Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below error message when start: "Fatal Error: No CreateCADIBroker entry point found. Not a CADI 2.0 model ?"
Anyone knows what this message means? Do I need any additional configuration to bring up the fastmodel over 64bit ubuntu?
Thanks, Lei
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
Dave
On 14 Sep 2012, at 08:40, Lei Wen wrote:
Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below error message when start: "Fatal Error: No CreateCADIBroker entry point found. Not a CADI 2.0 model ?"
Anyone knows what this message means? Do I need any additional configuration to bring up the fastmodel over 64bit ubuntu?
Thanks, Lei _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott dave.pigott@linaro.org wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.soboot-wrapper/linux-system-semi.axf -f params
And it is interested that while I just run the "./isim_system", there is no such error report out. But I don't know to use isim_system to launch this simulation...
Dave
On 14 Sep 2012, at 08:40, Lei Wen wrote:
Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below error
message when start:
"Fatal Error: No CreateCADIBroker entry point found. Not a CADI 2.0
model ?"
Anyone knows what this message means? Do I need any additional
configuration to bring up
the fastmodel over 64bit ubuntu?
Thanks, Lei _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Thanks, Lei
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott dave.pigott@linaro.org wrote: Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so boot-wrapper/linux-system-semi.axf -f params
And it is interested that while I just run the "./isim_system", there is no such error report out. But I don't know to use isim_system to launch this simulation...
Dave
On 14 Sep 2012, at 08:40, Lei Wen wrote:
Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below error message when start: "Fatal Error: No CreateCADIBroker entry point found. Not a CADI 2.0 model ?"
Anyone knows what this message means? Do I need any additional configuration to bring up the fastmodel over 64bit ubuntu?
Thanks, Lei _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Thanks, Lei
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei, We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
And it is interested that while I just run the "./isim_system", there is no such error report out. But I don't know to use isim_system to launch this simulation...
Dave On 14 Sep 2012, at 08:40, Lei Wen wrote: > Hi, > > Does anyone try to run the fastmodel over ubuntu 64bit machine? > I try to run the simulation over Cortex-A15 model, but get below error message when start: > "Fatal Error: No CreateCADIBroker entry point found. Not a CADI 2.0 model ?" > > Anyone knows what this message means? Do I need any additional configuration to bring up > the fastmodel over 64bit ubuntu? > > Thanks, > Lei > _______________________________________________ > linaro-dev mailing list > linaro-dev@lists.linaro.org <mailto:linaro-dev@lists.linaro.org> > http://lists.linaro.org/mailman/listinfo/linaro-devThanks, Lei
On Tue, Sep 18, 2012 at 04:58:21PM -0500, Andy Doan wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so
Should you be running model_shell64 instead?
http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
I believe that the pre-built RTSM binaries are integrated simulation builds (i.e., like isim_system). So the required command-line for the isim_system binary built by the fast models package should be similar to the examples we document for launching the RTSM binaries.
Cheers ---Dave
And it is interested that while I just run the "./isim_system", there is no such error report out. But I don't know to use isim_system to launch this simulation...
Dave
On 14 Sep 2012, at 08:40, Lei Wen wrote:
Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below
error message when start:
"Fatal Error: No CreateCADIBroker entry point found. Not a CADI
2.0 model ?"
Anyone knows what this message means? Do I need any additional
configuration to bring up
the fastmodel over 64bit ubuntu?
Thanks, Lei _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org mailto:linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Thanks, Lei
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Adding Amit and Vishal. They go down this road quite a bit.
On 19 September 2012 11:38, Dave Martin dave.martin@linaro.org wrote:
On Tue, Sep 18, 2012 at 04:58:21PM -0500, Andy Doan wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so
Should you be running model_shell64 instead?
http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
I believe that the pre-built RTSM binaries are integrated simulation builds (i.e., like isim_system). So the required command-line for the isim_system binary built by the fast models package should be similar to the examples we document for launching the RTSM binaries.
Cheers ---Dave
And it is interested that while I just run the "./isim_system", there is no such error report out. But I don't know to use isim_system to launch this simulation...
Dave
On 14 Sep 2012, at 08:40, Lei Wen wrote:
Hi,
Does anyone try to run the fastmodel over ubuntu 64bit machine? I try to run the simulation over Cortex-A15 model, but get below
error message when start:
"Fatal Error: No CreateCADIBroker entry point found. Not a CADI
2.0 model ?"
Anyone knows what this message means? Do I need any additional
configuration to bring up
the fastmodel over 64bit ubuntu?
Thanks, Lei _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org mailto:linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Thanks, Lei
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
Hi Dave,
On Thu, Sep 20, 2012 at 12:38 AM, Dave Martin dave.martin@linaro.orgwrote:
On Tue, Sep 18, 2012 at 04:58:21PM -0500, Andy Doan wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so
Should you be running model_shell64 instead?
The model_shell64 also complain for the same error as "Not a CADI 2.0 model"...
http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
I believe that the pre-built RTSM binaries are integrated simulation builds (i.e., like isim_system). So the required command-line for the isim_system binary built by the fast models package should be similar to the examples we document for launching the RTSM binaries.
Yes, it is similar. Thanks for the hint.
Cheers ---Dave
Thanks,
Lei
Dave,
On Thu, Sep 20, 2012 at 10:10 AM, Lei Wen adrian.wenl@gmail.com wrote:
Hi Dave,
On Thu, Sep 20, 2012 at 12:38 AM, Dave Martin dave.martin@linaro.orgwrote:
On Tue, Sep 18, 2012 at 04:58:21PM -0500, Andy Doan wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so
Should you be running model_shell64 instead?
The model_shell64 also complain for the same error as "Not a CADI 2.0 model"...
http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
I believe that the pre-built RTSM binaries are integrated simulation builds (i.e., like isim_system). So the required command-line for the isim_system binary built by the fast models package should be similar to the examples we document for launching the RTSM binaries.
Yes, it is similar. Thanks for the hint.
Here is another issue, although the simulation is ok for running A15x1, but when I am trying to run the A15x4_A7x4, the simulation cannot be launched up, with the same configuration I could run at the 32bit machine.
The screen snap shot could be found at: https://lh4.googleusercontent.com/-M8G9kBQztNI/UFqGMGqetNI/AAAAAAAAACg/_1JGg...
And the launch command I am using: ./isim_system -a coretile.cluster0.*=boot-wrapper/linux-system-semi.axf -a coretile.cluster1.*=boot-wrapper/linux-system-semi.axf -f params Where the params is a file, containing: motherboard.smsc_91c111.enabled=1 motherboard.hostbridge.userNetworking=1 coretile.dualclustersystemconfigurationblock.CFG_ACTIVECLUSTER=0x3 coretile.cluster0.cpu0.semihosting-cmd_line="--kernel uImage --fdt rtsm_ve-v2p-ca15x4-ca7x4.dtb -- console=ttyAMA0 mem=2048M root=/dev/nfs nfsroot=192.168.1.1:/srv/nfsroot/ rw ip=dhcp"
Any idea?
Thanks, Lei
On Thu, Sep 20, 2012 at 11:07 AM, Lei Wen adrian.wenl@gmail.com wrote:
Dave,
On Thu, Sep 20, 2012 at 10:10 AM, Lei Wen adrian.wenl@gmail.com wrote:
Hi Dave,
On Thu, Sep 20, 2012 at 12:38 AM, Dave Martin dave.martin@linaro.orgwrote:
On Tue, Sep 18, 2012 at 04:58:21PM -0500, Andy Doan wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org> wrote:
Hi Lei,
We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?
I launch the fast model by below command: model_shell cadi_system_Linux64-Release-GCC-4.4.so
Should you be running model_shell64 instead?
The model_shell64 also complain for the same error as "Not a CADI 2.0 model"...
http://cadi_system_Linux64-Release-GCC-4.4.so/ boot-wrapper/linux-system-semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/Linux64_RTSM_VE_Cortex-A15x4-A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.img -C coretile.cluster0.cpu0.semihosting-enable=1 -C motherboard.hostbridge.userNetworking=1 -C coretile.cache_state_modelled=0 -C motherboard.smsc_91c111.enabled=1
I also follow pages like:
https://wiki.linaro.org/Platform/Android/AndroidOnFastModels
I believe that the pre-built RTSM binaries are integrated simulation builds (i.e., like isim_system). So the required command-line for the isim_system binary built by the fast models package should be similar to the examples we document for launching the RTSM binaries.
Yes, it is similar. Thanks for the hint.
Here is another issue, although the simulation is ok for running A15x1, but when I am trying to run the A15x4_A7x4, the simulation cannot be launched up, with the same configuration I could run at the 32bit machine.
The screen snap shot could be found at:
https://lh4.googleusercontent.com/-M8G9kBQztNI/UFqGMGqetNI/AAAAAAAAACg/_1JGg...
And the launch command I am using: ./isim_system -a coretile.cluster0.*=boot-wrapper/linux-system-semi.axf -a coretile.cluster1.*=boot-wrapper/linux-system-semi.axf -f params Where the params is a file, containing: motherboard.smsc_91c111.enabled=1 motherboard.hostbridge.userNetworking=1 coretile.dualclustersystemconfigurationblock.CFG_ACTIVECLUSTER=0x3 coretile.cluster0.cpu0.semihosting-cmd_line="--kernel uImage --fdt rtsm_ve-v2p-ca15x4-ca7x4.dtb -- console=ttyAMA0 mem=2048M root=/dev/nfs nfsroot=192.168.1.1:/srv/nfsroot/ rw ip=dhcp"
Any idea?
Please ignore this mail, seems this issue is caused by bootwrapper, after applying the patch of "bootwrapper: Get it working for big.LITTLE models", the model is working fine for A15x4_A7x4 over 64bit machine using isim_system method.
However, the A15x4_A7x4 model is running quit slow for launching the kernel, about 1minute to showing the kernel booting screen. Whereas A15x1 model is launching quite fast, just in several seconds.
Anyone has any idea to speed up the A15x4_A7x4 model launching time?
Thanks, Lei
On 20 September 2012 10:33, Lei Wen adrian.wenl@gmail.com wrote:
However, the A15x4_A7x4 model is running quit slow for launching the kernel, about 1minute to showing the kernel booting screen. Whereas A15x1 model is launching quite fast, just in several seconds.
This is in general not very surprising, since the A15x4_A7x4 model is doing eight times as much work (it has eight CPUs to emulate)... If your use case allows you to hold some of the CPUs in reset rather than letting them run, this should make things a bit faster. Also for multi-CPU models make sure you have lots of RAM in your host system and that you are not starting to swap.
-- PMM
Hi Andy,
On Wed, Sep 19, 2012 at 5:58 AM, Andy Doan andy.doan@linaro.org wrote:
On 09/17/2012 04:01 AM, Dave Pigott wrote:
Hi Lei,
I'm copying Andy Doan directly on this, since he's had more experience of getting FM working in LAVA and can probably point you in the right direction.
Thanks
Dave
On 14 Sep 2012, at 14:35, Lei Wen wrote:
Hi Dave,
On Fri, Sep 14, 2012 at 3:58 PM, Dave Pigott <dave.pigott@linaro.org mailto:dave.pigott@linaro.org**> wrote:
Hi Lei, We run Fast Models on 64 bit all the time. How are you creating the model, and how are you trying to launch it?I launch the fast model by below command: model_shell cadi_system_Linux64-Release-**GCC-4.4.sohttp://cadi_system_Linux64-Release-GCC-4.4.so <http://cadi_system_Linux64-**Release-GCC-4.4.so/http://cadi_system_Linux64-Release-GCC-4.4.so/
boot-wrapper/linux-system-**semi.axf -f params
Looks like you do things different than we do. The way we launch things is with:
/opt/arm/RTSM_A15-A7x14_VE/**Linux64_RTSM_VE_Cortex-A15x4-**A7x4/RTSM_VE_Cortex-A15x4-A7x4
-a coretile.cluster0.*=img.axf -C motherboard.mmc.p_mmc_file=sd.**img -C coretile.cluster0.cpu0.**semihosting-enable=1 -C motherboard.hostbridge.**userNetworking=1 -C coretile.cache_state_modelled=**0 -C motherboard.smsc_91c111.**enabled=1
There is no such prebuilt binary in my installation... However I find the isim_system is similar to the model_shell, with the same input I set to model_shell, it could launch the simulation finally.
Still curious to know why model_shell don't work.
Thanks, Lei
-
Andy Doan -
Dave Martin -
Dave Pigott -
Lei Wen -
Peter Maydell -
Zach Pfeffer