Displaying 20 results from an estimated 200 matches similar to: "Spill hoisting on RAL: looking for some debugging ideas"
2012 Oct 25
2
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
>
> PHIElim and TwoAddress passes leave SSA form.
> May be a missed something in your code but %vreg48 seems to be there
> after PHI elimination. PHIElim tags those kind of registers as being
> PHIJoin regs, updating LiveVariables pass, so the regcoalescer is aware
> of them (some SSA info is still alive but the reg coalescer will
> invalidate that information after
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
2012 Oct 25
0
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
Hi Vincent,
On 24/10/2012 23:26, Vincent Lejeune wrote:
> Hi,
>
> I don't know if my llvm ir code is faulty, or if I spot a bug in the RegisterCoalescing Pass, so I'm posting my issue on the ML. Shader and print-before-all dump are given below.
>
> The interessing part is the vreg6/vreg48 reduction : before RegCoalescing, the machine code is :
>
> // BEFORE LOOP
>
2012 Oct 24
3
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
Hi,
I don't know if my llvm ir code is faulty, or if I spot a bug in the RegisterCoalescing Pass, so I'm posting my issue on the ML. Shader and print-before-all dump are given below.
The interessing part is the vreg6/vreg48 reduction : before RegCoalescing, the machine code is :
// BEFORE LOOP
... Some COPYs....
400B%vreg47<def> = COPY %vreg2<kill>; R600_Reg32:%vreg47,%vreg2
2017 Oct 13
2
Machine Scheduler on Power PC: Latency Limit and Register Pressure
Hi,
I've been looking at the Machine Scheduler on Power PC. I am looking only
at the pre-RA machine scheduler and I am running it in the default
bi-directional mode (so, both top down and bottom up queues are
considered). I've come across an example where the scheduler picks a poor
ordering for the instructions which results in very high register pressure
which results in spills.
2014 Aug 19
2
[LLVMdev] Help with definition of subregisters; spill, rematerialization and implicit uses
Hi Quentin,
On 08/15/14 19:01, Quentin Colombet wrote:
[...]
>> The question is: How should true subregister definitions be
>> expressed so that they do not interfere with each other? See the
>> detailed problem description below.
>
> We do have a limitation in our current liveness tracking for
> sub-register. Therefore, I am not sure that is possible.
>
>
2012 Oct 26
1
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
Vincent,
File a bug report so you can get a fix for it.
Ivan
On 25/10/2012 23:01, Vincent Lejeune wrote:
> Thank for your help. You're right, merging vreg32 and vreg48 is perfectly fine, sorry I missed that.
> I "brute force" debuged by adding MachineFunction dump after each join, I think I found the issue : it's when vreg32 and vreg10 are merged.
> vreg10 only
2012 Oct 25
0
[LLVMdev] RegisterCoalescing Pass seems to ignore part of CFG.
Thank for your help. You're right, merging vreg32 and vreg48 is perfectly fine, sorry I missed that.
I "brute force" debuged by adding MachineFunction dump after each join, I think I found the issue : it's when vreg32 and vreg10 are merged.
vreg10 only appears in BB#3, and the join only occurs in BB#3 apparently even if vreg32 lives in the 4 machine blocks
After joining, there
2008 May 14
0
NFS subdirectory on client is out of sync
Today a user asked me whether a file on one host can be different on
another host. I was busy composing an answer to tell that the /home
space on all clients are mounted using NFS from the file server. Any
host will therefor see the same file. The user pointed me to his file
and I copied this file from the client and compared this with the file
on the file server. To my surprise it turned out
2013 Feb 08
2
[LLVMdev] help with X86 DAG->DAG Instruction Selection
I have an llvm ir, which generates the following machine code using llc
(llvm 3.0 on win32) after # *** IR Dump After X86 DAG->DAG Instruction
Selection ***:
The first three lines and the last two lines alone together are used to
compute "sin" for some double number.
- line 1: move the stack pointer down 8
- line 2: copy the updated stack pointer to a base register
- line 3: copy a
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
2017 Oct 13
3
Machine Scheduler on Power PC: Latency Limit and Register Pressure
> On Oct 13, 2017, at 1:46 PM, Matthias Braun <matze at braunis.de> wrote:
>
> Yes, I've run into the problem myself that the Pending queue isn't even checked with the tryCandidate() logic and so takes priority over all other scheduling decisions.
>
> I personally would be open to changes in this area. To start the brainstorming I could imagine that we move nodes
2014 Aug 15
2
[LLVMdev] Help with definition of subregisters; spill, rematerialization and implicit uses
Hi,
I have a problem regarding sub-register definitions and LiveIntervals on
our target. When a subregister is defined, other parts of the register
are always left untouched - they are neither read or def:ed.
It however seems that Codegen treats subregister definitions as somehow
clobbering the whole register.
The SSA-code looks like this after isel:
(Reg0 and Reg1 are 16bit registers. Reg2,
2016 Jul 30
1
Instruction selection bug for vector store with FixedStack
Hello.
Could you please help me solve the following LLC bug happening at instruction
selection time:
ISEL: Starting pattern match on root node: t172: ch = store<ST64[FixedStack6]>
t0, t6, FrameIndex:i64<6>, undef:i64
Initial Opcode index to 157
Skipped scope entry (due to false predicate) at index 162, continuing at 236
Match failed at
2014 Nov 03
0
RFC: new variable battery.status (was: [PATCH] upscode2: Report ABM Status)
Hi,
Am 13.02.2012 um 18:58 schrieb Arnaud Quette:
> 2012/2/6 thomas schorpp <thomas.schorpp at googlemail.com>:
>> Hi,
>
> Hi Thomas,
>
>> I want the driver report the battery status from ABM charging controllers
>> -patch attached- :
>
> For now, I've tracked your patch here:
>
2020 Apr 07
2
[ARM] Register pressure with -mthumb forces register reload before each call
If I'm understanding what's going on in this test correctly, what's happening is:
* ARMTargetLowering::LowerCall prefers indirect calls when a function is called at least 3 times in minsize
* In thumb 1 (without -fno-omit-frame-pointer) we have effectively only 3 callee-saved registers (r4-r6)
* The function has three arguments, so those three plus the register we need to hold the
2019 Sep 09
2
LiveInterval error with 2 dead defs
Hi,
I’m hitting a machine verifier error in a trivial testcase which I don’t understand. There are 2 dead defs of the same register:
---
name: multiple_connected_compnents_dead
tracksRegLiveness: true
body: |
bb.0:
dead %0:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
dead %0:vgpr_32 = V_MOV_B32_e32 1, implicit $exec
...
The live intervals look OK to me with 1 valno
2013 Feb 08
0
[LLVMdev] help with X86 DAG->DAG Instruction Selection
Hi Peng,
Can you please open a bugzilla and attache the LL file ? Can you please reproduce it on ToT ?
Thanks,
Nadav
On Feb 7, 2013, at 9:08 PM, Peng Cheng <gm4cheng at gmail.com> wrote:
> I have an llvm ir, which generates the following machine code using llc (llvm 3.0 on win32) after # *** IR Dump After X86 DAG->DAG Instruction Selection ***:
>
> The first three lines
2018 Sep 11
2
linear-scan RA
> On Sep 10, 2018, at 5:25 PM, Matthias Braun <mbraun at apple.com> wrote:
>
>
>
>> On Sep 10, 2018, at 5:11 PM, Preston Briggs <preston.briggs at gmail.com <mailto:preston.briggs at gmail.com>> wrote:
>>
>> The phi instruction is irrelevant; just the way I think about things.
>> The question is if the allocator believes that t0 and t2
2012 Feb 13
3
RFC: new variable battery.status (was: [PATCH] upscode2: Report ABM Status)
2012/2/6 thomas schorpp <thomas.schorpp at googlemail.com>:
> Hi,
Hi Thomas,
> I want the driver report the battery status from ABM charging controllers
> -patch attached- :
thanks for your patch.
since it introduces a new variable (battery.status.abm), I prefer to
step back and think to something useful more widely.
thus, I'm thinking about a generic