[ThinLTO] Add HasLocal flag to GlobalValueSummaryInfo #164647
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
Add a flag to the GlobalValueSummaryInfo indicating whether the associated SummaryList (all summaries with the same GUID) contains any summaries with local linkage. This flag is set when building the index, so it is associated with the original linkage type before internalization and promotion. Consumers should check the withInternalizeAndPromote() flag on the index before using it. In most cases we expect a 1-1 mapping between a GUID and a summary with local linkage, because for locals the GUID is computed from the hash of "modulepath:name". However, there can be multiple locals with the same GUID if translation units are not compiled with enough path. And in rare but theoretically possible cases, there can be hash collisions on the underlying MD5 computation. So to be safe when looking for local summaries, analyses currently look through all summaries in the list. These lists can be extremely long in the case of large binaries with template function defs in widely used headers (i.e. linkonce_odr). A follow on change will use this flag to reduce ThinLTO analysis time in WPD by 5-6% for a large target (details in PR164046 which will be reworked to use this flag). Note that in the past we have tried to keep bits related to the GUID in the ValueInfo (which has a pointer to the associated GlobalValueSummaryInfo), via its PointerIntPair. However, we are out of bits there. This change does add a byte to every GlobalValueSummaryInfo instance, which I measured as a little under 0.90% overhead in a large target. However, it enables adding 7 bits of other per-GUID flags in the future without adding more overhead. Note that it was lower overhead to add this to the GlobalValueSummaryInfo than the ValueInfo, which tends to be copied into other maps.
|
@llvm/pr-subscribers-lto @llvm/pr-subscribers-llvm-ir Author: Teresa Johnson (teresajohnson) ChangesAdd a flag to the GlobalValueSummaryInfo indicating whether the In most cases we expect a 1-1 mapping between a GUID and a summary with A follow on change will use this flag to reduce ThinLTO analysis time in Note that in the past we have tried to keep bits related to the GUID in Full diff: https://github.com/llvm/llvm-project/pull/164647.diff 2 Files Affected:
diff --git a/llvm/include/llvm/IR/ModuleSummaryIndex.h b/llvm/include/llvm/IR/ModuleSummaryIndex.h
index 98df06ae43ad0..fb1f5c3bc0440 100644
--- a/llvm/include/llvm/IR/ModuleSummaryIndex.h
+++ b/llvm/include/llvm/IR/ModuleSummaryIndex.h
@@ -172,9 +172,11 @@ struct alignas(8) GlobalValueSummaryInfo {
/// Add a summary corresponding to a global value definition in a module with
/// the corresponding GUID.
- void addSummary(std::unique_ptr<GlobalValueSummary> Summary) {
- return SummaryList.push_back(std::move(Summary));
- }
+ inline void addSummary(std::unique_ptr<GlobalValueSummary> Summary);
+
+ /// Verify that the HasLocal flag is consistent with the SummaryList. Should
+ /// only be called prior to index-based internalization and promotion.
+ inline void verifyLocal() const;
private:
/// List of global value summary structures for a particular value held
@@ -183,6 +185,22 @@ struct alignas(8) GlobalValueSummaryInfo {
/// compiling without sufficient distinguishing path, or (theoretically) hash
/// collisions. Each summary is from a different module.
GlobalValueSummaryList SummaryList;
+
+ /// True if the SummaryList contains at least one summary with local linkage.
+ /// In most cases there should be only one, unless translation units with
+ /// same-named locals were compiled without distinguishing path. And generally
+ /// there should not be a mix of local and non-local summaries, because the
+ /// GUID for a local is computed with the path prepended and a ':' delimiter.
+ /// In extremely rare cases there could be a GUID hash collision. Having the
+ /// flag saves having to walk through all summaries to prove the existence or
+ /// not of any locals.
+ /// NOTE: this flag is set when the index is built. It does not reflect
+ /// index-based internalization and promotion decisions. Generally most
+ /// index-based analysis occurs before then, but any users should assert that
+ /// the withInternalizeAndPromote() flag is not set on the index.
+ /// TODO: Replace checks in various ThinLTO analyses that loop through all
+ /// summaries to handle the local case with a check of the flag.
+ bool HasLocal : 1;
};
/// Map from global value GUID to corresponding summary structures. Use a
@@ -219,6 +237,8 @@ struct ValueInfo {
return getRef()->second.getSummaryList();
}
+ void verifyLocal() const { getRef()->second.verifyLocal(); }
+
// Even if the index is built with GVs available, we may not have one for
// summary entries synthesized for profiled indirect call targets.
bool hasName() const { return !haveGVs() || getValue(); }
@@ -649,7 +669,23 @@ class GlobalValueSummary {
friend class ModuleSummaryIndex;
};
-GlobalValueSummaryInfo::GlobalValueSummaryInfo(bool HaveGVs) : U(HaveGVs) {}
+GlobalValueSummaryInfo::GlobalValueSummaryInfo(bool HaveGVs)
+ : U(HaveGVs), HasLocal(false) {}
+
+void GlobalValueSummaryInfo::addSummary(
+ std::unique_ptr<GlobalValueSummary> Summary) {
+ if (GlobalValue::isLocalLinkage(Summary->linkage()))
+ HasLocal = true;
+ return SummaryList.push_back(std::move(Summary));
+}
+
+void GlobalValueSummaryInfo::verifyLocal() const {
+ assert(HasLocal ==
+ llvm::any_of(SummaryList,
+ [](const std::unique_ptr<GlobalValueSummary> &Summary) {
+ return GlobalValue::isLocalLinkage(Summary->linkage());
+ }));
+}
/// Alias summary information.
class AliasSummary : public GlobalValueSummary {
diff --git a/llvm/lib/LTO/LTO.cpp b/llvm/lib/LTO/LTO.cpp
index 72ae064e46a0c..86780e19f22b8 100644
--- a/llvm/lib/LTO/LTO.cpp
+++ b/llvm/lib/LTO/LTO.cpp
@@ -477,6 +477,10 @@ static void thinLTOInternalizeAndPromoteGUID(
return !GlobalValue::isLocalLinkage(Summary->linkage());
});
+ // Before performing index-based internalization and promotion for this GUID,
+ // the local flag should be consistent with the summary list linkage types.
+ VI.verifyLocal();
+
for (auto &S : VI.getSummaryList()) {
// First see if we need to promote an internal value because it is not
// exported.
|
mingmingl-llvm
approved these changes
Oct 22, 2025
| /// In most cases there should be only one, unless translation units with | ||
| /// same-named locals were compiled without distinguishing path. And generally | ||
| /// there should not be a mix of local and non-local summaries, because the | ||
| /// GUID for a local is computed with the path prepended and a ':' delimiter. |
There was a problem hiding this comment.
There was a problem hiding this comment.
Thanks, forgot about that change. I'll update the comment and PR description accordingly
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/66/builds/21126 Here is the relevant piece of the build log for the reference |
dvbuka
pushed a commit
to dvbuka/llvm-project
that referenced
this pull request
Oct 27, 2025
Add a flag to the GlobalValueSummaryInfo indicating whether the associated SummaryList (all summaries with the same GUID) contains any summaries with local linkage. This flag is set when building the index, so it is associated with the original linkage type before internalization and promotion. Consumers should check the withInternalizeAndPromote() flag on the index before using it. In most cases we expect a 1-1 mapping between a GUID and a summary with local linkage, because for locals the GUID is computed from the hash of "modulepath;name". However, there can be multiple locals with the same GUID if translation units are not compiled with enough path. And in rare but theoretically possible cases, there can be hash collisions on the underlying MD5 computation. So to be safe when looking for local summaries, analyses currently look through all summaries in the list. These lists can be extremely long in the case of large binaries with template function defs in widely used headers (i.e. linkonce_odr). A follow on change will use this flag to reduce ThinLTO analysis time in WPD by 5-6% for a large target (details in PR164046 which will be reworked to use this flag). Note that in the past we have tried to keep bits related to the GUID in the ValueInfo (which has a pointer to the associated GlobalValueSummaryInfo), via its PointerIntPair. However, we are out of bits there. This change does add a byte to every GlobalValueSummaryInfo instance, which I measured as a little under 0.90% overhead in a large target. However, it enables adding 7 bits of other per-GUID flags in the future without adding more overhead. Note that it was lower overhead to add this to the GlobalValueSummaryInfo than the ValueInfo, which tends to be copied into other maps.
Lukacma
pushed a commit
to Lukacma/llvm-project
that referenced
this pull request
Oct 29, 2025
Add a flag to the GlobalValueSummaryInfo indicating whether the associated SummaryList (all summaries with the same GUID) contains any summaries with local linkage. This flag is set when building the index, so it is associated with the original linkage type before internalization and promotion. Consumers should check the withInternalizeAndPromote() flag on the index before using it. In most cases we expect a 1-1 mapping between a GUID and a summary with local linkage, because for locals the GUID is computed from the hash of "modulepath;name". However, there can be multiple locals with the same GUID if translation units are not compiled with enough path. And in rare but theoretically possible cases, there can be hash collisions on the underlying MD5 computation. So to be safe when looking for local summaries, analyses currently look through all summaries in the list. These lists can be extremely long in the case of large binaries with template function defs in widely used headers (i.e. linkonce_odr). A follow on change will use this flag to reduce ThinLTO analysis time in WPD by 5-6% for a large target (details in PR164046 which will be reworked to use this flag). Note that in the past we have tried to keep bits related to the GUID in the ValueInfo (which has a pointer to the associated GlobalValueSummaryInfo), via its PointerIntPair. However, we are out of bits there. This change does add a byte to every GlobalValueSummaryInfo instance, which I measured as a little under 0.90% overhead in a large target. However, it enables adding 7 bits of other per-GUID flags in the future without adding more overhead. Note that it was lower overhead to add this to the GlobalValueSummaryInfo than the ValueInfo, which tends to be copied into other maps.
aokblast
pushed a commit
to aokblast/llvm-project
that referenced
this pull request
Oct 30, 2025
Add a flag to the GlobalValueSummaryInfo indicating whether the associated SummaryList (all summaries with the same GUID) contains any summaries with local linkage. This flag is set when building the index, so it is associated with the original linkage type before internalization and promotion. Consumers should check the withInternalizeAndPromote() flag on the index before using it. In most cases we expect a 1-1 mapping between a GUID and a summary with local linkage, because for locals the GUID is computed from the hash of "modulepath;name". However, there can be multiple locals with the same GUID if translation units are not compiled with enough path. And in rare but theoretically possible cases, there can be hash collisions on the underlying MD5 computation. So to be safe when looking for local summaries, analyses currently look through all summaries in the list. These lists can be extremely long in the case of large binaries with template function defs in widely used headers (i.e. linkonce_odr). A follow on change will use this flag to reduce ThinLTO analysis time in WPD by 5-6% for a large target (details in PR164046 which will be reworked to use this flag). Note that in the past we have tried to keep bits related to the GUID in the ValueInfo (which has a pointer to the associated GlobalValueSummaryInfo), via its PointerIntPair. However, we are out of bits there. This change does add a byte to every GlobalValueSummaryInfo instance, which I measured as a little under 0.90% overhead in a large target. However, it enables adding 7 bits of other per-GUID flags in the future without adding more overhead. Note that it was lower overhead to add this to the GlobalValueSummaryInfo than the ValueInfo, which tends to be copied into other maps.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Add a flag to the GlobalValueSummaryInfo indicating whether the
associated SummaryList (all summaries with the same GUID) contains any
summaries with local linkage. This flag is set when building the index,
so it is associated with the original linkage type before
internalization and promotion. Consumers should check the
withInternalizeAndPromote() flag on the index before using it.
In most cases we expect a 1-1 mapping between a GUID and a summary with
local linkage, because for locals the GUID is computed from the hash of
"modulepath;name". However, there can be multiple locals with the same
GUID if translation units are not compiled with enough path. And in rare
but theoretically possible cases, there can be hash collisions on the
underlying MD5 computation. So to be safe when looking for local
summaries, analyses currently look through all summaries in the list.
These lists can be extremely long in the case of large binaries with
template function defs in widely used headers (i.e. linkonce_odr).
A follow on change will use this flag to reduce ThinLTO analysis time in
WPD by 5-6% for a large target (details in PR164046 which will be
reworked to use this flag).
Note that in the past we have tried to keep bits related to the GUID in
the ValueInfo (which has a pointer to the associated
GlobalValueSummaryInfo), via its PointerIntPair. However, we are out of
bits there. This change does add a byte to every GlobalValueSummaryInfo
instance, which I measured as a little under 0.90% overhead in a large
target. However, it enables adding 7 bits of other per-GUID flags in the
future without adding more overhead. Note that it was lower overhead to
add this to the GlobalValueSummaryInfo than the ValueInfo, which tends
to be copied into other maps.