Hi Ryan,
I was trying to rerun some tests at my end via LAVA on armv8 FVPs using 14.04 hwpack.
What I see is that LAVA fails prematurely stating that
Traceback (most recent call last): File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/actions/boot_control.py", line 105, in run client.boot_linaro_image() File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 508, in boot_linaro_image self._boot_linaro_image() File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 481, in _boot_linaro_image self.proc = self.target_device.power_on() File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 325, in power_on self._check_needed_files() File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 188, in _check_needed_files self.config.simulator_bl1) RuntimeError: No SECURE FLASHLOADER found, u'bl1.bin'
Looking at the 14.04 hwpack I see that now there are 2 sets of bl* images there one for FVPs and another for Juno.
LAVA default config has
simulator_bl1 = bl1.bin simulator_bl2 = [bB][lL]2.*.bin simulator_bl3 = [bB][lL]31.*.bin simulator_uefi = uefi_fvp-base.bin
fastmodel.py # SECURE FLASHLOADERs are needed only for base and cortex models if self._bl1 is None and self.config.simulator_bl1: raise RuntimeError('No SECURE FLASHLOADER found, %r' % self.config.simulator_bl1)
which I guess gets 'checked' by lava whether the needed files present and results in a runtime failure.
With the latest hardware packs, can you please confirm how does the bl images get unpacked?
Are they in some new folder structure because of platform variants are present?
Please advice.
For the moment I could not get LAVA to work with the latest hardware pack contents due to this issue.
Thanks Basil Eljuse...
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
Hi Basil,
On 28 April 2014 12:07, Basil Eljuse Basil.Eljuse@arm.com wrote:
Hi Ryan,
I was trying to rerun some tests at my end via LAVA on armv8 FVPs using 14.04 hwpack.
What I see is that LAVA fails prematurely stating that
Traceback (most recent call last):
File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/actions/boot_control.py", line 105, in run
client.boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 508, in boot_linaro_image
self._boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 481, in _boot_linaro_image
self.proc = self.target_device.power_on()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 325, in power_on
self._check_needed_files()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 188, in _check_needed_files
self.config.simulator_bl1)RuntimeError: No SECURE FLASHLOADER found, u'bl1.bin'
Looking at the 14.04 hwpack I see that now there are 2 sets of bl* images there one for FVPs and another for Juno.
LAVA default config has
simulator_bl1 = bl1.bin
simulator_bl2 = [bB][lL]2.*.bin
simulator_bl3 = [bB][lL]31.*.bin
simulator_uefi = uefi_fvp-base.bin
fastmodel.py
# SECURE FLASHLOADERs are needed only for base and cortex models
if self._bl1 is None and self.config.simulator_bl1:
raise RuntimeError('No SECURE FLASHLOADER found, %r' % self.config.simulator_bl1)which I guess gets ‘checked’ by lava whether the needed files present and results in a runtime failure.
With the latest hardware packs, can you please confirm how does the bl images get unpacked?
Are they in some new folder structure because of platform variants are present?
Due to confusion over the changes to TF, the experience isn't as smooth as I'd hoped.
In the disk image, they will be both in the boot partition and in the fvp sub-dir. They are also now prefixed with "fvp_". Hence why the instructions have the extra step to "ln -s fvp_bl1.bin bl1.bin", etc... or fvp_bl1.bin will not be able to load bl2.bin or bl31.bin via semihosting. In future, they should only be in the fvp sub-dir.
We are still using the old TF for FVP. 14.05 should get the latest version available now on github, assuming it passes my tests properly. Then we will also have the fvp_fip.bin file and semihosting will not be used at all.
Please advice.
For the moment I could not get LAVA to work with the latest hardware pack contents due to this issue.
Thanks
Basil Eljuse…
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
Hi
On 28 April 2014 14:46, Ryan Harkin ryan.harkin@linaro.org wrote:
Hi Basil,
On 28 April 2014 12:07, Basil Eljuse Basil.Eljuse@arm.com wrote:
Hi Ryan,
I was trying to rerun some tests at my end via LAVA on armv8 FVPs using 14.04 hwpack.
What I see is that LAVA fails prematurely stating that
Traceback (most recent call last):
File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/actions/boot_control.py", line 105, in run
client.boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 508, in boot_linaro_image
self._boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 481, in _boot_linaro_image
self.proc = self.target_device.power_on()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 325, in power_on
self._check_needed_files()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 188, in _check_needed_files
self.config.simulator_bl1)RuntimeError: No SECURE FLASHLOADER found, u'bl1.bin'
Looking at the 14.04 hwpack I see that now there are 2 sets of bl* images there one for FVPs and another for Juno.
LAVA default config has
simulator_bl1 = bl1.bin
simulator_bl2 = [bB][lL]2.*.bin
simulator_bl3 = [bB][lL]31.*.bin
simulator_uefi = uefi_fvp-base.bin
fastmodel.py
# SECURE FLASHLOADERs are needed only for base and cortex models
if self._bl1 is None and self.config.simulator_bl1:
raise RuntimeError('No SECURE FLASHLOADER found, %r' % self.config.simulator_bl1)which I guess gets ‘checked’ by lava whether the needed files present and results in a runtime failure.
With the latest hardware packs, can you please confirm how does the bl images get unpacked?
Are they in some new folder structure because of platform variants are present?
Due to confusion over the changes to TF, the experience isn't as smooth as I'd hoped.
In the disk image, they will be both in the boot partition and in the fvp sub-dir. They are also now prefixed with "fvp_". Hence why the instructions have the extra step to "ln -s fvp_bl1.bin bl1.bin", etc... or fvp_bl1.bin will not be able to load bl2.bin or bl31.bin via semihosting. In future, they should only be in the fvp sub-dir.
We are still using the old TF for FVP. 14.05 should get the latest version available now on github, assuming it passes my tests properly. Then we will also have the fvp_fip.bin file and semihosting will not be used at all.
Using the symlink workaround is fine for manual testing. It doesn't resolve the bug in LAVA. FYI, it's a known issue: https://bugs.launchpad.net/lava-dispatcher/+bug/1310495
Please advice.
For the moment I could not get LAVA to work with the latest hardware pack contents due to this issue.
Thanks
Basil Eljuse…
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
linaro-validation mailing list linaro-validation@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-validation
Thanks Fathi for the pointer.
For the time being we are planning to redefine the bl* file name definitions in LAVA default config file, and since we override hwpack bl* images with ours using the semihosting dir and overriding the FVP arguments with the right path, I am hoping this could be worked around.
Shall update if it goes well.
I believe this re-iterates the fact that we might need some easy configurability of these file names also via json files for LAVA.
Thanks Basil Eljuse...
-----Original Message----- From: Fathi Boudra [mailto:fathi.boudra@linaro.org] Sent: 28 April 2014 13:47 To: ryan.harkin@linaro.org Cc: Basil Eljuse; Linaro Validation; Dean Arnold Subject: Re: [Linaro-validation] Question on 14.04 hwpack bl* image names...
Hi
On 28 April 2014 14:46, Ryan Harkin ryan.harkin@linaro.org wrote:
Hi Basil,
On 28 April 2014 12:07, Basil Eljuse Basil.Eljuse@arm.com wrote:
Hi Ryan,
I was trying to rerun some tests at my end via LAVA on armv8 FVPs using 14.04 hwpack.
What I see is that LAVA fails prematurely stating that
Traceback (most recent call last):
File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/actions/boot_control.py", line 105, in run
client.boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 508, in boot_linaro_image
self._boot_linaro_image()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py", line 481, in _boot_linaro_image
self.proc = self.target_device.power_on()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 325, in power_on
self._check_needed_files()File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py", line 188, in _check_needed_files
self.config.simulator_bl1)RuntimeError: No SECURE FLASHLOADER found, u'bl1.bin'
Looking at the 14.04 hwpack I see that now there are 2 sets of bl* images there one for FVPs and another for Juno.
LAVA default config has
simulator_bl1 = bl1.bin
simulator_bl2 = [bB][lL]2.*.bin
simulator_bl3 = [bB][lL]31.*.bin
simulator_uefi = uefi_fvp-base.bin
fastmodel.py
# SECURE FLASHLOADERs are needed only for base and cortex models
if self._bl1 is None and self.config.simulator_bl1:
raise RuntimeError('No SECURE FLASHLOADER found, %r' % self.config.simulator_bl1)which I guess gets ‘checked’ by lava whether the needed files present and results in a runtime failure.
With the latest hardware packs, can you please confirm how does the bl images get unpacked?
Are they in some new folder structure because of platform variants are present?
Due to confusion over the changes to TF, the experience isn't as smooth as I'd hoped.
In the disk image, they will be both in the boot partition and in the fvp sub-dir. They are also now prefixed with "fvp_". Hence why the instructions have the extra step to "ln -s fvp_bl1.bin bl1.bin", etc... or fvp_bl1.bin will not be able to load bl2.bin or bl31.bin via semihosting. In future, they should only be in the fvp sub-dir.
We are still using the old TF for FVP. 14.05 should get the latest version available now on github, assuming it passes my tests properly. Then we will also have the fvp_fip.bin file and semihosting will not be used at all.
Using the symlink workaround is fine for manual testing. It doesn't resolve the bug in LAVA. FYI, it's a known issue: https://bugs.launchpad.net/lava-dispatcher/+bug/1310495
Please advice.
For the moment I could not get LAVA to work with the latest hardware pack contents due to this issue.
Thanks
Basil Eljuse…
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
linaro-validation mailing list linaro-validation@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-validation
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782
Hi Basil; We have an effort underway to add support in LAVA to find the files in their current hiding places and hope to have this included soon.
In the future, when an image changes, I'm hoping that all the downstream projects have time to adjust and all future Linaro images should work with a production released version of LAVA at the time the image releases. Sorry for the confusion on this release.
I've added an agenda item to our next sync so we can cover our progress and talk further on your needs for configurable options in JSON, and I'd like to discuss your experience with the file overlay support.
Thanks again Alan
On 28 April 2014 07:10, Basil Eljuse Basil.Eljuse@arm.com wrote:
Thanks Fathi for the pointer.
For the time being we are planning to redefine the bl* file name definitions in LAVA default config file, and since we override hwpack bl* images with ours using the semihosting dir and overriding the FVP arguments with the right path, I am hoping this could be worked around.
Shall update if it goes well.
I believe this re-iterates the fact that we might need some easy configurability of these file names also via json files for LAVA.
Thanks Basil Eljuse...
-----Original Message----- From: Fathi Boudra [mailto:fathi.boudra@linaro.org] Sent: 28 April 2014 13:47 To: ryan.harkin@linaro.org Cc: Basil Eljuse; Linaro Validation; Dean Arnold Subject: Re: [Linaro-validation] Question on 14.04 hwpack bl* image names...
Hi
On 28 April 2014 14:46, Ryan Harkin ryan.harkin@linaro.org wrote:
Hi Basil,
On 28 April 2014 12:07, Basil Eljuse Basil.Eljuse@arm.com wrote:
Hi Ryan,
I was trying to rerun some tests at my end via LAVA on armv8 FVPs using 14.04 hwpack.
What I see is that LAVA fails prematurely stating that
Traceback (most recent call last):
File
"/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/actions/boot_control.py",
line 105, in run
client.boot_linaro_image()File
"/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py",
line 508, in boot_linaro_image
self._boot_linaro_image()File
"/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/client/base.py",
line 481, in _boot_linaro_image
self.proc = self.target_device.power_on()File
"/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py",
line 325, in power_on
self._check_needed_files()File
"/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.03-17-g9381215/lava_dispatcher/device/fastmodel.py",
line 188, in _check_needed_files
self.config.simulator_bl1)RuntimeError: No SECURE FLASHLOADER found, u'bl1.bin'
Looking at the 14.04 hwpack I see that now there are 2 sets of bl*
images
there one for FVPs and another for Juno.
LAVA default config has
simulator_bl1 = bl1.bin
simulator_bl2 = [bB][lL]2.*.bin
simulator_bl3 = [bB][lL]31.*.bin
simulator_uefi = uefi_fvp-base.bin
fastmodel.py
# SECURE FLASHLOADERs are needed only for base and cortex models
if self._bl1 is None and self.config.simulator_bl1:
raise RuntimeError('No SECURE FLASHLOADER found, %r' % self.config.simulator_bl1)which I guess gets ‘checked’ by lava whether the needed files present
and
results in a runtime failure.
With the latest hardware packs, can you please confirm how does the bl images get unpacked?
Are they in some new folder structure because of platform variants are present?
Due to confusion over the changes to TF, the experience isn't as smooth
as
I'd hoped.
In the disk image, they will be both in the boot partition and in the fvp sub-dir. They are also now prefixed with "fvp_". Hence why the instructions have the extra step to "ln -s fvp_bl1.bin bl1.bin", etc...
or
fvp_bl1.bin will not be able to load bl2.bin or bl31.bin via semihosting. In future, they should only be in the fvp sub-dir.
We are still using the old TF for FVP. 14.05 should get the latest
version
available now on github, assuming it passes my tests properly. Then we
will
also have the fvp_fip.bin file and semihosting will not be used at all.
Using the symlink workaround is fine for manual testing. It doesn't resolve the bug in LAVA. FYI, it's a known issue: https://bugs.launchpad.net/lava-dispatcher/+bug/1310495
Please advice.
For the moment I could not get LAVA to work with the latest hardware
pack
contents due to this issue.
Thanks
Basil Eljuse…
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy
the
information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1
9NJ,
Registered in England & Wales, Company No: 2548782
linaro-validation mailing list linaro-validation@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-validation
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2557590 ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ, Registered in England & Wales, Company No: 2548782 _______________________________________________ linaro-validation mailing list linaro-validation@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-validation
linaro-validation@lists.linaro.org
-
Alan Bennett -
Basil Eljuse -
Fathi Boudra -
Ryan Harkin