Yep, recompiling fdkaac, liblame and libaacplus (along with building
liquidsoap v1.1.1 from source) seems to have fixed the problems.

Now all working nicely!

Thanks for the help.


On 3 October 2013 03:10, Romain Beauxis <to...@rastageeks.org> wrote:

> Have you fixed fdkaac by recompiling it?
>
> Romain
>
>
> 2013/10/2 Matt Camp <m...@noise.net.nz>
>
>> Thanks... recompiling from source seems to have fixed that problem.
>>
>> However, I've now found something new... it seems that whenever I compile
>> with fdkaac enabled, liquidsoap will immediately crash with 'Illegal
>> Instruction'.
>>
>> I only need do 'ocamlfind remove fdkaac' then recompile liquidsoap and it
>> goes back to working fine...
>>
>> Again, this is on the raspberry pi, using the libfdk-aac-dev:armhf
>> and libfdk-aac0:armhf from deb-multimedia.
>>
>> Any ideas? Stack trace is below if that helps (partial)
>>
>> access("/etc/ld.so.nohwcap", F_OK)      = -1 ENOENT (No such file or
>> directory)
>>
>> open("/lib/arm-linux-gnueabihf/librt.so.1", O_RDONLY) = 4
>>
>> read(4,
>> "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0(\0\1\0\0\0\300\26\0\0004\0\0\0"...,
>> 512) = 512
>>
>> lseek(4, 25312, SEEK_SET)               = 25312
>>
>> read(4,
>> "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
>> 1320) = 1320
>>
>> lseek(4, 24924, SEEK_SET)               = 24924
>>
>> read(4,
>> "A0\0\0\0aeabi\0\1&\0\0\0\0056\0\6\6\10\1\t\1\n\2\22\4\24\1\25"..., 49) = 49
>>
>> fstat64(4, {st_mode=S_IFREG|0644, st_size=26632, ...}) = 0
>>
>> mmap2(NULL, 57876, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 4, 0)
>> = 0xb6979000
>>
>> mprotect(0xb697f000, 28672, PROT_NONE)  = 0
>>
>> mmap2(0xb6986000, 8192, PROT_READ|PROT_WRITE,
>> MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 4, 0x5) = 0xb6986000
>>
>> close(4)                                = 0
>>
>> mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
>> = 0xb6f35000
>>
>> mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
>> = 0xb6f34000
>>
>> set_tls(0xb6f344c0, 0xb6f34ba8, 0xb6f3c048, 0xb6f344c0, 0xb6f3c048) = 0
>>
>> mprotect(0xb6986000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb69a4000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6acf000, 8192, PROT_READ)   = 0
>>
>> mprotect(0xb6b1e000, 8192, PROT_READ)   = 0
>>
>> mprotect(0xb6b42000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6b4d000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6b6a000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6bcb000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6c3c000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6cdf000, 4096, PROT_READ)   = 0
>>
>> mprotect(0xb6da3000, 12288, PROT_READ)  = 0
>>
>> mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
>> = 0xb6f33000
>>
>> mprotect(0xb6e77000, 16384, PROT_READ)  = 0
>>
>> mprotect(0xb6efc000, 8192, PROT_READ)   = 0
>>
>> mprotect(0x55000, 4096, PROT_READ)      = 0
>>
>> mprotect(0xb6f3b000, 4096, PROT_READ)   = 0
>>
>> munmap(0xb6f00000, 46890)               = 0
>>
>> set_tid_address(0xb6f34068)             = 3944
>>
>> set_robust_list(0xb6f34070, 0xc)        = 0
>>
>> futex(0xbee3c768, FUTEX_WAIT_BITSET_PRIVATE|FUTEX_CLOCK_REALTIME, 1,
>> NULL, b6b6b000) = -1 EAGAIN (Resource temporarily unavailable)
>>
>> rt_sigaction(SIGRTMIN, {0xb6b5420c, [], SA_SIGINFO|0x4000000}, NULL, 8) =
>> 0
>>
>> rt_sigaction(SIGRT_1, {0xb6b540b4, [], SA_RESTART|SA_SIGINFO|0x4000000},
>> NULL, 8) = 0
>>
>> rt_sigprocmask(SIG_UNBLOCK, [RTMIN RT_1], NULL, 8) = 0
>>
>> getrlimit(RLIMIT_STACK, {rlim_cur=8192*1024, rlim_max=RLIM_INFINITY}) = 0
>>
>> --- SIGILL (Illegal instruction) @ 0 (0) ---
>>
>> +++ killed by SIGILL +++
>>
>> Illegal instruction
>>
>>
>> On 2 October 2013 10:24, Daniel James <daniel.ja...@sourcefabric.org>wrote:
>>
>>> Hi Matt,
>>>
>>> > I used the liquidsoap v1.0.1 package supplied by debian, manually
>>> > recompiled and installed libaacplus, and then manually recompiled
>>> > ocaml-aacplus-0.2.1 (grabbed from src-package from debian-multimedia).
>>>
>>> I would suggest manually compiling Liquidsoap 1.1.1 also, it should tell
>>> you if the libraries are not available. Plus you'll be using the latest
>>> version.
>>>
>>> Cheers!
>>>
>>> Daniel
>>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> October Webinars: Code for Performance
>> Free Intel webinars can help you accelerate application performance.
>> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
>> from
>> the latest Intel processors and coprocessors. See abstracts and register >
>>
>> http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Savonet-users mailing list
>> Savonet-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/savonet-users
>>
>>
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
> from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
> _______________________________________________
> Savonet-users mailing list
> Savonet-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/savonet-users
>
>
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
_______________________________________________
Savonet-users mailing list
Savonet-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/savonet-users

Reply via email to