Re: userdef.exp regression for ppc?

Re: userdef.exp regression for ppc?

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 439 bytes --]

> It's not ringing any bells.
> 
> My last tests on native i686-pc-linux-gnu were working on 2002-05-29.
> 
> Is there anything obvious in gdb.log, like the debugger crashing
> or getting out of sync at the beginning of the tests?

Nothing like that but it is obviously broken - see attached. I think its 
abi related but nothing directly to do with C++ (no C++ changes were 
made and notice how all the new C++ tests pass! :-)).

Andrew


[-- Attachment #2: gdb.log --]
[-- Type: text/plain, Size: 2968 bytes --]

(gdb) PASS: gdb.c++/userdef.exp: up from marker1
print one + two
$1 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one + two
print one - two
$2 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one - two
print one * two
$3 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one * two
print one / two
$4 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one / two
print one % two
$5 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one % two
print one && two
$6 = 1
(gdb) PASS: gdb.c++/userdef.exp: print one && two
print one || two
$7 = 1
(gdb) PASS: gdb.c++/userdef.exp: print one || two
print one & two
$8 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one & two
print one | two
$9 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one | two
print one ^ two
$10 = {x = 2147479536, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one ^ two
print one < two
$11 = 1
(gdb) PASS: gdb.c++/userdef.exp: print one < two
print one <= two
$12 = 1
(gdb) PASS: gdb.c++/userdef.exp: print one <= two
print one > two
$13 = 0
(gdb) PASS: gdb.c++/userdef.exp: print one > two
print one >= two
$14 = 0
(gdb) PASS: gdb.c++/userdef.exp: print one >= two
print one == two
$15 = 0
(gdb) PASS: gdb.c++/userdef.exp: print one == two
print one != two
$16 = 1
(gdb) PASS: gdb.c++/userdef.exp: print one != two
print one << 31
$17 = {x = 25168188, y = 1099239424}
(gdb) PASS: gdb.c++/userdef.exp: print one << 31
print one >> 31
$18 = {x = 25168188, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one >> 31
print !one
$19 = 0
(gdb) PASS: gdb.c++/userdef.exp: print !one
print ~one
$20 = {x = 25168188, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print ~one
print -one
$21 = {x = 25168188, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print -one
print one++
$22 = {x = 25168188, y = 1099239425}
(gdb) FAIL: gdb.c++/userdef.exp: print one++
print ++one
$23 = {x = 25168189, y = 1099239425}
(gdb) FAIL: gdb.c++/userdef.exp: print ++one
print one--
$24 = {x = 25168189, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print one--
print --one
$25 = {x = 25168188, y = 1099239424}
(gdb) FAIL: gdb.c++/userdef.exp: print --one
print one += 7
$26 = {x = -2122322940, y = -1048251704}
(gdb) FAIL: gdb.c++/userdef.exp: print one += 7
print two = one
$27 = {x = 2, y = 3}
(gdb) FAIL: gdb.c++/userdef.exp: print two = one
break A1::'operator+'
Breakpoint 3 at 0x1800d5c: file /home/scratch/GDB/src/gdb/testsuite/gdb.c++/userdef.cc, line 58.
(gdb) PASS: gdb.c++/userdef.exp: break A1::'operator+'
break A1::'operator +'
Note: breakpoint 3 also set at pc 0x1800d5c.
Breakpoint 4 at 0x1800d5c: file /home/scratch/GDB/src/gdb/testsuite/gdb.c++/userdef.cc, line 58.
(gdb) PASS: gdb.c++/userdef.exp: break A1::'operator +'
testcase /home/scratch/GDB/src/gdb/testsuite/gdb.c++/userdef.exp completed in 1 seconds

Re: userdef.exp regression for ppc?

[Michael Elizabeth Chastain]

This looks like a bug in calling functions that return a structure
by value.  All the functions that return structures FAIL:

  print one + two
  $1 = {x = 2147479536, y = 1099239424}
  (gdb) FAIL: gdb.c++/userdef.exp: print one + two

The functions that return a scalar PASS:

  print one && two
  $6 = 1
  (gdb) PASS: gdb.c++/userdef.exp: print one && two

Look at the results in gdb.base/call-rt-st.exp and gdb.base/structs.exp.
I bet there is a lot of juicy log info there.

Michael C

Re: userdef.exp regression for ppc?

[Michael Snyder]

Michael Elizabeth Chastain wrote:
> 
> This looks like a bug in calling functions that return a structure
> by value.  All the functions that return structures FAIL:

That could be checked by looking at the results of structs.exp, 
callfuncs.exp, and to some extent call-ar-st and call-rt-st.

> 
>   print one + two
>   $1 = {x = 2147479536, y = 1099239424}
>   (gdb) FAIL: gdb.c++/userdef.exp: print one + two
> 
> The functions that return a scalar PASS:
> 
>   print one && two
>   $6 = 1
>   (gdb) PASS: gdb.c++/userdef.exp: print one && two
> 
> Look at the results in gdb.base/call-rt-st.exp and gdb.base/structs.exp.
> I bet there is a lot of juicy log info there.
> 
> Michael C

userdef.exp regression for ppc?

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 405 bytes --]

(yes I know, it's html)

I've just noticed an up-surge in userdef.exp test failures.  Any idea's? 
   userdef.exp hasn't been changed and nothing jumps out (except ppc 
netbsd changes :-).

NetBSD/PPC m/c with gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)

The attached gives the results for the last few days (cvs update -D ... 
gdb).  The gap in days is due to failed builds :-(

enjoy,
Andrew

[-- Attachment #2: diff.html --]
[-- Type: text/html, Size: 19394 bytes --]