This is the mail archive of the
gcc-prs@gcc.gnu.org
mailing list for the GCC project.
driver/9822: The "%c" spec option produces: gcc: spec failure: unrecognized spec option 'c'
- From: bradley-gnu at bradley dot lcs dot mit dot edu
- To: gcc-gnats at gcc dot gnu dot org
- Date: 24 Feb 2003 06:27:23 -0000
- Subject: driver/9822: The "%c" spec option produces: gcc: spec failure: unrecognized spec option 'c'
- Reply-to: bradley-gnu at bradley dot lcs dot mit dot edu
>Number: 9822
>Category: driver
>Synopsis: The "%c" spec option produces: gcc: spec failure: unrecognized spec option 'c'
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: unassigned
>State: open
>Class: doc-bug
>Submitter-Id: net
>Arrival-Date: Mon Feb 24 06:36:00 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator: bradley-gnu at bradley dot lcs dot mit dot edu
>Release: gcc-3.2.1
>Organization:
>Environment:
>Description:
I tried to write a spec file that used the %c option, but it is no longer accepted by the driver. The error is printed by line 5234 of gcc.c.
The gcc.info documentation says that %c is accepted, however:
> `%c'
> Process the `signed_char' spec. This is intended to be used to
> tell cpp whether a char is signed. It typically has the
> definition:
> %{funsigned-char:-D__CHAR_UNSIGNED__}
This is probably a documentation bug, rather than a code bug.
The documentation on the spec files seems quite out-of-date.
For example, I cannot figure out, from the documentation, what does this entry mean in the spec file.
%{fsigned-char&unsigned-char}
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: