similar to: [LLVMdev] embedded metadata

In LLVM IR metadata is used to attach auxiliary information with various IR constructs. Currently metadata information is represented using MDNode and MDString. The metadata can refer to LLVM values but these references are not counted as regular "uses" of these values because metadata is maintained 'on the side'. This ensures that the optimizer is not influenced by auxiliary

On 12/21/11 11:24 PM, Oksana Tkachuk wrote: > Hello, > Thank you very much for the pointers. > I am are able to create new MDNodes, filled with some constants, and > attach them > to llvm instructions. However, the metadata map is not getting updated > as expected. > For example, instead of the expected new entry If you look at PoolMDPass::runOnModule(), you'll see that

Hello, Thank you very much for the pointers. I am are able to create new MDNodes, filled with some constants, and attach them to llvm instructions. However, the metadata map is not getting updated as expected. For example, instead of the expected new entry !n = metadata !{some values} we are getting !n = metadata !{null} Do you know what might be wrong? Do we need to enter the MDNodes into the

Devang Patel wrote: > In LLVM IR metadata is used to attach auxiliary information with > various IR constructs. Currently metadata information is represented > using MDNode and MDString. The metadata can refer to LLVM values but > these references are not counted as regular "uses" of these values > because metadata is maintained 'on the side'. This ensures that the

> Dear llvm-ers, > I am trying to attach a customized metadata to llvm instructions. > Let's say, I want to attach some number to each instruction. > I am trying to use > > void Instruction::setMetadata(unsigned KindID, MDNode* Node) > > but I am not sure how to create a brand new instance of an MDNode. > Do you have any code samples demonstrating how to do that?

On Jul 27, 2009, at 10:10 AM, Devang Patel wrote: > In LLVM IR metadata is used to attach auxiliary information with > various IR constructs. Currently metadata information is represented > using MDNode and MDString. The metadata can refer to LLVM values but > these references are not counted as regular "uses" of these values > because metadata is maintained 'on the

Pekka Nikander wrote: >>>> Here is another version of the patch. This one includes also a small patch to llvm-gcc so that it generates the type metadata for structures and classes. This one also generates and parses correctly the metadata for .ll files. No .bc support yet. The biggest problem with this version is that it breaks when the compiler/linker performs type reductions, and

On Mon, Jul 27, 2009 at 1:13 PM, Dan Gohman<gohman at apple.com> wrote: > > On Jul 27, 2009, at 10:10 AM, Devang Patel wrote: > > >> In LLVM IR metadata is used to attach auxiliary information with >> various IR constructs. Currently metadata information is represented >> using MDNode and MDString. The metadata can refer to LLVM values but >> these

Dear llvm-ers, I am trying to attach a customized metadata to llvm instructions. Let's say, I want to attach some number to each instruction. I am trying to use void Instruction::setMetadata(unsigned KindID, MDNode* Node) but I am not sure how to create a brand new instance of an MDNode. Do you have any code samples demonstrating how to do that? Due to some tool dependencies, I am using

> On 2014-Nov-10, at 08:30, Chris Lattner <clattner at apple.com> wrote: > >> On Nov 9, 2014, at 5:02 PM, Duncan P. N. Exon Smith <dexonsmith at apple.com> wrote: >> In response to my recent commits (e.g., [1]) that changed API from >> `MDNode` to `Value`, Eric had a really interesting idea [2] -- split >> metadata entirely from the `Value` hierarchy, and

Hi All, Type uniquing for C++ is in. Some data for Xalan with -flto -g: 9.9MB raw dwarf size, peak memory usage at 2.8GB The raw dwarf size was 58MB, memory usage was 7GB back in May, 2013. Other efforts at size reduction helped, and type uniquing improved on top of those. Data on building clang with "-flto -g" after type uniquing: 3.4GB MDNodes after parsing all bc files, 7GB

>>> Here is another version of the patch. This one includes also a small patch to llvm-gcc so that it generates the type metadata for structures and classes. This one also generates and parses correctly the metadata for .ll files. No .bc support yet. The biggest problem with this version is that it breaks when the compiler/linker performs type reductions, and I don't understand

On Jun 21, 2013, at 10:18 AM, David Blaikie wrote: > On Fri, Jun 21, 2013 at 10:13 AM, Manman Ren <mren at apple.com> wrote: >> >> More details please :] >> What do you mean by "name the metadata"? Are you referring to the name field >> of the MDNode? > > Using named metadata rather than unnamed metadata. > > rather than having: > >

On Jun 21, 2013 1:19 AM, "Jeremy Lakeman" <Jeremy.Lakeman at gmail.com> wrote: > > Wouldn't it be simpler to name the metadata based on the hash of the content? Then you could use a normal reference to that metadata without needing to create a new type or teach the rest of llvm how to use it... Sounds reasonable (probably with some nice prefix to get these things a

On Fri, Jun 21, 2013 at 10:13 AM, Manman Ren <mren at apple.com> wrote: > > More details please :] > What do you mean by "name the metadata"? Are you referring to the name field > of the MDNode? Using named metadata rather than unnamed metadata. rather than having: !llvm.hardref = !{metadata !"foo.h::myClass", !3, metadata !"bar.h:myEnum", !4} ...

Hi Manman, Thanks for sending this summary and progress plans - it's great to see the impact your changes have had and ideas for future direction. Type uniquing for C++ is in. Some data for Xalan with -flto -g: > 9.9MB raw dwarf size, peak memory usage at 2.8GB > The raw dwarf size was 58MB, memory usage was 7GB back in May, 2013. > Other efforts at size reduction helped, and type

Wouldn't it be simpler to name the metadata based on the hash of the content? Then you could use a normal reference to that metadata without needing to create a new type or teach the rest of llvm how to use it... On Fri, Jun 21, 2013 at 3:22 PM, Manman Ren <mren at apple.com> wrote: > > A summary of options for issue #3: > 3> To actually access the MDNode referenced via

More details please :] What do you mean by "name the metadata"? Are you referring to the name field of the MDNode? Thanks, Manman On Jun 21, 2013, at 7:19 AM, David Blaikie wrote: > > On Jun 21, 2013 1:19 AM, "Jeremy Lakeman" <Jeremy.Lakeman at gmail.com> wrote: > > > > Wouldn't it be simpler to name the metadata based on the hash of the content?

On Tue, Nov 12, 2013 at 1:01 PM, David Blaikie <dblaikie at gmail.com> wrote: > Hi Manman, > > Thanks for sending this summary and progress plans - it's great to see the > impact your changes have had and ideas for future direction. > > Type uniquing for C++ is in. Some data for Xalan with -flto -g: >> 9.9MB raw dwarf size, peak memory usage at 2.8GB >> The

A summary of options for issue #3: 3> To actually access the MDNode referenced via the hash value, we need to perform a lookup from the hash value to find the corresponding MDNode. The questions are where to store this map and how to keep it up-to-date when a MDNode is replaced. --------------------- Option a) a map in DwarfDebug, AsmWriter and DIBuilder, modify access functions in DI classes