llvm-poolalloc compilation failed

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

llvm-poolalloc compilation failed

Ryan M. Lefever
I tried to compile the currently checked in version of llvm-poolalloc
with the curently checked in version of llvm and got the following errors:

llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
unsigned int, const llvm::TargetData&)':
Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
named 'getTypeAlignment'

There seem to be 3 methods in TargetData that are close to
getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
what method should be used.

Thanks,
Ryan
_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

Chris Lattner
On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
> I tried to compile the currently checked in version of llvm-poolalloc
> with the curently checked in version of llvm and got the following errors:
>
> llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
> Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
> unsigned int, const llvm::TargetData&)':
> Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
> named 'getTypeAlignment'

The getTypeAlignment method was renamed to getTypeAlignmentABI, it should
be safe to make that substitution.

-Chris

> There seem to be 3 methods in TargetData that are close to
> getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
> getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
> what method should be used.
>
> Thanks,
> Ryan
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>

-Chris

--
http://nondot.org/sabre/
http://llvm.org/
_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

Vikram S. Adve
This kind of regression is a fallout of having some code live in separate CVS modules as "projects."  It would help to have someone volunteer to maintain each project, which can be as little as making sure it continues to compile or as much as supporting it with answers to questions and bug fixes.  Here is the current list of projects:

llvm-poolalloc
llvm-java (currently unused AFAIK)
llvm-stacker
llvm-tv


Would anyone like to volunteer for one or more of these?


--Vikram

http://www.cs.uiuc.edu/~vadve

http://llvm.cs.uiuc.edu/



On Jan 23, 2007, at 10:53 PM, Chris Lattner wrote:

On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
I tried to compile the currently checked in version of llvm-poolalloc
with the curently checked in version of llvm and got the following errors:

llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
unsigned int, const llvm::TargetData&)':
Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
named 'getTypeAlignment'

The getTypeAlignment method was renamed to getTypeAlignmentABI, it should 
be safe to make that substitution.

-Chris

There seem to be 3 methods in TargetData that are close to
getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
what method should be used.

Thanks,
Ryan
_______________________________________________
LLVM Developers mailing list


-Chris

-- 
_______________________________________________
LLVM Developers mailing list


_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

John Criswell
Vikram S. Adve wrote:

> This kind of regression is a fallout of having some code live in
> separate CVS modules as "projects."  It would help to have someone
> volunteer to maintain each project, which can be as little as making
> sure it continues to compile or as much as supporting it with answers
> to questions and bug fixes.  Here is the current list of projects:

I've been using the llvm-poolalloc project in my work and have currently
been keeping it up to date.  However, mainline LLVM CVS has been moving
faster than I'm able to keep up; constantly updating llvm-poolalloc
every time mainline LLVM has changed has proved too time consuming.  
I've decided that it's better to update once mainline has stabalized
(and I've got more time).

So, for now, I'm using LLVM 1.9 (technically, the release_19 branch,
which is identical to LLVM 1.9 except that DSA has been moved from the
LLVM source base to the Pool Allocation project).  Using the release_19
branch of the llvm and llvm-poolalloc modules should give you something
that compiles and (hopefully) works.

To get it, download the llvm-gcc3 or llvm-gcc4 frontend from the website
and then grab the LLVM/PoolAllocation source using the following commands:

cvs -d :pserver:[hidden email]:/var/cvs/llvm login
<Hit return for an empty password>
cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm
cd llvm/projects
cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm-poolalloc

-- John T.

>
> llvm-poolalloc
> llvm-java (currently unused AFAIK)
> llvm-stacker
> llvm-tv
>
>
> Would anyone like to volunteer for one or more of these?
>
>
> --Vikram
>
> http://www.cs.uiuc.edu/~vadve <http://www.cs.uiuc.edu/%7Evadve>
>
> http://llvm.cs.uiuc.edu/
>
>
>
> On Jan 23, 2007, at 10:53 PM, Chris Lattner wrote:
>
>> On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
>>
>>> I tried to compile the currently checked in version of llvm-poolalloc
>>> with the curently checked in version of llvm and got the following
>>> errors:
>>>
>>> llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
>>> Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
>>> unsigned int, const llvm::TargetData&)':
>>> Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
>>> named 'getTypeAlignment'
>>
>>
>> The getTypeAlignment method was renamed to getTypeAlignmentABI, it
>> should
>> be safe to make that substitution.
>>
>> -Chris
>>
>>> There seem to be 3 methods in TargetData that are close to
>>> getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
>>> getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
>>> what method should be used.
>>>
>>> Thanks,
>>> Ryan
>>> _______________________________________________
>>> LLVM Developers mailing list
>>> [hidden email] <mailto:[hidden email]>        
>>> http://llvm.cs.uiuc.edu
>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>
>>
>> -Chris
>>
>> --
>> http://nondot.org/sabre/
>> http://llvm.org/
>> _______________________________________________
>> LLVM Developers mailing list
>> [hidden email] <mailto:[hidden email]>        
>> http://llvm.cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>LLVM Developers mailing list
>[hidden email]         http://llvm.cs.uiuc.edu
>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>  
>

_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

Reid Spencer-2
In reply to this post by Vikram S. Adve
Vikram,

I maintain llvm-stacker and keep it checked out in my build trees. I
will continue to update it if it gets broken.

Reid.

On Tue, 2007-01-23 at 22:31 -0600, Vikram S. Adve wrote:

> This kind of regression is a fallout of having some code live in
> separate CVS modules as "projects."  It would help to have someone
> volunteer to maintain each project, which can be as little as making
> sure it continues to compile or as much as supporting it with answers
> to questions and bug fixes.  Here is the current list of projects:
>
>
> llvm-poolalloc
> llvm-java (currently unused AFAIK)
> llvm-stacker
> llvm-tv
>
> Would anyone like to volunteer for one or more of these?
>
>
> --Vikram
>
> http://www.cs.uiuc.edu/~vadve
>
> http://llvm.cs.uiuc.edu/
>
>
>
>
> On Jan 23, 2007, at 10:53 PM, Chris Lattner wrote:
>
> > On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
> > > I tried to compile the currently checked in version of
> > > llvm-poolalloc
> > > with the curently checked in version of llvm and got the following
> > > errors:
> > >
> > >
> > > llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
> > > Heuristic.cpp: In function 'bool Wants8ByteAlignment(const
> > > llvm::Type*,
> > > unsigned int, const llvm::TargetData&)':
> > > Heuristic.cpp:77: error: 'const class llvm::TargetData' has no
> > > member
> > > named 'getTypeAlignment'
> >
> >
> > The getTypeAlignment method was renamed to getTypeAlignmentABI, it
> > should
> > be safe to make that substitution.
> >
> >
> > -Chris
> >
> >
> > > There seem to be 3 methods in TargetData that are close to
> > > getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref
> > > (),
> > > getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc
> > > know
> > > what method should be used.
> > >
> > >
> > > Thanks,
> > > Ryan
> > > _______________________________________________
> > > LLVM Developers mailing list
> > > [hidden email]         http://llvm.cs.uiuc.edu
> > > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> > >
> > >
> >
> >
> > -Chris
> >
> >
> > --
> > http://nondot.org/sabre/
> > http://llvm.org/
> > _______________________________________________
> > LLVM Developers mailing list
> > [hidden email]         http://llvm.cs.uiuc.edu
> > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>
>
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

Ryan M. Lefever
In reply to this post by John Criswell
Chris,

I performed the steps you gave below to check out llvm-poolalloc.  When
I want to get new updates made to llvm-poolalloc, do I perform the
following commands?

cvs -d :pserver:[hidden email]:/var/cvs/llvm login
<Hit return for an empty password>
cd llvm/projects
cvs -d :pserver:[hidden email]:/var/cvs/llvm update -r release_19
llvm-poolalloc

In other words, do I specify the -r option to perform an update?  Also,
how do I produce the doxygen for DSA?

Regards,
Ryan

John T. Criswell wrote:

> Vikram S. Adve wrote:
>
>
>>This kind of regression is a fallout of having some code live in
>>separate CVS modules as "projects."  It would help to have someone
>>volunteer to maintain each project, which can be as little as making
>>sure it continues to compile or as much as supporting it with answers
>>to questions and bug fixes.  Here is the current list of projects:
>
>
> I've been using the llvm-poolalloc project in my work and have currently
> been keeping it up to date.  However, mainline LLVM CVS has been moving
> faster than I'm able to keep up; constantly updating llvm-poolalloc
> every time mainline LLVM has changed has proved too time consuming.  
> I've decided that it's better to update once mainline has stabalized
> (and I've got more time).
>
> So, for now, I'm using LLVM 1.9 (technically, the release_19 branch,
> which is identical to LLVM 1.9 except that DSA has been moved from the
> LLVM source base to the Pool Allocation project).  Using the release_19
> branch of the llvm and llvm-poolalloc modules should give you something
> that compiles and (hopefully) works.
>
> To get it, download the llvm-gcc3 or llvm-gcc4 frontend from the website
> and then grab the LLVM/PoolAllocation source using the following commands:
>
> cvs -d :pserver:[hidden email]:/var/cvs/llvm login
> <Hit return for an empty password>
> cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm
> cd llvm/projects
> cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm-poolalloc
>
> -- John T.
>
>
>>llvm-poolalloc
>>llvm-java (currently unused AFAIK)
>>llvm-stacker
>>llvm-tv
>>
>>
>>Would anyone like to volunteer for one or more of these?
>>
>>
>>--Vikram
>>
>>http://www.cs.uiuc.edu/~vadve <http://www.cs.uiuc.edu/%7Evadve>
>>
>>http://llvm.cs.uiuc.edu/
>>
>>
>>
>>On Jan 23, 2007, at 10:53 PM, Chris Lattner wrote:
>>
>>
>>>On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
>>>
>>>
>>>>I tried to compile the currently checked in version of llvm-poolalloc
>>>>with the curently checked in version of llvm and got the following
>>>>errors:
>>>>
>>>>llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
>>>>Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
>>>>unsigned int, const llvm::TargetData&)':
>>>>Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
>>>>named 'getTypeAlignment'
>>>
>>>
>>>The getTypeAlignment method was renamed to getTypeAlignmentABI, it
>>>should
>>>be safe to make that substitution.
>>>
>>>-Chris
>>>
>>>
>>>>There seem to be 3 methods in TargetData that are close to
>>>>getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
>>>>getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
>>>>what method should be used.
>>>>
>>>>Thanks,
>>>>Ryan
>>>>_______________________________________________
>>>>LLVM Developers mailing list
>>>>[hidden email] <mailto:[hidden email]>        
>>>>http://llvm.cs.uiuc.edu
>>>>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>>>
>>>
>>>-Chris
>>>
>>>--
>>>http://nondot.org/sabre/
>>>http://llvm.org/
>>>_______________________________________________
>>>LLVM Developers mailing list
>>>[hidden email] <mailto:[hidden email]>        
>>>http://llvm.cs.uiuc.edu
>>>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>
>>
>>------------------------------------------------------------------------
>>
>>_______________________________________________
>>LLVM Developers mailing list
>>[hidden email]         http://llvm.cs.uiuc.edu
>>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>>
>>
>
>
> _______________________________________________
> LLVM Developers mailing list
> [hidden email]         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev

--
Ryan M. Lefever  [http://www.ews.uiuc.edu/~lefever]
_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: llvm-poolalloc compilation failed

Reid Spencer-2
On Tue, 2007-02-06 at 17:35 -0600, Ryan M. Lefever wrote:
> Chris,
>
> I performed the steps you gave below to check out llvm-poolalloc.  When
> I want to get new updates made to llvm-poolalloc, do I perform the
> following commands?
>
> cvs -d :pserver:[hidden email]:/var/cvs/llvm login
> <Hit return for an empty password>

If you've already checked it out, you shouldn't need to login again. CVS
will remember you rlogin.

> cd llvm/projects
> cvs -d :pserver:[hidden email]:/var/cvs/llvm update -r release_19 llvm-poolalloc

The -r option specifies the revision you want to check out. If you don't
specify it then you get the latest revision of each file (the "HEAD").
If you specify a symbolic name like release_19 then you get the version
of each file corresponding to the tag "release_19". So, assuming you
previously checked out with -rrelease_19, this will only update
differences made to that tag since you checked out (probably few or
none).

> In other words, do I specify the -r option to perform an update?  

It depends on what you want to do:

1. Update to latest version of release 1.9: cvs update
2. Restore version to that of 1.9: cvs update -r release_19
3. Update to head version (latest): cvs update -A

Note that cvs will remember your "sticky tag" (release_19) from your
previous checkout. Item #3 overcomes that by using the -A option to tell
it to forget the sticky tag and really update to the HEAD version (not
the latest version of the tag as in #1)



> Also,
> how do I produce the doxygen for DSA?

Sorry, don't know about that. Should be just "make install" in the docs
directory like LLVM, but I don't know how the DSA repository is set up.

>
> Regards,
> Ryan
>
> John T. Criswell wrote:
> > Vikram S. Adve wrote:
> >
> >
> >>This kind of regression is a fallout of having some code live in
> >>separate CVS modules as "projects."  It would help to have someone
> >>volunteer to maintain each project, which can be as little as making
> >>sure it continues to compile or as much as supporting it with answers
> >>to questions and bug fixes.  Here is the current list of projects:
> >
> >
> > I've been using the llvm-poolalloc project in my work and have currently
> > been keeping it up to date.  However, mainline LLVM CVS has been moving
> > faster than I'm able to keep up; constantly updating llvm-poolalloc
> > every time mainline LLVM has changed has proved too time consuming.  
> > I've decided that it's better to update once mainline has stabalized
> > (and I've got more time).
> >
> > So, for now, I'm using LLVM 1.9 (technically, the release_19 branch,
> > which is identical to LLVM 1.9 except that DSA has been moved from the
> > LLVM source base to the Pool Allocation project).  Using the release_19
> > branch of the llvm and llvm-poolalloc modules should give you something
> > that compiles and (hopefully) works.
> >
> > To get it, download the llvm-gcc3 or llvm-gcc4 frontend from the website
> > and then grab the LLVM/PoolAllocation source using the following commands:
> >
> > cvs -d :pserver:[hidden email]:/var/cvs/llvm login
> > <Hit return for an empty password>
> > cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm
> > cd llvm/projects
> > cvs -d :pserver:[hidden email]:/var/cvs/llvm co -r release_19 llvm-poolalloc
> >
> > -- John T.
> >
> >
> >>llvm-poolalloc
> >>llvm-java (currently unused AFAIK)
> >>llvm-stacker
> >>llvm-tv
> >>
> >>
> >>Would anyone like to volunteer for one or more of these?
> >>
> >>
> >>--Vikram
> >>
> >>http://www.cs.uiuc.edu/~vadve <http://www.cs.uiuc.edu/%7Evadve>
> >>
> >>http://llvm.cs.uiuc.edu/
> >>
> >>
> >>
> >>On Jan 23, 2007, at 10:53 PM, Chris Lattner wrote:
> >>
> >>
> >>>On Tue, 23 Jan 2007, Ryan M. Lefever wrote:
> >>>
> >>>
> >>>>I tried to compile the currently checked in version of llvm-poolalloc
> >>>>with the curently checked in version of llvm and got the following
> >>>>errors:
> >>>>
> >>>>llvm[4]: Compiling Heuristic.cpp for Debug build  (PIC)
> >>>>Heuristic.cpp: In function 'bool Wants8ByteAlignment(const llvm::Type*,
> >>>>unsigned int, const llvm::TargetData&)':
> >>>>Heuristic.cpp:77: error: 'const class llvm::TargetData' has no member
> >>>>named 'getTypeAlignment'
> >>>
> >>>
> >>>The getTypeAlignment method was renamed to getTypeAlignmentABI, it
> >>>should
> >>>be safe to make that substitution.
> >>>
> >>>-Chris
> >>>
> >>>
> >>>>There seem to be 3 methods in TargetData that are close to
> >>>>getTypeAlignment(): getTypeAlignmentABI (), getTypeAlignmentPref (),
> >>>>getTypeAlignmentShift ().  Does anyone working on llvm-poolalloc know
> >>>>what method should be used.
> >>>>
> >>>>Thanks,
> >>>>Ryan
> >>>>_______________________________________________
> >>>>LLVM Developers mailing list
> >>>>[hidden email] <mailto:[hidden email]>        
> >>>>http://llvm.cs.uiuc.edu
> >>>>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> >>>>
> >>>
> >>>-Chris
> >>>
> >>>--
> >>>http://nondot.org/sabre/
> >>>http://llvm.org/
> >>>_______________________________________________
> >>>LLVM Developers mailing list
> >>>[hidden email] <mailto:[hidden email]>        
> >>>http://llvm.cs.uiuc.edu
> >>>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> >>
> >>
> >>------------------------------------------------------------------------
> >>
> >>_______________________________________________
> >>LLVM Developers mailing list
> >>[hidden email]         http://llvm.cs.uiuc.edu
> >>http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
> >>
> >>
> >
> >
> > _______________________________________________
> > LLVM Developers mailing list
> > [hidden email]         http://llvm.cs.uiuc.edu
> > http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>

_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev