Lower multi-dimensional arrays reduction for add and mul operator.

Depends on D153448

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D153455

Lower 1d array reduction for add and mul operator. Multi-dimensional arrays and
other operator will follow.

Reviewed By: razvanlupusoru

Differential Revision: https://reviews.llvm.org/D153448

This patch uses castAs instead of getAs which will assert if the type doesn't match and adds nullptr check if needed.

Also this patch improves the codes and passes I.getData() instead of doing a lookup in dumpVarDefinitionName()
since we're iterating over the same map in LocalVariableMap::dumpContex().

Reviewed By: aaron.ballman, aaronpuchert

Differential Revision: https://reviews.llvm.org/D153033

This patch adds missing assignment operator to the class which has user-defined copy constructor.

Reviewed By: tahonermann, aaronpuchert

Differential Revision: https://reviews.llvm.org/D150931

Sanitizers allocate shadow and memory as MAP_NORESERVE.

User memory can stay this way and do not increase RSS as long as we
don't store there.

The shadow unpoisoning also can avoid RSS increase for zeroed pages.
However as soon we poison the shadow, we need the page in RSS.

To avoid unnececary RSS increase we should not poison memory just before
unpoisoning them.

Depends on D153497.

Reviewed By: thurston

Differential Revision: https://reviews.llvm.org/D153500

When the MCAssembler is non-null and the MCAsmLayout is null, we can fold A-B
when

* A and B are in the same fragment, or
* A's fragment suceeds B's fragment, and they are not separated by non-data fragments (D69411)

This patch allows folding when A's fragment precedes B's fragment so
that `9997b - . == 0` below can be evaluated as true:

```
nop
.arch_extension sec
9997:nop
// old behavior: error: expected absolute expression
.if 9997b - . == 0
.endif
```

Add a case to llvm/test/MC/ARM/directive-if-subtraction.s.
Note: for MCAsmStreamer, we cannot evaluate `.if . - 9997b == 0` at parse
time due to MCAsmStreamer::getAssemblerPtr returning nullptr (D45164).

Some Darwin tests check that this folding does not work. Add `.p2align 2` to
block some label difference folding or adjust the tests.

Reviewed By: nickdesaulniers

Differential Revision: https://reviews.llvm.org/D153096

Test cases for #62565.

These headers are currently broken when included from the offloading
languages like OpenMP, OpenCL, CUDA, and HIP. Turn this logic off so we
can compile these languages when the GPU libc is installed. I am
currently trying to remedy this and have made an RFC for it in libc,
see https://discourse.llvm.org/t/rfc-implementing-gpu-headers-in-the-llvm-c-library/71523.

Reviewed By: JonChesterfield

Differential Revision: https://reviews.llvm.org/D153578

Reviewed By: erichkeane, steakhal, tahonermann, shafik

Differential Revision: https://reviews.llvm.org/D150744

This appears to be just an accidental copy rather than move from a scratch
variable.
As well as doing redundant work, these copies introduce extra SAT variables
which make debugging harder (each Enviroment has a unique FC token).

Example flow condition before:
```
(B0:1 = V15)
(B1:1 = V8)
(B2:1 = V10)
(B3:1 = (V4 & (!V7 => V6)))
(V10 = (B3:1 & !V7))
(V12 = B1:1)
(V13 = B2:1)
(V15 = (V12 | V13))
(V3 = V2)
(V4 = V3)
(V8 = (B3:1 & !!V7))
B0:1
V2
```

after:
```
(B0:1 = (V9 | V10))
(B1:1 = (B3:1 & !!V6))
(B2:1 = (B3:1 & !V6))
(B3:1 = (V3 & (!V6 => V5)))
(V10 = B2:1)
(V3 = V2)
(V9 = B1:1)
B0:1
V2
```

(with labelling from D153488)

There are also some more copies that can be avoided here (when multiple blocks
without terminating statements are joined), but they're less trivial, so I'll
put those in another patch.

Differential Revision: https://reviews.llvm.org/D153491

Causes a clang crash (see crbug.com/1457256).

This reverts commit 01504933.

The _e64_dpp suffix can be added to an instruction to force the
AsmParser to encode it as VOP3 with DPP if possible on GFX11+. This has
been the behavior since GFX11 was introduced; this patch only updates
the documentation.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D153564

ComplexDeinterleavingPass.cpp:1849:3: error: default label in switch which covers all enumeration values

This reverts commit 116953b8.

As the extension list continues to grow it probably makes sense
to use a binary search rather than linear search. Sorting the strings
will make this possible.

This also avoids any question about where to add new strings in
the tables.

Reviewed By: asb

Differential Revision: https://reviews.llvm.org/D153170

For the secondary allocation we don't need poison and fill memory if we
skip quarantine, and we don't need to poison after quarantine. In both
cases the secondary allocator will unmap memory and unpoison the shadow
from get_allocator().Deallocate().

Depends on D153496.

Reviewed By: thurston

Differential Revision: https://reviews.llvm.org/D153497

The select-of-different-exp pattern appears in the device
libraries. I haven't seen the select-of-values case.

getExpr is missing a check to make sure the result is invertible.
This can lead to incorrect results, so return nullptr in those cases
like in other places in IVUsers.

Fixes #62660.

Reviewed By: qcolombet

Differential Revision: https://reviews.llvm.org/D153202

There is a gap between running opt -Oz and running opt -passes="OZ_PASSES" where OZ_PASSES is taken from running opt -Oz -print-pipeline-passes.

One of the reasons causing this is that -Oz uses non-default setting for LoopRotate but LoopRotate does not expose its settings when printing the pipeline.

This commit fixes this by exposing LoopRotates parameters.

Reviewed By: aeubanks

Differential Revision: https://reviews.llvm.org/D153437

Revert "[llvm-exegesis] Introduce Subprocess Executor Mode"

This reverts commit 5e9173c4.
This reverts commit 4d618b52.

Reverting the PID commit as it is currently breaking MinGW builds and
the way I'm checking for the presence of pid_t needs to be fixed and I
need to do some testing. The subprocess executor mode patch is a
dependent patch so also needs to be reverted and also needs some work as
it is currently failing tests where libpfm is installed and the kernel
version is less than 5.6.

Fix build failure on windows system with msvc toolchain

Reviewed By: ellis

Differential Revision: https://reviews.llvm.org/D153318

D152759 introduced the Android .zip so file support, but it only considered
POSIX path. The code also runs on Windows, so the path could be Windows path.
Support both patterns on Windows.

Differential Revision: https://reviews.llvm.org/D153390

Reviewed By: thurston

Differential Revision: https://reviews.llvm.org/D153496

Since the scheduling resources for reductions and ordered reductions now
account for LMUL and SEW, we can modify the Latency and ResourceCycles
for these resoruces.

* Most reductions take a total of approx `vl*SEW/DLEN + 5*(4 + log2(DLEN/SEW))`
  cycles.
* Ordered floating-point reductions take a total of approx `5*vl` cycles.

This commit re-commits 208fc34c. It was
failing because it used wrong version of SchedSEWSet.

Differential Revision: https://reviews.llvm.org/D153474

Differential Revision: https://reviews.llvm.org/D146148

This reverts commit 208fc34c.

Reverting because build failure.

Since the scheduling resources for reductions and ordered reductions now
account for LMUL and SEW, we can modify the Latency and ResourceCycles
for these resoruces.

* Most reductions take a total of approx `vl*SEW/DLEN + 5*(4 + log2(DLEN/SEW))`
  cycles.
* Ordered floating-point reductions take a total of approx `5*vl` cycles.

Differential Revision: https://reviews.llvm.org/D153474

* Unit-stride loads and stores can operate at the full bandwidth of the
memory pipe. The memory pipe is DLEN bits wide.

* Strided loads and stores operate at one element per cycle and should
be scheduled accordingly.

* Indexed loads and stores operate at one element per cycle, and they
stall the machine until all addresses have been generated, so they
cannot be scheduled.

* Unit stride seg2 load is number of DLEN parts

* seg3-8 are one segment per cycle, unless the segment is larger
than DLEN in which each segment takes multiple cycles.

Differential Revision: https://reviews.llvm.org/D153475

Also moves the wait-until-inbox-changes test into a shared method.

Reviewed By: jhuber6

Differential Revision: https://reviews.llvm.org/D153573

Almost NFC, as blocks over max quarantine size will trigger immediate
drain anyway. In followup patches we can optimize passthrough case.

Reviewed By: thurston

Differential Revision: https://reviews.llvm.org/D153495

The `__DATA,xray_instr_map` section has label differences like
`.quad Lxray_sled_0-Ltmp0` that is represented as a pair of UNSIGNED and SUBTRACTOR relocations.

LLVM integrated assembler attempts to rewrite A-B into A-B'+offset where B' can
be included in the symbol table. B' is called an atom and should be a
non-temporary symbol in the same section. However, since `xray_instr_map` does
not define a non-temporary symbol, the SUBTRACTOR relocation will have no
associated symbol, and its `r_extern` value will be 0. Therefore, we will see
linker errors like:

    error: SUBTRACTOR relocation must be extern at offset 0 of __DATA,xray_instr_map in a.o

To fix this issue, we need to define a non-temporary symbol in the section. We
can accomplish this by renaming `Lxray_sleds_start0` to `lxray_sleds_start0`
("L" to "l").

`lxray_sleds_start0` serves as the atom for this dead-strippable subsection.
With the `S_ATTR_LIVE_SUPPORT` attribute, `ld -dead_strip` will retain
subsections that reference live functions.

Special thanks to Oleksii Lozovskyi for reporting the issue and providing
initial analysis.

Differential Revision: https://reviews.llvm.org/D153239

Replace getAs with castAs and add assert if needed.
Differential revision: https://reviews.llvm.org/D153236

Adds the capability to recognize SelectInst that appear in the IR.
These instructions are generated during scalable vectorization for reduction
and when the code contains conditions inside the loop body or when
"-prefer-predicate-over-epilogue=predicate-dont-vectorize" is set.

Differential Revision: https://reviews.llvm.org/D152558

Signed-off-by: Jun Zhang <jun@junz.org>

Differential Revision: https://reviews.llvm.org/D153572

This makes the interface less error prone. The acquire was previously
forgotten. Release is currently missing if recv() is the last operation made
before close.

Reviewed By: jhuber6

Differential Revision: https://reviews.llvm.org/D153571