[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#998177: marked as done (gnat: segfault in tree_could_trap_p when C double parameter, -O2 -gnatn -gnatVa (regression))

Your message dated Fri, 20 Oct 2023 22:00:08 +0200
with message-id <ZTLcSFG_jsm2HISM@pegase>
and subject line close 998177
has caused the Debian Bug report #998177,
regarding gnat: segfault in tree_could_trap_p when C double parameter, -O2 -gnatn -gnatVa (regression)
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
998177: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=998177
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: gnat-11
Version: 11.2.0-10
Severity: normal

Hello.

The following valid Ada sources crash 'gcc-11 -O2 -gnatn -gnatVa' as shown below.
All goes well with gcc-10, with -O1, without inlining or without validity checks.

cat > p.ads <<EOF
package P is
   procedure P3;
end P;
EOF
cat > p.adb <<EOF
with Interfaces.C;
package body P is
   procedure P1 (A : Interfaces.C.double) is null;
   procedure P2 (A : Interfaces.C.double) is
   begin
      P1 (A);
   end P2;
   procedure P3 is
   begin
      P2 (0.0);
   end P3;
end P;
EOF
gdb /usr/lib/gcc/x86_64-linux-gnu/11/gnat1
(gdb) run -O2 -dumpbase p.adb -gnatn -gnatVa p.adb
Starting program: /usr/lib/gcc/x86_64-linux-gnu/11/gnat1 -O2 -dumpbase p.adb -gnatn -gnatVa p.adb
[...]
P.P1 P.P2 P.P3 System.Fat_Lflt.Attr_Long_Float.Valid
Analyzing compilation unit
Performing interprocedural optimizations
<*free_lang_data> {heap 2040k} <visibility> {heap 2040k} <build_ssa_passes> {heap 2040k} <opt_local_passes> {heap 2040k} <remove_symbols> {heap 2936k} <targetclone> {heap 2936k} <free-fnsummary> {heap 2936k}Streaming LTO
<whole-program> {heap 2936k} <profile_estimate> {heap 2936k} <icf> {heap 2936k} <devirt> {heap 2936k} <cp> {heap 2936k} <sra> {heap 2936k} <fnsummary> {heap 2936k} <inline> {heap 2936k} <pure-const> {heap 2936k} <modref> {heap 2936k} <fre\
e-fnsummary> {heap 2936k} <static-var> {heap 2936k} <single-use> {heap 2936k} <comdats> {heap 2936k}Assembling functions:
<simdclone> {heap 2936k} P.P3
Program received signal SIGSEGV, Segmentation fault.
0x000000000119a4f6 in tree_could_trap_p(tree_node*) ()

--- End Message ---
--- Begin Message --- 
Upstream has never been able to reproduce this issue (see the
forwarded bug).
Now, version 11.4.0-4 in Debian cannot reproduce it either.
We’ll never know.

--- End Message ---
Reply to: