search for: i226

Displaying 6 results from an estimated 6 matches for "i226".

Did you mean: 226
2012 Jul 02
4
[LLVMdev] [NVPTX] Backend failure in LegalizeDAG due to unimplemented expand in target lowering
...ely do not want to declare this type as illegal. The real issue is lack of complete support for this type. The PTX language places restrictions on what can be done with .pred registers, and it looks like the failure is here: kernelgen_hostcall.exit228: ; preds = %while.cond.i226 store i1 false, i1 addrspace(1)* undef, align 8 Ignoring for a second that you're storing to an undefined address (???), the back-end does not yet handle up-casting an i1 to an appropriate type for storage. The memory space is not bit-addressable, so a direct store of an i1 does not make se...
2012 Jul 03
0
[LLVMdev] [NVPTX] Backend failure in LegalizeDAG due to unimplemented expand in target lowering
...ely do not want to declare this type as illegal. The real issue is lack of complete support for this type. The PTX language places restrictions on what can be done with .pred registers, and it looks like the failure is here: kernelgen_hostcall.exit228: ; preds = %while.cond.i226 store i1 false, i1 addrspace(1)* undef, align 8 Ignoring for a second that you're storing to an undefined address (???), the back-end does not yet handle up-casting an i1 to an appropriate type for storage. The memory space is not bit-addressable, so a direct store of an i1 does not make se...
2012 Jul 08
1
[LLVMdev] [NVPTX] Backend failure in LegalizeDAG due to unimplemented expand in target lowering
...pe as illegal. The real > issue is lack of complete support for this type. The PTX language places > restrictions on what can be done with .pred registers, and it looks like the > failure is here: > > > > kernelgen_hostcall.exit228: ; preds = %while.cond.i226 > > store i1 false, i1 addrspace(1)* undef, align 8 > > > > Ignoring for a second that you're storing to an undefined address (???), the > back-end does not yet handle up-casting an i1 to an appropriate type for > storage. The memory space is not bit-addressable, so a...
2012 Jul 01
0
[LLVMdev] [NVPTX] Backend failure in LegalizeDAG due to unimplemented expand in target lowering
Hi Duncan, Sorry I don't understand your point, could you please explain a little bit more? Why i1 should be declared illegal? Operations on byte-wide types like char or bool are pretty legal, according to PTX spec: "Registers may be typed (signed integer, unsigned integer, floating point, predicate) or untyped. Register size is restricted; aside from predicate registers which are
2012 Jun 30
2
[LLVMdev] [NVPTX] Backend failure in LegalizeDAG due to unimplemented expand in target lowering
Hi Dmitry, >> did you declare i1 to be an illegal type? > > No. How? I think it will be considered illegal if you don't add it to any register class. Ciao, Duncan. > > 2012/6/30 Duncan Sands <baldrick at free.fr>: >> Hi Dmitry, >>> So instead of setOperationAction(ISD::STORE, MVT::i1, Expand); one >>> should probably do
2013 Feb 14
1
[LLVMdev] LiveIntervals analysis problem
...if.end.i243.i ], [ %scevgep.i.i, %if.end.i218.i ] %52 = load i16* %x.addr.023.i223.i, align 2, !tbaa !5 %tobool.i224.i = icmp slt i16 %52, 0 br i1 %tobool.i224.i, label %if.then.i229.i, label %if.end.i243.i if.then.i229.i: ; preds = %for.body.i225.i %conv1.i226.i = zext i16 %bits.024.i222.i to i32 %or.i227.i = or i32 %conv1.i226.i, 1 %conv2.i228.i = trunc i32 %or.i227.i to i16 br label %if.end.i243.i if.end.i243.i: ; preds = %if.then.i229.i, %for.body.i225.i %bits.1.i230.i = phi i16 [ %conv2.i228.i, %if.then.i22...