Re: [fpc-devel] Type Helper and Array

2016-01-05 Thread Sven Barth
Am 03.01.2016 14:47 schrieb "Ondrej Pokorny" : > > On 03.01.2016 14:13, Sven Barth wrote: >> >> >> Please report a bug with a simple example that should compile with the error message you get at http://bugs.freepascal.org as I don't see any difference between dynamic and static

Re: [fpc-devel] Fixing line info units (stabs, dwarf)

2016-01-05 Thread Denis Kozlov
On 4 January 2016 at 11:52, Michael Van Canneyt wrote: > Maybe it is possible to make it configurable ? > Yes, should be doable. I'll look into it. Denis ___ fpc-devel maillist - fpc-devel@lists.freepascal.org

Re: [fpc-devel] Type Helper and Array

2016-01-05 Thread Sven Barth
Am 03.01.2016 19:15 schrieb "Mathias" : > > Hello, > > I have created a bug report. > > http://mantis.freepascal.org/view.php?id=29321 Fixed. I forgot to mention that in my answer, but next time please provide a self contained, complete example that does not depend on the LCL

Re: [fpc-devel] Inconsistent use of Timeout in SimpleIPC

2016-01-05 Thread Denis Kozlov
On 5 January 2016 at 12:39, Michael Van Canneyt wrote: > I was in fact waiting for this part, rather than applying the patch as it > is. > But if you prefer we can do it in steps ? > I prefer to do this in steps, easier to control changes. Bug report with same patch files:

Re: [fpc-devel] Inconsistent use of Timeout in SimpleIPC

2016-01-05 Thread Michael Van Canneyt
On Tue, 5 Jan 2016, Denis Kozlov wrote: On 5 January 2016 at 12:39, Michael Van Canneyt wrote: I was in fact waiting for this part, rather than applying the patch as it is. But if you prefer we can do it in steps ? I prefer to do this in steps, easier to control changes. Bug report with

Re: [fpc-devel] Capability CPUARM_HAS_BLX_LABEL with ARM instruction set

2016-01-05 Thread Garry Wood
>>While it's true that BLX always changes instructions set, I've heard some >>reports indicating that some assembler/linker combinations on OSX (apparently >>clang based?) have been behaving >>strangely with this lately. >>GCC+binutils have always been following the "emit blx and let the linker