Iced Quinn

> would you perhaps be able to set a breakpoint at makeSynthMsgWithTags and step through the call to see where it is going wrong? it always segfaults on the first...

> how are you building Debug build, unistd.h patch. > GCC version gcc version 10.2.0 (Alpine 10.2.0)

> Dump of assembler code for function addMsgSlot(big_scpacket*, PyrSlot*): 0x00005555556477ee : push %rbx 0x00005555556477ef : sub $0x10050,%rsp => 0x00005555556477f6 : mov %rdi,0x8(%rsp) 0x00005555556477fb : mov %rsi,(%rsp) 0x00005555556477ff : mov %fs:0x28,%rax...

Trashed ~/.ccache, added -fstack-protector-all to build flags and did a rebuild. Does not appear to have changed anything.

https://stackoverflow.com/questions/10499400/sigsegv-when-entering-a-function

 

i tried `ulimit -s` to increase the stack size and it did not help, although i may have used it wrong. there is also the `getrlimit` and `setrlimit` posix calls...

well there either needs to be a mode for big packets to be on the heap, or some kind of `setrlimit` call or equivalent somewhere to grow the stack, otherwise...

made a little tester app finally. ```c #include #include #include int main() { struct rlimit lim; getrlimit(RLIMIT_STACK, &lim); printf("%lld x %lld\n", lim.rlim_cur, lim.rlim_max); } ``` which says ``` 8388608 x...

i have come back with more science. asked around and Kubuntu 18 as well as Slackware Current have stack sizes of 8mb as well. I thought it was quite odd...