This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

sim/frv not regeneratable with cgen and guile-1.8.5


I had to find an installation with guile-1.3.4 in order to
regenerate the frv sim.  Maybe a guile/scheme-savvy
person/maintainer can look into it?  I had to give up myself,
sorry.  CGEN and guile..lacks in the debuggability department.

Using guile 1.8.5 (guile-1.8.5-1.fc9.i386 in F 9) and latest
cgen CVS and "make stamp-cpu" in obj/sim/frv yields:

/bin/sh /home/hp/sim/src/sim/frv/../common/cgen.sh cpu-decode /home/hp/sim/src/sim/frv \
		"`if [ -f ../../guile/libguile/guile ]; then echo ../../guile/libguile/guile; else echo guile ; fi` -l /home/hp/sim/src/sim/frv/../../cgen/guile.scm -s" /home/hp/sim/src/sim/frv/../../cgen "-v" \
		frv "with-scache with-profile=fn with-generic-write with-parallel-only" frvbf "" frv,fr550,fr500,fr450,fr400,tomcat,simple "" \
		/home/hp/sim/src/sim/frv/../../cpu/frv.cpu "-S tmp-sem.c1"
Skipping slib/sort, already loaded.
Skipping slib/random, already loaded.
cgen -s /home/hp/sim/src/sim/frv/../../cgen/cgen-sim.scm -s /home/hp/sim/src/sim/frv/../../cgen -v -f "with-scache with-profile=fn with-generic-write with-parallel-only" -m frv,fr550,fr500,fr450,fr400,tomcat,simple -a /home/hp/sim/src/sim/frv/../../cpu/frv.cpu -i all -C tmp-cpu.h1 -U tmp-cpu.c1 -M tmp-mod.c1 -S tmp-sem.c1 -T tmp-dec.h1 -D tmp-dec.c1 -S tmp-sem.c1 
Setting option `with-scache' to "".
Setting option `with-profile' to "fn".
Setting option `with-generic-write' to "".
Setting option `with-parallel-only' to "".
Loading cpu description /home/hp/sim/src/sim/frv/../../cpu/frv.cpu
Including file /home/hp/sim/src/sim/frv/../../cpu/simplify.inc ...
Generating frvbf cpu.h ...
Analyzing instruction set ...
Done analysis.
Computing sformat argument buffers ...
Generating frvbf cpu.c ...
Generating frvbf model.c ...
Generating frvbf semantics.c ...
ERROR: In procedure string-length:
ERROR: Wrong type argument in position 1 (expecting string): SI
No backtrace available.
make[1]: *** [cgen-cpu-decode] Error 1

It looks like something is trigged with the (cgen-semantics.c)
machinery that isn't with (cgen-sem-switch.c); sims using the
latter, like sim/cris, aren't affected.

Those files hadn't been regenerated for a while: I regenerated
them to get a baseline for a sim/cgen/gcc problem I see and
considered the regeneration by itself to be worthwhile of
committing and the issue to be reported.  Of interest is that
the GPL notice still said version 3 and the copyright years
didn't regress. :) Besides the copyright header, the rest was
limited to emitting e.g. "MAKEDI (2147483647, 0xffffffff)"
instead of "9223372036854775807".

No regressions for sim/frv for neither "64-bit host"
(x86_64-unknown-linux-gnu) nor "32-bit host" (i686-pc-linux-gnu)
(there *are* FAILs for the latter, related to the issue I'm
fixing, but no regressions).  Regenerated with "make stamp-arch
stamp-cpu".  Patch elided.

sim/frv:
	* arch.c, arch.h, cpu.c, cpu.h, cpuall.h, decode.c, decode.h,
	model.c, sem.c: Regenerate.

brgds, H-P


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]