[LLVMdev] Possible issue with EXPANDING POST-RA PSEUDO INSTRS (original) (raw)

Sergei Larin slarin at codeaurora.org
Mon Jul 9 07:56:04 PDT 2012

Jakob,

It looks something like this:

%vreg15<def> = COMBINE_rr %vreg11, %vreg14<kill>; DoubleRegs:%vreg15

IntRegs:%vreg11,%vreg14 %vreg20 = CONST32GP_set ga:@inst3; IntRegs:%vreg20 STriw_GP ga:@inst2+4, 0, %vreg3; mem:ST4getelementptr inbounds (%struct.myStruct* @inst2, i32 0, i32 1) IntRegs:%vreg3 %vreg28 = CONST32GP_set ga:@inst2; IntRegs:%vreg28 %vreg17:subreg_hireg = COPY %vreg15:subreg_loreg; DoubleRegs:%vreg17,%vreg15

COMBINE_rr == combine two (32) regs into one (64) pair.

You are seeing whole live range of vreg15.

At this point: %vreg17:subreg_hireg = COPY %vreg15:subreg_loreg; DoubleRegs:%vreg17,%vreg15

We know that we are killing %vreg15:subreg_loreg, but after Stack Slot Coloring I got this:

%R15 = COPY %R4, %D2<imp-use,kill>, %D7<imp-use,kill>, %D7

The actual problem happens much later, in post RA scheduler - we use critical antidependency breaker, and that is messed up by the missing kill.

Thanks for your help.

Sergei

-- Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum.

-----Original Message----- From: Jakob Stoklund Olesen [mailto:stoklund at 2pi.dk] Sent: Sunday, July 08, 2012 8:27 PM To: Sergei Larin Cc: 'List' Subject: Re: [LLVMdev] Possible issue with EXPANDING POST-RA PSEUDO INSTRS

On Jul 8, 2012, at 3:42 PM, Sergei Larin <slarin at codeaurora.org> wrote: > ********** EXPANDING POST-RA PSEUDO INSTRS ********** > ********** Function: main > real copy: %R15 = COPY %R4, %D2<imp-use,kill>, %D7<imp-_ _use,kill>, > %D7 > replaced by: %R15 = TFR %R4, %D7 > > The R4 is a subreg of D2 double register (basically R0:R1==D0; > R4:R5==D2 etc.). After this copy D2 is dead, and is marked as such. > Register allocator did not mark R4 as , and when ExpandPostRA > copies operands, it loses the live range marker. > > My questions - is this a known issue? ...and should R4 has been marked > as in the original instruction? It's not a bug as such. It is OK for kill flags to be missing. It's weird that R4 isn't killed out of RA, though. What does it look like before rewriting? (-debug-only=regalloc). /jakob

More information about the llvm-dev mailing list