This change implements pseudo probe encoding and emission for CSSPGO. Please see RFC here for more context: https://groups.google.com/g/llvm-dev/c/1p1rdYbL93s
Pseudo probes are in the form of intrinsic calls on IR/MIR but they do not turn into any machine instructions. Instead they are emitted into the binary as a piece of data in standalone sections. The probe-specific sections are not needed to be loaded into memory at execution time, thus they do not incur a runtime overhead.
**ELF object emission**
The binary data to emit are organized as two ELF sections, i.e, the `.pseudo_probe_desc` section and the `.pseudo_probe` section. The `.pseudo_probe_desc` section stores a function descriptor for each function and the `.pseudo_probe` section stores the actual probes, each fo which corresponds to an IR basic block or an IR function callsite. A function descriptor is stored as a module-level metadata during the compilation and is serialized into the object file during object emission.
Both the probe descriptors and pseudo probes can be emitted into a separate ELF section per function to leverage the linker for deduplication. A `.pseudo_probe` section shares the same COMDAT group with the function code so that when the function is dead, the probes are dead and disposed too. On the contrary, a `.pseudo_probe_desc` section has its own COMDAT group. This is because even if a function is dead, its probes may be inlined into other functions and its descriptor is still needed by the profile generation tool.
The format of `.pseudo_probe_desc` section looks like:
```
.section .pseudo_probe_desc,"",@progbits
.quad 6309742469962978389 // Func GUID
.quad 4294967295 // Func Hash
.byte 9 // Length of func name
.ascii "_Z5funcAi" // Func name
.quad 7102633082150537521
.quad 138828622701
.byte 12
.ascii "_Z8funcLeafi"
.quad 446061515086924981
.quad 4294967295
.byte 9
.ascii "_Z5funcBi"
.quad -2016976694713209516
.quad 72617220756
.byte 7
.ascii "_Z3fibi"
```
For each `.pseudoprobe` section, the encoded binary data consists of a single function record corresponding to an outlined function (i.e, a function with a code entry in the `.text` section). A function record has the following format :
```
FUNCTION BODY (one for each outlined function present in the text section)
GUID (uint64)
GUID of the function
NPROBES (ULEB128)
Number of probes originating from this function.
NUM_INLINED_FUNCTIONS (ULEB128)
Number of callees inlined into this function, aka number of
first-level inlinees
PROBE RECORDS
A list of NPROBES entries. Each entry contains:
INDEX (ULEB128)
TYPE (uint4)
0 - block probe, 1 - indirect call, 2 - direct call
ATTRIBUTE (uint3)
reserved
ADDRESS_TYPE (uint1)
0 - code address, 1 - address delta
CODE_ADDRESS (uint64 or ULEB128)
code address or address delta, depending on ADDRESS_TYPE
INLINED FUNCTION RECORDS
A list of NUM_INLINED_FUNCTIONS entries describing each of the inlined
callees. Each record contains:
INLINE SITE
GUID of the inlinee (uint64)
ID of the callsite probe (ULEB128)
FUNCTION BODY
A FUNCTION BODY entry describing the inlined function.
```
To support building a context-sensitive profile, probes from inlinees are grouped by their inline contexts. An inline context is logically a call path through which a callee function lands in a caller function. The probe emitter builds an inline tree based on the debug metadata for each outlined function in the form of a trie tree. A tree root is the outlined function. Each tree edge stands for a callsite where inlining happens. Pseudo probes originating from an inlinee function are stored in a tree node and the tree path starting from the root all the way down to the tree node is the inline context of the probes. The emission happens on the whole tree top-down recursively. Probes of a tree node will be emitted altogether with their direct parent edge. Since a pseudo probe corresponds to a real code address, for size savings, the address is encoded as a delta from the previous probe except for the first probe. Variant-sized integer encoding, aka LEB128, is used for address delta and probe index.
**Assembling**
Pseudo probes can be printed as assembly directives alternatively. This allows for good assembly code readability and also provides a view of how optimizations and pseudo probes affect each other, especially helpful for diff time assembly analysis.
A pseudo probe directive has the following operands in order: function GUID, probe index, probe type, probe attributes and inline context. The directive is generated by the compiler and can be parsed by the assembler to form an encoded `.pseudoprobe` section in the object file.
A example assembly looks like:
```
foo2: # @foo2
# %bb.0: # %bb0
pushq %rax
testl %edi, %edi
.pseudoprobe 837061429793323041 1 0 0
je .LBB1_1
# %bb.2: # %bb2
.pseudoprobe 837061429793323041 6 2 0
callq foo
.pseudoprobe 837061429793323041 3 0 0
.pseudoprobe 837061429793323041 4 0 0
popq %rax
retq
.LBB1_1: # %bb1
.pseudoprobe 837061429793323041 5 1 0
callq *%rsi
.pseudoprobe 837061429793323041 2 0 0
.pseudoprobe 837061429793323041 4 0 0
popq %rax
retq
# -- End function
.section .pseudo_probe_desc,"",@progbits
.quad 6699318081062747564
.quad 72617220756
.byte 3
.ascii "foo"
.quad 837061429793323041
.quad 281547593931412
.byte 4
.ascii "foo2"
```
With inlining turned on, the assembly may look different around %bb2 with an inlined probe:
```
# %bb.2: # %bb2
.pseudoprobe 837061429793323041 3 0
.pseudoprobe 6699318081062747564 1 0 @ 837061429793323041:6
.pseudoprobe 837061429793323041 4 0
popq %rax
retq
```
**Disassembling**
We have a disassembling tool (llvm-profgen) that can display disassembly alongside with pseudo probes. So far it only supports ELF executable file.
An example disassembly looks like:
```
00000000002011a0 <foo2>:
2011a0: 50 push rax
2011a1: 85 ff test edi,edi
[Probe]: FUNC: foo2 Index: 1 Type: Block
2011a3: 74 02 je 2011a7 <foo2+0x7>
[Probe]: FUNC: foo2 Index: 3 Type: Block
[Probe]: FUNC: foo2 Index: 4 Type: Block
[Probe]: FUNC: foo Index: 1 Type: Block Inlined: @ foo2:6
2011a5: 58 pop rax
2011a6: c3 ret
[Probe]: FUNC: foo2 Index: 2 Type: Block
2011a7: bf 01 00 00 00 mov edi,0x1
[Probe]: FUNC: foo2 Index: 5 Type: IndirectCall
2011ac: ff d6 call rsi
[Probe]: FUNC: foo2 Index: 4 Type: Block
2011ae: 58 pop rax
2011af: c3 ret
```
Reviewed By: wmi
Differential Revision: https://reviews.llvm.org/D91878
The NPM runs loop passes over loops in forward program order, rather
than the legacy loop PM's reverse program order. This seems to produce
better results as shown here.
I verified that changing the loop order to reverse program order results
in the same IR with the NPM.
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D92817
If SETUNE isn't legal, UO can use the NOT of the SETO expansion.
Removes some complex isel patterns. Most of the test changes are
from using XORI instead of SEQZ.
Differential Revision: https://reviews.llvm.org/D92008
This makes it slightly easier to deal with custom attributes and
CallBase already provides hasFnAttr versions that support both AttrKind
and StringRef arguments in a similar fashion.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D92567
Ports 6e42a417bacb since it's now needed, and undo an accidental
deletion from d69762c404ded while here (this part is not needed to fix
the build, it's just in the vicinity).
Remove the OpenMP clause information from the OMPKinds.def file and use the
information from the new OMP.td file. There is now a single source of truth for the
directives and clauses.
To avoid generate lots of specific small code from tablegen, the macros previously
used in OMPKinds.def are generated almost as identical. This can be polished and
possibly removed in a further patch.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D92955
This patch changes performMSCATTERCombine to also promote the indices of
masked gathers where the element type is i8 or i16, and adds various tests
for gathers with illegal types.
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D91433
Remove target features crypto for Cortex-R82, because it doesn't have any, and
add LSE which was missing while we are at it.
This also removes crypto from the v8-R architecture description because that
aligns better with GCC and so far none of the R-cores have implemented crypto,
so is probably a more sensible default.
Differential Revision: https://reviews.llvm.org/D91994
We currently have problems with the way that low overhead loops are
specified, with LR being spilled between the t2LoopDec and the t2LoopEnd
forcing the entire loop to be reverted late in the backend. As they will
eventually become a single instruction, this patch introduces a
t2LoopEndDec which is the combination of the two, combined before
registry allocation to make sure this does not fail.
Unfortunately this instruction is a terminator that produces a value
(and also branches - it only produces the value around the branching
edge). So this needs some adjustment to phi elimination and the register
allocator to make sure that we do not spill this LR def around the loop
(needing to put a spill after the terminator). We treat the loop very
carefully, making sure that there is nothing else like calls that would
break it's ability to use LR. For that, this adds a
isUnspillableTerminator to opt in the new behaviour.
There is a chance that this could cause problems, and so I have added an
escape option incase. But I have not seen any problems in the testing
that I've tried, and not reverting Low overhead loops is important for
our performance. If this does work then we can hopefully do the same for
t2WhileLoopStart and t2DoLoopStart instructions.
This patch also contains the code needed to convert or revert the
t2LoopEndDec in the backend (which just needs a subs; bne) and the code
pre-ra to create them.
Differential Revision: https://reviews.llvm.org/D91358
When llvm-rc loads an external file, it looks for it relative to
a number of include directories and the current working directory.
If the path is considered absolute, llvm-rc tries to open the
filename as such, and doesn't try to open it relative to other
paths.
On Windows, a path name like "\dir\file" isn't considered absolute
as it lacks the drive name, but by appending it on top of the search
dirs, it's not found.
LLVM's sys::path::append just appends such a path (same with a properly
absolute posix path) after the paths it's supposed to be relative to.
This fix doesn't handle the case if the resource script and the
external file are on a different drive than the current working
directory; to fix that, we'd have to make LLVM's sys::path::append
handle appending fully absolute and partially absolute paths (ones
lacking a drive prefix but containing a root directory), or switch
to C++17's std::filesystem.
Differential Revision: https://reviews.llvm.org/D92558
Current interface of AddressMap assumes that relocations exist.
That is correct for not-linked object file but is not correct
for linked executable. This patch changes interface in such way
that AddressMap could be used not only with not-linked object files:
hasValidRelocationAt()
replaced with:
hasLiveMemoryLocation()
hasLiveAddressRange()
Differential Revision: https://reviews.llvm.org/D87723
Both ds_read_b128 and ds_read2_b64 are valid for 128bit 16-byte aligned
loads but the one that will be selected is determined either by the order in
tablegen or by the AddedComplexity attribute. Currently ds_read_b128 has
priority.
While ds_read2_b64 has lower alignment requirements, we cannot always
restrict ds_read_b128 to 16-byte alignment because of unaligned-access-mode
option. This was causing ds_read_b128 to be selected for 8-byte aligned
loads regardles of chosen access mode.
To resolve this we use two patterns for selecting ds_read_b128. One
requires alignment of 16-byte and the other requires
unaligned-access-mode option.
Same goes for ds_write2_b64 and ds_write_b128.
Differential Revision: https://reviews.llvm.org/D92767
The phi created in a low overhead loop gets created with a default
register class it seems. There are then copied inserted between the low
overhead loop pseudo instructions (which produce/consume GPRlr
instructions) and the phi holding the induction. This patch removes
those as a step towards attempting to make t2LoopDec and t2LoopEnd a
single instruction, and appears useful in it's own right as shown in the
tests.
Differential Revision: https://reviews.llvm.org/D91267
This patch implements amx programming model that discussed in llvm-dev
(http://lists.llvm.org/pipermail/llvm-dev/2020-August/144302.html).
Thank Hal for the good suggestion in the RA. The fast RA is not in the patch yet.
This patch implemeted 7 components.
1. The c interface to end user.
2. The AMX intrinsics in LLVM IR.
3. Transform load/store <256 x i32> to AMX intrinsics or split the
type into two <128 x i32>.
4. The Lowering from AMX intrinsics to AMX pseudo instruction.
5. Insert psuedo ldtilecfg and build the def-use between ldtilecfg to amx
intruction.
6. The register allocation for tile register.
7. Morph AMX pseudo instruction to AMX real instruction.
Change-Id: I935e1080916ffcb72af54c2c83faa8b2e97d5cb0
Differential Revision: https://reviews.llvm.org/D87981
llvm::Linker::linkModules() is a static member, so there is no need
to pass reference to llvm::Linker instance to loadArFile() function.
Reviewed By: tra
Differential Revision: https://reviews.llvm.org/D92918
There is an explicit option for the lexer to support this, but we crash
when `-preserve-comments` is enabled because it checks for
`getTok().getString().empty()` to detect the case. This doesn't
work currently because the lexer reports this case as a string of length
1, containing a null byte.
Change the lexer to instead report this case via an empty string, as the
null terminator isn't logically a part of the textual input, and the
check for `.empty()` seems natural and obvious in the calling code.
Reviewed By: niravd
Differential Revision: https://reviews.llvm.org/D92681
Rather than creating a series of associated calls and ensuring that
everything is lined up, use a table driven approach that ensures that
they two always stay in sync.
Avoids spurious newlines showing up in the output when emitting assembly
via MC.
Reviewed By: MaskRay, arsenm
Differential Revision: https://reviews.llvm.org/D92690
Pretty sure we meant to be checking signed 32 immediates here
rather than unsigned 32 bit. I suspect I messed this up because
in MathExtras.h we have isIntN and isUIntN so isIntN differs in
signedness depending on whether you're using APInt or plain integers.
This fixes a case where we didn't fold a constant created
by shrinkAndImmediate. Since shrinkAndImmediate doesn't topologically
sort constants it creates, we can fail to convert the Constant
to a TargetConstant. This leads to very strange behavior later.
Fixes PR48458.
-DENABLE_EXPERIMENTAL_NEW_PASS_MANAGER=on configured LLD and LLVMgold.so
will use the new pass manager by default. Add an option to
use the legacy pass manager. This will also be used by the Clang driver
when -fno-new-pass-manager (D92915) / -fno-experimental-new-pass-manager is set.
Reviewed By: aeubanks, tejohnson
Differential Revision: https://reviews.llvm.org/D92916
Revert part of https://reviews.llvm.org/D92084 to make it simpler to
start consuming the EndOfStatement token within AMDGPU's
ParseInstruction in a future patch. This also brings us back to what
every other target currently does.
A future change to move the position back to the end of the statement
would likely need to audit all of the AMDGPUOperand SMLoc ranges, and
determine the SMLoc for the last character of the last operand.
Reviewed By: dp
Differential Revision: https://reviews.llvm.org/D92960
Add builtins required to implement vcmla and rotated variants from
the ACLE
Reviewed By: t.p.northover
Differential Revision: https://reviews.llvm.org/D92929
Regenerated using:
./llvm/utils/update_llc_test_checks.py -u llvm/test/CodeGen/RISCV/*.ll
This has added comments to spill-related instructions and added @plt to
some symbols.
Differential Revision: https://reviews.llvm.org/D92841
