On Tue, Apr 25, 2017 at 10:53 PM, Nathan Rossi <nat...@nathanrossi.com> wrote:
> On 26 April 2017 at 04:11, Alistair Francis <alistai...@gmail.com> wrote:
>> On Tue, Apr 25, 2017 at 10:54 AM, Alistair Francis <alistai...@gmail.com> 
>> wrote:
>>> On Thu, Apr 20, 2017 at 3:35 AM, Nathan Rossi <nat...@nathanrossi.com> 
>>> wrote:
>>>> This change reworks how the meta-xilinx layer enables and provides the
>>>> custom version of QEMU based on Xilinx's fork of QEMU. The existing
>>>> implementation relied on the single sysroot which was changed in oe-core
>>>> such that now recipes have their own sysroots (RSS support).
>>>> Additionally oe-core now provides the QEMU binaries to the runqemu
>>>> script via the 'qemu-helper-native' recipes sysroot as opposed to the
>>>> image sysroot.
>>>>
>>>> These rework changes allow for a single machine to select the targeted
>>>> QEMU version as well as to provide the qemuboot config specific to the
>>>> targeted QEMU version. The selection of QEMU version is now handled by
>>>> PREFERRED_PROVIDER mechanics with the meta-xilinx layer providing an
>>>> additional recipe that is equivalent to qemu-helper-native and which
>>>> also provides said target allowing for the machine to select via the use
>>>> of PREFERRED_PROVIDER_qemu-helper-native. This recipe
>>>> (qemu-xilinx-helper-native) however instead provides the sysroot
>>>> populated with qemu-xilinx instead of qemu.
>>>>
>>>> Additionally the XILINX_QEMUBOOT variable is replaced with the
>>>> qemuboot-xilinx.bbclass, this provides the overrides for setting up
>>>> qemu-xilinx specific QB_* args. Additionally this bbclass points runqemu
>>>> at the qemu-xilinx-helper-native sysroot for QEMU binaries.
>>>>
>>>> These changes also work towards making the meta-xilinx layer better
>>>> handle multiple qemuboot.conf variants as well as handling different
>>>> QEMU versions.
>>>>
>>>> This change also removes the 'qemu-system-xilinx' MACHINE_FEATURES, this
>>>> is due to MACHINE_FEATURES no longer being available for native recipes.
>>>> Additionally there is no longer any logic that needs to know this any
>>>> way.
>>>>
>>>> Signed-off-by: Nathan Rossi <nat...@nathanrossi.com>
>>>> ---
>>>>  classes/qemuboot-xilinx.bbclass                    | 27 +++++++++++++++
>>>>  conf/machine/include/machine-xilinx-qemu.inc       | 39 
>>>> ++++------------------
>>>>  conf/machine/ml605-qemu-microblazeel.conf          |  1 +
>>>>  conf/machine/qemu-zynq7.conf                       |  1 +
>>>>  conf/machine/s3adsp1800-qemu-microblazeeb.conf     |  1 +
>>>>  conf/machine/zcu102-zynqmp.conf                    |  8 +++--
>>>>  .../qemu/qemu-xilinx-helper-native_1.0.bb          | 20 +++++++++++
>>>>  7 files changed, 61 insertions(+), 36 deletions(-)
>>>>  create mode 100644 classes/qemuboot-xilinx.bbclass
>>>>  create mode 100644 recipes-devtools/qemu/qemu-xilinx-helper-native_1.0.bb
>>>>
>>>> diff --git a/classes/qemuboot-xilinx.bbclass 
>>>> b/classes/qemuboot-xilinx.bbclass
>>>> new file mode 100644
>>>> index 0000000000..a2f5ef3eb6
>>>> --- /dev/null
>>>> +++ b/classes/qemuboot-xilinx.bbclass
>>>> @@ -0,0 +1,27 @@
>>>> +
>>>> +inherit qemuboot
>>>> +
>>>> +# enable the overrides for the context of the conf only
>>>> +OVERRIDES .= ":qemuboot-xilinx"
>>>> +
>>>> +# setup the target binary
>>>> +QB_SYSTEM_NAME_prepend = "qemu-xilinx/"
>>>> +
>>>> +# Default machine targets for Xilinx QEMU (FDT Generic)
>>>> +QB_MACHINE_aarch64 = "-machine arm-generic-fdt"
>>>> +QB_MACHINE_arm = "-machine arm-generic-fdt-plnx"
>>>
>>> Hey Nathan,
>>>
>>> This should be: "arm-generic-fdt-7series", that name is the old name
>>>
>>>> +QB_MACHINE_microblaze = "-machine microblaze-generic-fdt"
>>>
>>> This should be "microblaze-generic-fdt-plnx", I'm not sure why this
>>> wasn't always this.
>
> I just moved them from the .inc to the .bbclass. You should ask past
> Alistair why he used those values ;). They are currently not used, but
> i've updated the patch to make them the values you provided above.

Yeah, I know. Past Alistair didn't know what he was doing apparently.

Thanks for updating it.

>
>>>
>>>> +
>>>> +# defaults
>>>> +QB_DEFAULT_KERNEL ?= "none"
>>>
>>> For the newer QEMU builds (my other patch that hangs after ATF) we
>>> need to ensure that QB_DEFAULT_KERNEL is NULL when booting Xilinx's
>>> fork of QEMU. It appears that using the QEMU -kernel loader must
>>> somehow overwrite images that we want to boot which is why the boot
>>> fails.
>>>
>>> Adding a QB_DEFAULT_KERNEL_qemuboot-xilinx = "none" in
>>> machine-xilinx-qemu.inc fixes the issue, can you add that in somewhere
>>> in this patch?
>
> I did transfer that over in the qemuboot-xilinx.bbclass, however it
> was not taking effect due to "?=" not overriding the "?=" from
> qemuboot.bbclass. I have shifted the inherit such that it is after the
> QB_* setup in qemuboot-xilinx.bbclass so that setting via ?= takes
> precedence over the qemuboot.bbclass. Which results in it set to
> "none" when using qemuboot-xilinx.
>
> Updated version:
> https://github.com/nathanrossi/meta-xilinx/commit/0a02e2f114079d9cc056536ad2b16a9379d8e3d9

Great! Thanks, that should fix the issue I was seeing.

>
>>
>> Woops, forgot to mention that with with the kernel set to none the RAM
>> sizes don't line up. The old QEMU version will work, but new QEMU
>> releases default to only 2GB of RAM being created, while Linux expects
>> 4GB as the command line arguments aren't passed in. The QB_MEM
>> variable needs to be updated to 4GB of RAM for Linux to boot.
>
> I've created a separate patch which updates QB_MEM, since it is
> related to the 2017.1 QEMU updates.
>
> https://github.com/nathanrossi/meta-xilinx/commit/e5bb84593da7835a36d846994faee17547919084
>
> I've queued up this series, your QEMU version updates and the above
> patch in nrossi/next. If you are good with the above changes I will
> merge it all into master.

I'll check it today and let you know.

Thanks,

Alistair

>
> Thanks,
> Nathan
-- 
_______________________________________________
meta-xilinx mailing list
meta-xilinx@yoctoproject.org
https://lists.yoctoproject.org/listinfo/meta-xilinx

Reply via email to