Free-PDK

Can you explain the code in your pull request? If I read it correctly then - you add a check to a single MACRO which is only related partially to the problem - you check for SDCC version >= revision with breaking change - also you check if the NEW version __sdcc_external_startup is defined - and if so then emit a warning How should this warn people which do NOT define the new version __sdcc_external_startup? Can you really check in assembly for a function definition with .ifdef? And if so, why no...

Can you explain the code in your pull request? If I read it correctly then - you add a check to a single MACRO which is only related partially to the problem - you check for SDCC version >= revision with breaking change - also you check if the NEW version __sdcc_external_startup is defined - and if so then emit a warning How should this warn people which do NOT define the new version __sdcc_external_startup? Can you really check in assembly for a function definition with .ifdef? And if so, why no...

Can you explain the code in your pull request? If I read it correctly then - you add a check to a single MACRO which is only related partially to the problem - you check for SDCC version >= revision with breaking change - also you check if the NEW version __sdcc_external_startup is defined - and if so then emit a warning How should this warn people which do NOT define the new version __sdcc_external_startup? Can you really check in assembly for a function definition wif .ifdef? And if so, why no...

I see... Right now it is just a REALLY REALLY REALLY bad behaviour from SDCC. It took me some time (I had to look at the disassembly) to find out why every single existing program ever written for PDK was not working when using latest SDCC branch. Breaking every source code written for a platform is one thing... ... but not adding a simple warning in the compiler when the old function name is found but using a self defined function which is in 99.999% of all cases just WRONG for PDK is crazy. Note:...

PDK bad peephole rule

detection / call to _sdcc_external_startup is not working anymore

looks like the assembler emit function for swapc was called with plain op (without the extra options) and therefore the mandatory falg "PDK_OPCODE_ADR_IO" was not present. patch for pdk14 and pdk15 (pdk13 does not have swapc and pdk16 implementation is missing it): % svn diff Index: aspdk14/pdk14mch.c =================================================================== --- aspdk14/pdk14mch.c (revision 13761) +++ aspdk14/pdk14mch.c (working copy) @@ -222,7 +222,7 @@ case S_SWAPC: def.mask = 0x3F; -...

Great idea... I disabled the "exit 1" after the error message in file "support/cpp/gcc/config.gcc" line 3592: *) echo "*** Configuration ${target} not supported" 1>&2 # exit 1 ;; esac Besides a few error messages from gcc's configure the compilation of sdcc worked without a problem: Error messages (which can be ignored): *** Configuration aarch64-apple-darwin21.6.0 not supported grep: ./config/aarch64/aarch64-cores.def: No such file or directory clang: error: no such file or directory: './config/aarch64/aarch64-option-extensions.def'...