Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: 'build.sh release' broken for evbarm-eb and -el



On Sun, 25 Nov 2012, Alexander Nasonov wrote:

Paul Goyette wrote:
IMHO, we should actually fix the issue, rather than hide it.  But my
goal here is only to get the port (and as many others as possible)
to build, so I would defer to arm users to make a decision on how to
proceed.

I don't see any issue with this. Let say you want to assign __swp return
to some variable x:

__cpu_simple_lock_t x = __swp(val, ptr);

__swp will discard volatile but the initialization of x will bring it
back.

There are no reference types in C and hence it can't be missing '&' in
the return type. If there was an explicit 'volatile' in the return type,
it'd be probably an oversight but volatile is a part of typedef.

A very similar issue has been fixed for C++ code [1]. The code below
generates only one warning at line 5:

template<class T>
const T foo() { static T t; return t; }

typedef const int T;
const T bar() { return 0; }

int main()
{
       return foo<int>() + bar();
}

IMO, this warning if for information only and it doesn't mix with -Werror.

I'm not going to fix it, either way.

I've identified a problem that prevents the port from building, and there are at least two approaches to fixing it.

Someone who actually cares about evbarm ports can pick the approach and commit a fix.



-------------------------------------------------------------------------
| Paul Goyette     | PGP Key fingerprint:     | E-mail addresses:       |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com    |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net |
| Kernel Developer |                          | pgoyette at netbsd.org  |
-------------------------------------------------------------------------


Home | Main Index | Thread Index | Old Index