Use llvm::endianness::{big,little,native} (NFC)

Note that llvm::support::endianness has been renamed to
llvm::endianness while becoming an enum class as opposed to an enum.
This patch replaces llvm:

Use llvm::endianness::{big,little,native} (NFC)

Note that llvm::support::endianness has been renamed to
llvm::endianness while becoming an enum class as opposed to an enum.
This patch replaces llvm::support::{big,little,native} with
llvm::endianness::{big,little,native}.

show more ...

Cleanup LLVMDebugInfoCodeView headers

Major user-facing changes:

Many headers in llvm/DebugInfo/CodeView no longer include
llvm/Support/BinaryStreamReader.h or llvm/Support/BinaryStreamWriter.h,
th

Cleanup LLVMDebugInfoCodeView headers

Major user-facing changes:

Many headers in llvm/DebugInfo/CodeView no longer include
llvm/Support/BinaryStreamReader.h or llvm/Support/BinaryStreamWriter.h,
those headers may need to be included manually.

Several headers in llvm/DebugInfo/CodeView no longer include
llvm/DebugInfo/CodeView/EnumTables.h or llvm/DebugInfo/CodeView/CodeView.h,
those headers may need to be included manually.

Some statistics:
$ clang++ -E -Iinclude -I../llvm/include ../llvm/lib/DebugInfo/CodeView/*.cpp -std=c++14 -fno-rtti -fno-exceptions | wc -l
after: 2794466
before: 2832765

Discourse thread on the topic: https://discourse.llvm.org/t/include-what-you-use-include-cleanup/

Differential Revision: https://reviews.llvm.org/D119092

show more ...

Re-land "[PDB] Merge types in parallel when using ghashing"

Stored Error objects have to be checked, even if they are success
values.

This reverts commit 8d250ac3cd48d0f17f9314685a85e77895c05351.
R

Re-land "[PDB] Merge types in parallel when using ghashing"

Stored Error objects have to be checked, even if they are success
values.

This reverts commit 8d250ac3cd48d0f17f9314685a85e77895c05351.
Relands commit 49b3459930655d879b2dc190ff8fe11c38a8be5f..

Original commit message:
-----------------------------------------

This makes type merging much faster (-24% on chrome.dll) when multiple
threads are available, but it slightly increases the time to link (+10%)
when /threads:1 is passed. With only one more thread, the new type
merging is faster (-11%). The output PDB should be identical to what it
was before this change.

To give an idea, here is the /time output placed side by side:
BEFORE | AFTER
Input File Reading: 956 ms | 968 ms
Code Layout: 258 ms | 190 ms
Commit Output File: 6 ms | 7 ms
PDB Emission (Cumulative): 6691 ms | 4253 ms
Add Objects: 4341 ms | 2927 ms
Type Merging: 2814 ms | 1269 ms -55%!
Symbol Merging: 1509 ms | 1645 ms
Publics Stream Layout: 111 ms | 112 ms
TPI Stream Layout: 764 ms | 26 ms trivial
Commit to Disk: 1322 ms | 1036 ms -300ms
----------------------------------------- --------
Total Link Time: 8416 ms 5882 ms -30% overall

The main source of the additional overhead in the single-threaded case
is the need to iterate all .debug$T sections up front to check which
type records should go in the IPI stream. See fillIsItemIndexFromDebugT.
With changes to the .debug$H section, we could pre-calculate this info
and eliminate the need to do this walk up front. That should restore
single-threaded performance back to what it was before this change.

This change will cause LLD to be much more parallel than it used to, and
for users who do multiple links in parallel, it could regress
performance. However, when the user is only doing one link, it's a huge
improvement. In the future, we can use NT worker threads to avoid
oversaturating the machine with work, but for now, this is such an
improvement for the single-link use case that I think we should land
this as is.

Algorithm
----------

Before this change, we essentially used a
DenseMap<GloballyHashedType, TypeIndex> to check if a type has already
been seen, and if it hasn't been seen, insert it now and use the next
available type index for it in the destination type stream. DenseMap
does not support concurrent insertion, and even if it did, the linker
must be deterministic: it cannot produce different PDBs by using
different numbers of threads. The output type stream must be in the same
order regardless of the order of hash table insertions.

In order to create a hash table that supports concurrent insertion, the
table cells must be small enough that they can be updated atomically.
The algorithm I used for updating the table using linear probing is
described in this paper, "Concurrent Hash Tables: Fast and General(?)!":
https://dl.acm.org/doi/10.1145/3309206

The GHashCell in this change is essentially a pair of 32-bit integer
indices: <sourceIndex, typeIndex>. The sourceIndex is the index of the
TpiSource object, and it represents an input type stream. The typeIndex
is the index of the type in the stream. Together, we have something like
a ragged 2D array of ghashes, which can be looked up as:
tpiSources[tpiSrcIndex]->ghashes[typeIndex]

By using these side tables, we can omit the key data from the hash
table, and keep the table cell small. There is a cost to this: resolving
hash table collisions requires many more loads than simply looking at
the key in the same cache line as the insertion position. However, most
supported platforms should have a 64-bit CAS operation to update the
cell atomically.

To make the result of concurrent insertion deterministic, the cell
payloads must have a priority function. Defining one is pretty
straightforward: compare the two 32-bit numbers as a combined 64-bit
number. This means that types coming from inputs earlier on the command
line have a higher priority and are more likely to appear earlier in the
final PDB type stream than types from an input appearing later on the
link line.

After table insertion, the non-empty cells in the table can be copied
out of the main table and sorted by priority to determine the ordering
of the final type index stream. At this point, item and type records
must be separated, either by sorting or by splitting into two arrays,
and I chose sorting. This is why the GHashCell must contain the isItem
bit.

Once the final PDB TPI stream ordering is known, we need to compute a
mapping from source type index to PDB type index. To avoid starting over
from scratch and looking up every type again by its ghash, we save the
insertion position of every hash table insertion during the first
insertion phase. Because the table does not support rehashing, the
insertion position is stable. Using the array of insertion positions
indexed by source type index, we can replace the source type indices in
the ghash table cells with the PDB type indices.

Once the table cells have been updated to contain PDB type indices, the
mapping for each type source can be computed in parallel. Simply iterate
the list of cell positions and replace them with the PDB type index,
since the insertion positions are no longer needed.

Once we have a source to destination type index mapping for every type
source, there are no more data dependencies. We know which type records
are "unique" (not duplicates), and what their final type indices will
be. We can do the remapping in parallel, and accumulate type sizes and
type hashes in parallel by type source.

Lastly, TPI stream layout must be done serially. Accumulate all the type
records, sizes, and hashes, and add them to the PDB.

Differential Revision: https://reviews.llvm.org/D87805

show more ...

Revert "[PDB] Merge types in parallel when using ghashing"

This reverts commit 49b3459930655d879b2dc190ff8fe11c38a8be5f.

[PDB] Merge types in parallel when using ghashing

This makes type merging much faster (-24% on chrome.dll) when multiple
threads are available, but it slightly increases the time to link (+10%)
when

[PDB] Merge types in parallel when using ghashing

This makes type merging much faster (-24% on chrome.dll) when multiple
threads are available, but it slightly increases the time to link (+10%)
when /threads:1 is passed. With only one more thread, the new type
merging is faster (-11%). The output PDB should be identical to what it
was before this change.

To give an idea, here is the /time output placed side by side:
BEFORE | AFTER
Input File Reading: 956 ms | 968 ms
Code Layout: 258 ms | 190 ms
Commit Output File: 6 ms | 7 ms
PDB Emission (Cumulative): 6691 ms | 4253 ms
Add Objects: 4341 ms | 2927 ms
Type Merging: 2814 ms | 1269 ms -55%!
Symbol Merging: 1509 ms | 1645 ms
Publics Stream Layout: 111 ms | 112 ms
TPI Stream Layout: 764 ms | 26 ms trivial
Commit to Disk: 1322 ms | 1036 ms -300ms
----------------------------------------- --------
Total Link Time: 8416 ms 5882 ms -30% overall

The main source of the additional overhead in the single-threaded case
is the need to iterate all .debug$T sections up front to check which
type records should go in the IPI stream. See fillIsItemIndexFromDebugT.
With changes to the .debug$H section, we could pre-calculate this info
and eliminate the need to do this walk up front. That should restore
single-threaded performance back to what it was before this change.

This change will cause LLD to be much more parallel than it used to, and
for users who do multiple links in parallel, it could regress
performance. However, when the user is only doing one link, it's a huge
improvement. In the future, we can use NT worker threads to avoid
oversaturating the machine with work, but for now, this is such an
improvement for the single-link use case that I think we should land
this as is.

Algorithm
----------

Before this change, we essentially used a
DenseMap<GloballyHashedType, TypeIndex> to check if a type has already
been seen, and if it hasn't been seen, insert it now and use the next
available type index for it in the destination type stream. DenseMap
does not support concurrent insertion, and even if it did, the linker
must be deterministic: it cannot produce different PDBs by using
different numbers of threads. The output type stream must be in the same
order regardless of the order of hash table insertions.

In order to create a hash table that supports concurrent insertion, the
table cells must be small enough that they can be updated atomically.
The algorithm I used for updating the table using linear probing is
described in this paper, "Concurrent Hash Tables: Fast and General(?)!":
https://dl.acm.org/doi/10.1145/3309206

The GHashCell in this change is essentially a pair of 32-bit integer
indices: <sourceIndex, typeIndex>. The sourceIndex is the index of the
TpiSource object, and it represents an input type stream. The typeIndex
is the index of the type in the stream. Together, we have something like
a ragged 2D array of ghashes, which can be looked up as:
tpiSources[tpiSrcIndex]->ghashes[typeIndex]

By using these side tables, we can omit the key data from the hash
table, and keep the table cell small. There is a cost to this: resolving
hash table collisions requires many more loads than simply looking at
the key in the same cache line as the insertion position. However, most
supported platforms should have a 64-bit CAS operation to update the
cell atomically.

To make the result of concurrent insertion deterministic, the cell
payloads must have a priority function. Defining one is pretty
straightforward: compare the two 32-bit numbers as a combined 64-bit
number. This means that types coming from inputs earlier on the command
line have a higher priority and are more likely to appear earlier in the
final PDB type stream than types from an input appearing later on the
link line.

After table insertion, the non-empty cells in the table can be copied
out of the main table and sorted by priority to determine the ordering
of the final type index stream. At this point, item and type records
must be separated, either by sorting or by splitting into two arrays,
and I chose sorting. This is why the GHashCell must contain the isItem
bit.

Once the final PDB TPI stream ordering is known, we need to compute a
mapping from source type index to PDB type index. To avoid starting over
from scratch and looking up every type again by its ghash, we save the
insertion position of every hash table insertion during the first
insertion phase. Because the table does not support rehashing, the
insertion position is stable. Using the array of insertion positions
indexed by source type index, we can replace the source type indices in
the ghash table cells with the PDB type indices.

Once the table cells have been updated to contain PDB type indices, the
mapping for each type source can be computed in parallel. Simply iterate
the list of cell positions and replace them with the PDB type index,
since the insertion positions are no longer needed.

Once we have a source to destination type index mapping for every type
source, there are no more data dependencies. We know which type records
are "unique" (not duplicates), and what their final type indices will
be. We can do the remapping in parallel, and accumulate type sizes and
type hashes in parallel by type source.

Lastly, TPI stream layout must be done serially. Accumulate all the type
records, sizes, and hashes, and add them to the PDB.

Differential Revision: https://reviews.llvm.org/D87805

show more ...

Make llvm::StringRef to std::string conversions explicit.

This is how it should've been and brings it more in line with
std::string_view. There should be no functional change here.

This is mostly m

Make llvm::StringRef to std::string conversions explicit.

This is how it should've been and brings it more in line with
std::string_view. There should be no functional change here.

This is mostly mechanical from a custom clang-tidy check, with a lot of
manual fixups. It uncovers a lot of minor inefficiencies.

This doesn't actually modify StringRef yet, I'll do that in a follow-up.

show more ...

Update the file headers across all of the LLVM projects in the monorepo
to reflect the new license.

We understand that people may be surprised that we're moving the header
entirely to discuss the ne

Update the file headers across all of the LLVM projects in the monorepo
to reflect the new license.

We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.

Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.

llvm-svn: 351636

show more ...

[CodeView] Minimal support for S_UNAMESPACE records

Differential Revision: https://reviews.llvm.org/D50007

llvm-svn: 338417

[CodeView] Correctly compute the name of S_PROCREF symbols.

We have a function which switches on the type of a symbol record
to return a hardcoded offset into the record that contains the
symbol nam

[CodeView] Correctly compute the name of S_PROCREF symbols.

We have a function which switches on the type of a symbol record
to return a hardcoded offset into the record that contains the
symbol name. Not all symbols have names to begin with, and for
those records we return -1 for the offset.

Names are used for various things. Importantly for this particular
bug, a hash of the record name is used as a key for certain hash
tables which are serialied into the PDB file. One of these hash
tables is for the global symbol stream, which is basically a
collection of S_PROCREF symbols which contain the name of the
symbol, a module, and an address offset.

However, for S_PROCREF symbols, the function to return the offset
of the name was returning -1: basically it wasn't implemented.
As a result of this, all global symbols were hashing to the same
value, essentially it was as if every single global symbol's name
was the empty string.

This manifests in the VS debugger when you try to call a function
(global or member, doesn't matter) through the immediate window
and the debugger simply reports an error because it can't find the
function. This makes perfect sense, because it is hashing the name
for real, looking in the global symbol hash table, and there is only
1 entry there which corresponds to a symbol whose name is the empty
string.

Fixing this fixes the MSVC debugger in this case.

llvm-svn: 336024

show more ...

Fix line endings (CR/LF -> LF) introduced by rL329613

reviewer: zturner
llvm-svn: 329646

[Debuginfo][COFF] Minimal serialization support for precompiled types records

This change adds support for the LF_PRECOMP and LF_ENDPRECOMP records required
to read/write Microsoft precompiled types

[Debuginfo][COFF] Minimal serialization support for precompiled types records

This change adds support for the LF_PRECOMP and LF_ENDPRECOMP records required
to read/write Microsoft precompiled types .objs.
See https://en.wikipedia.org/wiki/Precompiled_header#Microsoft_Visual_C_and_C++

This also adds handling for the .debug$P section, which is actually a .debug$T
section in disguise, found only in precompiled .objs.

Differential Revision: https://reviews.llvm.org/D45283

llvm-svn: 329613

show more ...

[CodeView] Lower __restrict and other pointer qualifiers correctly

Qualifiers on a pointer or reference type may apply to either the
pointee or the pointer itself. Consider 'const char *' and 'char

[CodeView] Lower __restrict and other pointer qualifiers correctly

Qualifiers on a pointer or reference type may apply to either the
pointee or the pointer itself. Consider 'const char *' and 'char *
const'. In the first example, the pointee data may not be modified
without casts, and in the second example, the pointer may not be updated
to point to new data.

In the general case, qualifiers are applied to types with LF_MODIFIER
records, which support the usual const and volatile qualifiers as well
as the __unaligned extension qualifier.

However, LF_POINTER records, which are used for pointers, references,
and member pointers, have flags for qualifiers applying to the
*pointer*. In fact, this is the only way to represent the restrict
qualifier, which can only apply to pointers, and cannot qualify regular
data types.

This patch causes LLVM to correctly fold 'const' and 'volatile' pointer
qualifiers into the pointer record, as well as adding support for
'__restrict' qualifiers in the same place.

Based on a patch from Aaron Smith

Differential Revision: https://reviews.llvm.org/D43060

llvm-svn: 326260

show more ...

[codeview] Remove unused variable

llvm-svn: 326253

[LLD/PDB] Write actual records to the globals stream.

Previously we were writing an empty globals stream. Windows
tools interpret this as "private symbols are not present in
this PDB", even when th

[LLD/PDB] Write actual records to the globals stream.

Previously we were writing an empty globals stream. Windows
tools interpret this as "private symbols are not present in
this PDB", even when they are, so we need to fix this. Regardless,
without it we don't have information about global variables, so
we need to fix it anyway. This patch does that.

With this patch, the "lm" command in WinDbg correctly reports
that we have private symbols available, but the "dv" command
still refuses to display local variables.

Differential Revision: https://reviews.llvm.org/D36535

llvm-svn: 310743

show more ...