On 05/27/2014 03:19 PM, Thorsten Behrens wrote:
I wrote:
Sage Weil wrote:
If anybody is interested in helping with that effort, pull requests
are very welcome! :)
[snip]
As hinted at in the patch, something like boost::program_options would
be nice, but that's a chunk of work & I'd rather
I wrote:
> Sage Weil wrote:
> > If anybody is interested in helping with that effort, pull requests
> > are very welcome! :)
> >
[snip]
> As hinted at in the patch, something like boost::program_options would
> be nice, but that's a chunk of work & I'd rather hear feedback first
> this way or th
Sage Weil wrote:
> If anybody is interested in helping with that effort, pull requests
> are very welcome! :)
>
Here goes: https://github.com/ceph/ceph/pull/1814
I'm not sure though I like what I did, this macro magic is slightly
over-verbose, and even clang error messages are not overly helpful
As I said, I only uploaded the summary.
The reports take around 5 GB of space.
Please run the analyzer yourself for the individual reports.
On May 8, 2014 2:55:28 PM CEST, Milosz Tanski wrote:
>Daniel,
>
>It is orthogonal. Having said that I find the issues raised by them of
>higher value (logic
As I said, I only uploaded the summary.
The reports take around 5 GB of space.
Please run the analyzer yourself for the detailed individual reports.
On May 8, 2014 2:55:28 PM CEST, Milosz Tanski wrote:
>Daniel,
>
>It is orthogonal. Having said that I find the issues raised by them of
>higher val
Daniel,
It is orthogonal. Having said that I find the issues raised by them of
higher value (logic errors, bad comparisons) then warnings about
unused private fields, or various extensions that make the code easier
to write (0 len array, anon union).
A a side note Daniel it looks like the links
Using the Static Analyzer is orthogonal to enabling more diagnostics.
I uploaded my latest report results (only the summary, the specific reports
take around 5 GB) here:
> http://trvx.org/~daniel/ceph-devel/report.html
Quite interesting -- hopefully you soon get this from a Jenkins build, as
d
Couple of notes for anyone following along on ubuntu precise or
another older system:
* make sure you're using clang >=3.3 to get support for suppressing
all the warning flags in Daniel's command line.
* libcrypto++ 5.6.1 (the version in ubuntu) doesn't compile with
clang, unless you hack
I was
Indeed clangs warnings are much exhaustive. We've also had good
success with static analyzer based on clang
(http://clang-analyzer.llvm.org/). We've found a few logic issues in
our application using it as well... esp. a few bugs that where corner
cases had to due to very specific arguments passed c
On Tue, 6 May 2014, Daniel Hofmann wrote:
> Preamble: you might want to read the decent formatted version of this mail at:
> > https://gist.github.com/daniel-j-h/06f16015c89bec4fbb42
>
>
> Motivation
> --
>
> I tried to build Ceph with the Clang (3.4) compiler. The only issue
> preventin
Preamble: you might want to read the decent formatted version of this mail at:
> https://gist.github.com/daniel-j-h/06f16015c89bec4fbb42
Motivation
--
I tried to build Ceph with the Clang (3.4) compiler. The only issue
preventing the build to finish was a "VLA of non-POD type" usage whic
11 matches
Mail list logo
