| Age | Commit message (Collapse) | Author |
|---|
|
These variables are used by checkpkg task to figure out if there are
a new version of the upstream software.
Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This SRCREV fixes rmc build failure under musl
Signed-off-by: Todor Minchev <todor.minchev@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Signed-off-by: Todor Minchev <todor.minchev@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Since we can run ordinary intel-corei7-64 wic images under QEMU
it makes sense to add QEMU quirks enabling serial console
to the RMC DB.
This patch appends "console=ttyS0,115200" to kernel command
line in case the image is used to run a QEMU VM e.g. with the
command:
$ runqemu ovmf refkit-image-common wic intel-corei7-64 slirp serial nographic
Unfortunately different major versions of QEMU produce different
RMC fingerprints, so the QEMU versions 2.6.x and 2.8.x should be
treated as different boards.
Signed-off-by: Dmitry Rozhkov <dmitry.rozhkov@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This SRCREV adds fingerprint and database extraction functionality to RMC
Extract database:
rmc -E -d rmc.db -o output/directory/
Extract fingerprint:
rmc -E -f fingerprint_file
https://bugzilla.yoctoproject.org/show_bug.cgi?id=10092
Signed-off-by: Todor Minchev <todor.minchev@linux.intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
With some UEFI shells LoadOptionsSize is reported being > 0
but the corresponding LoadOptions does not contain any data
(the first element has value 0).
When that happens, the stub feature that allows .cmdline to be
replaced by what's in LoadOptions ends up copying nothing/random
data to the kernel cmdline resulting in different kinds of boot
problems.
To fix this, add a check to see if LoadOptions contains data
before replacing the .cmdline.
Upstream-Status: Accepted [https://github.com/systemd/systemd/pull/5467]
Fixes [YOCTO #11078].
Signed-off-by: Mikko Ylinen <mikko.ylinen@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The OE-Core LICENSE is mostly MIT, but should not be used as a checksum
file for a purely MIT licensed package.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The command line setting for Intel 570x added 'noxsave' which
disabled xsave/xrstor to save and restore extended registers.
The systems run fine without the setting so drop noxsave to
gain optimized context switches and get more CPU features
flags (XSAVE*/AVX/MPX) enabled.
Signed-off-by: Mikko Ylinen <mikko.ylinen@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The same patches from LSM8 can be applied to LMS7 for
enabling world build with MUSL in meta-intel
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Thermald uses the argp.h interface of glibc, which is not available
in MUSL, so we need to add a depends on argp-standalone.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The new recipe specific sysroot exposed a missing dependecny on
the dbus-binding-tool
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
To be able to detect a Minnowboard Turbot, its fingerprint
is needed in the RMC database.
This initial commit adds the fingerprint and a simple cmdline
entry (KBOOTPARAM) for the board.
Signed-off-by: Mikko Ylinen <mikko.ylinen@intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
systemd-boot's EFI stub can be built in an EFI executable
with the kernel, cmdline, and initrd.
This commit enables the EFI stub code to use the RMC database
and appends the board specific cmdline (KBOOTPARAM) to the
built-in cmdline.
Signed-off-by: Mikko Ylinen <mikko.ylinen@intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
This is the latest release version.
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
Thermald uses the command glib-genmarshal to build, which is from
glib-2.0-native.
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
Add two patches to make lms8 build with musl libc (both generic fixes,
not musl specific).
Signed-off-by: Jussi Kukkonen <jussi.kukkonen@intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
This change updates the locked systemd-boot revision to v232
release which is the current in OE.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
Use the variable RMC_BOARD_DATA_DIRS, to disable, append to, or
override the default board data in meta-intel with boards' data
provided by users.
Ideally, users should get the updated database in the new built
image after an incremental build.
Examples of RMC database output (db):
RMC_BOARD_DATA_DIRS = "" (disable db generation)
RMC_BOARD_DATA_DIRS_append = " top_dir" (db of defaults & user's)
RMC_BOARD_DATA_DIRS = "top_dir" (db for user's , no defaults)
RMC_BOARD_DATA_DIRS = " " (same as "")
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
OE-Core recently renamed the recipe to have PV in the name
and the bbappend wildcard was added to make the transition
smooth.
The renaming is now complete so rename the bbappend again
to match the new versioned recipe only.
The aggresive systemd-boot% wildcard matches systemd-bootchart
too giving unexpected results.
Depends on OE-Core 8fe1e5197f6f94a49693de09f4eb9394df531cc8.
Signed-off-by: Mikko Ylinen <mikko.ylinen@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Instead of hardcoding /usr in the recipe, and hardcoding $prefix/lib in the
Makefile, override all of the paths with the distro configuration.
[ YOCTO #10800 ]
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
OE-Core is in process of renaming the recipe
to have PV in recipe name. Cover that case
for bbappend
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
|
|
We once added a patch in gnu-efi for rmc's sake. The latest rmc
gets rid of this dependency on gnu-efi.
The removed patch is still a reasonable change and merged into
the upstream gnu-efi project. It should come with the next update
in the gnu-efi recipe in OE.
At this point, however, we don't see any urgency to keep it in
meta-intel any more.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This is a whole package of rmc work in meta-intel to reflect
some major changes in the upstream project:
In rmc.bb recipe, EFI_ARCH, path of EFI header files, and
dependency on gnu-efi are removed with the updated revision.
In systemd-boot, patches to integrate with rmc are re-worked
mainly because of new APIs. Size of patches are smaller than
the previous implementation. Notice we still use multiple APIs
instead of calling an one-step interface multiple times, to get
some potential runtime performance benefit. (rmc tool in user
space is changed to use single API in the upstream project.)
Fixes [YOCTO #10086]
Fixes [YOCTO #10671]
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Since the rmc is used standalone from systemd-boot and does not
use libc which provides the stack_chk_fail code that causes an
undefined symbol
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Intel common BSP is using ttyS0 and tty0 for consoles by default. By Adding
support for mohonpeak platforms to rmc, these console port can customized to
ttyS1 and tty1 by detecting the fingerprint file at runtime.
It will use their own boot.conf/install.conf for booting/installing.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This re-enables the video framebuffer for displaying cosole output
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Makes meta-intel behave well in multi-BSP scene
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Add the new intel-x86-common to the meta-data in order to limit the
changes when the meta-intel layer is included without using
meta-intel machines.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Add the new MACHINEOVERRIDE to the SRC_URI in order to limit the changes when the
meta-intel layer is included without using meta-intel machines.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
By adding these entries, we can get the serial console output on the correct ttyS2 port
the default port for meta-intel is ttyS0, but we can fix this with the rmc configuration
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
On install we want to keep rmc's boot.conf. Also add POSTINSTALL.sh
that removes ttyS0 from inittab since it uses ttyS2 instead.
Signed-off-by: California Sullivan <california.l.sullivan@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
This is an experimental RMC sample for quark. At runtime, a
boot option with board name will be available for user to
boot the Galileo.
Note: The supported image type is hddimg instead of the direct-
boot image type usually used on quark. Use a USB stick flashed
with the image to boot the Galileo Gen 2 board.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
A particular intention of this change is to remove the serial
setting in kernel command line that is hardcoded in core2 and
corei7 conf files.
However we cannot use KBOOTPARAM because it is designed as a
fragment appended to hardcoded part at runtime. This is why
we need to create more files in this patch to use cmdline
in bootloader entry files to bypass whatever from the build.
Once we have a clean base (no any preset console setting)
in generic configurations, this change can be dropped.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
New changes in the upstream rmc project captured by this SRCREV:
2e38d05 Merge branch 'rmc-dev' (Aug 30 2016)
a1a4a26 Makefile: remove static linking in Linux build
b22169a doc: update README file
29e943d Merge branch 'rmc-dev' (Aug 24 2016)
fc2fb1e rmc: fix allocating buffer issue in read_file() (32 bit)
38b0553 rmc: fix wrong error message
7ee67a0 test: Provide test scripts and sample data
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
The QA sanity checker in OE fails when an ELF binary has a symbol
table without GNU hash.
There is no issue so far because rmc is statically linked with C
libraries and the symbol table is stripped out in build.
But once we compile rmc with dynamic linking, the symbol table in the
dynamic section stays. Thus it fails the QA checker because the
default hash style is sysv.
To have some peace of mind, we specify "both" for the hash style to
the linker.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
This change shows how we can get rid of an error message caused
by board-specific design at runtime, and still keep the rest
in system generic.
The new added hook removes line(s) in inittab which triggers the
error.
README of rmc is also updated accordingly.
Fixes [YOCTO #5351].
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
Add Broxton-m fingerprint and global kernel bootparam fragment
into rmc.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
Provide a README for RMC feature. Also check in fingerprints and
configuration data for several boards as examples for users.
They can be used for validation too.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
RMC Feature is based on RMC project, systemd-boot, EFI installer
to enable a single generic image, built for multiple platforms,
automatically applies customization and quirks specific to a type
of boards at runtime.
In another word, you will see a single image behaves differently
and intelligently according to the type of board it is running on.
To Enable this feature: add the two lines in conf file:
DISTRO_FEATURES_append = " rmc"
EFI_PROVIDER = "rmc-systemd-boot"
Based on Saul Wold's initial work on a feature switch, RMC patches
in systemd-boot, installer and gnu-efi aren't built unless the
feature is enabled.
For a supported board, this feature can :
() show and boot with board-specific boot entries in boot menu
in live-boot and post-installation.
() apply a kernel cmdline fragment to the end of cmdline to boot
Linux kernel. This is effective for any boot entry user chooses in
boot menu.
() create directory and deploy files only for the type of the
running board to target's file systems. What left on target after
installation is just same as the result from installing a conventional
image customized for a single type of hardware.
To add support of new boards, a new variable RMC_BOARD_DATA_DIRS
is the interface to developers. How-to information will be provided
with examples in following patches.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
Invoke RMC APIs in this bootloader to query board-specific data
from RMC database(DB) file on ESP. Data can be boot entries or a
global kernel boot command line fragment specific to a type of
board supported in RMC DB.
Bootloader queries a file blob named BOOTENTRY.CONFIG from RMC
DB first. In success, bootloader parses BOOTENTRY.CONFIG to get
name of each boot entry file associated to the type of running
board, and then tries to load the entry into internal config data
structure. Once any entry is loaded from RMC DB, bootloader skips
loading conf files on ESP.
BOOTENTRY.CONFIG has a very simple format - every line is a boot
entry file's name. For example, to specify two boot entries in it:
boot.conf
install.conf
Bootloader also seeks another file named KBOOTPARAM in RMC dB.
when it can obtain this file associated to the type of running
board, it appends what in file to the end of kernel command
line before it boots up kernel. The appending is effective on
every boot entry, so it is called "global" cmdline fragment.
When Bootloader doesn't get config, an entry or cmdline fragment
for the type of board, it simply moves to the next step.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
UEFI spec has different GUIDs for 32 and 64 bit SMBIOS
entry point structure. This change adds definition for
64 bit GUID, so that software linked with gnu-efi can
have it for x86_64 architecture.
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
RMC recipe fetch RMC project and build it more than once in
build time:
RMC tool is built for host architecture (native). The tool for
host is used to generate RMC database in build time.
RMC tool is also built for target architecture, so that scripts
in user space can call RMC tool on a running target. Developers
can also boot a target and run rmc tool to obtain fingerprint
for a new board type.
RMC libraries are compiled for both of UEFI context and user
space. They are always linked in RMC tool and can be linked
into an EFI bootloader. The recipes don't install libraries
for target's user space until we have a new client needs it.
The rmc-db.bbclass provides functions to generate rmc database
file for other software components to reuse.
We absorb a patch from Tom Zanussi to update source location with
the public link. We could put this change in another commit, but
leaving the replaced internal link in this commit could cause
trouble when people bisect the project but don't have access to
the internal location:
----------------------
rmc: Update to use public repo
The repo the rmc recipe was pointing to was private - it's now public
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
common/recipes-bsp/rmc/rmc.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/recipes-bsp/rmc/rmc.inc b/common/recipes-bsp/rmc/rmc.inc
index c046e2e..bdf930d 100644
--- a/common/recipes-bsp/rmc/rmc.inc
+++ b/common/recipes-bsp/rmc/rmc.inc
@@ -15,7 +15,7 @@ LICENSE = "MIT"
LIC_FILES_CHKSUM = "file://COPYING;md5=bcdd376d27b26bde6afadd67aa3c8b07"
-SRC_URI = "git://git@git.yoctoproject.org/rmc;protocol=ssh"
+SRC_URI = "git://git.yoctoproject.org/rmc"
...
----------------------
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Reviewed-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
This marks the conf.xml files as changable so that the package management
software won't overwrite them.
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
This user space thermal daemon utility is used for thermal management.
Signed-off-by: Yong Li <yong.b.li@intel.com>
Acked-by: Saul Wold <sgw@linux.intel.com>
Tested-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
This change is ported from the fix on lms8 for
the same issue:
commit 6dc3746443523a02f72bf5142cfbe3a800d32f4a
Author: Saul Wold <sgw@linux.intel.com>
Date: Mon May 16 10:01:49 2016 -0700
lms8: Add patch to fix gcc6 C++ whitespace
This adds a patch to lms8 to fix an error cause by the newer
C++11 standard being enabled in GCC6 that requires additional
whitespace around User-Defined literals.
[YOCTO #9640]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Fixes [YOCTO #9785]
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
The new machconfig probes USB keyboard and touch screen, and
then sets HAVE_* variables according to detection.
Detectable devices:
USB HID keyboards (Generic Desktop)
USB HID touch screens (Digitizer)
Note:
The intention is to have a way to provide initial formfactor
settings in a boot procedure. That means supported keyboard
and touch screen must be connected before machconfig runs.
Any new connection or disconnection won't be detected until
machconfig is executed again.
Limitation:
There could be some USB HID devices presents more than one
usage in a single descriptor. We will add support once such
device emerges.
Some platforms may have _virtual_ devices provided by BIOS.
It will cause false detection when they are presented as
types we supported. We can add black list logic when it
becomes a big concern.
Fixes [YOCTO #9205]
Signed-off-by: Jianxun Zhang <jianxun.zhang@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
This adds a patch to lms8 to fix an error cause by the newer
C++11 standard being enabled in GCC6 that requires additional
whitespace around User-Defined literals.
[YOCTO #9640]
Signed-off-by: Saul Wold <sgw@linux.intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
The gma500-gfx-check we used had a dependency to meta-poky via a bbappend file
this is unnacceptable, but this check is still needed on genericx86 and
intel-core2-32, so this patch makes copy of the gma500-gfx-check to meta-intel
to avoid the dependency issue
[YOCTO #9181]
Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
The gma500_gfx driver will match certain devices on which it causes incorrect functionality,
we want to avoid inserting this module, basicallly blacklist it for specific hardware,
but still allow it to work on other hardware by default; usually when we have an already working system,
using udev rules would do the job, but since we are building it, it is impossible to blacklist
a driver when a certain udev rule matches, since rootfs isn't writeable at this point during boot time,
the solution is to use modprobe install, which runs a certain command instead of inserting a matching module
this command needs to insert the module manually afterwards and have a flag --ignore-install
so it doesn't create an infinite loop executing itself everytime it wants to insert the module,
busybox's modprobe doesnt provide this functionality, so a small hack had to be used to avoid
the infite loop instead.
Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
Aníbal Limón
Todor Minchev
Dmitry Rozhkov
Mikko Ylinen
Saul Wold
Mikko Ylinen
California Sullivan
Jussi Kukkonen
Jianxun Zhang
Ross Burton
Khem Raj
Rahul Kumar Gupta
Yong Li
Alejandro Hernandez