| Age | Commit message (Collapse) | Author |
|---|
|
The existing way of managing clustered nodes was done through adding
weak edges between the neighbouring cluster nodes, which is a sort of
ordered queue. And this will be later recorded as `NextClusterPred` or
`NextClusterSucc` in `ScheduleDAGMI`.
But actually the instruction may be picked not in the exact order of the
queue. For example, we have a queue of cluster nodes A B C. But during
scheduling, node B might be picked first, then it will be very likely
that we only cluster B and C for Top-Down scheduling (leaving A alone).
Another issue is:
```
if (!ReorderWhileClustering && SUa->NodeNum > SUb->NodeNum)
std::swap(SUa, SUb);
if (!DAG->addEdge(SUb, SDep(SUa, SDep::Cluster)))
```
may break the cluster queue.
For example, we want to cluster nodes (order as in `MemOpRecords`): 1 3
2. 1(SUa) will be pred of 3(SUb) normally. But when it comes to (3, 2),
As 3(SUa) > 2(SUb), we would reorder the two nodes, which makes 2 be
pred of 3. This makes both 1 and 2 become preds of 3, but there is no
edge between 1 and 2. Thus we get a broken cluster chain.
To fix both issues, we introduce an unordered set in the change. This
could help improve clustering in some hard case.
One key reason the change causes so many test check changes is: As the
cluster candidates are not ordered now, the candidates might be picked
in different order from before.
The most affected targets are: AMDGPU, AArch64, RISCV.
For RISCV, it seems to me most are just minor instruction reorder, don't
see obvious regression.
For AArch64, there were some combining of ldr into ldp being affected.
With two cases being regressed and two being improved. This has more
deeper reason that machine scheduler cannot cluster them well both
before and after the change, and the load combine algorithm later is
also not smart enough.
For AMDGPU, some cases have more v_dual instructions used while some are
regressed. It seems less critical. Seems like test `v_vselect_v32bf16`
gets more buffer_load being claused.
|
|
(#102915)
A long time ago (back in 2009) there was a commit 52d4d8244b7c39c5cd
that changed the scheduler to not dirty height/depth when adding or
removing SUnit predecessors when the latency on the edge was zero. That
commit message is claiming that the depth or height isn't affected when
the latency is zero.
As a matter of fact, the depth/height can change even with a zero
latency on the edge. If for example adding a new SUnit A, with zero
latency, but as a predecessor to a SUnit B, then both height of A and
depth of B should be marked as dirty. If for example B has a greater
height than A, then the height of A needs to be adjusted even if the
latency is zero.
I think this has been wrong for many years. Downstream we have had
commit 52d4d8244b7c39c5cd reverted since back in 2016. There is no
motivating lit test for 52d4d8244b7c39c5cd (only an incomplete C level
reproducer in https://github.com/llvm/llvm-project/issues/3613).
After commit 13d04fa560e156797c21f1 there finally appeared an upstream
lit test that shows that we get better code if marking height/depth as
dirty (llvm/test/CodeGen/AArch64/abds.ll).
|
|
path consistently (#93001)
Patch co-authored by AtariDreams (gfunni234@gmail.com).
Fixes #38037.
[AMDGPU] Update test results to fix build (#92982)
|
|
critical path consistently (#92368)" (#92992)
This reverts commit f0d1ae83b046ff0eeffc72a103f7c8b01ad33787.
This commit breaks tests: http://45.33.8.238/linux/138572/step_11.txt
|
|
path consistently (#92368)
Fix https://github.com/llvm/llvm-project/issues/38037
|
|
Identified with misc-include-cleaner.
|
|
In SUnit::removePred, N->WeakSuccsLeft is reduced but WeakSuccsLeft is checked.
Reviewed By: kerbowa
Differential Revision: https://reviews.llvm.org/D151311
|
|
This change fixes static code analysis errors
Reviewed By: skan
Differential Revision: https://reviews.llvm.org/D149506
|
|
This reverts commit 22b23a5213b57ce1834f5b50fbbf8a50297efc8a.
This commit caused the following two build bots to start failing:
https://lab.llvm.org/buildbot/#/builders/216/builds/20322
https://lab.llvm.org/buildbot/#/builders/123/builds/18511
|
|
This change fixes static code analysis errors
Reviewed By: skan
Differential Revision: https://reviews.llvm.org/D148912
|
|
In SUnit::removePred edges are removed from the Preds and Succs lists before
updating the bookkeeping. This could result in incorrect values for
NumPreds/SuccsLeft and cause WeakPreds/SuccsLeft to underflow, since the
incorrect SDep will be used to update these values.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D142325
|
|
|
|
|
|
This reverts commit 0345d88de654259ae90494bf9b015416e2cccacb.
Google internal backend uses EntrySU, we are looking into removing
dependency on it.
Differential Revision: https://reviews.llvm.org/D88018
|
|
EntrySU doesn't seem to be used at all when building the ScheduleDAG.
Differential Revision: https://reviews.llvm.org/D87867
|
|
In some cases ScheduleDAGRRList has to add new nodes to resolve problems
with interfering physical registers. When new nodes are added, it
completely re-computes the topological order, which can take a long
time, but is unnecessary. We only add nodes one by one, and initially
they do not have any predecessors. So we can just insert them at the end
of the vector. Later we add predecessors, but the helper function
properly updates the topological order much more efficiently. With this
change, the compile time for the program below drops from 300s to 30s on
my machine.
define i11129 @test1() {
%L1 = load i11129, i11129* undef
%B30 = ashr i11129 %L1, %L1
store i11129 %B30, i11129* undef
ret i11129 %L1
}
This should be generally beneficial, as we can skip a large amount of
work. Theoretically there are some scenarios where we might not safe
much, e.g. when we add a dependency between the first and last node.
Then we would have to shift all nodes. But we still do not have to spend
the time re-computing the initial order.
Reviewers: MatzeB, atrick, efriedma, niravd, paquette
Reviewed By: paquette
Differential Revision: https://reviews.llvm.org/D59722
|
|
Currently there is a single point in ScheduleDAGRRList, where we
actually query the topological order (besides init code). Currently we
are recomputing the order after adding a node (which does not have
predecessors) and then we add predecessors edge-by-edge.
We can avoid adding edges one-by-one after we added a new node. In that case, we can
just rebuild the order from scratch after adding the edges to the DAG
and avoid all the updates to the ordering.
Also, we can delay updating the DAG until we query the DAG, if we keep a
list of added edges. Depending on the number of updates, we can either
apply them when needed or recompute the order from scratch.
This brings down the geomean compile time for of CTMark with -O1 down 0.3% on X86,
with no regressions.
Reviewers: MatzeB, atrick, efriedma, niravd, paquette
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D60125
llvm-svn: 358583
|
|
This is helpful to measure the impact of D60125 on maintaining
topological orders.
Reviewers: MatzeB, atrick, efriedma, niravd
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D60187
llvm-svn: 358058
|
|
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
|
|
- Instead of having both `SUnit::dump(ScheduleDAG*)` and
`ScheduleDAG::dumpNode(ScheduleDAG*)`, just keep the latter around.
- Add `ScheduleDAG::dump()` and avoid code duplication in several
places. Implement it for different ScheduleDAG variants.
- Add `ScheduleDAG::dumpNodeName()` in favor of the `SUnit::print()`
functions. They were only ever used for debug dumping and putting the
function into ScheduleDAG is consistent with the `dumpNode()` change.
llvm-svn: 342520
|
|
See r331124 for how I made a list of files missing the include.
I then ran this Python script:
for f in open('filelist.txt'):
f = f.strip()
fl = open(f).readlines()
found = False
for i in xrange(len(fl)):
p = '#include "llvm/'
if not fl[i].startswith(p):
continue
if fl[i][len(p):] > 'Config':
fl.insert(i, '#include "llvm/Config/llvm-config.h"\n')
found = True
break
if not found:
print 'not found', f
else:
open(f, 'w').write(''.join(fl))
and then looked through everything with `svn diff | diffstat -l | xargs -n 1000 gvim -p`
and tried to fix include ordering and whatnot.
No intended behavior change.
llvm-svn: 331184
|
|
LLVM Coding Standards:
Function names should be verb phrases (as they represent actions), and
command-like function should be imperative. The name should be camel
case, and start with a lower case letter (e.g. openFile() or isFoo()).
Differential Revision: https://reviews.llvm.org/D40416
llvm-svn: 319168
|
|
All these headers already depend on CodeGen headers so moving them into
CodeGen fixes the layering (since CodeGen depends on Target, not the
other way around).
llvm-svn: 318490
|
|
This header includes CodeGen headers, and is not, itself, included by
any Target headers, so move it into CodeGen to match the layering of its
implementation.
llvm-svn: 317647
|
|
causing link errors for several people.
Error LNK2019 unresolved external symbol "public: void __cdecl `anonymous namespace'::MatchableInfo::dump(void)const " (?dump@MatchableInfo@?A0xf4f1c304@@QEBAXXZ) referenced in function "public: void __cdecl `anonymous namespace'::AsmMatcherEmitter::run(class llvm::raw_ostream &)" (?run@AsmMatcherEmitter@?A0xf4f1c304@@QEAAXAEAVraw_ostream@llvm@@@Z) llvm-tblgen D:\llvm\2017\utils\TableGen\AsmMatcherEmitter.obj 1
llvm-svn: 315854
|
|
Summary:
Add LLVM_FORCE_ENABLE_DUMP cmake option, and use it along with
LLVM_ENABLE_ASSERTIONS to set LLVM_ENABLE_DUMP.
Remove NDEBUG and only use LLVM_ENABLE_DUMP to enable dump methods.
Move definition of LLVM_ENABLE_DUMP from config.h to llvm-config.h so
it'll be picked up by public headers.
Differential Revision: https://reviews.llvm.org/D38406
llvm-svn: 315590
|
|
Add SDep printer to make debugging sessions more productive.
Differential revision: https://reviews.llvm.org/D35144
llvm-svn: 307799
|
|
I did this a long time ago with a janky python script, but now
clang-format has built-in support for this. I fed clang-format every
line with a #include and let it re-sort things according to the precise
LLVM rules for include ordering baked into clang-format these days.
I've reverted a number of files where the results of sorting includes
isn't healthy. Either places where we have legacy code relying on
particular include ordering (where possible, I'll fix these separately)
or where we have particular formatting around #include lines that
I didn't want to disturb in this patch.
This patch is *entirely* mechanical. If you get merge conflicts or
anything, just ignore the changes in this patch and run clang-format
over your #include lines in the files.
Sorry for any noise here, but it is important to keep these things
stable. I was seeing an increasing number of patches with irrelevant
re-ordering of #include lines because clang-format was used. This patch
at least isolates that churn, makes it easy to skip when resolving
conflicts, and gets us to a clean baseline (again).
llvm-svn: 304787
|
|
Patch by Axel Davy (axel.davy@normalesup.org)
Differential revision: https://reviews.llvm.org/D30626
llvm-svn: 298896
|
|
other minor fixes (NFC).
llvm-svn: 295893
|
|
- Fix doxygen comments (do not repeat documented name, remove definition
comment if there is already one at the declaration, add \p, ...)
- Add some const modifiers
- Use range based for
llvm-svn: 295688
|
|
We had various variants of defining dump() functions in LLVM. Normalize
them (this should just consistently implement the things discussed in
http://lists.llvm.org/pipermail/cfe-dev/2014-January/034323.html
For reference:
- Public headers should just declare the dump() method but not use
LLVM_DUMP_METHOD or #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
- The definition of a dump method should look like this:
#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
LLVM_DUMP_METHOD void MyClass::dump() {
// print stuff to dbgs()...
}
#endif
llvm-svn: 293359
|
|
This method seems to have had a troubled life. This patch proposes that it
replaces the recently added helper function dumpSUIdentifier. This way, the
method can be used in other files using the SUnit class.
Differential revision: https://reviews.llvm.org/D28488
llvm-svn: 291520
|
|
support to skipping a node."
Revert accidentally committed change.
This reverts commit r286655.
llvm-svn: 286656
|
|
The DAG mutators in the scheduler cannot really remove DAG nodes as
additional anlysis information such as ScheduleDAGToplogicalSort are
already computed at this point and rely on a fixed number of DAG nodes.
Alleviate the missing removal with a new flag: Setting the new skip
flag on a node ignores it during scheduling.
llvm-svn: 286655
|
|
llvm-svn: 286652
|
|
It is less confusing to have the same names in the debug print as the
enum members.
llvm-svn: 282273
|
|
If the result of the find is only used to compare against end(), just
use is_contained instead.
No functionality change is intended.
llvm-svn: 278469
|
|
llvm-svn: 252340
|
|
off of the MachineFunction or with the version that takes a
Function reference as an argument.
llvm-svn: 227185
|
|
information and update all callers. No functional change.
llvm-svn: 214781
|
|
define below all header includes in the lib/CodeGen/... tree. While the
current modules implementation doesn't check for this kind of ODR
violation yet, it is likely to grow support for it in the future. It
also removes one layer of macro pollution across all the included
headers.
Other sub-trees will follow.
llvm-svn: 206837
|
|
instead of comparing to nullptr.
llvm-svn: 206142
|
|
Remove the old functions.
llvm-svn: 202636
|
|
Sweep the codebase for common typos. Includes some changes to visible function
names that were misspelt.
llvm-svn: 200018
|
|
avoid specifying the vector size unnecessarily.
llvm-svn: 185512
|
|
llvm-svn: 176316
|
|
We avoided computing DAG height/depth during Node printing because it
shouldn't depend on an otherwise valid DAG. But this has become far
too annoying for the common case of a valid DAG where we want to see
valid values. If doing the computation on-the-fly turns out to be a
problem in practice, then I'll add a mode to the diagnostics to only
force it when we're likely to have a valid DAG, otherwise explicitly
print INVALID instead of bogus numbers. For now, just go for it all
the time.
llvm-svn: 176314
|
|
No functionality change.
llvm-svn: 175366
|
|
Allow schedulers to order DAG edges by critical path. This makes
DFS-based heuristics more stable and effective.
llvm-svn: 173317
|
