Reviewed By: MaskRay

differential Revision: https://reviews.llvm.org/D125544

Added handling rounding behavior in 32-bits for when possible. This
avoids kernel compilation generating scalarized code on platforms where
64-bit vectors are not available.

As the 48-bit lowering requires 64-bit anyway, we added a full 64-bit
solution simplifying the old path.

Reviewed By: dcaballe, mravishankar

Differential Revision: https://reviews.llvm.org/D125583

This reverts commit 79a66ec9.

The stronger asserts served their purpose; I stumbled across another bug.  Will reapply once this one is also fixed.

The bug appears to be a variant of a previous one:
* We mutate an instruction in one block.
* That mutation changes the phase3 results of another block.

This is very similiar to a previous issue, except cross block instead of within a single block.

Before this fix, the bufferization implementation made the incorrect assumption that the values yielded from the "before" region must match with the values yielded from the "after" region.

Differential Revision: https://reviews.llvm.org/D125835

This diff adjusts binaryAnd to take advantage of the analysis
based on KnownBits.

Differential revision: https://reviews.llvm.org/D125603

Test plan:
1/ ninja check-llvm
2/ ninja check-llvm-unit

Addresses the warnings emitted by Apple Clang 13.1.6 (Xcode 13.3.1).
Tip @tschuett issue #55404.

Reviewed By: rafauler

Differential Revision: https://reviews.llvm.org/D125733

Split up the BinaryLoop header and move BinaryDominatorTree into its own header,
preparing it for a standalone use.

Reviewed By: rafauler

Differential Revision: https://reviews.llvm.org/D125664

After adding sprintf, snprintf is simple. The functions are very
similar. The tests only cover the behavior of the max length since the
sprintf tests should cover the other behavior.

Reviewed By: lntue

Differential Revision: https://reviews.llvm.org/D125826

Sub-expressions that are logical operators are not spelled out
separately in basic blocks, so we need to manually visit them when we
encounter them. We do this in both the `TerminatorVisitor`
(conditionally) and the `TransferVisitor` (unconditionally), which can
cause cause an expression to be visited twice when the binary
operators are nested 2+ times.

This changes the visit in `TransferVisitor` to check if it has been
evaluated before trying to visit the sub-expression.

Differential Revision: https://reviews.llvm.org/D125821

[trace][intelpt] Support system-wide tracing [6] - Break IntelPTCollector into smaller files and minor refactor

IntelPTCollector is very big and has 3 classes in it. It's actually cleaner if each one has its own file. This also gives more visibility to the developer about the different kinds of "tracers" that we have.

Besides that, I'm now restricting the creation of the BinaryData chunks to GetState() instead of having it in different places, which is not very clean, because the gdb-remote protocol should be as restricted as possible.

Differential Revision: https://reviews.llvm.org/D125047

[trace][intelpt] Support system-wide tracing [5] - Disable/enable per-core tracing based on the process state

When tracing on per-core mode, we are tracing all processes, which means
that after hitting a breakpoint, our process will stop running (thus
producing no more tracing data) but other processes will continue
writing to our trace buffers. This causes a big data loss for our trace.
As a way to remediate this, I'm adding some logic to pause and unpause
tracing based on the target's state. The earlier we do it the better,
however, I'm not adding the trigger at the earliest possible point for
simplicity of this diff. Later we can improve that part.

Differential Revision: https://reviews.llvm.org/D124962

This diffs implements per-core tracing on lldb-server. It also includes tests that ensure that tracing can be initiated from the client and that the jLLDBGetState ppacket returns the list of trace buffers per core.

This doesn't include any decoder changes.

Finally, this makes some little changes here and there improving the existing code.

A specific piece of code that can't reliably be tested is when tracing
per core fails due to permissions. In this case we add a
troubleshooting message and this is the manual test:

```
/proc/sys/kernel/perf_event_paranoid set to 1

(lldb) process trace start --per-core-tracing                                         error: perf event syscall failed: Permission denied
 You might need that /proc/sys/kernel/perf_event_paranoid has a value of 0 or -1.
``

Differential Revision: https://reviews.llvm.org/D124858

Differential Revision: https://reviews.llvm.org/D125279

This reverts commit ccdb56ac.

Still seeing windows failures on GN bots: http://45.33.8.238/win/58316/step_9.txt

Unfortunately I can't debug these at all - it's a bare unsymbolized
stacktrace, and I can't reproduce the failure.

We always had global and scratch loads to LDS in the gfx9,
but did not handle it. These were available via the 'lds'
encoding bit. In gfx940 this bit was reused as 'svs' which
resulted in new '_lds' opcodes effectively pushing this
bit into the opcode, but functionally it is the same. These
instructions are also available on gfx10.

Differential Revision: https://reviews.llvm.org/D125126

This reverts commit 76ddbb1c.

Weaken the guard for whether a sub-expression has been evaluated to
only check for the storage location, instead of checking for the
value. It should be sufficient to check for the storage location, as
we don't necessarily guarantee that a value will be set for the
location (although this is currently true right now).

Differential Revision: https://reviews.llvm.org/D125823

This patch adds a transform to the local prepass in InsertVSETVLI which canonicalizes an AVL of a register from another vsetvli into immediate or VLMAX when VTYPE is the same. In this patch, I chose to be conservative and avoid arbitrary vreg forwarding due to profitability concerns about possibility overlapping live ranges.

This has the effect of eliminating vsetvli instructions in loops which are walking either VLMAX or a constant number of lanes per iteration.

Differential Revision: https://reviews.llvm.org/D125812

This adds the sprintf entrypoint, as well as unit tests. Currently
sprintf only supports %%, %s, and %c, but the other conversions are on
the way.

Reviewed By: sivachandra, lntue

Differential Revision: https://reviews.llvm.org/D125573

shuffle (cast X), (cast Y), Mask --> cast (shuffle X, Y, Mask)

This extends the transform added with 0353c2c9.

If the casts are to a larger element type, the transform
reduces shuffle bit width, so that should be a win for
most codegen (if not, it can be inverted).

- add missing benchmark for lex/preprocess steps
- name benchmarks after the function they're benchmarking, when appropriate
- remove unergonomic "run" prefixes from benchmark names
- give a useful error message if --grammar or --source are missing
- Use realistic example of how to run, run all benchmarks by default.
  (for someone who doesn't know the commands, this is the most useful action)
- Improve typos/wording in comment
- clean up unused vars
- avoid "parseable stream" name, which isn't a great name & not one I expected
  to escape from ClangPseudoMain

Differential Revision: https://reviews.llvm.org/D125312

This diff updates the LLVMIR dialect Fastmath flags attribute to use recently
added features of `BitEnum` attributes. Specifically, this diff uses the bit
enum "group" case to represent the `fast` value as an alias for a combination
of other values (`ninf`, `nnan`, ...), instead of using a separate integer
value. (This is in line with LLVM's fastmath flags representation.) This diff
also leverages the `printBitEnumPrimaryGroups` `tblgen` field for concise
enum printing.

The `BitEnum` features were developed for an upcoming diff that adds `fastmath`
support to the arithmetic dialect. This diff simply applies some of the relevant
new features to the LLVM dialect attribute.

Reviewed By: ftynse, Mogball

Differential Revision: https://reviews.llvm.org/D124720

This is a followup to D124715, which changed the default, and it anticipates
future patches raising the priority of Low (which is currently equal to
Background on Windows & Linux).
The main point is to allow users to restore the old behavior, which e.g.
allows efficiency cores to remain idle.

I did consider making this a config setting, this is a more complicated change:
 - needs to touch queue priorities as well as thread priorities
 - we don't know the priority until evaluating the config inside the task
 - users would want the ability to prioritize background indexing tasks relative
   to each other without necessarily affecting thread priority, so using one
   option for both may be confusing
I don't really have a use case, so I prefer the simpler thing.

Differential Revision: https://reviews.llvm.org/D125673

These asserts are believed to hold after several recent miscompiles have been fixed.  If you see an assertion failure on this change, please toggle the default back and make sure you file a bug with a reproducer.  We may have as yet uncaught miscompiles lurking in this code.

Differential Revision: https://reviews.llvm.org/D125271

llvm's json parser supports uint64_t, so let's better use it for the
packets being sent between lldb and lldb-server instead of using int64_t
as an intermediate type, which might be error-prone.

MC layer support for SOP(scalar alu operations) including encoding
support for s_delay_alu and s_sendmsg_rtn.

Contributors:
Jay Foad <jay.foad@amd.com>

Patch 7/N for upstreaming of AMDGPU gfx11 architecture.

Depends on D125319

Reviewed By: #amdgpu, arsenm

Differential Revision: https://reviews.llvm.org/D125498

A wait is only needed after the DMA before LDS can be read.

Its a lot easier to quickly search for the combine when it actually contains the name of the opcode it combines.

Differential Revision: https://reviews.llvm.org/D124884

Previously, GEPOp relies on `findKnownStructIndices` to check if a GEP
index should be static. The truth is, `findKnownStructIndices` can only
tell you a GEP index _might_ be indexing into a struct (which should use
a static GEP index). But GEPOp::build and GEPOp::verify are falsely
taking this information as a certain answer, which creates many false
alarms like the one depicted in
`test/Target/LLVMIR/Import/dynamic-gep-index.ll`.

The solution presented here adopts a new verification scheme: When we're
recursively checking the child element types of a struct type, instead
of checking every child types, we only check the one dictated by the
(static) GEP index value. We also combine "refinement" logics --
refine/promote struct index mlir::Value into constants -- into the very
verification process since they have lots of logics in common. The
resulting code is more concise and less brittle.

We also hide GEPOp::findKnownStructIndices since most of the
aforementioned logics are already encapsulated within GEPOp::build and
GEPOp::verify, we found little reason for findKnownStructIndices (or the
new findStructIndices) to be public.

Differential Revision: https://reviews.llvm.org/D124935

Fix a typo error in the comment in DivergenceAnalysis.h

Reviewed By: asbirlea

Differential Revision: https://reviews.llvm.org/D125808