[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Generating code coverage reports
From: |
Ludovic Courtès |
Subject: |
Re: Generating code coverage reports |
Date: |
Mon, 03 Nov 2008 14:06:28 +0100 |
User-agent: |
Gnus/5.11 (Gnus v5.11) Emacs/22.3 (gnu/linux) |
Hi Simon,
Simon Josefsson <address@hidden> writes:
> Hi Ludovic. Thanks, I pushed this patch. Adding anything to LDFLAGS
> doesn't seem to be required though?
According to the doc (info "(gcc)Debugging Options") it is required.
See also this example (with GCC 4.2.4):
$ cat > t.c <<EOF
int main (int argc, char *argv[]) { return 0; }
EOF
$ gcc --coverage -c t.c
$ gcc t.o
t.o: In function `global constructors keyed to 0_main':
t.c:(.text+0x43): undefined reference to `__gcov_init'
t.o:(.data+0x24): undefined reference to `__gcov_merge_add'
collect2: ld returned 1 exit status
$ gcc --coverage t.o
[ success ]
> Btw, it is better/worse to compile with -O2? It may be better for
> profiling optimizations, but the use-case here is to generate code
> coverage reports for the self tests. So I would assume -O2 is harmful
> here. Whether -g is useful or not is another question.
I don't think `-O' matters. I don't think `-g' is required to get
useful source location information.
Thanks,
Ludo'.