Peter Eisentraut wrote:
> Michelle Caisse wrote:
>> gcov gets confused when source files are generated. I eliminated
>> src/backend/bootstrap and ../parser from coverage analysis to avoid
>> errors of this type.
>
> The problem with those files is that the source file contains lines like this:
>
> #line 1042 "y.tab.c"
>
> but that source file does not exist, as it is renamed to gram.c.

This problem is now fixed, so the workaround in the coverage patch can
be dropped.


--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Michelle Caisse wrote:
> I've attached a patch that allows the generation of code coverage
> statistics. To test it, apply the patch, then:
>
> autoconf
> ./configure --enable-coverage
> make
> make check (or execute any other application against the database to see
> the coverage of that app)
> make coverage
> make coverage_out

"make clean" does not work for me; it doesn't remove the .gcda and .gcno
files. Apparently the problem is that $(enable_coverage) is not
defined, so that part of common.mk is not called.

Note: one thing to keep in mind is directories like src/port. There are
some .gcda and .gcno files in there too, but even if common.mk is fixed,
they will not be cleaned because src/port does not seem to use
common.mk.

Another thing that I'm unsure about is the coverage_out target. It does
work, but is it running the coverage target each time it is invoked?
Because if so, it's removing all of ./coverage and creating it again ...
is this the desired behavior?

This patch is missing a installation.sgml patch, at a minimum. I think
it would be useful to mention that we support gcov, and the make targets
we have, in some other part of the documentation. I can't readily find
a good spot, but I think a new file to be inserted in the internals
chapter would be appropriate.

Two hunks no longer apply, but that's OK because they were working
around a problem that no longer exists. Other than the minor gripes
above, the patch looks OK to me.

--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Thanks, I'll take a look at these issues.

-- Michelle

Alvaro Herrera wrote:

>Michelle Caisse wrote:
>
>
>>I've attached a patch that allows the generation of code coverage
>>statistics. To test it, apply the patch, then:
>>
>>autoconf
>>./configure --enable-coverage
>>make
>>make check (or execute any other application against the database to see
>>the coverage of that app)
>>make coverage
>>make coverage_out
>>
>>
>
>"make clean" does not work for me; it doesn't remove the .gcda and .gcno
>files. Apparently the problem is that $(enable_coverage) is not
>defined, so that part of common.mk is not called.
>
>Note: one thing to keep in mind is directories like src/port. There are
>some .gcda and .gcno files in there too, but even if common.mk is fixed,
>they will not be cleaned because src/port does not seem to use
>common.mk.
>
>Another thing that I'm unsure about is the coverage_out target. It does
>work, but is it running the coverage target each time it is invoked?
>Because if so, it's removing all of ./coverage and creating it again ...
>is this the desired behavior?
>
>This patch is missing a installation.sgml patch, at a minimum. I think
>it would be useful to mention that we support gcov, and the make targets
>we have, in some other part of the documentation. I can't readily find
>a good spot, but I think a new file to be inserted in the internals
>chapter would be appropriate.
>
>Two hunks no longer apply, but that's OK because they were working
>around a problem that no longer exists. Other than the minor gripes
>above, the patch looks OK to me.
>
>
>

--
Michelle Caisse Sun Microsystems
California, U.S. http://sun.com/postgresql