_ARM_SMC_WATCHDOG b/redhat/configs/common/generic/CONFIG_ARM_SMC_WATCHDOG
new file mode 100644
index 000000000000..5292578ce994
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_ARM_SMC_WATCHDOG
@@ -0,0 +1 @@
+# CONFIG_ARM_SMC_WATCHDOG is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_ARM_SMC_WATCHDOG b/redhat/configs/pending-common/generic/CONFIG_ARM_SMC_WATCHDOG
deleted file mode 100644
index af24ce5fab9c..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_ARM_SMC_WATCHDOG
+++ /dev/null
@@ -1,21 +0,0 @@
-# CONFIG_ARM_SMC_WATCHDOG:
-#
-# Say Y here to include support for a watchdog timer
-# implemented by the EL3 Secure Monitor on ARM platforms.
-# Requires firmware support.
-# To compile this driver as a module, choose M here: the
-# module will be called arm_smc_wdt.
-#
-# Symbol: ARM_SMC_WATCHDOG [=n]
-# Type : tristate
-# Defined at drivers/watchdog/Kconfig:871
-# Prompt: ARM Secure Monitor Call based watchdog support
-# Depends on: WATCHDOG [=y] && (ARM || ARM64 [=y]) && OF [=y] && HAVE_ARM_SMCCC [=y]
-# Location:
-# -> Device Drivers
-# -> Watchdog Timer Support (WATCHDOG [=y])
-# Selects: WATCHDOG_CORE [=y]
-#
-#
-#
-# CONFIG_ARM_SMC_WATCHDOG is not set
--
2.26.2
3 years
[OS-BUILD PATCH] [redhat] New configs in lib/Kconfig.debug
by GitLab Bridge on behalf of jeremycline
From: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
Hi,
As part of the ongoing rebase effort, the following configuration
options need to be reviewed.
As a reminder, the ARK configuration flow involves moving unreviewed
configuration options from the pending directory to the ark directory.
In the diff below, options are removed from the pending directory and
added to the ark hierarchy. The final options that need to be ACKed
are the files that are being added to the ark hierarchy.
If the value for a file that is added should be changed, please reply
with a better option.
CONFIG_CSD_LOCK_WAIT_DEBUG:
This option enables debug prints when CPUs are slow to respond
to the smp_call_function*() IPI wrappers. These debug prints
include the IPI handler function currently executing (if any)
and relevant stack traces.
Symbol: CSD_LOCK_WAIT_DEBUG [=n]
Type : bool
Defined at lib/Kconfig.debug:1380
Prompt: Debugging for csd_lock_wait(), called from smp_call_function*()
Depends on: DEBUG_KERNEL [=y] && 64BIT [=y]
Location:
-> Kernel hacking
-> Lock Debugging (spinlocks, mutexes, etc...)
---
CONFIG_SCF_TORTURE_TEST:
This option provides a kernel module that runs torture tests
on the smp_call_function() family of primitives. The kernel
module may be built after the fact on the running kernel to
be tested, if desired.
Symbol: SCF_TORTURE_TEST [=n]
Type : tristate
Defined at lib/Kconfig.debug:1370
Prompt: torture tests for smp_call_function*()
Depends on: DEBUG_KERNEL [=y]
Location:
-> Kernel hacking
-> Lock Debugging (spinlocks, mutexes, etc...)
Selects: TORTURE_TEST [=n]
---
Cc: Prarit Bhargava <prarit(a)redhat.com>
Signed-off-by: Fedora Kernel Team <kernel-team(a)fedoraproject.org>
---
.../common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG | 1 +
.../common/generic/CONFIG_SCF_TORTURE_TEST | 1 +
.../generic/CONFIG_CSD_LOCK_WAIT_DEBUG | 19 ------------------
.../generic/CONFIG_SCF_TORTURE_TEST | 20 -------------------
4 files changed, 2 insertions(+), 39 deletions(-)
create mode 100644 redhat/configs/common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
create mode 100644 redhat/configs/common/generic/CONFIG_SCF_TORTURE_TEST
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
delete mode 100644 redhat/configs/pending-common/generic/CONFIG_SCF_TORTURE_TEST
diff --git a/redhat/configs/common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG b/redhat/configs/common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
new file mode 100644
index 000000000000..086e41bbfcf8
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
@@ -0,0 +1 @@
+# CONFIG_CSD_LOCK_WAIT_DEBUG is not set
diff --git a/redhat/configs/common/generic/CONFIG_SCF_TORTURE_TEST b/redhat/configs/common/generic/CONFIG_SCF_TORTURE_TEST
new file mode 100644
index 000000000000..53ebe3b18974
--- /dev/null
+++ b/redhat/configs/common/generic/CONFIG_SCF_TORTURE_TEST
@@ -0,0 +1 @@
+# CONFIG_SCF_TORTURE_TEST is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG b/redhat/configs/pending-common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
deleted file mode 100644
index d2e1af1bda65..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_CSD_LOCK_WAIT_DEBUG
+++ /dev/null
@@ -1,19 +0,0 @@
-# CONFIG_CSD_LOCK_WAIT_DEBUG:
-#
-# This option enables debug prints when CPUs are slow to respond
-# to the smp_call_function*() IPI wrappers. These debug prints
-# include the IPI handler function currently executing (if any)
-# and relevant stack traces.
-#
-# Symbol: CSD_LOCK_WAIT_DEBUG [=n]
-# Type : bool
-# Defined at lib/Kconfig.debug:1380
-# Prompt: Debugging for csd_lock_wait(), called from smp_call_function*()
-# Depends on: DEBUG_KERNEL [=y] && 64BIT [=y]
-# Location:
-# -> Kernel hacking
-# -> Lock Debugging (spinlocks, mutexes, etc...)
-#
-#
-#
-# CONFIG_CSD_LOCK_WAIT_DEBUG is not set
diff --git a/redhat/configs/pending-common/generic/CONFIG_SCF_TORTURE_TEST b/redhat/configs/pending-common/generic/CONFIG_SCF_TORTURE_TEST
deleted file mode 100644
index d22be82b0415..000000000000
--- a/redhat/configs/pending-common/generic/CONFIG_SCF_TORTURE_TEST
+++ /dev/null
@@ -1,20 +0,0 @@
-# CONFIG_SCF_TORTURE_TEST:
-#
-# This option provides a kernel module that runs torture tests
-# on the smp_call_function() family of primitives. The kernel
-# module may be built after the fact on the running kernel to
-# be tested, if desired.
-#
-# Symbol: SCF_TORTURE_TEST [=n]
-# Type : tristate
-# Defined at lib/Kconfig.debug:1370
-# Prompt: torture tests for smp_call_function*()
-# Depends on: DEBUG_KERNEL [=y]
-# Location:
-# -> Kernel hacking
-# -> Lock Debugging (spinlocks, mutexes, etc...)
-# Selects: TORTURE_TEST [=n]
-#
-#
-#
-# CONFIG_SCF_TORTURE_TEST is not set
--
GitLab
3 years