David,

no, the problem does not seem the same:
The cog & stack vm that I compiled both end the same, on
testOutOfMemorySignal with
[edwin at ottopedi ~/Smalltalk53/opensmalltalk-vm]$ ./squeak.sh
Killed
[edwin at ottopedi ~/Smalltalk53/opensmalltalk-vm]$
It happens in the cog as in the stack vm.

When discarding the allocation tests,

the stack vm just hangs reporting 1834 test run, but no way to see which
test is failing. I have to kill the vm.

the cog vm ends in a core dump, seemingly during the execution of
#testUnwindDebugger.
On the screen  I see:

C stack backtrace & registers:
don't know how to derive register state from a ucontext_t on this platform
*0x0 <???> at ???
0x25c04a <reportStackState+0xda> at
/usr/home/edwin/Smalltalk53/opensmalltalk-vm/building/linux64x64/squeak.cog.spur/build/squeak
0x25e237 <sigsegv+0x137> at
/usr/home/edwin/Smalltalk53/opensmalltalk-vm/building/linux64x64/squeak.cog.spur/build/squeak
0x800414c6e <_pthread_sigmask+0x53e> at /lib/libthr.so.3
0x0 <???> at ???

I'm sorry, but this is as far as I can go. I can only suppose that FreeBSD,
as configured on my laptop is not compatible enough with Linux, but I'm not
smart enough to find out where the problem is,

Thanks for looking at this, sorry I cannot do more.

Kind regards

Edwin Ancaer


Op vr 14 jan. 2022 om 21:12 schreef stes at PANDORA.BE <stes at telenet.be>:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
>
> Hi,
>
> Out of curiosity , I just tested on cog-spur on OpenIndiana 2021.10;
>
> Open Smalltalk Cog[Spur] VM [CoInterpreterPrimitives
> VMMaker.oscog-eem.3133]
> Unix built on Jan  8 2022 16:23:01 Compiler: 7.5.0  (gcc 7.5)
>
> That is the version that is in the OpenIndiana official repository :
>
>
> http://pkg.openindiana.org/hipster/en/search.shtml?token=cog-spur&action=Search
>
> I select in the cog VM 64bit -> Test Runner
>
>   -> KernelTests-Objects
>   -> AllocationTest (3 tests)
>
> When I run those 3 tests, they pass:
> testOneGigAllocation, testOneMegAllocation, testOutOfMemorySignal
>
> So - currently - on OpenIndiana the 3 tests pass.
>
> No crash, but in the Unix terminal window,
> on stderr/stdout the following is printed:
>
> sqAllocateMemorySegmentOfSizeAboveAllocatedSizeInto mmap: Resource
> temporarily unavailable
> sqAllocateMemorySegmentOfSizeAboveAllocatedSizeInto mmap: Resource
> temporarily unavailable
>
> Do you have the same problem in the stack-spur and cog-spur VM's ?
>
> stack-spur may behave a bit different than the cog VM.
>
> Regards,
> David Stes
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2
>
> iQEcBAEBCAAGBQJh4diVAAoJENdFDkXGicizx14IAJL+FtmANbqpI4dcuq/I3wAq
> UHHXbJLI+1+M+bOV0J/FqorNoC+RmXnLwUJvRc9XVgUXvNN0CLeGwmxwSABMTyor
> fleu996rGIPVdUIlPoW1AKfAPJW9honH+J+iZHgv/BUbUd8SxaWyLtWUQDv7MRns
> M1HB4LLBXOT2tbMO2ATmnYrJ50RUxQnklViYck31cjJ8n/6yRs+NyyEFCedIi5d3
> 0+8XlQ1uUg5CZaNRrNEDzH6S6tkfBFx/71HP/Utx+mqY2r9ckAaOEQnm7t7KHIXh
> ElgkxWtPKdsCXgq3iTtIpo+OybLLwolVZc4ux6dhMGXecGLH8+eSB3GVPlBeeoo=
> =hTag
> -----END PGP SIGNATURE-----
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20220115/1f1a78dc/attachment.html>