Introduce new override var-som #15
Conversation
|
@kraj thanks for your patches. We are discussing this and will follow up with you. |
There are few places where this layer need to apply tweaks which might break other ( non-variscite ) IMX machines, especially places where its using forked SRC_URIs from varigit. Therefore add an override which is available across all variscite SOMs. Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
These patches regress other imx8 SOCs ( non-variscite ) similarily having local do_compile is also specific to var-soms This helps distros like yoe distro which target multiple IMX based machines from different SOC/SOM makers. Moreover this also takes this layer a step closer towards Yocto compatible status. Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Helps building it with master branch of meta-freescale as well Signed-off-by: Khem Raj <raj.khem@gmail.com>
|
I have rebased it on latest kirkstone and also ensured that it works with meta-freescale/master branch |
| git://github.com/varigit/imx-sc-firmware.git;protocol=https;branch=${SCFW_BRANCH}; \ | ||
| https://developer.arm.com/-/media/Files/downloads/gnu-rm/8-2018q4/gcc-arm-none-eabi-8-2018-q4-major-linux.tar.bz2;name=gcc-arm-none-eabi \ | ||
| " | ||
|
|
||
| SRC_URI[gcc-arm-none-eabi.md5sum] = "f55f90d483ddb3bcf4dae5882c2094cd" | ||
| SRC_URI[gcc-arm-none-eabi.sha256sum] = "fb31fbdfe08406ece43eef5df623c0b2deb8b53e405e2c878300f7a1f303ee52" | ||
|
|
||
| unset do_compile[noexec] |
There was a problem hiding this comment.
Hi @kraj,
I am looking at finally integrating your changes into our langdale-var01 branch. One question, why did you make this change moving our compile functionality into do_deploy()?
There was a problem hiding this comment.
This makes it work in conjunction with latest meta-freescale and yocto, at yoe distro we use rolling release and always use master branches of all layers.
There was a problem hiding this comment.
Ok, but I think we should take a different approach rather than moving the task because our statements are compiling it. I see that leaving unset do_compile[noexec] would likely make your version run the default do_compile task (oe_runmake). What if instead we keep what you did in your earlier commit in this series e03c114 (correcting the MACHINEOVERRIDES typo)
do_compile:var-som() {
and then use a python anonymous function to dynamically delete the do_compile[noexec] flag in cases of var-som
python() {
machine_overrides = d.getVar('MACHINEOVERRIDES').split(':')
for override in machine_overrides:
if override == "var-som":
d.delVarFlag("do_compile", "noexec")
}
What do you think?
|
@kraj thanks for this pull request and suggstion. We've cherry-picked your patches starting in langdale: 87bbc4a...0e891a2 |
No description provided.