From: Wheeler, F. W (Research) <wh...@cr...> - 2003-08-05 13:05:06
|
This test opens a window and does not seem to finish until the user closes the window. That's probably not a good idea. contrib/brl/bmvl/brct/tests/brct_test I would expect this to cause dashboard builds to hang, but they have not. Hmm. Fred Wheeler |
From: Amitha P. <pe...@cs...> - 2003-08-05 13:25:11
|
On Tue 05 Aug 2003, Wheeler, Frederick W (Research) wrote: > This test opens a window and does not seem to finish until the user closes > the window. That's probably not a good idea. I agree. I think the display dependent tests should only be run manually from time to time, and not automatically every night. > I would expect this to cause dashboard builds to hang, but they have not. > Hmm. Is this perhaps Dart killing processes after a time-out? Amitha. |
From: Kongbin K. <kk...@le...> - 2003-08-05 14:51:09
|
The test is already removed. Kongbin On Tue, 5 Aug 2003, Wheeler, Frederick W (Research) wrote: > This test opens a window and does not seem to finish until the user closes > the window. That's probably not a good idea. > > contrib/brl/bmvl/brct/tests/brct_test > > I would expect this to cause dashboard builds to hang, but they have not. > Hmm. > > Fred Wheeler > > > ------------------------------------------------------- > This SF.Net email sponsored by: Free pre-built ASP.NET sites including > Data Reports, E-commerce, Portals, and Forums are available now. > Download today and enter to win an XBOX or Visual Studio .NET. > http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 > _______________________________________________ > Vxl-maintainers mailing list > Vxl...@li... > https://lists.sourceforge.net/lists/listinfo/vxl-maintainers > |
From: Peter V. <Pet...@es...> - 2003-08-08 10:04:43
|
Dear all, The following fragment (in core/vgui/internals/vgui_generic_vil_image_view.cxx) blows up the automatic instantiator of SGI's native compiler (see error message below): --------------------------------- template< vil_pixel_format fmt > struct extract { static view_data process( vil_image_view_base const& other ) { if( other.pixel_format() == fmt ) { typedef typename vil_pixel_format_type_of<fmt>::type pixel_type; return cast_and_extract( other, (pixel_type*)0 ); } else { return extract< vil_pixel_format(fmt+1) >::process( other ); } } }; -------------------------------- Any suggestions on how to solve this problem? I'm afraid we'll have to fall back to explicit enumeration... -- Peter. ---------------------- "internals/vgui_generic_vil_image_view.cxx", line 131: error(1435): excessive recursion at instantiation of function "<unnamed>::extract<VIL_PIXEL_FORMAT_RGB_INT_16>::process" return extract< vil_pixel_format(fmt+1) >::process( other ); ^ detected during: instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_RGB_UINT_16>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_RGB_INT_32>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_RGB_UINT_32>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<(vil_pixel_format)14>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<(vil_pixel_format)13>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_BOOL>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<(vil_pixel_format)11>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_DOUBLE>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_FLOAT>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_SBYTE>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_BYTE>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_INT_16>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_UINT_16>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_INT_32>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<VIL_PIXEL_FORMAT_UINT_32>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<(vil_pixel_format)2>::process(const vil_image_view_base &)" at line 131 instantiation of "<unnamed>::view_data <unnamed>::extract<(vil_pixel_format)1>::process(const vil_image_view_base &)" at line 163 ------------------------------------------------------- |
From: Amitha P. <pe...@cs...> - 2003-08-08 13:55:33
|
On Fri 08 Aug 2003, Peter Vanroose wrote: > The following fragment (in core/vgui/internals/vgui_generic_vil_image_view.cxx) > blows up the automatic instantiator of SGI's native compiler (see error > message below): [fragment omitted] It seems that the MIPSpro compiler is limited to a depth of 65. I compiled the program below with CC -LANG:std test.cxx The first few lines of the output are cc-1435 CC: ERROR File = test.cxx, Line = 8 There is excessive recursion at instantiation of function "depth<(my_enum)65>::value". static int value() { return depth< my_enum(n+1) >::value(); } I'll try to think of a way to reduce the depth, or come up with some other workaround. I don't like explicit listing of all the formats because that is prone to inconsistency. It may be the only portable solution, though. And I guess the vil file formats will rarely change, so a bold comment may be sufficient. Amitha. The test program: #include <iostream> enum my_enum { the_end = 100 }; template<my_enum n> struct depth { static int value() { return depth< my_enum(n+1) >::value(); } }; template<> struct depth<the_end> { static int value() { return the_end; } }; int main() { std::cout << "val=" << depth< my_enum(1) >::value() << "\n"; return 0; }; |
From: Amitha P. <pe...@cs...> - 2003-08-08 19:01:59
|
On Fri 08 Aug 2003, Amitha Perera wrote: > I'll try to think of a way to reduce the depth, or come up with some > other workaround. I've committed a workaround that breaks the recursion in steps. It's not very elegant, but it maintains the compiler generation property of the recursive template solution. It fixes the depth problems with gcc 2.95. Peter: does this workaround fix the problems with the MIPSpro compilers? Amitha. |
From: Peter V. <Pet...@es...> - 2003-08-09 13:09:40
|
> Peter: does this workaround fix the problems with the MIPSpro > compilers? Yes, it works! Thanks. Also, test_generic_image_view runs and passes all tests. -- Peter. |