Re: New build of Happy Halibut (8.0.1) available for testing

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, 17 Sep 2019 10:59:41 +0900
jonghwan Choi <jhbird.choi@xxxxxxxxx> wrote:

> Hi all
> 
> We found this problem only happening with amd cpu.
> This problem does not occur on intel cpu.
> 
> But if you remove --enable-rdrand from the json recipe, this problem
> doesn't happen with the amd cpu.
> 
> http://199.19.78.20/node/125681
> 
> Thank you for your kind reply.

Hi Jonghwan,

That is a great feedback and a difficult bug to find! It recalls me an
other issue linked to random number kernel generator blocking
because of entropy starvation:
https://jira.automotivelinux.org/browse/SPEC-1655

Have you push the change to the json recipe? I don't see it in gerrit.

Best regards
José Bollo


> 
> 
> Best Regards!
> 
> On Tue, Sep 10, 2019 at 8:29 PM José Bollo <jose.bollo@xxxxxxx> wrote:
> 
> > On Tue, 10 Sep 2019 08:03:14 +0900
> > jonghwan Choi <jhbird.choi@xxxxxxxxx> wrote:
> >  
> > > Hi  José Bollo  
> >
> > Hi, see below.
> >  
> > > If I understand correctly, the program afb-genskel doesn't return
> > > and hangs. This is strange because it works on my configuration.
> > > Can you check the path of the program used (ls -l /proc/PID/exe)
> > > to be sure that afb-genskel of yocto is used.  
> > > -> I checked for additional information in / proc / pid / x.  
> > >
> > > 1. Host PC
> > >
> > > *jhbirdchoi@jhbirdchoi-MS-7B79*:
> > > *~/work/AGL/halibut/build/tmp/log/cooker/raspberrypi3*$
> > > cat /proc/cpuinfo
> > >
> > > processor: 0
> > >
> > > vendor_id: AuthenticAMD
> > >
> > > cpu family: 23
> > >
> > > model: 113
> > >
> > > model name: AMD Ryzen 9 3900X 12-Core Processor
> > >
> > >
> > > 2. *jhbirdchoi@jhbirdchoi-MS-7B79*:*~*$ ls -l /proc/25877/exe
> > >
> > > lrwxrwxrwx 1 jhbirdchoi jhbirdchoi 0  9월 10 07:51
> > > */proc/25877/exe* ->  
> > >  
> > */home/jhbirdchoi/work/AGL/halibut/build/tmp/work/armv7vet2hf-neon-vfpv4-agl-linux-gnueabi/agl-service-can-low-level/8.0.1-r0/recipe-sysroot-native/usr/bin/afb-genskel*  
> > >
> > >
> > > *3. **jhbirdchoi@jhbirdchoi-MS-7B79*:*~*$ file
> > >  
> > /home/jhbirdchoi/work/AGL/halibut/build/tmp/work/armv7vet2hf-neon-vfpv4-agl-linux-gnueabi/agl-service-can-low-level/8.0.1-r0/recipe-sysroot-native/usr/bin/afb-genskel  
> > >
> > >  
> > /home/jhbirdchoi/work/AGL/halibut/build/tmp/work/armv7vet2hf-neon-vfpv4-agl-linux-gnueabi/agl-service-can-low-level/8.0.1-r0/recipe-sysroot-native/usr/bin/afb-genskel:  
> > > ELF 64-bit LSB shared object, x86-64, version 1 (SYSV),
> > > dynamically linked, interpreter /home/jh, for GNU/Linux 3.2.0,
> > > BuildID[sha1]=0d335cc8ae9fc2e83516ff34841a86bcd657f5c9, stripped
> > >
> > >
> > > 4. *jhbirdchoi@jhbirdchoi-MS-7B79*:*~*$ cat /proc/25877/cmdline
> > >
> > >  
> > afb-genskel-3--cpp/home/jhbirdchoi/work/AGL/halibut/build/tmp/work/armv7vet2hf-neon-vfpv4-agl-linux-gnueabi/agl-service-can-low-level/8.0.1-r0/git/low-can-binding/binding/low-can-apidef.json  
> > >
> > >
> > > 5. *jhbirdchoi@jhbirdchoi-MS-7B79*:*~*$ sudo cat /proc/25877/stack
> > >
> > > [<0>] exit_to_usermode_loop+0x61/0x100
> > >
> > > [<0>] prepare_exit_to_usermode+0x83/0x90
> > >
> > > [<0>] retint_user+0x8/0x8
> > >
> > > [<0>] 0xffffffffffffffff  
> >
> > Thanks for the investigation. The run progam is the expected one and
> > the command line is looking good too.
> >
> > The stack is very strange however. Looks like the program terminated
> > but glic hangs.
> >
> > Best wishes
> > José Bollo
> >  
> > > Let's try again after updating linux kernel version or ubuntu.
> > >
> > > Kind Regards!
> > >
> > > thanks.
> > > On Mon, Sep 9, 2019 at 4:46 PM José Bollo <jose.bollo@xxxxxxx>
> > > wrote: 
> > > > On 09/09/2019 09:16, jonghwan Choi wrote:  
> > > > > Hi all  
> > > >
> > > > Hi Jonghwan,
> > > >
> > > > [[snip]]
> > > >  
> > > > > Using top produces the following output:
> > > > >
> > > > >
> > > > >    PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM
> > > > > TIME+ COMMAND
> > > > >
> > > > >
> > > > > *11073 ultra+  20   0    4.3m   0.1m   0.0m R 100.0  0.0
> > > > > 5297:04 afb-genskel
> > > > >                 *
> > > > >
> > > > > *12943 ultra+  20   0    4.3m   0.1m   0.0m R 100.0  0.0
> > > > > 5297:19 afb-genskel        *  
> > > >
> > > > If I understand correctly, the program afb-genskel doesn't
> > > > return and hangs. This is strange because it works on my
> > > > configuration. Can you check the path of the program used (ls
> > > > -l /proc/PID/exe) to be sure that afb-genskel of yocto is used.
> > > >
> > > > Best regards
> > > > José Bollo
> > > >  
> > > > > Is there anything wrong with the host pc?
> > > > >
> > > > > Kind Regards!  
> > > >
> > > >  
> >
> >  

_______________________________________________
automotive-discussions mailing list
automotive-discussions@xxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linuxfoundation.org/mailman/listinfo/automotive-discussions




[Index of Archives]     [LARTC]     [Bugtraq]     [Yosemite Forum]     [Photo]

  Powered by Linux