Displaying 9 results from an estimated 9 matches for "vreg34".
Did you mean:
vreg3
2017 Apr 24
3
Debugging UNREACHABLE "Couldn't join subrange" in RegisterCoalescer (out-of-tree backend)
...;8" "target-cpu"="generic" "unsafe-fp-math"="false" "use-soft-float"="false" }
Here's the tail-end of the log, with debugging turned on:
$llc bugpoint.reduced.simplified.bc -debug
...
208B %vreg13:sub_64_1<def> = COPY %vreg34:sub_64_1; VecRegs:%vreg13,%vreg34
Considering merging to VecRegs with %vreg34 in %vreg13
RHS = %vreg34 [160r,240r:0)[240r,384B:1)[400B,480r:1)[480r,496r:2)[496r,672r:3) 0 at 160r 1 at 240r 2 at 480r 3 at 496r L00000020 [240r,384B:1)[400B,672r:1) 0 at x 1 at 240r L00000010...
2012 Oct 25
3
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
Hi Vincent,
On 25/10/2012 18:14, Vincent Lejeune wrote:
> When examining the debug output of regalloc, it seems that joining 32bits reg also joins 128 parent reg.
>
> If I look at the :
> %vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34 R600_Reg128:%vreg6
>
> instructions ; it gets joined to :
> 928B%vreg34<def> = COPY %vreg48:sel_y;
>
> when vreg6 and vreg48 are joined. It's right.
>
> But joining the following copy
>
> 912B%vreg32:sel_x<de...
2012 Oct 25
0
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
When examining the debug output of regalloc, it seems that joining 32bits reg also joins 128 parent reg.
If I look at the :
%vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34 R600_Reg128:%vreg6
instructions ; it gets joined to :
928B%vreg34<def> = COPY %vreg48:sel_y;
when vreg6 and vreg48 are joined. It's right.
But joining the following copy
912B%vreg32:sel_x<def,read-undef> = COPY %vreg48:sel_x;...
2012 Oct 24
3
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
...Reg32:%vreg30
560BJUMP <BB#3>, pred:%PREDICATE_BIT
576BJUMP <BB#2>, pred:%noreg
// LOOP BODY
896B%vreg31<def> = COPY %vreg6:sel_x; R600_Reg32:%vreg31 R600_Reg128:%vreg6
912B%vreg32:sel_x<def,read-undef> = COPY %vreg31<kill>; R600_Reg128:%vreg32 R600_Reg32:%vreg31
928B%vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34 R600_Reg128:%vreg6
944B%vreg35<def> = COPY %vreg32<kill>; R600_Reg128:%vreg35,%vreg32
960B%vreg35:sel_y<def> = COPY %vreg34<kill>; R600_Reg128:%vreg35 R600_Reg32:%vreg34
976B%vreg36<def> = COPY %vreg35<kill>; R60...
2012 Oct 25
0
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
...#3>, pred:%PREDICATE_BIT
> 576BJUMP <BB#2>, pred:%noreg
>
> // LOOP BODY
> 896B%vreg31<def> = COPY %vreg6:sel_x; R600_Reg32:%vreg31 R600_Reg128:%vreg6
> 912B%vreg32:sel_x<def,read-undef> = COPY %vreg31<kill>; R600_Reg128:%vreg32 R600_Reg32:%vreg31
> 928B%vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34 R600_Reg128:%vreg6
> 944B%vreg35<def> = COPY %vreg32<kill>; R600_Reg128:%vreg35,%vreg32
> 960B%vreg35:sel_y<def> = COPY %vreg34<kill>; R600_Reg128:%vreg35 R600_Reg32:%vreg34
> 976B%vreg36<def> = COPY %vreg35&l...
2012 Oct 25
2
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
...t;imp-use,kill>
BB#3:# derived from
896B%vreg31<def> = COPY %vreg6:sel_x; R600_Reg32:%vreg31 R600_Reg128:%vreg6
register: %vreg31 +[896r,912r:0)
912B%vreg32:sel_x<def,read-undef> = COPY %vreg31<kill>; R600_Reg128:%vreg32 R600_Reg32:%vreg31
register: %vreg32 +[912r,944r:0)
928B%vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34 R600_Reg128:%vreg6
register: %vreg34 +[928r,960r:0)
944B%vreg35<def> = COPY %vreg32<kill>; R600_Reg128:%vreg35,%vreg32
register: %vreg35 +[944r,976r:0)
960B%vreg35:sel_y<def> = COPY %vreg34<kill>; R600_Reg128:%vreg35 R600_Re...
2012 Oct 25
0
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
...gisterCoalescing Pass seems to ignore part of CFG.
>
> Hi Vincent,
>
> On 25/10/2012 18:14, Vincent Lejeune wrote:
>> When examining the debug output of regalloc, it seems that joining 32bits
> reg also joins 128 parent reg.
>>
>> If I look at the :
>> %vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34
> R600_Reg128:%vreg6
>>
>> instructions ; it gets joined to :
>> 928B%vreg34<def> = COPY %vreg48:sel_y;
>>
>> when vreg6 and vreg48 are joined. It's right.
>>
>> But joining the followi...
2016 Jun 08
2
Instruction Itineraries: question about operand latencies
I overrode getInstrLatency and did some printing to see what is available
there. It looks like the registers are still virtual at that point when
getInstrLatency is called - is that correct? (we needed to make some
decisions based on actual registers that have been assigned since some
registers are reserved as address space pointers and we could vary the
latency based on which address space
2012 Oct 26
1
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
...re part of CFG.
>>
>> Hi Vincent,
>>
>> On 25/10/2012 18:14, Vincent Lejeune wrote:
>>> When examining the debug output of regalloc, it seems that joining 32bits
>> reg also joins 128 parent reg.
>>>
>>> If I look at the :
>>> %vreg34<def> = COPY %vreg6:sel_y; R600_Reg32:%vreg34
>> R600_Reg128:%vreg6
>>>
>>> instructions ; it gets joined to :
>>> 928B%vreg34<def> = COPY %vreg48:sel_y;
>>>
>>> when vreg6 and vreg48 are joined. It's right.
>>>
>&g...