similar to: [LLVMdev] Metadata

On Wed, Feb 10, 2010 at 12:02 AM, David Greene <dag at cray.com> wrote: > Since 2.7 is getting close to code freeze, I'd like to see if I can get in our > changes to track non-temporal memory operations into trunk. > > As discussed earlier, I was hoping to do this via metadata.  It's pretty clear > how to attach the data to Instructions, but after that, I'm not

On Feb 10, 2010, at 9:27 AM, Devang Patel wrote: > On Wed, Feb 10, 2010 at 12:02 AM, David Greene <dag at cray.com> wrote: >> Since 2.7 is getting close to code freeze, I'd like to see if I can get in our >> changes to track non-temporal memory operations into trunk. >> >> As discussed earlier, I was hoping to do this via metadata. It's pretty clear

On Wednesday 10 February 2010 11:46:25 Chris Lattner wrote: > > There is not any mechanism to attach metadata with SDNode or > > MachineInstrs today. > > For nontemporal stores, you wouldn't want to do this anyway. In > selectiondag builder, you'd want to check the store instruction to see if > it has the metadata on it, and if so make a different ISD opcode or

On Thursday 11 February 2010 14:44:23 Dan Gohman wrote: > > Then we can't use it to hold a non-temporal flag. The operand might be > > non-temporal in one context but it may not be in another. > > Sharing only happens when two instructions have the "same" memory > reference info. You just need to make sure that the non-temporal > flag is significant.

While hacking around in the SelectionDAG build code, I've made the isVolatile, (new) isNonTemporal and Alignment parameters to SelectionDAG::getLoad/getStore and friends non-default. I've already caught one bug in the XCore backend by doing this: if (Offset % 4 == 0) { // We've managed to infer better alignment information than the load // already has. Use an aligned

On Feb 11, 2010, at 12:07 PM, David Greene wrote: > On Thursday 11 February 2010 14:02:13 Dan Gohman wrote: > >>>> Putting a bit (or multiple bits) in MachineMemOperand for this >>>> would also make sense. >>> >>> Is there any chance a MachineMemOperand will be shared by multiple >>> instructions? >> >> Yes. > > Then

On Thursday 11 February 2010 17:40:24 David Greene wrote: > While hacking around in the SelectionDAG build code, I've made the > isVolatile, (new) isNonTemporal and Alignment parameters to > SelectionDAG::getLoad/getStore and friends non-default. > > I've already caught one bug in the XCore backend by doing this: > > if (Offset % 4 == 0) { > // We've

On Thu, Feb 11, 2010 at 10:54 PM, David Greene <dag at cray.com> wrote: > On Thursday 11 February 2010 16:31:16 Garrison Venn wrote: >> I don't think the lang ref metadata grammar section >> (http://llvm.org/docs/LangRef.html#metadata) has been fully updated with >> how !dbg metadata is used in >> http://llvm.org/docs/SourceLevelDebugging.html. At least to me

On Thu, Feb 11, 2010 at 8:50 PM, David Greene <dag at cray.com> wrote: > > So I tried a bunch of things: > > %r8 = load <2 x double>* %r6, align 16, metadata !"nontemporal" > %r8 = load <2 x double>* %r6, align 16, metadata !nontemporal > %r8 = load <2 x double>* %r6, align 16, !{ metadata !"nontemporal" } > %r8 = load <2 x

hi Devang and thanks for the tips, i finally managed to fit all the pieces together into something that seems to work. It's probably not the best (or even correct!) way of doing it but here's a brief overview for reference: An instruction in the LLVM IR gets converted into an SDNode in the DAG then later into a MachineInstr. I'd already attached my own MDNodes to IR instructions I

Hello. Quentin, retrieving the LLVM IR instruction from which an SDNode originates is useful during the instruction selection phase. For example, I need to recover the LLVM IR variable which is used to fill an entire vector with the ISD::BUILD_VECTOR target-independent SDNode . From the recovered LLVM IR variable I can walk on the use-def-chains in order to get the most complete

Hi Roger, On Mon, Aug 23, 2010 at 4:01 PM, Roger Wang <innit42 at gmail.com> wrote: > Dear all, > > I'd like to find the memory location of certain instructions in a > compiled/linked binary. During the IR phase, I tag instructions I'm > interested in with LLVM'-2.7's new metadata (MDNodes with an identifiable > ID). I'd now like to propagate that data

Andrew Trick <atrick at apple.com> writes: >> Ok, but that doesn't answer the question. Is SchedulerDAG going away? >> If so, what's the timeframe for that? 3.2? > > SchedulerDAG is used for both SD scheduling and MI scheduling. It's not going away. Oh! That's good news! > SD scheduling is not going away in 3.2--it will be the first release with MI

On Thursday 11 February 2010 14:05:21 David Greene wrote: > Either ParseLoad and probably other instructions need to look for metadata > explicitly or ParseOptionalCommaAlign needs to know about general metadata. > > My inkling is to fix ParseOptionalCommaAlign. Sound reasonable? Well, that's a rat's nest. I backed up and thought maybe I have the metadata syntax wrong. So

Dear all, I'd like to find the memory location of certain instructions in a compiled/linked binary. During the IR phase, I tag instructions I'm interested in with LLVM'-2.7's new metadata (MDNodes with an identifiable ID). I'd now like to propagate that data to the assembly via a custom DWARF tag I attach to each X86 instruction created from a tagged IR instruction. This will

On Thursday 11 February 2010 13:31:58 David Greene wrote: > > Putting a bit (or multiple bits) in MachineMemOperand for this > > would also make sense. > > Is there any chance a MachineMemOperand will be shared by multiple > instructions? So I tried to do this: %r8 = load <2 x double>* %r6, align 16, !"nontemporal" and the assembler doesn't like it.

Right now the MemSDNode keeps a volatile bit in the SubclassData to mark volatile memory operations. We have some changes we'd like to push back that adds a NonTemporal flag to MemSDNode to mark instructions where movnt (on x86) and other goodness can happen (we'll also add the TableGen patterns to properly select movnt). In our tree we simply added another flag to the MemSDNode

On Thursday 11 February 2010 16:31:16 Garrison Venn wrote: > I don't think the lang ref metadata grammar section > (http://llvm.org/docs/LangRef.html#metadata) has been fully updated with > how !dbg metadata is used in > http://llvm.org/docs/SourceLevelDebugging.html. At least to me it is not > clear. In my mind I translate the phrase "LLVM IR allows metadata to be >

Hello. I would like to access the LLVM IR Instruction from which an SDNode (from SelectionDAG) originates. For this I have modified: - llvm/lib/CodeGen/SelectionDAGISel.cpp, SelectionDAGISel::SelectBasicBlock(), namely I put SDB->clear() at the beginning of the method in order to avoid clearing NodeMap after creating the initial SelectionDAG from LLVM IR, since I want to access

I don't think the lang ref metadata grammar section (http://llvm.org/docs/LangRef.html#metadata) has been fully updated with how !dbg metadata is used in http://llvm.org/docs/SourceLevelDebugging.html. At least to me it is not clear. In my mind I translate the phrase "LLVM IR allows metadata to be attached to instructions " into a grammar depicted in SourceLevelDebugging.html. I