| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Update the dpdk recipe from v16.04 to v16.07. Renamed and modified
the patches:
dpdk-16.04-add-sysroot-option-within-app-makefile.patch
dpdk-16.04-dpdk-fix-installation-warning-and-issue.patch
accordingly since these patches don't apply cleanly on top of
dpdk v16.07.
Fixed compilation by adding examples/ethtool/lib directory in
DEPDIRS-y and by adding lib path in LDLIBS.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
Remove the maintainer name for HASWELL-WC and VALLEYISLAND
from the maintainer list, since these two layers is retired on
meta-intel master branch.
Signed-off-by: Ong Chun Weng <chun.weng.ong@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The meta-haswell-wc BSP layer is retired in future YP releases.The
plan is to consolidate all Intel SoC BSP into intel common BSP.
Since the features provided by meta-haswell-wc can be obtained
from Intel Common BSP with newer kernel version, this layer can
be retired from master.
Signed-off-by: Ong Chun Weng <chun.weng.ong@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
This layer is retired from meta-intel master branch in future
YP releases because there are no immediate requirements for
support in newer LTSI kernel.
Signed-off-by: Ong Chun Weng <chun.weng.ong@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
| |
Crystalforest BSP is no longer using the firmware files provided by these
recipes. These files were added to support qat. Now qat packages its own
calgary and canterbury firmware as part of its tarball. So removing these
from MACHINE_EXTRA_RRECOMMENDS.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
openssl-qat is different from other packages in the way that it
has a patch present in a zip file that is unpacked and applied
on top of openssl. This results in patch errors when doing an
incremental build using bitbake.
To resolve, splitted do_patch into two - one to unpack the qat patch
and apply it on openssl. And, other to apply the patches added in
SRC_URI.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
zlib-qat is different from other packages in the way
that it has a patch present in a zip file that is
unpacked and applied on top of zlib. This results in
errors when doing an incremental build using bitbake.
To resolve, splitted do_patch into two - one to unpack
the qat patch and apply it on zlib-1.2.8. And, other to
apply the patches added in SRC_URI.
Added flags "-q -o" flags to unzip command to quietly overwrite if
files/folders already exist to avoid build from breaking.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
| |
Fix for:
ERROR: QA Issue: No GNU_HASH in the elf binary
by adding LDFLAGS to TARGET_CC_ARCH.
To be sure that the LDFLAGS variable is being passed to the linker command.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
Fix for:
ERROR: QA Issue: No GNU_HASH in the elf binary
by adding LDFLAGS to TARGET_CC_ARCH.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
Ensure that the correct CFLAGS are passed to LD when compiling rte libs
as shared.
For building the dpdk with config CONFIG_RTE_BUILD_SHARED_LIB=y
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
| |
Using makefile for installing targets.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
| |
fix for:
ERROR: QA Issue: No GNU_HASH in the elf binary
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
| |
Corrected the identation of code to match the block structure to avoid
error/warning for build.
Since yocto is using -Wall and -Werror flags so this warning is treated
as error.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
| |
base_contains is simply a compatibility wrapper and may warn in the future, so
replace all instances with bb.utils.contains().
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Tom Zanussi <tom.zanussi@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
update the dpdk recipe from v2.2.0 to v16.04. Rename the patches
according to the recipe version.
Now dpdk is following 'year-month' format for version numbering.
dpdk v16.04 required "CONFIG_RTE_IP_FRAG=y" in order to build
ip_fragmentation application.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add recipe for qat16 v2.6.0-65 and retire qat16 v2.3.0-34 with respective
patches
* Use the version directly in SRC_URI instead of ${PV} since version of qatmux
(v2.6.0-60) and qat16 (v2.6.0-65) are no longer same.
* Added an install target to makefiles to help install ingredients at a proper
path.
* Ensure that compiler linker and archiver will override only when it is not
defined by the environment.
* Override the MACHINE name for do_install.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move the inclusion of SRC_URI:
https://01.org/sites/default/files/page/qatmux.l.${PV}.tgz;name=qat
from qat16.inc to version specific qat16 bb files. Since qatmux version
and qat16 version is no more same. So can't use ${PV} variable for next
version.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
The openssl recipe currently relies on EXTRA_OEMAKE having been set to
"-e MAKEFLAGS=" in bitbake.conf to operate. It is necessary to make this
explicit so that the default in bitbake.conf can be changed.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
'disable-static' option is there in EXTRA_OECONF variable. No such option
is supported by configure script. So better to pass required option
directly without using variable.
This recipe currently relies on EXTRA_OEMAKE having been set to
"-e MAKEFLAGS=" in bitbake.conf to operate. It is necessary to make this
explicit so that the default in bitbake.conf can be changed.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The qat recipe currently relies on EXTRA_OEMAKE having been set to
"-e MAKEFLAGS=" in bitbake.conf to operate. It is necessary to make this
explicit so that the default in bitbake.conf can be changed.
The prototype of a func used in QAT16 is changed in kernel v4.4
Also, make sure that we point to the correct include path to the local
headers.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This recipe adds the openssl-qat engine that accelerates some of the libcrypto
algorithms via the Intel QAT implemented on Intel communication chipsets.
QAT sample patch v0.4.9-009 is based on async branch of openssl-1.0.1m.
openssl & zlib shim layers both install the qat_mem.ko driver. Make openssl
shim dependent on zlib-qat and ensure that only zlib-qat installs this
kernel module.
This fix change in future when qat_mem module has been forked into
a separate package.
The recipe is for platforms like grantley and crystalforest
using the common BSP intel-corei7-64.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This will rename the patches according to the dpdk version.
modify the .inc and .bb according to modified name of patches.
change the commit msg for patch:
dpdk-2.2.0-add-sysroot-option-within-app-makefile.patch
since it have mentioned the version not in use.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Removing the older version recipe dpdk v2.0.0 and dpdk v1.8.0
since newer version added so it is no longer required.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
This will upgrade the recipe dpdk v2.1.0 to dpdk v2.2.0.
The recipe is for platforms like grantley and crystalforest
using the common BSP intel-corei7-64.
Signed-off-by: Rahul Kumar Gupta <rahul.kumarxx.gupta@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Backported fixes from upstream dpdk sources to
ensure dpdk 1.8.0 compiles against kernel 4.1.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Backported fixes from upstream dpdk sources
to ensure dpdk 2.0.0 compiles against kernel 4.1.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Update the PREFERRED_VERSION of linux-yocto-3.14 to 4.1 for
mohonpeak32 and mohonpeak64 machines.
Signed-off-by: Wu Chia Chuan <chia.chuan.wu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
| |
Update the PREFERRED_VERSION of linux-yocto-3.14 to 4.1 for crystalforest BSP.
Signed-off-by: Wu Chia Chuan <chia.chuan.wu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
This is in support of BSP consolidation of all Intel platforms in to meta-isg common layer.
The "MAINTAINERS file" and "meta-isg/MAINTAINERS file" are updated to reflect the changed.
Signed-off-by: Wu Chia Chuan <chia.chuan.wu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Remove meta-romley BSP layer from meta-intel master branch
which is targeting for YP 2.0.
A heads-up email was sent to the meta-intel mailing list for
requesting any feedback regarding retirement of these BSP.
Since there is no concern, we assume that it is agreed upon
and thus this patch to retire the BSP.
The "MAINTAINERS file" and "conf/machine/README file" are updated to
reflect removal of the BSP.
We expect intel-corei7-64 machine is continued to be used to support
the platform and dpdk recipe under meta-romley is now supported under
meta-isg/common/recipes-extended/dpdk.
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Mark common intel-core-i7 as the COMPATIBLE_MACHINE
for zlib-qat recipe.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Mark intel common BSP as the compatible machine for qat
recipes.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
This adds the recipe for qat16 v2.5.0-80 for grantley platforms
using the common BSP intel-corei7-64.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Defines strict_str* have become obsolete and removed entirely
starting from kernel v3.18. This results in errors like:
error: implicit declaration of function 'strict_strtoull'
[-Werror=implicit-function-declaration]
Replace strict_strtoull with kstrtoull to ensure qat 2.3.0-34
compiles.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
gcc5 has changed its default standard to gnu11 from gnu89.
These two standards have different inline semantics. This
results in errors like:
undefined reference to `icp_adf_getQueueNext'
undefined reference to `icp_adf_updateQueueTail'
undefined reference to `icp_adf_pollQueue'
undefined reference to `icp_adf_queueDataToSend'
undefined reference to `icp_adf_getQueueMemory'
Use the -fgnu89-inline command line option as suggested in [1].
Also, make sure that all the CFLAGS values are being picked up
by the compiler.
[1] https://gcc.gnu.org/gcc-5/porting_to.html
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
* Move common meta-data from version specific recipe to inc.
* Remove PV in the recipe.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Move qat and zlib-qat recipes from meta-crystalforest to
meta-isg.
This allows us to utilize these common components across
all the supported platforms.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
This adds the recipe for dpdk v2.1.0 for grantley platforms
based on common intel-corei7-64 BSP.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Add intel-corei7-64 as a compatible machine in addition
to crystalforest.
This is in line with the move to intel common BSP for all
supported platforms.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move the inclusion of patch:
dpdk-1.8.0-and-2.0.0-examples-add-config-variable-to-enable-disable-dpdk.patch
from dpdk.inc to version specific dpdk bb files where it needs to be applied.
This patch doesn't apply cleanly on top of dpdk v2.1.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Update the PREFERRED_VERSION of linux-yocto-3.14 to 3.19
for haswell-walnut canyon.
Signed-off-by: Ng, Wei Tee <wei.tee.ng@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
Add linux-firmware as a whole for both valleyisland-32.conf and
valleyisland-64.conf in order to enable the Marvell 8897 card, as well
as other devices requiring firmware blobs.
Signed-off-by: Ng Wei Tee <wei.tee.ng@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make sure that we check for the right qat package when
enabling dpdk_qat example.
The qat package name has been changed to "virtual/qat"
to enable switching between different/future versions of qat
for supported platforms.
The preferred version of "virtual/qat" can be set in machine
conf when some other version, besides qat16, becomes available
and needs to be supported.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Make sure that the dpdk_qat example app builds okay
when enabled using PACKAGECONFIG ?= " dpdk_qat".
Added a patch to dpdk_qat makefile to point to correct location
of qat libs and headers.
Signed-off-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
| |
dpdk v2.0.0 required "CONFIG_RTE_IP_FRAG=y" in order to build
ip_fragmentation application.
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch fixes issue in build configuration variables that were
not correctly parsed by do_configure() task.The environment variables
are not evaluated because the config files under dpdk/config/ folder
are not parsed by Makefile but instead by cpp compiler.
Reported-by: Anuj Mittal <anujx.mittal@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Tested-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
This patch tidy up dpdk.inc and removing unnecessary patches.
Integrating common recipe from dpdk_1.8.0.bb and dpdk_2.0.0.bb.
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
This patch makes reference to early dpdk meta-data implementation
and extends it to support DPDK v2.0.0.
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
applications
Introduce a patch to enable/disable the vhost and ip_fragmentation examples applications
under common_linuxapp config file. The flag introduce is controlled by PACKAGECONFIG.
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|
|
|
|
|
|
|
|
|
| |
For patches which are common to DPDK v1.8 and DPDK v2.0, they are renamed to
following file naming format for clarity dpdk-1.8.0-and-2.0.0*.patch.
Signed-off-by: WU CHIA CHUAN <chia.chuan.wu@intel.com>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: Saul Wold <sgw@linux.intel.com>
|