john o goyo <jog37@riddermarkfarm.ca> writes:
Greetings, Gaius.
On 2024-04-02 18:56, Gaius Mulley wrote:
Hi John,
I've git pushed the m2 trace flag -fm2-debug-trace=all,quad,line,token
to help track down the ICE
regards,
Gaius
I tried tracing all. Output is voluminous, so I piped it to a file.
Here is an extract therefrom:
./MCPublic.def:39: Statset = SET OF Compilerstatus;
./MCPublic.def:39: Statset = SET OF Compilerstatus;
./MCPublic.def:39: ^^^^^^^
./MCPublic.def:39:4051
./MCPublic.def:39: Statset = SET OF Cotoken: 3
token: 4
token: 5
token: 6
token: 7
[...]
token: 4085
token: 4086
token: 4087
token: 4088
token: 3
token: 4
token: 5
token: 6
token: 7
[...]
token: 1358
token: 1359
token: 1360cc1gm2: internal compiler error: not implemented yet
0x6d35cf m2linemap_internal_error
/home/build/opt/gcc/src/gcc-git/gcc/m2/gm2-gcc/m2linemap.cc:253
As this does not seem to help me, I searched for occurrences of "not
implemented yet" and found them in
gcc/m2/gm2-compiler/SymbolTable.mod. Adding some procedure context,
the problem seems to be in the procedure GetScope. The function
GetPsym returns something not found in the CASE statement.
Hi John,
the compiler is falling over at token 1360 (or just after). Does the
top of the file have the tokens listed? If it does what was the
token sequence, say 1340..1360 ?
It would be interesting to see what declarations (and their
dependencies) it was using (or could be using from that region of code)
regards,
Gaius