When using `BranchProbabilityPrinterPass`, if a BB has no name, we get pretty unusable information like `edge -> has probability...` (i.e. we have no idea what the vertices of that edge are).
This patch uses `printAsOperand`, which uses the same naming scheme as `Function::dump`, so for example during debugging sessions, the IR obtained from a function and the names used by `BranchProbabilityPrinterPass` will match.
A shortcoming is that `printAsOperand` will result in the numbering algorithm re-running for every edge and every vertex (when `BranchProbabilityPrinterPass` is run on a function). If, for the given scenario, this is a problem, we can revisit this subsequently.
Another nuance is that the entry basic block will be numbered, which may be slightly confusing when it's anonymous, but it's easily identifiable - the first edge would have it as source (and the number should be easily recognizable)
- Change `BranchProbabilityPrinterPass` output to match expectations of `update_analyze_test_checks.py`.
- Add `Branch Probability Analysis` to list of supported analyses.
- Process `llvm/test/Analysis/BranchProbabilityInfo/basic.ll` with `update_analyze_test_checks.py` as proof of concept. Leaving the other tests unchanged to reduce the amount of churn.
Recommits the changes from https://reviews.llvm.org/D148216.
Explicitly named globals are now matched literally, instead of emitting
a capture group for the name. This resolves#70047.
Metadata and annotations, on the other hand, are captured and matched
against by default, since their identifiers are not stable.
The reasons for revert (#63746) have been fixed:
The first issue, that of duplicated checkers, has already been resolved
in #70050.
This PR resolves the second issue listed in #63746, regarding the order
of named and unnamed globals. This is fixed by recording the index of
substrings containing global values, and sorting the checks according to
that index before emitting them. This results in global value checks
being emitted in the order they were seen instead of being grouped
separately.
SCEV expressions may contain multiple {{ or }} in the debug output,
which needs escaping.
See
llvm/test/Analysis/LoopAccessAnalysis/loops-with-indirect-reads-and-writes.ll
for a test that needs escaping.
Previously the kill flags of the source register were unconditionally
cleared when a `str` pair was merged, which results in suboptimal
register allocation and inhibits some renaming opportunities which may
allow further merging `str`.
update_analyze_test_checks.py is an invaluable tool in updating tests.
Unfortunately, it only supports output from the CostModel,
ScalarEvolution, and LoopVectorize analyses. Many LoopAccessAnalysis
tests use hand-crafted CHECK lines, and it is moreover tedious to
generate these CHECK lines, as the output fom the analysis is not
stable, and requires the test-writer to hand-craft FileCheck matches.
Alleviate this pain, and support output from:
$ opt -passes='print<loop-accesses>'
This patch includes several non-trivial changes including:
- Preserving whitespace at the beginning of the line, so that the LAA
output can be properly indented.
- Regexes matching the unstable output, which is basically a pointer
address hex.
- Separating is_analyze from preserve_names clearly, as the former was
formerly used as an overload for the latter.
To demonstate the utility of this patch, several tests in
LoopAccessAnalysis have been auto-generated by
update_analyze_test_checks.py.
Previously when using `-p` a.k.a. `--preserve-names` existing lines for
checking globals were not recognised as such, leading to the line being
kept while also being emitted again, resulting in duplicated CHECK
lines.
This resolves#70048.
update_analyze_test_checks.py currently outputs a warning when updating
a script with the run line:
$ opt -passes='print<scalar-evolution>'
saying that the script doesn't support its output, when it indeed does,
as evidenced by several tests in test/Analysis/ScalarEvolution generated
by this script. There is even a test for update_analyze_test_checks that
makes sure that SCEV output is supported. Hence, squelch the warning.
While at it, rename the update_analyze_test_checks test from basic.ll to
a more explicit scev.ll.
Following on from D139466 which added support for dead vreg defs, this
patch adds support for "undef" defs of subregs.
Use this to regenerate checks for amx-greedy-ra-spill-shape.ll which
previously required manual tweaks to the autogenerated checks to fix an
EXPENSIVE_CHECKS failure; see commit
8b7c1fbd9647a5a6ef246a6b5b2543ea0f5a2337
When (post-SSA) MIR has multiple defs of the same vreg,
update_mir_test_checks would use different variable names for each def
like this, where DEF and DEF1 both refer to %0:
```
%0:gr32 = IMPLICIT_DEF
%0:gr32 = IMPLICIT_DEF
-->
; CHECK: [[DEF:%[0-9]+]]:gr32 = IMPLICIT_DEF
; CHECK-NEXT: [[DEF1:%[0-9]+]]:gr32 = IMPLICIT_DEF
```
This should be harmless, but it messed up the way that mangle_vreg
counts the number of names in vreg_map to come up with a new numeric
suffix, such that you could get the same variable name for different
vregs, like this, where DEF2 refers to both %0 and %2:
```
%0:gr32 = IMPLICIT_DEF
%1:gr32 = IMPLICIT_DEF
%0:gr32 = IMPLICIT_DEF
%2:gr32 = IMPLICIT_DEF
-->
; CHECK: [[DEF:%[0-9]+]]:gr32 = IMPLICIT_DEF
; CHECK-NEXT: [[DEF1:%[0-9]+]]:gr32 = IMPLICIT_DEF
; CHECK-NEXT: [[DEF2:%[0-9]+]]:gr32 = IMPLICIT_DEF
; CHECK-NEXT: [[DEF2:%[0-9]+]]:gr32 = IMPLICIT_DEF
```
Fix this by always using the same variable name for the same vreg.
If the function argument block contains patterns, we split argument
matching into a separate SAME line, because LABEL labels may not contain
pattern matches.
Until now, in this case we moved the parenthesis opening the argument block
into the second line.
This generates incorrect labels in case function names are not prefix-free.
For example, for a function `foo` we generated:
CHECK-LABEL: foo
CHECK-SAME: (<args of foo>)
If the output also contains a function `foo.specialzied`, then the label for
`foo` can match `foo.specialized`, depending on output order.
This patch moves opening parenthesis to the first line, breaking common prefixes:
CHECK-LABEL: foo(
CHECK-SAME: <args of foo>)
Bump the UTC version to 3, and only move the parenthesis for version 3 and later.
Differential Revision: https://reviews.llvm.org/D158497
Without this we cannot update various clang OpenMP tests as the UTC_ARGS
version of -global-value-regex is simply ignored. The handling of the
flag should be changed to be in line with others, I left TODOs for now.
Refresh of the generic scheduling model to use A510 instead of A55.
Main benefits are to the little core, and introducing SVE scheduling information.
Changes tested on various OoO cores, no performance degradation is seen.
Differential Revision: https://reviews.llvm.org/D156799
BuildMI automatically adds the implicit operands of the
instruction. This meant we couldn''t set the dead flag on
dead implicit defs in that case.
Fix it by introducing an opcode to mark a given implicit
def as dead.
Fixes#64565
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D157515
PromoteAlloca now uses SSAUpdater, it doesn't need SROA to clean-up after it anymore.
Internal testing shows no noticeable performance impact.
Reviewed By: #amdgpu, arsenm
Differential Revision: https://reviews.llvm.org/D156398
Traps will not read/write the program state but they need an effect for
preservation, similar to `llvm.assume`. We really want a new memory kind
for that (see TODO), but for now `inaccessiblemem: write` is better than
any possible effect.
Differential Revision: https://reviews.llvm.org/D156476
Both the pattern for finding the clang version metadata, and the emitted
checker, are now more robust, to handle a vendor prefix.
Differential Revision: https://reviews.llvm.org/D154520
This prevents update_cc_tests.py from emitting hard-coded identifiers
for metadata (global variable checkers still check hard-coded
identifiers). Instead it emits regex checkers that match even if the
identifiers change. Also adds a new mode for --check-globals: instead of
simply being on or off, it now has the options 'none', 'smart' and
'all', with 'none' and 'all' corresponding to the previous modes.
The 'smart' mode only emits checks for global definitions referenced
in the IR or other metadata that itself has a definition checker
emitted, making the rule transitive. It does not emit checks for
attribute sets, since that is better checked by --check-attributes. This
mode is made the new default. To make the change in default mode
backwards compatible a version bump is introduced (to v3), and the
default remains 'none' in v1 & v2.
This will result in metadata checks being emitted more often, so filters
are added to not check absolute file paths and compiler version git
hashes.
rdar://105239218
D135579 added support for fixedStack, but did not cope with the output
of -simplify-mir which does not include the fixedStack section by
default.
Differential Revision: https://reviews.llvm.org/D152896
Consider only targets where `MCAsmInfo::ExceptionsType == ExceptionHandling::None`
and that support CFI (when `MCAsmInfo::UsesCFIForDebug` is set to true):
currently, only AMDGPU.
This patch enables the emission of CFI information in the .eh_frame
section when the uwtable attribute is present on a function.
Before, we could generate CFI information for debugging puproses only.
This patch prepares AMDGPU to support collecting GPU stack traces in the future.
I did a first implementation (https://reviews.llvm.org/D139024)
but at the time I had not realized that no other platform used
`UsesCFIForDebug`.
Reviewed By: scott.linder
Differential Revision: https://reviews.llvm.org/D151806
Derive the mustprogress attribute based on the willreturn attribute
or the fact that all callers are mustprogress.
Differential Revision: https://reviews.llvm.org/D94740
This is a follow-up to b71edfaa4ec3c998aadb35255ce2f60bba2940b0
since I forgot the lit.local.cfg files in that one.
Reformatting is done with `black`.
If you end up having problems merging this commit because you
have made changes to a python file, the best way to handle that
is to run git checkout --ours <yourfile> and then reformat it
with black.
If you run into any problems, post to discourse about it and
we will try to help.
RFC Thread below:
https://discourse.llvm.org/t/rfc-document-and-standardize-python-code-style
Reviewed By: barannikov88, kwk
Differential Revision: https://reviews.llvm.org/D150762
Implement emission of DWARF CFI instructions for MSP430. This includes descriptions of stack frame layout and location of callee-saved registers that could be used for backtracing.
Differential Revision: https://reviews.llvm.org/D146966
Integrate the BranchRelaxation pass to help with relaxing out-of-range
conditional branches.
This is mostly of concern for SPARCv9, which uses conditional branches with
much smaller range than its v8 counterparts.
(Some large autogenerated code, such as the ones generated by TableGen, already
hits this limitation when building in Release)
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D142458
This patch enables --function-signature by default under --version 2
and makes --version 2 the default. This means that all newly created
tests will check the function signature, while leaving old tests alone.
There's two motivations for this change:
* Without --function-signature, the generated check lines may fail
in a very hard to understand way if the test both includes a
function definition and a call to that function. (Though we could
address this by making the CHECK-LABEL stricter, without checking
the full signature.)
* This actually checks that uses of the arguments in the function
body use the correct argument, instead of matching against any
variable.
This is a replacement for D139006 and D140212 based on the
--version mechanism.
I did not include an opt-out flag --no-function-signature because
I'm not sure we need it. Would be happy to include it though,
if desired.
Differential Revision: https://reviews.llvm.org/D145149
We have a number of pending changes to update_test_checks.py
(and friends) that are essentially blocked on test churn:
If the output of UTC for an existing flag combination changes,
then the next time a test is regenerated, it will contain many
spurious changes. This makes changes to UTC default
behavior essentially impossible.
Examples of such changes are:
* D133943/D142373 want --function-signature to also check the
return type/attributes.
* D139006/D140212 want to make --function-signature the default
behavior.
* D142452 wants to add wildcards for block labels.
This patch tries to resolve this issue by adding a --version
argument, which works as follows:
* When regenerating an old test, the default version is 1.
* When generating a new test, the default version is the newest.
When an explicit version is specified, that of course wins.
This means that any currently existing tests will keep using
--version 1 format, while any new tests will automatically embed
--version N where N is the latest version, and then keep using
that test format from then on.
This patch only implements the --version flag without bumping
the default version, so it does not have any visible behavior
change by itself.
Differential Revision: https://reviews.llvm.org/D142473
Materialize zeros by copying from %g0, which is now marked as constant.
This makes it possible for some common operations (like integer negation) to be
performed in fewer instructions.
This continues @arichardson's patch at D132561.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D138887
This reverts commit a888825aeef8d6592c6cf5f4e5854cc39af49633.
This changes the default output of UTC, and as such introduces
spurious changes whenever existing tests are regenerated.
I've indicated in https://reviews.llvm.org/D139006#3989954 how
this can be implemented without causing test churn.
Previously, the label also matched function calls with the function
name, which caused tests to fail because the label matched on the wrong
line.
Add the `define` prefix, so only function defines are matched.
Differential Revision: https://reviews.llvm.org/D139006
As a reminder, UTC_ARGS is used by lit test cases to specify which
arguments need to be passed to update_XXXX_test_checks.py to be
auto-updated properly.
The support is achieved by relying on common.itertests, which is what
other test
updaters use to iterate over test files.
This commit also changes how the --llc-binary option is saved in args.
It used to be saved as "llc", but it is here changed to the standard
"llc_binary" to make use of an existing ignore mechanism for specific
arguments. Without that change, the option would not be ignored and
would appear in UTC_ARGS. This would be different from what e.g.
update_llc_test_checks does. As update_mir_test_checks.py now supports
UTC_ARGS, it became important to ensure the option is ignored.
Differential Revision: https://reviews.llvm.org/D135580
Over the past day or so, i've took a large swing at our tests,
and reduced the number of tests that were still using the old syntax
from ~1800 to just 200.
Left to handle: (as it is seen in this patch)
* Transforms/LSR
* Transforms/CGP
* Transforms/TypePromotion
* Transforms/HardwareLoops
* Analysis/*
* some misc.
I think this is the right point to start actively refusing
to honor the old syntax, except for the old tests,
to prevent the old syntax from creeping back in.
Thus, let's add temporary default-off flag,
and if it is not passed refuse to accept old syntax.
The tests that still need porting are annotated with this flag.
Reviewed By: aeubanks
Differential Revision: https://reviews.llvm.org/D139647
Generation of CHECK lines for fixedStack can be enabled with --print-fixed-stack.
This is particularly useful for tests which need to inspect how the
stack looks, e.g. for ABI tests.
See the other stacked revision building on top of this one which enables UTC_ARGS (in a similar fashion to other test updaters in utils/).
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D135579
