I'm having issues with reproducible builds on kirkstone when comparing a build driven by bitbake compared to a build driven from a yocto sdk. Can I expect reproducible build between bitbake builds and sdk builds? Some produced binaries have minor differences, but larger c++ binaries show quite huge diff. One common difference is that the ELF header seems to have different OS/ABI between the two builds. Can binutils on different host give such a result?

diffoscope

岫岳 File has been modified after NT_GNU_BUILD_ID has been applied.
念岸岸 readelf --wide --file-header {}
岫 @@ -1,20 +1,20 @@
岫  ELF Header:
岫 -  Magic:   7f 45 4c 46 01 01 01 03 00 00 00 00 00 00 00 00
岫 +  Magic:   7f 45 4c 46 01 01 01 00 00 00 00 00 00 00 00 00
岫    Class:                             ELF32
岫    Data:                              2's complement, little endian
岫    Version:                           1 (current)
岫 -  OS/ABI:                            UNIX - GNU
岫 +  OS/ABI:                            UNIX - System V
岫    ABI Version:                       0
岫    Type:                              DYN (Position-Independent Executable file)
岫    Machine:                           ARM
岫    Version:                           0x1
岫    Entry point address:               0x1f04
岫    Start of program headers:          52 (bytes into file)
岫 -  Start of section headers:          41316 (bytes into file)
岫 +  Start of section headers:          41328 (bytes into file)


Thanks,
Kenth