[APInt] improve initialization performance #106945
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
|
@llvm/pr-subscribers-llvm-support Author: Princeton Ferro (Prince781) ChangesFull diff: https://github.com/llvm/llvm-project/pull/106945.diff 1 Files Affected:
diff --git a/llvm/lib/Support/APInt.cpp b/llvm/lib/Support/APInt.cpp
index fe22e9ba04b6f5..b6f8a6abc9304a 100644
--- a/llvm/lib/Support/APInt.cpp
+++ b/llvm/lib/Support/APInt.cpp
@@ -31,14 +31,6 @@ using namespace llvm;
#define DEBUG_TYPE "apint"
-/// A utility function for allocating memory, checking for allocation failures,
-/// and ensuring the contents are zeroed.
-inline static uint64_t* getClearedMemory(unsigned numWords) {
- uint64_t *result = new uint64_t[numWords];
- memset(result, 0, numWords * sizeof(uint64_t));
- return result;
-}
-
/// A utility function for allocating memory and checking for allocation
/// failure. The content is not zeroed.
inline static uint64_t* getMemory(unsigned numWords) {
@@ -74,11 +66,14 @@ inline static unsigned getDigit(char cdigit, uint8_t radix) {
void APInt::initSlowCase(uint64_t val, bool isSigned) {
- U.pVal = getClearedMemory(getNumWords());
- U.pVal[0] = val;
- if (isSigned && int64_t(val) < 0)
- for (unsigned i = 1; i < getNumWords(); ++i)
- U.pVal[i] = WORDTYPE_MAX;
+ if (isSigned && int64_t(val) < 0) {
+ U.pVal = new WordType[getNumWords()];
+ U.pVal[0] = val;
+ memset(&U.pVal[1], 0xFF, sizeof(WordType) * (getNumWords() - 1));
+ } else {
+ U.pVal = new WordType[getNumWords()]();
+ U.pVal[0] = val;
+ }
clearUnusedBits();
}
@@ -93,7 +88,7 @@ void APInt::initFromArray(ArrayRef<uint64_t> bigVal) {
U.VAL = bigVal[0];
else {
// Get memory, cleared to 0
- U.pVal = getClearedMemory(getNumWords());
+ U.pVal = new WordType[getNumWords()]();
// Calculate the number of words to copy
unsigned words = std::min<unsigned>(bigVal.size(), getNumWords());
// Copy the words from bigVal to pVal
@@ -2105,7 +2100,7 @@ void APInt::fromString(unsigned numbits, StringRef str, uint8_t radix) {
if (isSingleWord())
U.VAL = 0;
else
- U.pVal = getClearedMemory(getNumWords());
+ U.pVal = new WordType[getNumWords()]();
// Figure out if we can shift instead of multiply
unsigned shift = (radix == 16 ? 4 : radix == 8 ? 3 : radix == 2 ? 1 : 0);
|
topperc
reviewed
Sep 2, 2024
nikic
previously requested changes
Sep 2, 2024
nikic
dismissed
their stale review
Did not notice the use of default initialization.
Prince781
force-pushed
the
dev/pferro/apint-initialization
branch
from
69c293b to
5743145
Compare
|
Small update to make default-initialization unsubtle. |
Prince781
force-pushed
the
dev/pferro/apint-initialization
branch
from
5743145 to
db5f904
Compare
|
Okay, have addressed all comments so far. |
Prince781
force-pushed
the
dev/pferro/apint-initialization
branch
2 times, most recently
from
f71e7d7 to
55df297
Compare
Prince781
changed the title
Prince781
force-pushed
the
dev/pferro/apint-initialization
branch
from
55df297 to
ce6e723
Compare
Prince781
changed the title
|
Updated again. Change is smaller. |
We improve initialization speed by avoiding unnecessary memsets and transforming a for-loop to a memset. The two cases handled: 1. When int64_t(val) < 0, zeroing out is redundant as the subsequent for-loop will initialize to val .. 0xFFFFF ..... Instead we should only create an uninitialized buffer, and transform the slow for-loop into a memset to initialize the higher words to 0xFF. 2. In the other case, first we create an uninitialized array (new int64_t[]) and then we zero it out with memset. But this can be combined in one operation with new int64_t[](), which default-initializes the array.
Prince781
force-pushed
the
dev/pferro/apint-initialization
branch
from
ce6e723 to
d88fc1b
Compare
|
@Prince781 Congratulations on having your first Pull Request (PR) merged into the LLVM Project! Your changes will be combined with recent changes from other authors, then tested by our build bots. If there is a problem with a build, you may receive a report in an email or a comment on this PR. Please check whether problems have been caused by your change specifically, as the builds can include changes from many authors. It is not uncommon for your change to be included in a build that fails due to someone else's changes, or infrastructure issues. How to do this, and the rest of the post-merge process, is covered in detail here. If your change does cause a problem, it may be reverted, or you can revert it yourself. This is a normal part of LLVM development. You can fix your changes and open a new PR to merge them again. If you don't get any reports, no action is required from you. Your changes are working as expected, well done! |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/73/builds/4929 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/72/builds/2934 Here is the relevant piece of the build log for the reference |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The purpose is to save an extra memset in both cases:
int64_t(val) < 0, zeroing out is redundant as the subsequent for-loop will initialize toval .. 0xFFFFF ..... Instead we should only create an uninitialized buffer, and transform the slow for-loop into a memset to initialize the higher words to0xFF.new int64_t[]) and then we zero it out withmemset. But this can be combined in one operation withnew int64_t[](), which default-initializes the array.On one example where use of APInt was heavy, this improved compile time by 1%.