This is the mail archive of the gdb-testers@sources.redhat.com 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]

Re: gdb, native i686-pc-linux-gnu


On Fri, Feb 27, 2004 at 04:54:55PM -0500, Michael Chastain wrote:
> . Highlights of This Spin
> 
>   I added gdb drow_intercu-20040221-branch.  There is one regression
>   in gdb.cp/namespace.exp, one significant improvement, and one
>   improvement which looks accidental.

Thanks.

>       . gdb.cp/namespace.exp: ptype CClass::NestedClass
>           PASS -> FAIL
> 
>           This happened with gcc gcc-3_4-branch, gcc HEAD, and gcc
>           tree-ssa-20020619-branch with dwarf-2.  This looks like a
>           regression in gdb drow_intercu-20040221-branch.
> 
>             # gdb.log gdb HEAD
>             ptype CClass::NestedClass
>             type = class C::CClass::NestedClass {
>               public:
>                 int y;
>             }
>             (gdb) PASS: gdb.cp/namespace.exp: ptype CClass::NestedClass
> 
>             # gdb.log gdb drow_intercu-20040221-branch
>             (gdb) FAIL: gdb.cp/namespace.exp: ptype CClass::NestedClass
>             ptype NestedClass
>             No symbol "NestedClass" in current context.

i.e. not with GCC 3.3, which I tested with.  Thanks, will investigate.

>       . gdb.cp/templates.exp: ptype Baz
>           FAIL -> KFAIL
> 
>           This happened with gcc 3.2-7-rh, gcc 3.3.3, and gcc
>           gcc-3_3-branch with dwarf-2.
> 
>             # gdb.log gdb HEAD
>             ptype Baz
>             type = class Baz<int, (char)115> {
> 
>             # gdb.log gdb drow_intercu-20040221-branch
>             ptype Baz
>             type = class Baz<int, 115> {
> 
>           This looks like more like a very recent bug in gdb HEAD rather
>           than a bug fix in gdb drow_intercu-20040221-branch.

It's a deliberate change in the output of the demangler.

>       . gdb.mi/mi-var-display.exp: assing to variable bar
>         gdb.mi/mi-var-display.exp: create local variable bar
>         gdb.mi/mi-var-display.exp: eval variable bar
>         gdb.mi/mi-var-display.exp: eval variable bar with new format
>         gdb.mi/mi-var-display.exp: eval variable bar with new value
>         gdb.mi/mi-var-display.exp: info type variable bar
>         gdb.mi/mi2-var-display.exp: assing to variable bar
>         gdb.mi/mi2-var-display.exp: create local variable bar
>         gdb.mi/mi2-var-display.exp: eval variable bar
>         gdb.mi/mi2-var-display.exp: eval variable bar with new format
>         gdb.mi/mi2-var-display.exp: eval variable bar with new value
>         gdb.mi/mi2-var-display.exp: info type variable bar
>           FAIL -> PASS
> 
>           gdb drow_intercu-20040221-branch is better than gdb HEAD.
>           This happened with gcc 2.95.3 -gdwarf-2.

That's really bizarre.  I have no idea why.


-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer


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