diff options
| author | Tom Tromey <tom@tromey.com> | 2017-08-30 16:12:56 -0600 |
|---|---|---|
| committer | Tom Tromey <tom@tromey.com> | 2017-09-06 11:11:03 -0600 |
| commit | 5aec60eb2f6f0905bfc76f5949fd5d55c6a10f10 (patch) | |
| tree | ff5a162d3206c6dec256da07b66554dea91f6830 /gdb/testsuite/gdb.base/sizeof.exp | |
| parent | a102602bc56e418f9794d653ef103a65f08a5a49 (diff) | |
| download | fsf-binutils-gdb-5aec60eb2f6f0905bfc76f5949fd5d55c6a10f10.zip fsf-binutils-gdb-5aec60eb2f6f0905bfc76f5949fd5d55c6a10f10.tar.gz fsf-binutils-gdb-5aec60eb2f6f0905bfc76f5949fd5d55c6a10f10.tar.bz2 | |
Cast char constant to int in sizeof.exp
PR gdb/22010 concerns a regression I introduced with the scalar
printing changes. The bug is that this code in sizeof.exp:
set signof_byte [get_integer_valueof "'\\377'" -1]
can incorrectly compute sizeof_byte. One underlying problem here is
that gdb's C parser doesn't treat a char constant as an int (this is
PR 19973).
However, it seems good to have an immediate fix for the regression.
The simplest is to cast to an int here.
testsuite/ChangeLog
2017-09-05 Tom Tromey <tom@tromey.com>
PR gdb/22010:
* gdb.base/sizeof.exp (check_valueof): Cast char constant to int.
Diffstat (limited to 'gdb/testsuite/gdb.base/sizeof.exp')
| -rw-r--r-- | gdb/testsuite/gdb.base/sizeof.exp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/gdb/testsuite/gdb.base/sizeof.exp b/gdb/testsuite/gdb.base/sizeof.exp index 5d89407..13d36f8 100644 --- a/gdb/testsuite/gdb.base/sizeof.exp +++ b/gdb/testsuite/gdb.base/sizeof.exp @@ -86,7 +86,7 @@ proc check_valueof { exp val } { # Check that GDB and the target agree over the sign of a character. -set signof_byte [get_integer_valueof "'\\377'" -1] +set signof_byte [get_integer_valueof "(int) '\\377'" -1] set signof_char [get_integer_valueof "(int) (char) -1" -1] set signof_signed_char [get_integer_valueof "(int) (signed char) -1" -1] set signof_unsigned_char [get_integer_valueof "(int) (unsigned char) -1" -1] |