[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug ld/609] New: --fatal-warnings behaves badly with any other printing
Tomer dot Levi at nsc dot com
[Bug ld/609] New: --fatal-warnings behaves badly with any other printing option
16 Dec 2004 11:25:12 -0000
The linker's --fatal-warnings option behaves badly when invoked with any other
printing option (e.g., --verbose) and especially with tracing (--trace). It's
quite easy to see why:
Almost all printing (warnings, but also tracing and verbosing) is done
using 'info_msg' function, which then calls 'vfinfo' to do the job. 'vfinfo'
wrongly assumes it is called to print only warnings, thus if requested to
treat warnings as errors (--fata-warnings is specified), don't create an
executable (config.make_executable = FALSE). So 'ld --fatal-warnings --
verbose' will never create an executable (even if no warnings occured), and
even won't tell the user about it. The behavior is different (and not in a
better way) with '--fatal-warnings --trace', because of (uncleared) code
in 'main' function that will pring the following message, even if no errors
occured :"link errors found, deleting executable".
To reproduce the problem, take a simple "Hello, world" example and use the
latest development linker to reproduce this problem.
Summary: --fatal-warnings behaves badly with any other printing
AssignedTo: unassigned at sources dot redhat dot com
ReportedBy: Tomer dot Levi at nsc dot com
CC: bug-binutils at gnu dot org
GCC host triplet: i686-pc-linux-gnu
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.
|[Prev in Thread]
||[Next in Thread]|
- [Bug ld/609] New: --fatal-warnings behaves badly with any other printing option,
Tomer dot Levi at nsc dot com <=