[prev in list] [next in list] [prev in thread] [next in thread]
List: openembedded-core
Subject: Re: [OE-core] [PATCH] oeqa/selftest/imagefeatures: Add testcase for fitImage
From: "Usama Arif" <usama.arif () arm ! com>
Date: 2020-07-29 14:47:44
Message-ID: c5e2f0fb-6353-77e6-3cca-37fd9cc84d54 () arm ! com
[Download RAW message or body]
On 29/07/2020 15:41, Usama Arif wrote:
> This testcase generates the Image Tree Source and
> the corresponding fitImage containing a kernel and
> a ramdisk. It then checks if the these files exist
> and if the right fields are present in the right
> order in the Image Tree Source.
>
> Tested with: oe-selftest -r imagefeatures.ImageFeatures.test_fit_image
>
> Signed-off-by: Usama Arif <usama.arif@arm.com>
> Cc: Richard Purdie <richard.purdie@linuxfoundation.org>
> ---
> meta/lib/oeqa/selftest/cases/imagefeatures.py | 74 +++++++++++++++++++
> 1 file changed, 74 insertions(+)
>
> diff --git a/meta/lib/oeqa/selftest/cases/imagefeatures.py \
> b/meta/lib/oeqa/selftest/cases/imagefeatures.py index dea519e6df..f7a2533746 100644
> --- a/meta/lib/oeqa/selftest/cases/imagefeatures.py
> +++ b/meta/lib/oeqa/selftest/cases/imagefeatures.py
> @@ -263,6 +263,80 @@ PNBLACKLIST[busybox] = "Don't build this"
>
> bitbake("--graphviz core-image-sato")
>
> + def test_fit_image(self):
> + """
> + Summary: Check if FIT image and Image Tree Source (its) are built
> + and the Image Tree Source has the correct fields.
> + Expected: 1. fitImage and fitImage-its can be built
> + 2. The type, load address, entrypoint address and
> + default values of kernel and ramdisk are as expected
> + in the Image Tree Source. Not all the fields are tested,
> + only the key fields that wont vary between different
> + architectures.
> + Product: oe-core
> + Author: Usama Arif <usama.arif@arm.com>
> + """
> + config = """
> +# Enable creation of fitImage
> +KERNEL_IMAGETYPE = "Image"
> +KERNEL_IMAGETYPES += " fitImage "
> +KERNEL_CLASSES = " kernel-fitimage "
> +
> +# RAM disk variables including load address and entrypoint for kernel and RAM disk
> +IMAGE_FSTYPES += "cpio.gz"
> +INITRAMFS_IMAGE = "core-image-minimal"
> +UBOOT_RD_LOADADDRESS = "0x88000000"
> +UBOOT_RD_ENTRYPOINT = "0x88000000"
> +UBOOT_LOADADDRESS = "0x80080000"
> +UBOOT_ENTRYPOINT = "0x80080000"
> +"""
> + self.write_config(config)
> +
> + # fitImage is created as part of linux recipe
> + bitbake("virtual/kernel")
> +
> + image_type = "core-image-minimal"
> + deploy_dir_image = get_bb_var('DEPLOY_DIR_IMAGE')
> + machine = get_bb_var('MACHINE')
> + fitimage_its_path = os.path.join(deploy_dir_image,
> + "fitImage-its-%s-%s-%s" % (image_type, machine, machine))
> + fitimage_path = os.path.join(deploy_dir_image,
> + "fitImage-%s-%s-%s" % (image_type, machine, machine))
> +
> + self.assertTrue(os.path.exists(fitimage_its_path),
> + "%s image tree source doesn't exist" % (fitimage_its_path))
> + self.assertTrue(os.path.exists(fitimage_path),
> + "%s FIT image doesn't exist" % (fitimage_path))
> +
> + # Check that the type, load address, entrypoint address and default
> + # values for kernel and ramdisk in Image Tree Source are as expected.
> + # The order of fields in the below array is important. Not all the
> + # fields are tested, only the key fields that wont vary between
> + # different architectures.
> + its_field_check = ['type = "kernel";',
> + 'load = <0x80080000>;',
> + 'entry = <0x80080000>;',
> + 'type = "ramdisk";',
> + 'load = <0x88000000>;',
> + 'entry = <0x88000000>;',
> + 'default = "conf@1";',
> + 'kernel = "kernel@1";',
> + 'ramdisk = "ramdisk@1";'
> + ]
> +
> + with open(fitimage_its_path) as its_file:
> + field_index = 0
> + for line in its_file:
> + if field_index == len(its_field_check):
> + break
> + if its_field_check[field_index] in line:
> + field_index +=1
> +
> + if field_index != len(its_field_check): # if its equal, the test passed
> + self.assertTrue(field_index == len(its_field_check),
> + "Fields in Image Tree Source File %s did not match, error in \
> finding %s" + % (fitimage_its_path, its_field_check[field_index]))
> +
> def test_image_gen_debugfs(self):
> """
> Summary: Check debugfs generation
>
Hi
This test assumes that the case for a single kernel and single
RAM disk with no dtb is supported by kernel-fitimage.bbclass.
https://lists.openembedded.org/g/openembedded-core/message/140810 adds
support for this combination of usecase and the test wont pass without
the usecase being supported.
Thanks,
Usama
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#141098): https://lists.openembedded.org/g/openembedded-core/message/141098
Mute This Topic: https://lists.openembedded.org/mt/75865813/4454766
Group Owner: openembedded-core+owner@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [openembedded-core@marc.info]
-=-=-=-=-=-=-=-=-=-=-=-
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic