From: Mathieu M. <mat...@gm...> - 2013-01-10 10:05:45
|
Here is the status of VXL 1.17 compilation on debian buildds systems: https://buildd.debian.org/status/package.php?p=vxl&suite=experimental VXL does not seems to support anything except x86 and x86_64 based CPUs. Thanks for comments, |
From: Mathieu M. <mat...@gm...> - 2013-01-10 16:12:11
|
On Thu, Jan 10, 2013 at 11:05 AM, Mathieu Malaterre <mat...@gm...> wrote: > Here is the status of VXL 1.17 compilation on debian buildds systems: > > https://buildd.debian.org/status/package.php?p=vxl&suite=experimental > > VXL does not seems to support anything except x86 and x86_64 based CPUs. > > Thanks for comments, My question is not very clear. What I am asking is: " as a packager of VXL, should I mark this package as linux-i386 and linux-amd64 only package ? " It is considered a regression, for a package to loose supported arch in Debian, so I would need at least an ACK from the -maint team to help this transition. Thanks. -- Mathieu |
From: Matthew L. <mat...@ki...> - 2013-01-10 16:27:52
|
Mathieu, Is it the case that a previous release of VXL built on all of these platforms? If so, then I agree that this is a regression and is probably something that can be fixed with a reasonable amount of effort. However, I was not aware that VXL was ever tested on, for example, ARM architectures. If it has never worked on these platforms then I expect the effort to make it work could be more significant and it would be better to just mark the package as linux-i386 and linux-amd64 only. Personally I would love to see VXL run on ARM, but I don't have the time to help make that happen. I don't think we have any ARM builds on the dashboard so developers are likely unaware of any issues with compiling on ARM. Maybe I should purchase a Raspberry Pi and turn it into a VXL dashboard build … :) --Matt On Jan 10, 2013, at 11:11 AM, Mathieu Malaterre <mat...@gm...> wrote: > On Thu, Jan 10, 2013 at 11:05 AM, Mathieu Malaterre > <mat...@gm...> wrote: >> Here is the status of VXL 1.17 compilation on debian buildds systems: >> >> https://buildd.debian.org/status/package.php?p=vxl&suite=experimental >> >> VXL does not seems to support anything except x86 and x86_64 based CPUs. >> >> Thanks for comments, > > My question is not very clear. What I am asking is: > > " as a packager of VXL, should I mark this package as linux-i386 and > linux-amd64 only package ? " > > It is considered a regression, for a package to loose supported arch > in Debian, so I would need at least an ACK from the -maint team to > help this transition. > > Thanks. > -- > Mathieu > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Vxl-maintainers mailing list > Vxl...@li... > https://lists.sourceforge.net/lists/listinfo/vxl-maintainers |
From: Mathieu M. <mat...@gm...> - 2013-01-10 16:35:18
|
This patch (taken from ITK) make most issues go away: http://patch-tracker.debian.org/patch/series/view/vxl/1.14.0-18/sse2.patch But it was never included in VXL. Maybe there is a good reason for not including it. As a side note, some x86 machine do not support SSE2 anyway. So VXL is by designed only support on x86 + SSE2. Thanks, On Thu, Jan 10, 2013 at 5:27 PM, Matthew Leotta <mat...@ki...> wrote: > Mathieu, > > Is it the case that a previous release of VXL built on all of these platforms? If so, then I agree that this is a regression and is probably something that can be fixed with a reasonable amount of effort. However, I was not aware that VXL was ever tested on, for example, ARM architectures. If it has never worked on these platforms then I expect the effort to make it work could be more significant and it would be better to just mark the package as linux-i386 and linux-amd64 only. > > Personally I would love to see VXL run on ARM, but I don't have the time to help make that happen. I don't think we have any ARM builds on the dashboard so developers are likely unaware of any issues with compiling on ARM. > > Maybe I should purchase a Raspberry Pi and turn it into a VXL dashboard build … :) > > --Matt > > > On Jan 10, 2013, at 11:11 AM, Mathieu Malaterre <mat...@gm...> wrote: > >> On Thu, Jan 10, 2013 at 11:05 AM, Mathieu Malaterre >> <mat...@gm...> wrote: >>> Here is the status of VXL 1.17 compilation on debian buildds systems: >>> >>> https://buildd.debian.org/status/package.php?p=vxl&suite=experimental >>> >>> VXL does not seems to support anything except x86 and x86_64 based CPUs. >>> >>> Thanks for comments, >> >> My question is not very clear. What I am asking is: >> >> " as a packager of VXL, should I mark this package as linux-i386 and >> linux-amd64 only package ? " >> >> It is considered a regression, for a package to loose supported arch >> in Debian, so I would need at least an ACK from the -maint team to >> help this transition. >> >> Thanks. >> -- >> Mathieu >> >> ------------------------------------------------------------------------------ >> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, >> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current >> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft >> MVPs and experts. ON SALE this month only -- learn more at: >> http://p.sf.net/sfu/learnmore_122712 >> _______________________________________________ >> Vxl-maintainers mailing list >> Vxl...@li... >> https://lists.sourceforge.net/lists/listinfo/vxl-maintainers > -- Mathieu |
From: Matt L. <mat...@ki...> - 2013-01-28 17:59:44
|
All, I did end up buying a Raspberry Pi and I am considering using it for a dashboard build (or maybe just cross-compiling for Raspberry Pi on my desktop as a dashboard build). Either way, the patch from ITK that Mathieu has shared does seem to help and looks okay to me. If there are no objections I will contribute this patch back to the upstream VXL source. --Matt On Thu, Jan 10, 2013 at 11:34 AM, Mathieu Malaterre < mat...@gm...> wrote: > This patch (taken from ITK) make most issues go away: > > http://patch-tracker.debian.org/patch/series/view/vxl/1.14.0-18/sse2.patch > > But it was never included in VXL. Maybe there is a good reason for not > including it. As a side note, some x86 machine do not support SSE2 > anyway. So VXL is by designed only support on x86 + SSE2. > > Thanks, > > On Thu, Jan 10, 2013 at 5:27 PM, Matthew Leotta <mat...@ki...> > wrote: > > Mathieu, > > > > Is it the case that a previous release of VXL built on all of these > platforms? If so, then I agree that this is a regression and is probably > something that can be fixed with a reasonable amount of effort. However, I > was not aware that VXL was ever tested on, for example, ARM architectures. > If it has never worked on these platforms then I expect the effort to make > it work could be more significant and it would be better to just mark the > package as linux-i386 and linux-amd64 only. > > > > Personally I would love to see VXL run on ARM, but I don't have the time > to help make that happen. I don't think we have any ARM builds on the > dashboard so developers are likely unaware of any issues with compiling on > ARM. > > > > Maybe I should purchase a Raspberry Pi and turn it into a VXL dashboard > build … :) > > > > --Matt > > > > > > On Jan 10, 2013, at 11:11 AM, Mathieu Malaterre < > mat...@gm...> wrote: > > > >> On Thu, Jan 10, 2013 at 11:05 AM, Mathieu Malaterre > >> <mat...@gm...> wrote: > >>> Here is the status of VXL 1.17 compilation on debian buildds systems: > >>> > >>> https://buildd.debian.org/status/package.php?p=vxl&suite=experimental > >>> > >>> VXL does not seems to support anything except x86 and x86_64 based > CPUs. > >>> > >>> Thanks for comments, > >> > >> My question is not very clear. What I am asking is: > >> > >> " as a packager of VXL, should I mark this package as linux-i386 and > >> linux-amd64 only package ? " > >> > >> It is considered a regression, for a package to loose supported arch > >> in Debian, so I would need at least an ACK from the -maint team to > >> help this transition. > >> > >> Thanks. > >> -- > >> Mathieu > >> > >> > ------------------------------------------------------------------------------ > >> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > >> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > >> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > >> MVPs and experts. ON SALE this month only -- learn more at: > >> http://p.sf.net/sfu/learnmore_122712 > >> _______________________________________________ > >> Vxl-maintainers mailing list > >> Vxl...@li... > >> https://lists.sourceforge.net/lists/listinfo/vxl-maintainers > > > > > > -- > Mathieu > |
From: Brooksby, G. W (GE G. Research) <bro...@ge...> - 2013-01-10 18:39:29
|
Mathieu, Matt, With limited effort I have been able to build a large portion of VxL on an ARM system (Beagleboard) running Ubuntu 10.04. I've also cross-compiled VxL on Android and I know others have done it for iOS. So I would definitely not tag VxL as x86 only. Discussion about this happened last Sept. on the vxl-users list. My original posting about this was from May 2011. See http://sourceforge.net/mailarchive/message.php?msg_id=27440979 . My original fix was to never define GCC_USE_FAST_IMPL to 1. A better fix for this particular problem would be to identify an ARM processor via __arm__ (in GCC). That said, I do not have a dashboard build running. A Raspberry Pi could provide a dashboard build, but you may find a Beagleboard more friendly due to more flexible I/O, albeit at a higher price point. Glen Brooksby -----Original Message----- From: Matthew Leotta [mailto:mat...@ki...] Sent: Thursday, January 10, 2013 11:28 AM To: Mathieu Malaterre Cc: vxl maintainers Subject: Re: [Vxl-maintainers] VXL on non x86* based CPUs Mathieu, Is it the case that a previous release of VXL built on all of these platforms? If so, then I agree that this is a regression and is probably something that can be fixed with a reasonable amount of effort. However, I was not aware that VXL was ever tested on, for example, ARM architectures. If it has never worked on these platforms then I expect the effort to make it work could be more significant and it would be better to just mark the package as linux-i386 and linux-amd64 only. Personally I would love to see VXL run on ARM, but I don't have the time to help make that happen. I don't think we have any ARM builds on the dashboard so developers are likely unaware of any issues with compiling on ARM. Maybe I should purchase a Raspberry Pi and turn it into a VXL dashboard build ... :) --Matt On Jan 10, 2013, at 11:11 AM, Mathieu Malaterre <mat...@gm...> wrote: > On Thu, Jan 10, 2013 at 11:05 AM, Mathieu Malaterre > <mat...@gm...> wrote: >> Here is the status of VXL 1.17 compilation on debian buildds systems: >> >> https://buildd.debian.org/status/package.php?p=vxl&suite=experimental >> >> VXL does not seems to support anything except x86 and x86_64 based CPUs. >> >> Thanks for comments, > > My question is not very clear. What I am asking is: > > " as a packager of VXL, should I mark this package as linux-i386 and > linux-amd64 only package ? " > > It is considered a regression, for a package to loose supported arch > in Debian, so I would need at least an ACK from the -maint team to > help this transition. > > Thanks. > -- > Mathieu > > ---------------------------------------------------------------------- > -------- Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, > HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your > skills current with LearnDevNow - 3,200 step-by-step video tutorials > by Microsoft MVPs and experts. ON SALE this month only -- learn more > at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Vxl-maintainers mailing list > Vxl...@li... > https://lists.sourceforge.net/lists/listinfo/vxl-maintainers ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnmore_122712 _______________________________________________ Vxl-maintainers mailing list Vxl...@li... https://lists.sourceforge.net/lists/listinfo/vxl-maintainers |
From: Sean M. <se...@ro...> - 2013-01-28 20:40:50
|
On Thu, 10 Jan 2013 11:05:18 +0100, Mathieu Malaterre said: >Here is the status of VXL 1.17 compilation on debian buildds systems: > >https://buildd.debian.org/status/package.php?p=vxl&suite=experimental > >VXL does not seems to support anything except x86 and x86_64 based CPUs. I just tried building on my PowerMac G5: <http://open.cdash.org/viewBuildError.php?buildid=2789402> The main failure is: "core/vnl/vnl_math.h:256: error: unknown register name 'st' in 'asm'" which is partly addressed here: <http://patch-tracker.debian.org/patch/series/view/vxl/1.14.0-18/sse2.patch> If someone reviews and applies that to svn trunk, I can try again and make this dashboard nightly. Cheers, -- ____________________________________________________________ Sean McBride, B. Eng se...@ro... Rogue Research www.rogue-research.com Mac Software Developer Montréal, Québec, Canada |
From: Matt L. <mat...@ki...> - 2013-01-28 20:50:04
|
Sean, The patch looks good to me. However, it touches both vnl and vil, so I'd like to give Gehua, Peter, and Ian a chance to comment before I apply it. They are the lead maintainers of vnl and vil. If I don't hear anything by tomorrow I will apply the patch. --Matt On Mon, Jan 28, 2013 at 3:40 PM, Sean McBride <se...@ro...>wrote: > On Thu, 10 Jan 2013 11:05:18 +0100, Mathieu Malaterre said: > > >Here is the status of VXL 1.17 compilation on debian buildds systems: > > > >https://buildd.debian.org/status/package.php?p=vxl&suite=experimental > > > >VXL does not seems to support anything except x86 and x86_64 based CPUs. > > I just tried building on my PowerMac G5: > <http://open.cdash.org/viewBuildError.php?buildid=2789402> > > The main failure is: > "core/vnl/vnl_math.h:256: error: unknown register name 'st' in 'asm'" > > which is partly addressed here: > < > http://patch-tracker.debian.org/patch/series/view/vxl/1.14.0-18/sse2.patch > > > > If someone reviews and applies that to svn trunk, I can try again and make > this dashboard nightly. > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng se...@ro... > Rogue Research www.rogue-research.com > Mac Software Developer Montréal, Québec, Canada > > > > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnnow-d2d > _______________________________________________ > Vxl-maintainers mailing list > Vxl...@li... > https://lists.sourceforge.net/lists/listinfo/vxl-maintainers > |
From: Sean M. <se...@ro...> - 2013-01-28 20:57:41
|
On Mon, 28 Jan 2013 15:49:53 -0500, Matt Leotta said: >The patch looks good to me. However, it touches both vnl and vil, so I'd >like to give Gehua, Peter, and Ian a chance to comment before I apply it. > They are the lead maintainers of vnl and vil. If I don't hear anything by >tomorrow I will apply the patch. OK. The patch isn't quite perfect... I'd remove: 1) the remaining "&& (!defined(__APPLE__) || !defined(__ppc__)" 2) the two "with the following exception: PPC with Mac OS X" comments Cheers, -- ____________________________________________________________ Sean McBride, B. Eng se...@ro... Rogue Research www.rogue-research.com Mac Software Developer Montréal, Québec, Canada |
From: Ian S. <sc...@im...> - 2013-01-28 21:56:57
|
On 28/01/2013 20:49, Matt Leotta wrote: > Sean, > > The patch looks good to me. However, it touches both vnl and vil, so > I'd like to give Gehua, Peter, and Ian a chance to comment before I > apply it. They are the lead maintainers of vnl and vil. If I don't > hear anything by tomorrow I will apply the patch. > Hi Matt, I've no significant problem with this patch, so long as it is well tested. Ian. |
From: Peter V. <pet...@ya...> - 2013-01-28 22:07:19
|
The same holds for me. Have no experience with this platform. -- Peter. ________________________________ Från: Ian Scott <sc...@im...> Till: Matt Leotta <mat...@ki...>; Vxl-maintainers <vxl...@li...> Skickat: måndag, 28 januari 2013 22:56 Ämne: Re: [Vxl-maintainers] VXL on non x86* based CPUs I've no significant problem with this patch, so long as it is well tested. Ian. |
From: Matt L. <mat...@ki...> - 2013-01-28 22:03:05
|
Sean, I will make your suggested changes before applying. That makes sense to me. It should not matter if it is Apple or not, only if it is Intel or not, and PPC on Mac OS X is not an exception to the rule because PPC is not Intel. Thanks, Matt On Mon, Jan 28, 2013 at 3:57 PM, Sean McBride <se...@ro...>wrote: > On Mon, 28 Jan 2013 15:49:53 -0500, Matt Leotta said: > > >The patch looks good to me. However, it touches both vnl and vil, so I'd > >like to give Gehua, Peter, and Ian a chance to comment before I apply it. > > They are the lead maintainers of vnl and vil. If I don't hear anything > by > >tomorrow I will apply the patch. > > OK. The patch isn't quite perfect... I'd remove: > > 1) the remaining "&& (!defined(__APPLE__) || !defined(__ppc__)" > > 2) the two "with the following exception: PPC with Mac OS X" comments > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng se...@ro... > Rogue Research www.rogue-research.com > Mac Software Developer Montréal, Québec, Canada > > > |
From: Matthew L. <mat...@ki...> - 2013-01-29 15:35:03
|
The patch has been applied and modified as Sean suggested. --Matt On Jan 28, 2013, at 5:02 PM, Matt Leotta <mat...@ki...> wrote: > Sean, > > I will make your suggested changes before applying. That makes sense to me. It should not matter if it is Apple or not, only if it is Intel or not, and PPC on Mac OS X is not an exception to the rule because PPC is not Intel. > > Thanks, > Matt > > > On Mon, Jan 28, 2013 at 3:57 PM, Sean McBride <se...@ro...> wrote: > On Mon, 28 Jan 2013 15:49:53 -0500, Matt Leotta said: > > >The patch looks good to me. However, it touches both vnl and vil, so I'd > >like to give Gehua, Peter, and Ian a chance to comment before I apply it. > > They are the lead maintainers of vnl and vil. If I don't hear anything by > >tomorrow I will apply the patch. > > OK. The patch isn't quite perfect... I'd remove: > > 1) the remaining "&& (!defined(__APPLE__) || !defined(__ppc__)" > > 2) the two "with the following exception: PPC with Mac OS X" comments > > Cheers, > > -- > ____________________________________________________________ > Sean McBride, B. Eng se...@ro... > Rogue Research www.rogue-research.com > Mac Software Developer Montréal, Québec, Canada > > > |
From: Sean M. <se...@ro...> - 2013-01-30 15:29:12
|
On Tue, 29 Jan 2013 10:34:50 -0500, Matthew Leotta said: >The patch has been applied and modified as Sean suggested. Cool. So I've enabled my OS X 10.5 / ppc64 nightly build: <http://open.cdash.org/viewBuildError.php?buildid=2791983> Only 4 errors today. All 'duplicate symbol' link errors. My x86_64 builds have been red for days too... Cheers, -- ____________________________________________________________ Sean McBride, B. Eng se...@ro... Rogue Research www.rogue-research.com Mac Software Developer Montréal, Québec, Canada |
From: Peter V. <pet...@ya...> - 2013-01-30 16:03:39
|
Sean, I have rmoved the offending duplicate instantiations from svn; your OS X 10.5 / ppc64 nightly build should now run OK. -- Peter. Från: Sean McBride <se...@ro...> Till: Matthew Leotta <mat...@ki...>; VXL <vxl...@li...> Skickat: onsdag, 30 januari 2013 16:29 Ämne: Re: [Vxl-maintainers] VXL on non x86* based CPUs On Tue, 29 Jan 2013 10:34:50 -0500, Matthew Leotta said: >The patch has been applied and modified as Sean suggested. Cool. So I've enabled my OS X 10.5 / ppc64 nightly build: <http://open.cdash.org/viewBuildError.php?buildid=2791983> Only 4 errors today. All 'duplicate symbol' link errors. My x86_64 builds have been red for days too... -- |
From: Sean M. <se...@ro...> - 2013-01-30 18:30:42
|
On Wed, 30 Jan 2013 16:03:30 +0000, Peter Vanroose said: >I have rmoved the offending duplicate instantiations from svn; your OS X >10.5 / ppc64 nightly build should now run OK. Thanks, it builds now! Although it's an old and probably uninteresting platform at this point, the dashboard will hopefully help keep vxl less x86-reliant. Can you fix my clang dashboards now? :) Thanks, -- ____________________________________________________________ Sean McBride, B. Eng se...@ro... Rogue Research www.rogue-research.com Mac Software Developer Montréal, Québec, Canada |
From: Peter V. <pet...@ya...> - 2013-01-30 18:54:51
|
Sean, Over the last few days I've been making some corrections to several vxl files as to make the clang dashboards compile error-free. What remains is out-of-bounds for me ... :-( -- Peter. Från: Sean McBride <se...@ro...> Till: Peter Vanroose <pe...@va...>; Peter Vanroose <pet...@ya...> Kopia: VXL <vxl...@li...> Skickat: onsdag, 30 januari 2013 19:30 Ämne: Re: Re VXL on non x86* based CPUs [...] Can you fix my clang dashboards now? :) -- |
From: Peter V. <pet...@ya...> - 2013-02-06 12:59:25
|
Dear all, Since a few days the dashboard is showing the following errors; seems like some class redesign took place but not all old uses were fixed; could somebody have a look at this? contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose.cxx:460:100: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose.cxx:496:9: error: unknown type name 'boxm2_volm_score_out_max'; did you mean 'boxm2_volm_score_out'? contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose_cam.cxx:102:100: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe.cxx:96:31: error: no matching constructor for initialization of 'volm_query' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe.cxx:159:25: error: no matching constructor for initialization of 'boxm2_volm_matcher_p0' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe_per_gpu.cxx:20:10: fatal error: 'boxm2/volm/boxm2_volm_matcher_order.h' file not found contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_integration.cxx:152:98: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_integration.cxx:280:27: error: no matching constructor for initialization of 'boxm2_volm_matcher_p1' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_rationale.cxx:266:100: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:68:43: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:69:45: error: no member named 'point_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:70:43: error: no member named 'top_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:71:46: error: no member named 'bottom_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:75:100: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:43: error: no member named 'point_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:86: error: no member named 'top_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:130: error: no member named 'bottom_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:172:98: error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:257:25: error: no matching constructor for initialization of 'boxm2_volm_matcher_p1' contrib/brl/bseg/boxm2/volm/tests/test_volm_obj_based_matcher.cxx:8:10: fatal error: 'boxm2/volm/boxm2_volm_obj_based_matcher.h' file not found contrib/brl/bseg/bstm/ocl/algo/bstm_ocl_render_expected_image_function.cxx:141:15: error: no member named 'release_current_event' in 'bocl_kernel' -- Peter. -- |
From: Yi D. <yid...@gm...> - 2013-02-06 15:59:36
|
Dear all, The compiling errors in bseg/boxm2/volm/ have been fixed, please update accordingly. Best Yi On 2/6/2013 7:59 AM, Peter Vanroose wrote: > Dear all, > > Since a few days the dashboard is showing the following errors; seems > like some class redesign took place but not all old uses were fixed; > could somebody have a look at this? > > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose.cxx:460:100: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose.cxx:496:9: > error: unknown type name 'boxm2_volm_score_out_max'; did you mean > 'boxm2_volm_score_out'? > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_compose_cam.cxx:102:100: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe.cxx:96:31: error: > no matching constructor for initialization of 'volm_query' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe.cxx:159:25: > error: no matching constructor for initialization of > 'boxm2_volm_matcher_p0' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_exe_per_gpu.cxx:20:10: > fatal error: 'boxm2/volm/boxm2_volm_matcher_order.h' file not found > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_integration.cxx:152:98: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_integration.cxx:280:27: > error: no matching constructor for initialization of > 'boxm2_volm_matcher_p1' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_matcher_rationale.cxx:266:100: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:68:43: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:69:45: > error: no member named 'point_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:70:43: > error: no member named 'top_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:71:46: > error: no member named 'bottom_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/exe/boxm2_volumetric_visualize_top_indices.cxx:75:100: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:43: > error: no member named 'point_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:86: > error: no member named 'top_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:170:130: > error: no member named 'bottom_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:172:98: > error: no member named 'cap_angle' in 'boxm2_volm_wr3db_index_params' > contrib/brl/bseg/boxm2/volm/tests/test_volm_matcher_p1.cxx:257:25: > error: no matching constructor for initialization of > 'boxm2_volm_matcher_p1' > contrib/brl/bseg/boxm2/volm/tests/test_volm_obj_based_matcher.cxx:8:10: fatal > error: 'boxm2/volm/boxm2_volm_obj_based_matcher.h' file not found > contrib/brl/bseg/bstm/ocl/algo/bstm_ocl_render_expected_image_function.cxx:141:15: > error: no member named 'release_current_event' in 'bocl_kernel' > > -- Peter. > > > > > > > > > > > > > > > > -- > > > > ------------------------------------------------------------------------------ > Free Next-Gen Firewall Hardware Offer > Buy your Sophos next-gen firewall before the end March 2013 > and get the hardware for free! Learn more. > http://p.sf.net/sfu/sophos-d2d-feb > > > _______________________________________________ > Vxl-maintainers mailing list > Vxl...@li... > https://lists.sourceforge.net/lists/listinfo/vxl-maintainers |
From: Peter V. <pet...@ya...> - 2013-02-07 13:41:39
|
Thanks, Yi! Just one error left, it seems: contrib/brl/bseg/bstm/ocl/algo/bstm_ocl_render_expected_image_function.cxx:141:15: error: no member named 'release_current_event' in 'bocl_kernel' -- Peter. -- |