PR #66334 tried to renumber slot indexes before register allocation, but
the numbering was still affected by list entries for instructions which
had been erased. Fix this to make the register allocator's live range
length heuristics even less dependent on the history of how instructions
have been added to and removed from SlotIndexes's maps.
SIInsertWaitcnts inserts waitcnt instructions to resolve data
dependencies. The GFX10+ vscnt (VMEM store count) counter is never used
in this way. It is only used to resolve memory dependencies, and that is
handled by SIMemoryLegalizer. Hence there is no need to conservatively
wait for vscnt to be 0 on function entry and before returns.
Differential Revision: https://reviews.llvm.org/D153537
There is a failure with this pass in the case when target register class for a subregister isn't known from instruction description (for ex. COPY).
Currently in this situation the RC is obtained using TargetRegisterInfo::getSubRegisterClass but in general it's not working.
In order to fix this two things should be done:
1. Stop processing a subregister if the target register class is unknown (conservative approach)
2. Improve deduction of subregister' target register class (i.e by processing COPY chain)
I was going to implement point 1 but my tests use implicit operands for S_NOP and they don't have associated target register class and all tests fail.
Therefore I decided to turn off the pass now, implement point 1 and fix my tests.
Reviewed By: arsenm, #amdgpu
Differential Revision: https://reviews.llvm.org/D152291
The main purpose of this is to simplify register pressure tracking as after the pass there is no need
to track subreg liveness anymore.
On the other hand this pass creates more possibilites for the subreg unaware code, as many of the subregs
becomes ordinary registers.
Intersting sideeffect: spill-vgpr.ll has lost a lot of spills.
Reviewed By: #amdgpu, arsenm
Differential Revision: https://reviews.llvm.org/D139732
Occupancy is expressed as waves per SIMD. This means that we need to
take into account the number of SIMDs per "CU" or, to be more precise,
the number of SIMDs over which a workgroup may be distributed.
getOccupancyWithLocalMemSize was wrong because it didn't take SIMDs
into account at all.
At the same time, we need to take into account that WGP mode offers
access to a larger total amount of LDS, since this can affect how
non-power-of-two LDS allocations are rounded. To make this work
consistently, we distinguish between (available) local memory size and
addressable local memory size (which is always limited by 64kB on
gfx10+, even with WGP mode).
This change results in a massive amount of test churn. A lot of it is
caused by the fact that the default work group size is 1024, which means
that (due to rounding effects) the default occupancy on older hardware
is 8 instead of 10, which affects scheduling via register pressure
estimates. I've adjusted most tests by just running the UTC tools, but
in some cases I manually changed the work group size to 32 or 64 to make
sure that work group size chunkiness has no effect.
Differential Revision: https://reviews.llvm.org/D139468
The full complement of physical VGPRs for GFX11 is 50% more than GFX10.
Some subtargets have this, others stay the same as GFX10. This affects
occupancy calculations.
Differential Revision: https://reviews.llvm.org/D134522
Only do this for 16 and 32 register tuples, although we might want to
extend to 8 tuples.
It's incredibly expensive to spill these, and doing so majorly
interferes with the ability to allocate anything else in the function.
The lit tests show mostly sizeable improvements with a handful of tiny
regressions with large vectors.
Add GFX11 test coverage to a bunch of tests where it was easy to do so,
mostly because the checks are autogenerated and/or GFX11 can share the
same checks as GFX10.
Differential Revision: https://reviews.llvm.org/D129295
Using a BufferSize of one for memory ProcResources will result in better
ILP since it more accurately models the dependencies between memory ops
and their consumers on an in-order processor. After this change, the
scheduler will treat the data edges from loads as blocking so that
stalls are guaranteed when waiting for data to be retreaved from memory.
Since we don't actually track waitcnt here, this should do a better job
at modeling their behavior.
Practically, this means that the scheduler will trigger the 'STALL'
heuristic more often.
This type of change needs to be evaluated experimentally. Preliminary
results are positive.
Fixes: SWDEV-282962
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D114777
This has a couple of benefits:
1. It can sometimes fix clusters that got broken apart when the register
allocator inserted a copy.
2. Post-RA scheduling does not have to worry about increasing register
pressure, which in some cases gives it more freedom to reorder
instructions.
Testing on a collection of 10,000 graphics shaders compiled for gfx1010
showed:
- The average length of each run of one or more load instructions
increased by about 1%.
- The number of runs of two or more load instructions increased by
about 4%.
Differential Revision: https://reviews.llvm.org/D111646
This has a couple of benefits:
1. It can sometimes fix clusters that got broken apart when the register
allocator inserted a copy.
2. Post-RA scheduling does not have to worry about increasing register
pressure, which in some cases gives it more freedom to reorder
instructions.
Testing on a collection of 10,000 graphics shaders compiled for gfx1010
showed:
- The average length of each run of one or more load instructions
increased by about 1%.
- The number of runs of two or more load instructions increased by
about 4%.
Use GCNHazardRecognizer in postra sched.
Updated tests for the new schedules.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D109536
Change-Id: Ia86ba2ae168f12fb34b4d8efdab491f84d936cde
the compilation time and there is no case for which we see any improvement in
performance. This patch removes this pass and its associated test cases from
the tree.
Differential Revision: https://reviews.llvm.org/D101313
Change-Id: I0599169a7609c19a887f8d847a71e664030cc141
This restores previous behaviour and is a step toward removing
unbundling entirely.
Reviewed By: foad, rampitec
Differential Revision: https://reviews.llvm.org/D99061
Previously we would use a bundle to hint the register allocator to not
overwrite the pointers in a sequence of loads to avoid breaking soft
clauses. This bundling was based on a fuzzy register pressure
heuristic, so we could not guarantee using more registers than are
really available. This would result in register allocator failing on
unsatisfiable bundles. Use a kill to artificially extend the live
ranges, so we can always succeed at register allocation even if it
means extra spills in the worst case.
This seems to capture most of the benefit of the bundle while avoiding
most of the risk presented by the bundle. However the lit tests do
show a handful of regressions. In some cases with sequences of
volatile loads, unused load components end up getting reallocated to
the next load which forces a wait between. There are also a few small
scheduling regressions where a hazard used to be avoided, and one
spill torture test which for some reason nearly doubles the stack
usage. There is also a bit of noise from leftover kills (it may make
sense for post-RA pseudos to strip all of these out).
Support for XNACK and SRAMECC is not static on some GPUs. We must be able
to differentiate between different scenarios for these dynamic subtarget
features.
The possible settings are:
- Unsupported: The GPU has no support for XNACK/SRAMECC.
- Any: Preference is unspecified. Use conservative settings that can run anywhere.
- Off: Request support for XNACK/SRAMECC Off
- On: Request support for XNACK/SRAMECC On
GCNSubtarget will track the four options based on the following criteria. If
the subtarget does not support XNACK/SRAMECC we say the setting is
"Unsupported". If no subtarget features for XNACK/SRAMECC are requested we
must support "Any" mode. If the subtarget features XNACK/SRAMECC exist in the
feature string when initializing the subtarget, the settings are "On/Off".
The defaults are updated to be conservatively correct, meaning if no setting
for XNACK or SRAMECC is explicitly requested, defaults will be used which
generate code that can be run anywhere. This corresponds to the "Any" setting.
Differential Revision: https://reviews.llvm.org/D85882
It is possible for copies or spills to be inserted in the middle of indirect
addressing sequences which use VGPR indexing. Spills to accvgprs could be
effected by the indexing mode.
Add new pseudo instructions that are expanded after register allocation to avoid
the problematic spill or copy placement.
Differential Revision: https://reviews.llvm.org/D91048
Passes that are run after the post-RA scheduler may insert instructions like
waitcnt which eliminate the need for certain noops. After this patch the
scheduler is still aware of possible latency from hazards but noops will
not be inserted until the dedicated hazard recognizer pass is run.
Depends on D89753.
Reviewed By: foad
Differential Revision: https://reviews.llvm.org/D89754
Change waitcnt insertion to check the memory operand tokens to see if
flat memory operations access VMEM in the same way it does to check if
accessing LDS. This avoids adding waitcnt for counters for address
spaces that are not accessed.
In addition, only generate the pessimistic waitcnt 0 if a flat memory
operation appears to access both VMEM and LDS.
This benefits flat memory operations that explicitly specify the
address space as GLOBAL or LOCAL.
Differential Revision: https://reviews.llvm.org/D89618
This reverts commit ca907bfb57d8ad3ec3bcc2cff2abab7b1b933af6.
According to michel.daenzer,
> This completely broke the Mesa radeonsi driver on Navi 14. Xorg +
> xterm come up with major corruption & psychedelic colours.
When memory operations are outstanding on function calls, either the
caller or the callee can insert a waitcnt to ensure that all reads are
finished.
Calls need some time to be executed, so if the callee inserts the
waitcnt, filling the instruction buffer and waiting for memory will be
interleaved, hiding some latency. This comes at the cost of having a
waitcnt inside functions that may not be needed as no memory operations
are outstanding.
For function calls, this is already implemented. The same principal
applies to returns: If the caller inserts a waitcnt after the call, the
callee does not have to wait and the return and memory operation can be
run in parallel.
This commit implements waiting in the caller after returning from a
function call.
Differential Revision: https://reviews.llvm.org/D87674
For AMDGPU, vectors with elements < 32 bits should be indexed in
32-bit elements and the desired bits extracted from there. For
elements > 64-bits, these should be reduce to 64/32 elements to enable
the normal dynamic indexing paths.
In the dynamic index cases, this produces shorter code most of the
time. This does immediately regress the constant index cases, but this
should be fixed once we have the most basic of shift combines.
The element size > 64 case is pretty much ported from the exisiting
DAG implementation for extract element promote. The increasing element
size case is new.
