applyLoopGuards() already combines conditions from multiple nested
guards. However, it cannot use multiple conditions on the same guard,
combined using and/or. Add support for this by recursing into either
`and` or `or`, depending on the direction of the branch.
Differential Revision: https://reviews.llvm.org/D101692
To the best of my knowledge, all instructions are modelled,
and have reasonable values to them; flipping the switch
doesn't cause any diff for MCA tests, so either we're good,
or we have test coverage gaps.
I'm not really sure why no other X86 sched model is marked as complete.
Symbols explicitly exported via command-line options `--exported_symbol SYM` and `--exported_symbols_list FILE` must be defined. Before this fix, lazy symbols defined in archives would be left to languish. We now force them to be included in the linked output.
Differential Revision: https://reviews.llvm.org/D102100
This patch lifts the restriction on the number of read/write registers for a
move elimination candidate. With this patch, move elimination candidates with
exactly two reads and two writes are treated like register swap operations for
the purpose of move elimination.
This patch currently doesn't affect any upstream model. However, it should help
unblock the progress on PR50258.
This fixes a long standing issue where the triple is not always set
consistently in all configurations. This change also moves the
back-deployment Lit features to using the proper target triple
instead of using something ad-hoc.
This will be necessary for using from scratch Lit configuration files
in both normal testing and back-deployment testing.
Differential Revision: https://reviews.llvm.org/D102012
For `AffineLoopFusion` pass, add `memref` dialect as a dependent
dialect. Since the fusion pass can create `memref::AllocOp`s, the
dialect must be registered in its dependent dialects.
The missing dependency was not discovered until now because the above
said op creation happes only when the input already has
`memref::AllocOp`s in it, and all dialects in the input are
automatically added to the context.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D102104
This appears to miscompile google benchmark's GetCacheSizesFromKVFS()
when compiling with -fstrict-vtable-pointers.
Runnable reproducer: https://godbolt.org/z/f9ovKqTzb
The "f.fail()" crashes with BUS error, it is compiled into testb,
and the adress it is testing is non-sensical.
This reverts commit 4c89bcadf6.
Currently we model i16 bswap as very high cost (`10`),
which doesn't seem right, with all other being at `1`.
Regardless of `MOVBE`, i16 reg-reg bswap is lowered into
(an extending move plus) rot-by-8:
https://godbolt.org/z/8jrq7fMTj
I think it should at worst have throughput of `1`:
Since i32/i64 already have cost of `1`,
`MOVBE` doesn't improve their costs any further.
BUT, `MOVBE` must have at least a single memory operand,
with other being a register. Which means, if we have
a bswap of load, iff load has a single use,
we'll fold bswap into load.
Likewise, if we have store of a bswap, iff bswap
has a single use, we'll fold bswap into store.
So i think we should treat such a bswap as free,
unless of course we know that for the particular CPU
they are performing badly.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D101924
This resolves issues when the CMake in use on the host is too old to
configure libc++ properly, but Xcode has a sufficiently recent version.
It is technically possible for the reverse issue to happen, where the
Xcode version would be too old and the user-installed version would be
better, however in the context of our build bots, we use AppleClang on
Apple platforms, and the CMake shipped with Xcode should work with the
AppleClang shipped alongside that Xcode.
Differential Revision: https://reviews.llvm.org/D102083
Motivation: we have passes with lot of rewrites and when one one them segfaults or asserts, it is very hard to find waht exactly pattern failed without debug info.
Differential Revision: https://reviews.llvm.org/D101443
Printing pass manager invocations is fairly verbose and not super
useful.
This allows us to remove DebugLogging from pass managers and PassBuilder
since all logging (aside from analysis managers) goes through
instrumentation now.
This has the downside of never being able to print the top level pass
manager via instrumentation, but that seems like a minor downside.
Reviewed By: ychen
Differential Revision: https://reviews.llvm.org/D101797
UnwindTable::parseRows() may return successfully if the CFIProgram has either
no CFI instructions or only DW_CFA_nop instructions and the UnwindRow return
argument will be empty. But currently, the callers are not checking for this case
which is leading to incorrect dumps in the unwind tables in such cases i.e.
CFA=unspecified
Reviewed By: clayborg
Differential Revision: https://reviews.llvm.org/D101892
The current design uses a unique entry for each argument/result attribute, with the name of the entry being something like "arg0". This provides for a somewhat sparse design, but ends up being much more expensive (from a runtime perspective) in-practice. The design requires building a string every time we lookup the dictionary for a specific arg/result, and also requires N attribute lookups when collecting all of the arg/result attribute dictionaries.
This revision restructures the design to instead have an ArrayAttr that contains all of the attribute dictionaries for arguments and another for results. This design reduces the number of attribute name lookups to 1, and allows for O(1) lookup for individual element dictionaries. The major downside is that we can end up with larger memory usage, as the ArrayAttr contains an entry for each element even if that element has no attributes. If the memory usage becomes too problematic, we can experiment with a more sparse structure that still provides a lot of the wins in this revision.
This dropped the compilation time of a somewhat large TensorFlow model from ~650 seconds to ~400 seconds.
Differential Revision: https://reviews.llvm.org/D102035
At 61 or over, I see messages like
File "...\Python\Python39\lib\multiprocessing\connection.py", line 816, in _exhaustive_wait
res = _winapi.WaitForMultipleObjects(L, False, timeout)
ValueError: need at most 63 handles, got a sequence of length 64
60 seems to work for me.
If this causes issues for anybody else, feel free to revert.
This provides information when the user hovers over a part of the source .mlir file. This revision adds the following hover behavior:
* Operation:
- Shows the generic form.
* Operation Result:
- Shows the parent operation name, result number(s), and type(s).
* Block:
- Shows the parent operation name, block number, predecessors, and successors.
* Block Argument:
- Shows the parent operation name, parent block, argument number, and type.
Differential Revision: https://reviews.llvm.org/D101113
This reverts commit d319005a37.
Causing messages like:
File "...\Python\Python39\lib\multiprocessing\connection.py", line 816, in _exhaustive_wait
res = _winapi.WaitForMultipleObjects(L, False, timeout)
ValueError: need at most 63 handles, got a sequence of length 74
We never bothered to have a separate set of combines for -O0 in the prelegalizer
before. This results in some minor performance hits for a mode where performance
isn't a concern (although not regressing code size significantly is still preferable).
This also removes the CSE option since we don't need it for -O0.
Through experiments, I've arrived at a set of combines that gets the most code
size improvement at -O0, while reducing the amount of time spent in the combiner
by around 35% give or take.
Differential Revision: https://reviews.llvm.org/D102038
We're trying to move DebugLogging into instrumentation, rather than
being part of PassManagers/AnalysisManagers.
Reviewed By: ychen
Differential Revision: https://reviews.llvm.org/D102093
Roman Lebedev