Sam James
Sam James
I got another failure with GCC 14 (which makes `-Wincompatible-pointer-types`) an error by default, but I'm not familiar enough with this codebase to fix it correctly: ``` dir.c:203:44: error: passing...
My best guess would be musl's malloc. Try shoving in mimalloc or jemalloc and see if it gets better?
Can you fix the issues CI reported please?
My memory of those bugs from last year is completely gone, but wonder if a fix needed to be backported to 2.19.x from 3.x?
Trying it now. It's possible I was just very unlucky. I can't hit it very often, and when I ran in a loop, for 3.3, all I hit was the...
No luck! I double checked the patch was applied. ``` Test run 7/100000 complete ran 6959 tests in 1.76 sec all tests ok tls: /usr/lib/gcc/x86_64-pc-linux-gnu/13/include/g++-v13/bits/stl_vector.h:1125: std::vector::reference std::vector::operator[](size_type) [with _Tp =...
Maybe I've made a mistake (or missing something obvious): ``` (gdb) frame 6 #6 Botan::TLS::TLS_CBC_HMAC_AEAD_Encryption::finish (this=0x7f1026808580, buffer=std::vector of length 0, capacity 0, offset=5) at src/lib/tls/tls_cbc/tls_cbc.cpp:209 209 copy_mem(&buffer[offset], msg().data(), msg_size); (gdb)...
I'd completely forgot about this but not seeing it anymore -- let's assume those patches fixed it.
Huh, I hit this one before in https://github.com/randombit/botan/issues/3623#issuecomment-1632113780.
Sorry, I'd forgot about this. I haven't hit it recently with that included. Thanks!