| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
|
| |
Motivation: mitigate linker input size increase from swisstable optimizations.
Note: the changes in raw_hash_set.h are fixing build errors that happened when adding the explicit instantiations. The change in unchecked_deref is because set iterators have const reference access whereas map iterators have mutable reference access and the function is never actually called for sets (it's used in raw_hash_map) so it wasn't needed before. I'm not sure why the soo_slot/soo_iterator problems didn't cause compile errors earlier.
PiperOrigin-RevId: 615174043
Change-Id: Iac5eb2332a76e9b70021156fbb2b8def47a5391d
|
|
|
|
|
| |
PiperOrigin-RevId: 615160537
Change-Id: I29070c898104c55e6563eed0eef7397441bef1d7
|
|
|
|
|
| |
PiperOrigin-RevId: 615131303
Change-Id: I68fcbdd943594983c67f8e07810b05d5fa9a6f2e
|
|
|
|
|
| |
PiperOrigin-RevId: 615090942
Change-Id: I7cc20a0129dcfbbddedd9e6d816bb6234bff14b3
|
|
|
|
|
|
|
|
|
| |
GCC did not support <filesystem> until GCC8.
Fixes #1635
PiperOrigin-RevId: 615051227
Change-Id: If7cd5802ead40805b1ff1c3bdfc10ba6d2858ef0
|
|
|
|
|
| |
PiperOrigin-RevId: 615047295
Change-Id: I86e5de8543dc1ebee6d5a526846e9121b241b0fa
|
|
|
|
|
|
|
| |
BasicMocking.MocksNotTriggeredForIncorrectTypes test.
PiperOrigin-RevId: 615030645
Change-Id: Ia0ff3e649c6155c44d4772ce32fbcd4c5b273cc7
|
|
|
|
|
| |
PiperOrigin-RevId: 615017130
Change-Id: I73277de8ece31d6a35b47dbdb205b473324b74a2
|
|
|
|
|
| |
PiperOrigin-RevId: 614701769
Change-Id: I7c2143dd467e376eb4936ef894f3413bba681419
|
|
|
|
|
|
|
| |
std:: equivalents
PiperOrigin-RevId: 614687225
Change-Id: I07421db08ee9c221e561f42e3bf8345fb5321401
|
|
|
|
|
| |
PiperOrigin-RevId: 614670264
Change-Id: Id6b1cffa4e104a82234130977235f1eb1021968f
|
|
|
|
|
|
|
|
|
|
|
| |
Addition and subtraction operators std::array::iterator are defined only for
ptrdiff_t, which is signed, instead of size_t, which is unsigned. Therefore,
adding the index variable to ar.begin() will trigger -Wsign-conversion if
std::array::iterator is not a raw pointer because the index variable will be
implicitly converted from size_t (an unsigned type) to ptrdiff_t (a signed
type). To fix this, we explicitly static_cast index to a ptrdiff_t.
PiperOrigin-RevId: 613662928
Change-Id: I5e06c2261d7b8f167fae7bb6acece076257f8579
|
|
|
|
|
| |
PiperOrigin-RevId: 613590317
Change-Id: I69f095681102e5492916085ada0eed085a75765b
|
|
|
|
|
| |
PiperOrigin-RevId: 613326708
Change-Id: I6e5ca195f208b8da0d21d70b5a035bfdc64f866d
|
|
|
|
|
| |
PiperOrigin-RevId: 613305668
Change-Id: Ifc247f48ea476745eaaf0dd41dbdab8404a6cafb
|
|
|
|
|
|
|
|
|
| |
Details:
- We use the space for control/slots pointers as the inline buffer.
- We use a max inline capacity of 1 to make the implementation much simpler and to avoid having to randomize the iteration order for inline tables.
- For iteration of inline tables, we introduce the kSooControl buffer which just has 1 full control byte followed by 1 sentinel control byte so that incrementing yields an end() iterator. We don't access kSooControl during lookups - only iteration.
PiperOrigin-RevId: 613253492
Change-Id: Id98ff11842f8bef27ac7ed88138dc03b46ce4fa6
|
|
|
|
|
| |
PiperOrigin-RevId: 613141617
Change-Id: Ife68495d4a5951b734db7407fb99686cf45a5133
|
|
|
|
|
| |
PiperOrigin-RevId: 612560213
Change-Id: Id75dfd1222a0bed8ec72ce21e4a97b1d09fc9eaa
|
|
|
|
|
| |
PiperOrigin-RevId: 612509928
Change-Id: I90de2e6bd229bf5cf71a27e9c491bc2794e9265f
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Imported from GitHub PR https://github.com/abseil/abseil-cpp/pull/1632
Prior art for the `vector::erase` optimization:
https://github.com/AmadeusITGroup/amc/blob/efcb7be/include/amc/vectorcommon.hpp#L176-L180 https://github.com/bloomberg/bde/blob/e15f05be6/groups/bsl/bslalg/bslalg_arrayprimitives.h#L3787-L3799 https://github.com/facebook/folly/blob/d24bf04/folly/FBVector.h#L1254-L1262 https://github.com/qt/qtbase/blob/fbfee2d/src/corelib/tools/qarraydataops.h#L856-L861
Merge 6ce011079ccf945ae95434ce45ea6c5e3a088af8 into 55d28d4b3b82f9a47b3fa9b811b675a032820621
Merging this change closes #1632
COPYBARA_INTEGRATE_REVIEW=https://github.com/abseil/abseil-cpp/pull/1632 from Quuxplusone:trivial-erase 6ce011079ccf945ae95434ce45ea6c5e3a088af8
PiperOrigin-RevId: 612278964
Change-Id: I327ace8a38292b4610c6be031cc334e77c76fd35
|
|
|
|
|
| |
PiperOrigin-RevId: 612201313
Change-Id: Ia9e7f146f5e1ecaffcb15de694049b716db38d02
|
|
|
|
|
|
|
|
| |
* It is stated in prose that the methods on `absl::NoDestructor` can never return a null pointer
* Now this is annotated via nullability annotations as well
PiperOrigin-RevId: 611619075
Change-Id: I078a5628430b7b221c009a5dcb849efe30409c28
|
|
|
|
|
| |
PiperOrigin-RevId: 611321915
Change-Id: Ia63e3e2f66981d9e4f5b97f6df276b8195459ba2
|
|
|
|
|
|
|
| |
Instead of casting an int to the enum type where the int does not have an associated enum value, cast the enum to its underlying type. This should be no functional change but make some linters happier.
PiperOrigin-RevId: 611172311
Change-Id: I9ae10f8fa2029014236f60a90ee2ab2273c66fa5
|
|
|
|
|
| |
PiperOrigin-RevId: 611151721
Change-Id: Ic7b342770fb3e1aab8e54ff125f1529534eedc3c
|
|
|
|
|
|
|
|
| |
in range of range-based for loop
Fixes: #1629
PiperOrigin-RevId: 611131201
Change-Id: I787731e00207b544ee16055e6e0d323a5094a433
|
|
|
|
|
|
|
| |
Breaks unit tests
PiperOrigin-RevId: 611077608
Change-Id: Ie024c299af75407e9798309f8cbf76abf60f9ee2
|
|
|
|
|
|
|
|
| |
Having DLOG(FATAL) be [[noreturn]] in debug builds makes dead-code
warnings impossible to satisfy between the release and debug build.
PiperOrigin-RevId: 610851706
Change-Id: I07104d6687e2b1a8472ee3ea876d5fd74a70574e
|
|
|
|
|
|
|
| |
comparisons in `IDivFastPath`.
PiperOrigin-RevId: 610758911
Change-Id: I3d0fa2f52a3bd75bdd2b5c365d79878b4160bd29
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
`time_internal::IDivDuration`, by splitting the slow path to a separate function.
With that change, the compiler can inline the fast path. This is specially important in the context of `Duration::operator%=`, because it allows proving that the return value is unused, therefore avoiding expensive multiplies and divides (e.g. `*q = num_hi / den_hi;`).
```
name old cpu/op new cpu/op delta
BM_Duration_Modulo 23.1ns ± 0% 22.5ns ± 0% -2.42% (p=0.000 n=20+16)
BM_Duration_Modulo_FastPath 7.05ns ± 0% 4.85ns ± 0% -31.17% (p=0.000 n=20+20)
name old time/op new time/op delta
BM_Duration_Modulo 23.1ns ± 0% 22.6ns ± 0% -2.43% (p=0.000 n=20+16)
BM_Duration_Modulo_FastPath 7.06ns ± 0% 4.86ns ± 0% -31.18% (p=0.000 n=20+20)
name old INSTRUCTIONS/op new INSTRUCTIONS/op delta
BM_Duration_Modulo 188 ± 0% 178 ± 0% -5.32% (p=0.000 n=20+20)
BM_Duration_Modulo_FastPath 84.0 ± 0% 62.0 ± 0% -26.19% (p=0.000 n=20+20)
name old CYCLES/op new CYCLES/op delta
BM_Duration_Modulo 73.8 ± 0% 72.1 ± 0% -2.27% (p=0.000 n=19+20)
BM_Duration_Modulo_FastPath 22.5 ± 0% 15.5 ± 0% -31.13% (p=0.000 n=19+20)
```
Note: We don't need to expose `absl::time_internal::IDivDuration` at all given that we have a public `absl::IDivDuration`.
PiperOrigin-RevId: 610710635
Change-Id: Ief7c3d5b1c000b397d931e9249edcaef96e7151e
|
|
|
|
|
| |
PiperOrigin-RevId: 610672905
Change-Id: I8e4ae98bb6e6421540a3073e3e6cb7de60fd32a9
|
|
|
|
|
| |
PiperOrigin-RevId: 610398878
Change-Id: Iec7570af855b61ad49faba1d29fdf41674a9bd50
|
|
|
|
|
|
|
| |
The current version violates the Google C++ style guide - see https://google.github.io/styleguide/cppguide.html#Formatting_Looping_Branching. This code does not fit into the historical exception that "the curly braces for the controlled statement or the line breaks inside the curly braces may be omitted if as a result the entire statement appears on either a single line (in which case there is a space between the closing parenthesis and the controlled statement) or on two lines (in which case there is a line break after the closing parenthesis and there are no braces)."
PiperOrigin-RevId: 609789188
Change-Id: Id7ae9596e454dac5581d19939564c07670077f92
|
|
|
|
|
|
|
| |
unnecessary probing before resize that is beneficial for small tables the most.
PiperOrigin-RevId: 609547787
Change-Id: If6584919b4c93945ea078b1c1a9f57b355dce924
|
|
|
|
|
|
|
| |
This resolves a couple of Clang Tidy performance warnings.
PiperOrigin-RevId: 609394317
Change-Id: Ibebc7e3f7121355b8660284e18c110bb9171d61c
|
|
|
|
|
| |
PiperOrigin-RevId: 609351346
Change-Id: I40f44d42efe65fd8ca1a441e69a0769c87938b20
|
|
|
|
|
| |
PiperOrigin-RevId: 609064443
Change-Id: Ifaeb53ae2d50bcef1ef6f67e662173b59c3fb8a2
|
|
|
|
|
|
|
| |
return type of insert.
PiperOrigin-RevId: 609058024
Change-Id: I2f7cc2daf862e7e2d23acd6dd3fe85cb1945d5f0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Imported from GitHub PR https://github.com/abseil/abseil-cpp/pull/1618
I noticed while working on #1615 that `inlined_vector` could use the trivial relocatability trait here, too.
Here the memcpy codepath already exists; we just have to opt in to using it.
Merge 567a1dd9b6b3352f649e900b24834b59e39cfa14 into a7012a5bfcf26a41b9dd32d4c429004773503dd6
Merging this change closes #1618
COPYBARA_INTEGRATE_REVIEW=https://github.com/abseil/abseil-cpp/pull/1618 from Quuxplusone:trivial-swap 567a1dd9b6b3352f649e900b24834b59e39cfa14
PiperOrigin-RevId: 609019296
Change-Id: I4055ab790245752179e405b490fcd479e7389726
|
|
|
|
|
| |
PiperOrigin-RevId: 608947694
Change-Id: Ie53a91c4d78dcb80c57227616b488ec64b23c588
|
|
|
|
|
| |
PiperOrigin-RevId: 608770171
Change-Id: Icca54086037e42826c272f04374aeb33d060ace5
|
|
|
|
|
|
|
|
| |
raw_hash_set::find(key).
Motivation: the implementation becomes more complicated with small object optimization.
PiperOrigin-RevId: 608742838
Change-Id: I55fc42321b1967f9c7bbee49817a2f2d4ee44b56
|
|
|
|
|
| |
PiperOrigin-RevId: 608705933
Change-Id: I9c11d8d8b3d9057bdf215863210f450332cf56cb
|
|
|
|
|
| |
PiperOrigin-RevId: 608661989
Change-Id: Ibfd94f8b2d23fd232bf93904ed68e11a400b3644
|
|
|
|
|
|
|
| |
This often indicates a bug from adding synchronization logic but not using it.
PiperOrigin-RevId: 608600311
Change-Id: I48bc3ae18b1e9a00b5e2152f33b486cd6be324a1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Imported from GitHub PR https://github.com/abseil/abseil-cpp/pull/1625
Trivial relocatability also requires that the type not do anything weird with its assignment operator; update the type-trait to reflect this. (This is the definition used by BSL, Folly, HPX, Thrust, Parlay, Amadeus, and P1144.)
This is important if we want to use `absl::is_trivially_relocatable` as a gate for memcpy optimizations in `inlined_vector::erase` and/or `inlined_vector::swap`, because in those cases relocation is used to replace part of a sequence involving assignment; the optimization requires an assignment operator that behaves value-semantically.
Clang's builtin currently fails to check the assignment operator, so we stop using it entirely for now. We already refused to use it on Win32, Win64, and Apple, for various unrelated reasons. I'm working on giving Clang's builtin the behavior that would let us re-enable it here.
Assume that any compiler providing both `__cpp_impl_trivially_relocatable` and a builtin `__is_trivially_relocatable(T)` will use the appropriate (P1144) definition for its builtin. Right now there's only one such compiler (the P1144 reference implementation, which forks Clang), so this is largely a moot point, but I'm being optimistic.
Merge d943abdbabc1b7080aa5f0a2fff3e724135164dc into 34604d5b1f6ae14c65b3992478b59f7108051979
Merging this change closes #1625
COPYBARA_INTEGRATE_REVIEW=https://github.com/abseil/abseil-cpp/pull/1625 from Quuxplusone:trivially-relocatable d943abdbabc1b7080aa5f0a2fff3e724135164dc
PiperOrigin-RevId: 607977323
Change-Id: I6436a60326c6d1064bdd71ec2e15b86b7a29efd4
|
|
|
|
|
|
|
|
| |
It can be used instead of `MaskEmptyOrDeleted` in case of inserting to empty table.
`MaskNonFull` requires less operations, in particular it eliminates `_mm_set1_epi8` and `_mm_cmpgt_epi8` operations.
PiperOrigin-RevId: 607587394
Change-Id: Ia48f922d1ca6de38cc91e7ab0d608c45f8f2c446
|
|
|
|
|
|
|
|
|
| |
substitutions
These were described in https://github.com/itanium-cxx-abi/cxx-abi/pull/85 and implemented by LLVM.
PiperOrigin-RevId: 607555558
Change-Id: I9991ac88c1fcf63b25b93d93977a83ca343cdb5d
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Imported from GitHub PR https://github.com/abseil/abseil-cpp/pull/1617
The intrinsics used aren't available on `x86_64` processors while running in 32-bit mode. See:
- list of 64-bit intrinsics (https://learn.microsoft.com/en-us/cpp/intrinsics/x64-amd64-intrinsics-list?view=msvc-170)
- list of 32-bit intrinsics (https://learn.microsoft.com/en-us/cpp/intrinsics/x86-intrinsics-list?view=msvc-170)
- list of predefined MSVC macros (https://learn.microsoft.com/en-us/cpp/preprocessor/predefined-macros?view=msvc-170)
The error message in question:
```console
F:\dev\opentrack-depends\onnxruntime-build\msvc\_deps\abseil_cpp-src\absl/crc/internal/crc32_x86_arm_combined_simd.h(145,32): error C3861: '_mm_crc32_u64': identifier not found
return static_cast<uint32_t>(_mm_crc32_u64(crc, v));
^
F:\dev\opentrack-depends\onnxruntime-build\msvc\_deps\abseil_cpp-src\absl/crc/internal/crc32_x86_arm_combined_simd.h(193,50): error C3861: '_mm_cvtsi128_si64': identifier not found
inline int64_t V128_Low64(const V128 l) { return _mm_cvtsi128_si64(l); }
```
Merge 06f5832108a2b01e0a900db51e1c870f7069a1f2 into 797501d12ea767dabdc8d36674e083869e62ee7d
Merging this change closes #1617
COPYBARA_INTEGRATE_REVIEW=https://github.com/abseil/abseil-cpp/pull/1617 from sthalik:pr/fix-msvc-32-bit-avx 06f5832108a2b01e0a900db51e1c870f7069a1f2
PiperOrigin-RevId: 607483370
Change-Id: Id2a6f6dd33c2707fe7ffe134e7335916f3fb9da3
|
|
|
|
|
| |
PiperOrigin-RevId: 607095897
Change-Id: I077857f0b4c5cf8bbc62bfc8e51a20bb6845296c
|