From: Borut R. <bor...@si...> - 2011-01-15 19:10:33
|
I've found the following: http://www.programare.org/viewtopic.php?t=6348. Unfortunately it is in Romanian language, but Google translate gives the following translation: "itm must be const_iterator, iterator not for the reference to a map that's passing print_map () is const. begin () const iterator gives a guy in case, and you can not assign it a non-const, which would allow you to change the map stuff." If I understand it correctly, iterator should be replaced by const_iterator. Does it make sense? Is it applicable for our case? Borut On 01/15/2011 07:11 PM, Maarten Brock wrote: > Hi, > > So, I've installed Visual C++ 2010 Express and also > Boost 1.44. Then I tried to compile SDCC #6109 again. It > still fails, but different. Aside from some more > warnings about conversions I get: > > 3>c:\program > files\boost\boost_1_44\boost\tuple\detail\tuple_basic.hpp(646): > error C2679: binary '=' : no operator found which takes a right-hand > operand of type 'const std::_Tree_const_iterator<_Mytree>' (or there > is no acceptable conversion) > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,true>> > 3> ] > 3> c:\program files\microsoft visual studio > 10.0\vc\include\xtree(337): could be > 'std::_Tree_const_iterator<_Mytree> > &std::_Tree_const_iterator<_Mytree>::operator =(const > std::_Tree_const_iterator<_Mytree> &)' > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,false>> > 3> ] > 3> while trying to match the argument list > '(std::_Tree_const_iterator<_Mytree>, const > std::_Tree_const_iterator<_Mytree>)' > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,false>> > 3> ] > 3> and > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,true>> > 3> ] > 3> c:\document\c-sources\sdcc-2010\src\z80\ralloc2.cc(397) > : see reference to function template instantiation > 'boost::tuples::tuple<T0,T1,T2,T3,T4,T5,T6,T7,T8,T9> > &boost::tuples::tuple<T0,T1,T2,T3,T4,T5,T6,T7,T8,T9>::operator > =<std::_Tree_const_iterator<_Mytree>,std::_Tree_const_iterator<_Mytr > ee>>(const std::pair<_Ty1,_Ty2> &)' being compiled > 3> with > 3> [ > 3> > T0=std::_Tree_const_iterator<std::_Tree_val<std::_Tmap_traits<int,un > signed short,std::less<int>,std::allocator<std::pair<const > int,unsigned short>>,false>>> &, > 3> > T1=std::_Tree_const_iterator<std::_Tree_val<std::_Tmap_traits<int,un > signed short,std::less<int>,std::allocator<std::pair<const > int,unsigned short>>,false>>> &, > 3> T2=boost::tuples::null_type, > 3> T3=boost::tuples::null_type, > 3> T4=boost::tuples::null_type, > 3> T5=boost::tuples::null_type, > 3> T6=boost::tuples::null_type, > 3> T7=boost::tuples::null_type, > 3> T8=boost::tuples::null_type, > 3> T9=boost::tuples::null_type, > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,true>>, > 3> > _Ty1=std::_Tree_const_iterator<std::_Tree_val<std::_Tmap_traits<int, > unsigned short,std::less<int>,std::allocator<std::pair<const > int,unsigned short>>,true>>>, > 3> > _Ty2=std::_Tree_const_iterator<std::_Tree_val<std::_Tmap_traits<int, > unsigned short,std::less<int>,std::allocator<std::pair<const > int,unsigned short>>,true>>> > 3> ] > 3> c:\document\c-sources\sdcc-2010\src\z80\ralloc2.cc(475) > : see reference to function template instantiation 'float > assign_cost<G_t,I_t>(const assignment&,unsigned short,const G_t > &,const I_t&)' being compiled > 3> with > 3> [ > 3> G_t=cfg_t, > 3> I_t=con2_t > 3> ] > 3>c:\program > files\boost\boost_1_44\boost\tuple\detail\tuple_basic.hpp(647): > error C2679: binary '=' : no operator found which takes a right-hand > operand of type 'const std::_Tree_const_iterator<_Mytree>' (or there > is no acceptable conversion) > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,true>> > 3> ] > 3> c:\program files\microsoft visual studio > 10.0\vc\include\xtree(337): could be > 'std::_Tree_const_iterator<_Mytree> > &std::_Tree_const_iterator<_Mytree>::operator =(const > std::_Tree_const_iterator<_Mytree> &)' > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,false>> > 3> ] > 3> while trying to match the argument list > '(std::_Tree_const_iterator<_Mytree>, const > std::_Tree_const_iterator<_Mytree>)' > 3> with > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,false>> > 3> ] > 3> and > 3> [ > 3> > _Mytree=std::_Tree_val<std::_Tmap_traits<int,unsigned > short,std::less<int>,std::allocator<std::pair<const int,unsigned > short>>,true>> > 3> ] > > These errors indicate to me that Boost itself may not be MSVC > compatible even though they claim it to be. Or am I totally wrong? > > I also had a problem with mcs51 because now the /Zm2000 setting > creates heap overflow instead of fixing it. Removing it was enough. > > I also attach the license that I had to accept during installation. > I see no limitation in it for our use, but I may have overlokked > something. > > Hope this helps, > Maarten > >>>> Unfortunately my C++ is not good enough to solve the _Traits tree errors. >>> You mean MSVC 6.0? >> I haven't tried MSVC 6.0. I'm currently using Visual C++ >> 2008 Express. >> >> I just tried to download 2010 Express but Internet >> Explorer failed to do it. Chrome can get it though. I'll >> try and see what 2010 thinks of ralloc2.cc. >> >> Maarten >> >> >> ------------------------------------------------------------------------------ >> Protect Your Site and Customers from Malware Attacks >> Learn about various malware tactics and how to avoid them. Understand >> malware threats, the impact they can have on your business, and how you >> can protect your company and customers by using code signing. >> http://p.sf.net/sfu/oracle-sfdevnl >> _______________________________________________ >> sdcc-devel mailing list >> sdc...@li... >> https://lists.sourceforge.net/lists/listinfo/sdcc-devel >> > > > The following section of this message contains a file attachment > prepared for transmission using the Internet MIME message format. > If you are using Pegasus Mail, or any other MIME-compliant system, > you should be able to save it or view it from within your mailer. > If you cannot, please ask your system administrator for assistance. > > ---- File information ----------- > File: Microsoft Visual C++ 2010 Express License.txt > Date: 15 Jan 2011, 13:43 > Size: 13939 bytes. > Type: Text > > > ------------------------------------------------------------------------------ > Protect Your Site and Customers from Malware Attacks > Learn about various malware tactics and how to avoid them. Understand > malware threats, the impact they can have on your business, and how you > can protect your company and customers by using code signing. > http://p.sf.net/sfu/oracle-sfdevnl > > > _______________________________________________ > sdcc-devel mailing list > sdc...@li... > https://lists.sourceforge.net/lists/listinfo/sdcc-devel |
From: Philipp K. K. <pk...@sp...> - 2011-01-16 17:04:33
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 15.01.2011 20:10, schrieb Borut Razem: > I've found the following: > http://www.programare.org/viewtopic.php?t=6348. Unfortunately it is in > Romanian language, but Google translate gives the following translation: > > "itm must be const_iterator, iterator not for the reference to a map > that's passing print_map () is const. begin () const iterator gives a > guy in case, and you can not assign it a non-const, which would allow > you to change the map stuff." > > If I understand it correctly, iterator should be replaced by > const_iterator. Does it make sense? Is it applicable for our case? It makes sense: Someone used a const container, got an iterator to an emlement inside, and tried to assign it to an iterator instead of a const_iterator. However it would seem strange that MSVC++ finds such an error when g++ does not, not even with -Wall. I have seen g++ give an error in this case before. Thus I doubt it is applicable for our case. Philipp -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk0zJRcACgkQbtUV+xsoLprXswCeJthcvuEJWtw8CcXgZDAmUPGM J38An2C62pTUVVgWNvftmTIksy1r0U2D =q9s1 -----END PGP SIGNATURE----- |
From: Maarten B. <sou...@ds...> - 2011-01-16 18:27:55
|
Hi, > Who will fix the MSVC compilation problem? Phillip, will you do it, or > you need help from me or Maarten, since we both have MSVC installed? I don't think I can fix the problem. I don't think I ever used C++ templates and I have no idea what these error messages are about. But I can help with the testing. > What about the MSVC project files upgrade? I would go for it. Maarten, > if I understand correctly, you already have the upgraded project files. > Can you commit them to svn or you have concerns? I propose to go with > the Visual Studio 2010 Express. I can commit them, but I do have some concerns. I got some warnings during the conversion and I'm not sure I fixed the settings the right way now. Btw. Why do we put the output in bin_vc ? What's wrong with the usual Debug and Release output dirs? And the same goes for the asxxx projects. Why not rename them to sdasxxx? Maarten |
From: Borut R. <bor...@si...> - 2011-01-16 18:39:56
|
On 01/16/2011 07:27 PM, Maarten Brock wrote: > Hi, > >> Who will fix the MSVC compilation problem? Phillip, will you do it, or >> you need help from me or Maarten, since we both have MSVC installed? > I don't think I can fix the problem. I don't think I > ever used C++ templates and I have no idea what these > error messages are about. But I can help with the > testing. > >> What about the MSVC project files upgrade? I would go for it. Maarten, >> if I understand correctly, you already have the upgraded project files. >> Can you commit them to svn or you have concerns? I propose to go with >> the Visual Studio 2010 Express. > I can commit them, but I do have some concerns. I got > some warnings during the conversion and I'm not sure I > fixed the settings the right way now. Btw. Why do we put > the output in bin_vc ? This is in case the same snapshot is used for cygwin (or migw) and MSVC build: cygwing binaries go to the bin directory, MSVC binaries goes to the bin_vc directory. > What's wrong with the usual Debug > and Release output dirs? And the same goes for the asxxx > projects. Why not rename them to sdasxxx? This is probably the remaining from before the asxxx to sdas renaming era. Go ahead and rebane them. Borut |
From: Borut R. <bor...@si...> - 2011-01-16 18:44:05
|
> Go ahead and rebane them. I meant "Go ahead and rename them." :-[ Borut On 01/16/2011 07:39 PM, Borut Razem wrote: > On 01/16/2011 07:27 PM, Maarten Brock wrote: >> Hi, >> >>> Who will fix the MSVC compilation problem? Phillip, will you do it, or >>> you need help from me or Maarten, since we both have MSVC installed? >> I don't think I can fix the problem. I don't think I >> ever used C++ templates and I have no idea what these >> error messages are about. But I can help with the >> testing. >> >>> What about the MSVC project files upgrade? I would go for it. Maarten, >>> if I understand correctly, you already have the upgraded project files. >>> Can you commit them to svn or you have concerns? I propose to go with >>> the Visual Studio 2010 Express. >> I can commit them, but I do have some concerns. I got >> some warnings during the conversion and I'm not sure I >> fixed the settings the right way now. Btw. Why do we put >> the output in bin_vc ? > This is in case the same snapshot is used for cygwin (or migw) and MSVC > build: cygwing binaries go to the bin directory, MSVC binaries goes to > the bin_vc directory. > >> What's wrong with the usual Debug >> and Release output dirs? And the same goes for the asxxx >> projects. Why not rename them to sdasxxx? > This is probably the remaining from before the asxxx to sdas renaming > era. Go ahead and rebane them. > > Borut |
From: Maarten B. <sou...@ds...> - 2011-01-17 21:36:30
Attachments:
ralloc2.txt
|
> Am 15.01.2011 11:45, schrieb Maarten Brock: > > Some suggestions. > > > > * Rename tree_dec.hh to tree_dec.hpp > > * Remove #include "sdccconf.h" > > * Remove #include "SDCCy.h" > > * Use new/delete for the dynamic allocated arrays > > Done. You might want to try the current optralloc branch. I made > multiple other changes which are likely to fix some of the other errors > you got as well. > > Philipp Thank you, I've switched my copy. I also removed "sdccconf.h" once again as it still does not exist when building with MSVC. It is handled by SDCCglobl.h which you already have. Unfortunately, the assignment errors are still there. See the attachment with the full report. Maarten |
From: Borut R. <bor...@si...> - 2011-01-17 21:42:20
|
Hi Maarten, can you commit the MSVC project files to the optralloc branch, so that I can use them? Borut On 01/17/2011 10:36 PM, Maarten Brock wrote: >> Am 15.01.2011 11:45, schrieb Maarten Brock: >>> Some suggestions. >>> >>> * Rename tree_dec.hh to tree_dec.hpp >>> * Remove #include "sdccconf.h" >>> * Remove #include "SDCCy.h" >>> * Use new/delete for the dynamic allocated arrays >> Done. You might want to try the current optralloc branch. I made >> multiple other changes which are likely to fix some of the other errors >> you got as well. >> >> Philipp > Thank you, I've switched my copy. > > I also removed "sdccconf.h" once again as it still does > not exist when building with MSVC. It is handled by > SDCCglobl.h which you already have. > > Unfortunately, the assignment errors are still there. > See the attachment with the full report. > > Maarten > |
From: Philipp K. K. <pk...@sp...> - 2011-01-17 23:07:05
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 17.01.2011 22:36, schrieb Maarten Brock: > Unfortunately, the assignment errors are still there. > See the attachment with the full report. I had used a std::map iterator where I should have used a std::multimap iterator. Current svn should have all the errors from your report fixed. Philipp -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk00y5AACgkQbtUV+xsoLpoUHwCgmegewcbNrJqDtl+G2MpkozwV 3wkAoOngvXXv6k1ArDMKTkolRTpLmgZt =rdiN -----END PGP SIGNATURE----- |
From: Maarten B. <sou...@ds...> - 2011-01-18 22:01:52
|
> Am 17.01.2011 22:36, schrieb Maarten Brock: > >> Unfortunately, the assignment errors are still there. >> See the attachment with the full report. > > I had used a std::map iterator where I should have used a std::multimap > iterator. > Current svn should have all the errors from your report fixed. > > Philipp Thank you! Now it compiles. It still generates warnings though, but I can live with that. If you want to merge into the trunk now, I have no objections assuming the regression tests still pass. I'm not sure if I'll get around to updating the MSVC 2010 projects this week and next week I'm going on a skiing holiday in Austria. Can this wait two weeks? Greets, Maarten |
From: Philipp K. K. <pk...@sp...> - 2011-01-18 22:32:48
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Am 18.01.2011 23:01, schrieb Maarten Brock: > > Thank you! Now it compiles. It still generates warnings though, but I can > live with that. If you want to merge into the trunk now, I have no > objections assuming the regression tests still pass. Could you send me the warnings? There could be some that I would want to fix among them. > I'm not sure if I'll get around to updating the MSVC 2010 projects this > week and next week I'm going on a skiing holiday in Austria. Can this wait > two weeks? It can wait. Philipp -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk02FQoACgkQbtUV+xsoLpo0CgCfTnhcKFPHBIfuzlH5CGnfUPAP 0aAAn0A1i4f8vmy09WAO8hXnXjs66uFl =nNcv -----END PGP SIGNATURE----- |
From: Borut R. <bor...@si...> - 2011-01-18 22:29:58
|
On 01/18/2011 11:01 PM, Maarten Brock wrote: >> Am 17.01.2011 22:36, schrieb Maarten Brock: >> >>> Unfortunately, the assignment errors are still there. >>> See the attachment with the full report. >> I had used a std::map iterator where I should have used a std::multimap >> iterator. >> Current svn should have all the errors from your report fixed. >> >> Philipp > Thank you! Now it compiles. It still generates warnings though, but I can > live with that. If you want to merge into the trunk now, I have no > objections assuming the regression tests still pass. Great! > I'm not sure if I'll get around to updating the MSVC 2010 projects this > week and next week I'm going on a skiing holiday in Austria. Can this wait > two weeks? I would rather wait with the merge until MSVC 2010 compilation is functional. Philipp is still quite active making changes in the optralloc branch, so is probably better to have it on a separate branch not to stabilize the sdcc snapshot builds. I converted MSVC 6.0 project to MSVC 2010, but when try to compile sdcc, I get the following errors: ------ Build started: Project: mcs51, Configuration: Debug Win32 ------ rtrack.c c1 : fatal error C1060: compiler is out of heap space ralloc.c c1 : fatal error C1060: compiler is out of heap space peep.c c1 : fatal error C1060: compiler is out of heap space main.c c1 : fatal error C1060: compiler is out of heap space gen.c c1 : fatal error C1060: compiler is out of heap space Generating Code... This makes me a little bit nervous an it is an other reason why to wait wit the merge. Do you have the same problems? Do you know the solution? I already tried to allocate more memory to the WinXP virtual machine (from 512 MB to cca. 800 MB), but it did not help... Borut |
From: Maarten B. <sou...@ds...> - 2011-01-19 09:48:47
|
Borut, >> I'm not sure if I'll get around to updating the MSVC 2010 projects this >> week and next week I'm going on a skiing holiday in Austria. Can this >> wait two weeks? > > I would rather wait with the merge until MSVC 2010 compilation is > functional. Philipp is still quite active making changes in the > optralloc branch, so is probably better to have it on a separate branch > not to stabilize the sdcc snapshot builds. I converted MSVC 6.0 project > to MSVC 2010, but when try to compile sdcc, I get the following errors: > > ------ Build started: Project: mcs51, Configuration: Debug Win32 ------ > rtrack.c > c1 : fatal error C1060: compiler is out of heap space > ralloc.c > c1 : fatal error C1060: compiler is out of heap space > peep.c > c1 : fatal error C1060: compiler is out of heap space > main.c > c1 : fatal error C1060: compiler is out of heap space > gen.c > c1 : fatal error C1060: compiler is out of heap space > Generating Code... > > This makes me a little bit nervous an it is an other reason why to wait > wit the merge. > Do you have the same problems? Do you know the solution? I already tried > to allocate more memory to the WinXP virtual machine (from 512 MB to > cca. 800 MB), but it did not help... I had the same errors and in a previous email I already wrote that the solution was to remove the /Zm option. Previously it was needed to compile source with a large peeph.rul include. Now it is no longer necessary and creates problems instead. The other problem I have is with sdld_all which cannot find some files anymore but also seems without special build commands. Maarten |
From: Maarten B. <sou...@ds...> - 2011-01-19 22:13:36
Attachments:
ralloc2.txt
|
> Am 18.01.2011 23:01, schrieb Maarten Brock: > > > > Thank you! Now it compiles. It still generates warnings though, but I can > > live with that. If you want to merge into the trunk now, I have no > > objections assuming the regression tests still pass. > > Could you send me the warnings? There could be some that I would want to > fix among them. > > Philipp The output messages are in the attached file. Hope it helps. Maarten |
From: Borut R. <bor...@gm...> - 2011-01-20 16:17:38
|
Hi Maarten, I noticed that you are using boost library version 1_44, while the latest version is 1_45_0. Is there a reason why you are not using the most recent one? Borut On 01/19/2011 11:13 PM, Maarten Brock wrote: >> Am 18.01.2011 23:01, schrieb Maarten Brock: >>> Thank you! Now it compiles. It still generates warnings though, but I can >>> live with that. If you want to merge into the trunk now, I have no >>> objections assuming the regression tests still pass. >> Could you send me the warnings? There could be some that I would want to >> fix among them. >> >> Philipp > The output messages are in the attached file. Hope it > helps. > > Maarten |
From: Borut R. <bor...@gm...> - 2011-01-20 18:10:53
|
Hi Maarten, I get the following errors when compiling ralloc2.cc. Obviously you don't have such a problem. Have you already solve it? Borut ralloc2.cc n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(94) : error C2380: type(s) preceding 'operand' (constructor with return type, or illegal redefinition of current class-name?) n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : warning C4832: token '.' is illegal after UDT 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see declaration of 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2273: 'function-style cast' : illegal as right side of '->' operator n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2228: left of '.symOperand' must have class/struct/union n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2227: left of '->key' must point to class/struct/union/generic type n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : warning C4832: token '.' is illegal after UDT 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see declaration of 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2273: 'function-style cast' : illegal as right side of '->' operator n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2228: left of '.symOperand' must have class/struct/union n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2227: left of '->key' must point to class/struct/union/generic type n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : warning C4832: token '.' is illegal after UDT 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see declaration of 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2273: 'function-style cast' : illegal as right side of '->' operator n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2228: left of '.symOperand' must have class/struct/union n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2227: left of '->nRegs' must point to class/struct/union/generic type n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : warning C4832: token '.' is illegal after UDT 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see declaration of 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2273: 'function-style cast' : illegal as right side of '->' operator n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2228: left of '.symOperand' must have class/struct/union n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2227: left of '->key' must point to class/struct/union/generic type n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : warning C4832: token '.' is illegal after UDT 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see declaration of 'operand' n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2273: 'function-style cast' : illegal as right side of '->' operator n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2228: left of '.symOperand' must have class/struct/union n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2227: left of '->key' must point to class/struct/union/generic type n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(284) : warning C4018: '<' : signed/unsigned mismatch n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(487) : warning C4018: '<=' : signed/unsigned mismatch n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(562) : error C2679: binary '=' : no operator found which takes a right-hand operand of type 'std::_Tree<_Traits>::const_iterator' (or there is no acceptable conversion) with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] c:\program files\microsoft visual studio 9.0\vc\include\xtree(498): could be 'std::_Tree<_Traits>::iterator &std::_Tree<_Traits>::iterator::operator =(const std::_Tree<_Traits>::iterator &)' with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] while trying to match the argument list '(std::_Tree<_Traits>::iterator, std::_Tree<_Traits>::const_iterator)' with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(562) : error C2679: binary '=' : no operator found which takes a right-hand operand of type 'std::_Tree<_Traits>::const_iterator' (or there is no acceptable conversion) with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] c:\program files\microsoft visual studio 9.0\vc\include\xtree(498): could be 'std::_Tree<_Traits>::iterator &std::_Tree<_Traits>::iterator::operator =(const std::_Tree<_Traits>::iterator &)' with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] while trying to match the argument list '(std::_Tree<_Traits>::iterator, std::_Tree<_Traits>::const_iterator)' with [ _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> ] Creating browse information file... Microsoft Browse Information Maintenance Utility Version 9.00.21022 Copyright (C) Microsoft Corporation. All rights reserved. BSCMAKE: error BK1506 : cannot open file '.\Debug\gen.sbr': No such file or directory Build log was saved at "file://N:\svn_snapshots\sdcc\optralloc\src\z80\Debug\BuildLog.htm" z80 - 19 error(s), 7 warning(s) On 01/20/2011 05:17 PM, Borut Ražem wrote: > Hi Maarten, > > I noticed that you are using boost library version 1_44, while the > latest version is 1_45_0. Is there a reason why you are not using the > most recent one? > > Borut > > On 01/19/2011 11:13 PM, Maarten Brock wrote: >>> Am 18.01.2011 23:01, schrieb Maarten Brock: >>>> Thank you! Now it compiles. It still generates warnings though, but >>>> I can >>>> live with that. If you want to merge into the trunk now, I have no >>>> objections assuming the regression tests still pass. >>> Could you send me the warnings? There could be some that I would >>> want to >>> fix among them. >>> >>> Philipp >> The output messages are in the attached file. Hope it >> helps. >> >> Maarten |
From: Maarten B. <sou...@ds...> - 2011-01-20 21:46:30
|
Hi Borut, I'm using Boost 1.44 because there was no installer for 1.45 yet last week. I guess there still isn't. I did not want to additionaly get into the trouble of compiling the Boost libraries themselves. When I use SDCC for a project I usually do the same and use the latest release instead of a nightly snapshot. About those errors. Yes, I saw them and I think I mentioned them previously too. I fixed it by renaming the 'operand' union inside the 'operand' struct typedef to 'svt' since it contains the fields symOperand, valOperand and typeOperand. I dared not make it an anonymous union. Then I also had to adapt all macros and use them some more often here and there and adapt some source in itself. If you like I can try to commit this to the branch tomorrow before I leave. Any objections to renaming operand to svt? Greets, Maarten > Hi Maarten, > > I get the following errors when compiling ralloc2.cc. Obviously you > don't have such a problem. Have you already solve it? > > Borut > > ralloc2.cc > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(94) : error C2380: > type(s) preceding 'operand' (constructor with return type, or illegal > redefinition of current class-name?) > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : warning C4832: > token '.' is illegal after UDT 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see > declaration of 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2273: > 'function-style cast' : illegal as right side of '->' operator > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2228: > left of '.symOperand' must have class/struct/union > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(190) : error C2227: > left of '->key' must point to class/struct/union/generic type > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : warning C4832: > token '.' is illegal after UDT 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see > declaration of 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2273: > 'function-style cast' : illegal as right side of '->' operator > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2228: > left of '.symOperand' must have class/struct/union > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(192) : error C2227: > left of '->key' must point to class/struct/union/generic type > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : warning C4832: > token '.' is illegal after UDT 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see > declaration of 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2273: > 'function-style cast' : illegal as right side of '->' operator > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2228: > left of '.symOperand' must have class/struct/union > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(193) : error C2227: > left of '->nRegs' must point to class/struct/union/generic type > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : warning C4832: > token '.' is illegal after UDT 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see > declaration of 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2273: > 'function-style cast' : illegal as right side of '->' operator > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2228: > left of '.symOperand' must have class/struct/union > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2227: > left of '->key' must point to class/struct/union/generic type > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : warning C4832: > token '.' is illegal after UDT 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccicode.h(75) : see > declaration of 'operand' > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2273: > 'function-style cast' : illegal as right side of '->' operator > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2228: > left of '.symOperand' must have class/struct/union > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(194) : error C2227: > left of '->key' must point to class/struct/union/generic type > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(284) : warning C4018: > '<' : signed/unsigned mismatch > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(487) : warning C4018: > '<=' : signed/unsigned mismatch > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(562) : error C2679: > binary '=' : no operator found which takes a right-hand operand of type > 'std::_Tree<_Traits>::const_iterator' (or there is no acceptable conversion) > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > c:\program files\microsoft visual studio > 9.0\vc\include\xtree(498): could be 'std::_Tree<_Traits>::iterator > &std::_Tree<_Traits>::iterator::operator =(const > std::_Tree<_Traits>::iterator &)' > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > while trying to match the argument list > '(std::_Tree<_Traits>::iterator, std::_Tree<_Traits>::const_iterator)' > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(562) : error C2679: > binary '=' : no operator found which takes a right-hand operand of type > 'std::_Tree<_Traits>::const_iterator' (or there is no acceptable conversion) > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > c:\program files\microsoft visual studio > 9.0\vc\include\xtree(498): could be 'std::_Tree<_Traits>::iterator > &std::_Tree<_Traits>::iterator::operator =(const > std::_Tree<_Traits>::iterator &)' > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > while trying to match the argument list > '(std::_Tree<_Traits>::iterator, std::_Tree<_Traits>::const_iterator)' > with > [ > > _Traits=std::_Tset_traits<var_t,std::less<var_t>,std::allocator<var_t>,false> > ] > Creating browse information file... > Microsoft Browse Information Maintenance Utility Version 9.00.21022 > Copyright (C) Microsoft Corporation. All rights reserved. > BSCMAKE: error BK1506 : cannot open file '.\Debug\gen.sbr': No such file > or directory > Build log was saved at > "file://N:\svn_snapshots\sdcc\optralloc\src\z80\Debug\BuildLog.htm" > z80 - 19 error(s), 7 warning(s) > > > On 01/20/2011 05:17 PM, Borut Razem wrote: > > Hi Maarten, > > > > I noticed that you are using boost library version 1_44, while the > > latest version is 1_45_0. Is there a reason why you are not using the > > most recent one? > > > > Borut > > > > On 01/19/2011 11:13 PM, Maarten Brock wrote: > >>> Am 18.01.2011 23:01, schrieb Maarten Brock: > >>>> Thank you! Now it compiles. It still generates warnings though, but > >>>> I can > >>>> live with that. If you want to merge into the trunk now, I have no > >>>> objections assuming the regression tests still pass. > >>> Could you send me the warnings? There could be some that I would > >>> want to > >>> fix among them. > >>> > >>> Philipp > >> The output messages are in the attached file. Hope it > >> helps. > >> > >> Maarten > > > ------------------------------------------------------------------------------ > Protect Your Site and Customers from Malware Attacks > Learn about various malware tactics and how to avoid them. Understand > malware threats, the impact they can have on your business, and how you > can protect your company and customers by using code signing. > http://p.sf.net/sfu/oracle-sfdevnl > _______________________________________________ > sdcc-devel mailing list > sdc...@li... > https://lists.sourceforge.net/lists/listinfo/sdcc-devel > |
From: Borut R. <bor...@gm...> - 2011-01-21 08:17:35
|
On 01/20/2011 10:46 PM, Maarten Brock wrote: > Hi Borut, > > I'm using Boost 1.44 because there was no installer for 1.45 yet last > week. I guess there still isn't. I did not want to additionaly get into > the trouble of compiling the Boost libraries themselves. OK. I didn't use the installer but I installed (actually only unpacked) the .tar.bz2. I didn't compile it since I think that the compilation is not necessary since the majority (at least the stuff we need) is implemented as C++ templates in header files. > When I use SDCC for a project I usually do the same and use the latest > release instead of a nightly snapshot. > > About those errors. Yes, I saw them and I think I mentioned them > previously too. Sorry if I overlooked it :-[ > I fixed it by renaming the 'operand' union inside the > 'operand' struct typedef to 'svt' since it contains the fields > symOperand, valOperand and typeOperand. I dared not make it an anonymous > union. Then I also had to adapt all macros and use them some more often > here and there and adapt some source in itself. > I already started to do exactly the same thing, only renaming 'operand' to 'oprnd'. Then I saw the mail that you are able to compile it without errors.... > If you like I can try to commit this to the branch tomorrow before I > leave. Any objections to renaming operand to svt? > Yes please, commit it before you leave, if possible. I don't have the objections about renaming. Can you please commit also the project files, even if you think that they are not prefect yet? So that I can take a look, compare them with mine, ... I assume that you'll commit everything to the optalloc branch. Thanks and happy skiing, Borut |
From: Maarten B. <sou...@ds...> - 2011-01-21 11:37:49
|
Borut, > On 01/20/2011 10:46 PM, Maarten Brock wrote: > > Hi Borut, > > > > I'm using Boost 1.44 because there was no installer for 1.45 yet last > > week. I guess there still isn't. I did not want to additionaly get into > > the trouble of compiling the Boost libraries themselves. > > OK. I didn't use the installer but I installed (actually only unpacked) > the .tar.bz2. I didn't compile it since I think that the compilation is > not necessary since the majority (at least the stuff we need) is > implemented as C++ templates in header files. You do need some library from boost. Setting the include and lib paths was enough for me. > > I fixed it by renaming the 'operand' union inside the > > 'operand' struct typedef to 'svt' since it contains the fields > > symOperand, valOperand and typeOperand. I dared not make it an anonymous > > union. Then I also had to adapt all macros and use them some more often > > here and there and adapt some source in itself. > > > > I already started to do exactly the same thing, only renaming 'operand' > to 'oprnd'. Then I saw the mail that you are able to compile it without > errors.... > > > If you like I can try to commit this to the branch tomorrow before I > > leave. Any objections to renaming operand to svt? > > > > Yes please, commit it before you leave, if possible. I don't have the > objections about renaming. Can you please commit also the project files, > even if you think that they are not prefect yet? So that I can take a > look, compare them with mine, ... > > I assume that you'll commit everything to the optalloc branch. I have commited the renaming of operand to svt to the optralloc branch. I have also commited the sdcc solution and accompanying project files. > Thanks and happy skiing, > Borut Thank you. Maarten |
From: Borut R. <bor...@gm...> - 2011-01-21 15:05:48
|
Maarten, sorry to bother you so much, but I'm still getting the error: n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(237): error C2679: binary '=' : no operator found which takes a right-hand operand of type 'unsigned int' (or there is no acceptable conversion) do you have an idea what I'm doing wrong? Where did you get the boost Windows installer? Is this the one: http://www.boostpro.com/download/? Borut Generating Code... ralloc2.cc n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(284): warning C4018: '<' : signed/unsigned mismatch n:\svn_snapshots\sdcc\optralloc\src\sdccralloc.hpp(487): warning C4018: '<=' : signed/unsigned mismatch n:\svn_snapshots\sdcc\optralloc\src\z80\ralloc2.cc(364): warning C4018: '<' : signed/unsigned mismatch n:\svn_snapshots\sdcc\optralloc\src\z80\ralloc2.cc(412) : see reference to function template instantiation 'void tree_dec_ralloc<tree_dec_t,cfg_t,con_t>(T_t &,const G_t &,const I_t &)' being compiled with [ T_t=tree_dec_t, G_t=cfg_t, I_t=con_t ] n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(237): error C2679: binary '=' : no operator found which takes a right-hand operand of type 'unsigned int' (or there is no acceptable conversion) n:\boost_1_45_0\boost\range\irange.hpp(91): could be 'boost::range_detail::integer_iterator<Integer> &boost::range_detail::integer_iterator<Integer>::operator =(const boost::range_detail::integer_iterator<Integer> &)' with [ Integer=unsigned int ] while trying to match the argument list '(boost::range_detail::integer_iterator<Integer>, unsigned int)' with [ Integer=unsigned int ] n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(262) : see reference to function template instantiation 'void add_vertices_to_tree_decomposition<T_t,std::reverse_iterator<_RanIt>,boost::adjacency_list<OutEdgeListS,VertexListS,DirectedS>>(T_t &,const v_t,const v_t,G_t &,std::vector<_Ty,_Ax> &)' being compiled with [ T_t=tree_dec_t, _RanIt=std::_List_const_iterator<std::_List_val<unsigned int,std::allocator<char32_t>>>, OutEdgeListS=boost::vecS, VertexListS=boost::vecS, DirectedS=boost::undirectedS, v_t=std::reverse_iterator<std::_List_const_iterator<std::_List_val<unsigned int,std::allocator<char32_t>>>>, G_t=boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS>, _Ty=bool, _Ax=std::allocator<bool> ] n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(272) : see reference to function template instantiation 'void tree_decomposition_from_elimination_ordering<T_t,G_t>(T_t &,const std::list<_Ty> &,const G_t &)' being compiled with [ T_t=tree_dec_t, G_t=cfg_t, _Ty=unsigned int ] n:\svn_snapshots\sdcc\optralloc\src\z80\ralloc2.cc(403) : see reference to function template instantiation 'void thorup_tree_decomposition<tree_dec_t,cfg_t>(T_t &,const G_t &)' being compiled with [ T_t=tree_dec_t, G_t=cfg_t ] |
From: Philipp K. K. <pk...@sp...> - 2011-01-21 15:22:19
|
Am 21.01.2011 16:05, schrieb Borut Razem: > Maarten, > > sorry to bother you so much, but I'm still getting the error: > > n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(237): error C2679: > binary '=' : no operator found which takes a right-hand operand of type > 'unsigned int' (or there is no acceptable conversion) I noticed I forgot the boost:: prefix in that line; it's fixed in current svn. Philipp |
From: Borut R. <bor...@gm...> - 2011-01-21 17:25:40
|
On 01/21/2011 04:23 PM, Philipp Klaus Krause wrote: > Am 21.01.2011 16:05, schrieb Borut Razem: >> Maarten, >> >> sorry to bother you so much, but I'm still getting the error: >> >> n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(237): error C2679: >> binary '=' : no operator found which takes a right-hand operand of type >> 'unsigned int' (or there is no acceptable conversion) > I noticed I forgot the boost:: prefix in that line; it's fixed in > current svn. Unfortunately the error is still there: n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(237): error C2679: binary '=' : no operator found which takes a right-hand operand of type 'unsigned int' (or there is no acceptable conversion) n:\boost_1_45_0\boost\range\irange.hpp(91): could be 'boost::range_detail::integer_iterator<Integer> &boost::range_detail::integer_iterator<Integer>::operator =(const boost::range_detail::integer_iterator<Integer> &)' with [ Integer=unsigned int ] while trying to match the argument list '(boost::range_detail::integer_iterator<Integer>, unsigned int)' with [ Integer=unsigned int ] n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(262) : see reference to function template instantiation 'void add_vertices_to_tree_decomposition<T_t,std::reverse_iterator<_RanIt>,boost::adjacency_list<OutEdgeListS,VertexListS,DirectedS>>(T_t &,const v_t,const v_t,G_t &,std::vector<_Ty,_Ax> &)' being compiled with [ T_t=tree_dec_t, _RanIt=std::_List_const_iterator<std::_List_val<unsigned int,std::allocator<char32_t>>>, OutEdgeListS=boost::vecS, VertexListS=boost::vecS, DirectedS=boost::undirectedS, v_t=std::reverse_iterator<std::_List_const_iterator<std::_List_val<unsigned int,std::allocator<char32_t>>>>, G_t=boost::adjacency_list<boost::vecS,boost::vecS,boost::undirectedS>, _Ty=bool, _Ax=std::allocator<bool> ] n:\svn_snapshots\sdcc\optralloc\src\sdcctree_dec.hpp(272) : see reference to function template instantiation 'void tree_decomposition_from_elimination_ordering<T_t,G_t>(T_t &,const std::list<_Ty> &,const G_t &)' being compiled with [ T_t=tree_dec_t, G_t=cfg_t, _Ty=unsigned int ] n:\svn_snapshots\sdcc\optralloc\src\z80\ralloc2.cc(403) : see reference to function template instantiation 'void thorup_tree_decomposition<tree_dec_t,cfg_t>(T_t &,const G_t &)' being compiled with [ T_t=tree_dec_t, G_t=cfg_t ] Borut |
From: Maarten B. <sou...@ds...> - 2011-01-21 15:55:19
|
> Where did you get the boost Windows installer? Is this the one: > http://www.boostpro.com/download/? Yes, that's the one. > > Borut |