xusheng6 wrote:
> > or do we already have some infra that I can use?
>
> `lldb/test/API/functionalities/gdb_remote_client/TestStopPCs.py` is an
> example of that infrastructure. The `threadStopInfo` is where you would
> insert the swbreak hwbreak key.
>
> In fact, if i386 is an architecture w
https://github.com/xusheng6 updated
https://github.com/llvm/llvm-project/pull/102873
>From 73c98df4baef99f96d9c67113ba2ed0d972e5a04 Mon Sep 17 00:00:00 2001
From: Xusheng
Date: Mon, 20 Mar 2023 20:24:11 +0800
Subject: [PATCH 1/6] [lldb] Claim to support swbreak and hwbreak packets when
debuggi
@@ -0,0 +1,459 @@
+//===-- DILAST.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Ap
labath wrote:
It's not just this bot. I also found failures on
[aarch64](https://lab.llvm.org/buildbot/#/builders/59/builds/2999) and
[arm](https://lab.llvm.org/buildbot/#/builders/18/builds/2079). I don't see it
on the mac bots, but I don't know how to look further into the history for
thos
labath wrote:
> I guess what I was proposing was more about deleting code than adding it. We
> must always have at most _one_ index, but we have a lot of code allowing for
> the situation of two indices (manual & {apple, dwarf}).
Okay. I think I understand now what you meant -- which was to ha
clayborg wrote:
We should also document in the headerdoc this new `const SymbolContext *sc =
nullptr` argument and say something like:
```
/// \param[in] sc If the value is NULL, then all modules will be searched in
/// order. If the value is a valid pointer and if a module is specified in th
@@ -351,8 +351,8 @@ CommandObjectDisassemble::GetNameRanges(CommandReturnObject
&result) {
// Find functions matching the given name.
SymbolContextList sc_list;
- GetTarget().GetImages().FindFunctions(name, eFunctionNameTypeAuto,
-
@@ -1764,8 +1764,8 @@ lldb::SBSymbolContextList SBTarget::FindFunctions(const
char *name,
function_options.include_inlines = true;
FunctionNameType mask = static_cast(name_type_mask);
- target_sp->GetImages().FindFunctions(ConstString(name), mask,
-
@@ -295,7 +295,7 @@ class ModuleList {
/// \see Module::FindFunctions ()
void FindFunctions(const RegularExpression &name,
const ModuleFunctionSearchOptions &options,
- SymbolContextList &sc_list);
+ const SymbolC
@@ -285,7 +285,7 @@ class ModuleList {
/// \see Module::FindFunctions ()
void FindFunctions(ConstString name, lldb::FunctionNameType name_type_mask,
const ModuleFunctionSearchOptions &options,
- SymbolContextList &sc_list) const;
+
@@ -355,6 +355,7 @@ class ModuleList {
void FindSymbolsWithNameAndType(ConstString name,
lldb::SymbolType symbol_type,
+ const SymbolContext &sc,
SymbolContextList &sc_list)
https://github.com/clayborg edited
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/clayborg commented:
Thanks for trying this, and yes, this is exactly what I was thinking!
After seeing all of the code changes that this created, I suggested in
ModuleList.h that we add a the symbol context as a default parameter to the
functions as `const SymbolContext *sc
@@ -0,0 +1,459 @@
+//===-- DILAST.h *- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Ap
clayborg wrote:
> I agree that passing some sort of "hint" object to prioritize the search
> would be a more complete solution, but since this is @DmT021's first
> contribution to LLVM I think we should merge this change if there are no
> concerns with the current implementation, and if @DmT02
augusto2112 wrote:
I agree that passing some sort of "hint" object to prioritize the search would
be a more complete solution, but since this is @DmT021's first contribution to
LLVM I think we should merge this change if there are no concerns with the
current implementation, and if @DmT021 wan
https://github.com/Jlalond closed
https://github.com/llvm/llvm-project/pull/101272
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
https://github.com/clayborg edited
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/clayborg commented:
While this does solve the problem, it means other clients that want to do the
same thing will need to create this same kind of code.
I would propose we modify `ModuleList::FindSymbolsWithNameAndType(...)` to take
an extra parameter `SymbolContext &sc` whi
DmT021 wrote:
> Out of curiosity, could you provide some context around where/how this came
> up?
I just noticed this by accident in the `dwarf all` log when I was trying to
figure out why LLDB was evaluating expressions slowly in my project.
Here's a swift forum thread with some logs.
https:
https://github.com/s-barannikov closed
https://github.com/llvm/llvm-project/pull/102839
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
Author: Sergei Barannikov
Date: 2024-08-12T21:46:47+03:00
New Revision: e26b42c70d37cbea47016984e1d2bfc347fb8818
URL:
https://github.com/llvm/llvm-project/commit/e26b42c70d37cbea47016984e1d2bfc347fb8818
DIFF:
https://github.com/llvm/llvm-project/commit/e26b42c70d37cbea47016984e1d2bfc347fb8818.d
jeffreytan81 wrote:
@labath, unfortunately the failing log does not provide much information to
troubleshoot. Is it possible to gather more information of the failure? Like is
it only this `lldb-x86_64-debian` bot timeout or including other bots? And is
it consistently failing or occasionally
https://github.com/augusto2112 approved this pull request.
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/Michael137 edited
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/Michael137 edited
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/Michael137 commented:
Generally, makes sense to me. Out of curiosity, could you provide some context
around where/how this came up?
The ideal version of this would be a version of
`FindFunctions`/`FindSymbolsWithNameAndType` that took a `ModuleSP` hint (aka
where to search
https://github.com/clayborg closed
https://github.com/llvm/llvm-project/pull/102570
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
Author: Greg Clayton
Date: 2024-08-12T10:57:04-07:00
New Revision: f1e2886261281e788e4faae406c24e787c1dbdd0
URL:
https://github.com/llvm/llvm-project/commit/f1e2886261281e788e4faae406c24e787c1dbdd0
DIFF:
https://github.com/llvm/llvm-project/commit/f1e2886261281e788e4faae406c24e787c1dbdd0.diff
https://github.com/Jlalond edited
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Dmitrii Galimzianov (DmT021)
Changes
When we search for a symbol, we first check if it is in the module_sp of the
current SymbolContext, and if not, we check in the target's modules. However,
the target's ModuleList also includes the alrea
https://github.com/DmT021 ready_for_review
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/DmT021 updated
https://github.com/llvm/llvm-project/pull/102835
>From 45c91e403a36360727a5cd51e1a84f86027a77bc Mon Sep 17 00:00:00 2001
From: Dmitrii Galimzianov
Date: Mon, 12 Aug 2024 01:09:33 +0200
Subject: [PATCH] Remove redundant symbol lookups in
IRExecutionUnit::FindIn
@@ -258,6 +258,7 @@ class TargetProperties : public Properties {
bool GetDebugUtilityExpression() const;
+
clayborg wrote:
remove blank line
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mail
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -408,3 +410,23 @@ llvm::json::Value DebuggerStats::ReportStatistics(
return std::move(global_stats);
}
+
+llvm::json::Value SummaryStatistics::ToJSON() const {
+ json::Object body {{
+ {"invocationCount", GetSummaryCount()},
+ {"totalTime", GetTotalTime()},
+
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
@@ -918,3 +920,24 @@ def test_order_of_options_do_not_matter(self):
debug_stats_1,
f"The order of options '{options[0]}' and '{options[1]}'
should not matter",
)
+
+def test_summary_statistics_providers(self):
+"""
+
@@ -1,7 +1,13 @@
// Test that the lldb command `statistics` works.
+#include
+
+void foo() {
+ std::string str = "hello world";
clayborg wrote:
We want to test that this works for:
- built in C++ summary formatters (std::string)
- python summary formatters
- s
@@ -1,7 +1,13 @@
// Test that the lldb command `statistics` works.
+#include
+
+void foo() {
+ std::string str = "hello world";
clayborg wrote:
We should also test a type that uses a template as some summary providers match
based on a regex like:
```
std::vec
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
https://github.com/clayborg requested changes to this pull request.
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
clayborg wrote:
Do
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -258,6 +258,7 @@ class TargetProperties : public Properties {
bool GetDebugUtilityExpression() const;
+
bulbazord wrote:
nit: spurious newline
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits
@@ -3193,6 +3193,14 @@ bool Target::SetSectionUnloaded(const lldb::SectionSP
§ion_sp,
void Target::ClearAllLoadedSections() { m_section_load_history.Clear(); }
+lldb_private::SummaryStatistics&
Target::GetSummaryStatisticsForProvider(lldb_private::ConstString
summary_provi
https://github.com/bulbazord commented:
Makes sense to me and the change looks good. Feel free to move this out of
draft mode :)
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.l
@@ -799,20 +803,21 @@ IRExecutionUnit::FindInSymbols(const
std::vector &names,
sc_list);
if (auto load_addr = resolver.Resolve(sc_list))
return *load_addr;
-}
-if (sc.target_sp) {
- SymbolContextList sc_list;
-
augusto2112 wrote:
@DmT021 I think you can update this patch and set it ready for review then :)
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
https://github.com/bulbazord commented:
Thanks!
https://github.com/llvm/llvm-project/pull/102793
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
DmT021 wrote:
> Could you run `ninja check-lldb` from your build folder to make sure this
> change doesn't break anything?
Already did. All tests are ok.
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@list
clayborg wrote:
The main issue that this covers is when you have N .o files compiled with
.debug_names, and M .o files compiled without it, and then you link N+M into a
final executable. The linker will simply concatenate all of the .o files
section together, but the linker doesn't know anythi
@@ -785,6 +785,10 @@ IRExecutionUnit::FindInSymbols(const
std::vector &names,
return LLDB_INVALID_ADDRESS;
}
+ ModuleList images = target->GetImages();
DmT021 wrote:
It was already checked 4 lines above
```
if (!target) {
// We shouldn't be doin
@@ -785,6 +785,10 @@ IRExecutionUnit::FindInSymbols(const
std::vector &names,
return LLDB_INVALID_ADDRESS;
}
+ ModuleList images = target->GetImages();
augusto2112 wrote:
Oh, looks like target is already checked right before this, disregard this
comm
@@ -785,6 +785,10 @@ IRExecutionUnit::FindInSymbols(const
std::vector &names,
return LLDB_INVALID_ADDRESS;
}
+ ModuleList images = target->GetImages();
augusto2112 wrote:
Check if `target` is null before accessing it.
https://github.com/llvm/llvm-pro
@@ -799,20 +803,21 @@ IRExecutionUnit::FindInSymbols(const
std::vector &names,
sc_list);
if (auto load_addr = resolver.Resolve(sc_list))
return *load_addr;
-}
-if (sc.target_sp) {
- SymbolContextList sc_list;
-
https://github.com/augusto2112 edited
https://github.com/llvm/llvm-project/pull/102835
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/augusto2112 requested changes to this pull request.
I think overall this looks good! Just some small things to fix.
Could you run `ninja check-lldb` from your build folder to make sure this
change doesn't break anything?
https://github.com/llvm/llvm-project/pull/102835
_
felipepiovezan wrote:
> The current setup makes sense to me, but I guess that's expected as I'm the
> one who created it. I can also imagine something like what you propose, but
> it doesn't seem like a clear win to me. These objects are owned by
> SymbolFileDWARF, and we probably don't want t
labath wrote:
The current setup makes sense to me, but I guess that's expected as I'm the one
who created it. I can also imagine something like what you propose, but it
doesn't seem like a clear win to me. These objects are owned by
SymbolFileDWARF, and we probably don't want to have it do the
https://github.com/Jlalond edited
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
Jlalond wrote:
@labath those did get omitted due to github marking them as outdated due to
moving the code around, I mixed that. I appreciate your long review on this
one, I learned a lot from your feedback.
I'm going to let CI run and merge when it's completed
https://github.com/llvm/llvm-pr
Author: Pavel Labath
Date: 2024-08-12T18:43:16+02:00
New Revision: a0c57a0f3c6b44ce8f2c7222d0932df85883cf06
URL:
https://github.com/llvm/llvm-project/commit/a0c57a0f3c6b44ce8f2c7222d0932df85883cf06
DIFF:
https://github.com/llvm/llvm-project/commit/a0c57a0f3c6b44ce8f2c7222d0932df85883cf06.diff
Author: Pavel Labath
Date: 2024-08-12T18:41:28+02:00
New Revision: b3ed1d92112e0f455f8ef0888ef4c5d0ca29096d
URL:
https://github.com/llvm/llvm-project/commit/b3ed1d92112e0f455f8ef0888ef4c5d0ca29096d
DIFF:
https://github.com/llvm/llvm-project/commit/b3ed1d92112e0f455f8ef0888ef4c5d0ca29096d.diff
labath wrote:
This appears to be the same issue that David reported. I'm going to revert this
while I investigate. I doubt it's a compiler thing as the test does not
actually compile anything. It's more likely that something in the code is flaky
(and I think I have an idea what could that be).
Author: Pavel Labath
Date: 2024-08-12T18:36:18+02:00
New Revision: 38b67c54ed858f60c0caebcfba4b61f9326684ca
URL:
https://github.com/llvm/llvm-project/commit/38b67c54ed858f60c0caebcfba4b61f9326684ca
DIFF:
https://github.com/llvm/llvm-project/commit/38b67c54ed858f60c0caebcfba4b61f9326684ca.diff
vogelsgesang wrote:
@walter-erquinigo I took a first stab at implementing the DAP proposal. Seemed
rather straightforward. I would love to hear if I missed anything.
Also, I am not sure how to implement `valueLocation`, i.e., a the source
location referecend by the value. E.g., for a function
https://github.com/vogelsgesang created
https://github.com/llvm/llvm-project/pull/102928
`declarationLocation` is about to become part of the upstream debug adapter
protocol (see microsoft/debug-adapter-protocol#343). This is a draft
implementation, to be finalized and merged after the corresp
labath wrote:
> So the main idea is: if either ObjectFile or PostMortemProcess can return a
> reference to the data in the Peek calls, then should as long as they have
> this data mapped into the LLDB address space and can hand out a pointer to
> the data. If they can't, we should fall back to
labath wrote:
It just so happens that I got a `memory find` bug today, which made me realize
just how broken the current behavior is -- it will effectively abort the search
as soon as it runs into any unreadable block of memory (plus it also reads one
byte at a time, which makes it excruciatin
walter-erquinigo wrote:
> FYI @walter-erquinigo: There is a proposal under discussion to add
> first-class support for `declarationLocation` (and also `valueLocation`) to
> the debug adapter protocol. See
> [microsoft/debug-adapter-protocol#343](https://github.com/microsoft/debug-adapter-proto
felipepiovezan wrote:
> The fallback index will only index those units that aren't covered by the
> debug_names index, so if debug_names covers everything, the fallback is a noop
Got it, this is what I was hoping that would happen! Thanks for explaining it.
(future ideas) This makes me wonder
vogelsgesang wrote:
FYI @walter-erquinigo: There is a proposal under discussion to add first-class
support for `declarationLocation` (and also `valueLocation`) to the debug
adapter protocol. See
https://github.com/microsoft/debug-adapter-protocol/issues/343
https://github.com/llvm/llvm-projec
adrian-prantl wrote:
@labath This test fails in two configs on the matrix bot:
https://green.lab.llvm.org/job/llvm.org/view/LLDB/job/lldb-cmake-matrix/602/testReport/junit/lldb-shell/SymbolFile_DWARF_x86/Test_Clang_17_0_6___dwp_hash_collision_s/
https://green.lab.llvm.org/job/llvm.org/view/LLDB/j
Michael137 wrote:
Stats around this area could definitely be interesting to explore!
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
labath wrote:
Sort of yes (*), but only so much as it is necessary to make it correct, and
the level of slowdown depends on how much non-indexed code you have. The
fallback index will only index those units that aren't covered by the
debug_names index, so if debug_names covers everything, the
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
https://github.com/Michael137 approved this pull request.
Lifetimes look correct to me, thanks!
https://github.com/llvm/llvm-project/pull/102839
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/l
@@ -1,7 +1,13 @@
// Test that the lldb command `statistics` works.
+#include
+
+void foo() {
+ std::string str = "hello world";
Michael137 wrote:
instead of relying on `std::string` formatter should we add a type summary for
a dummy type? And test against tha
@@ -918,3 +920,24 @@ def test_order_of_options_do_not_matter(self):
debug_stats_1,
f"The order of options '{options[0]}' and '{options[1]}'
should not matter",
)
+
+def test_summary_statistics_providers(self):
+"""
+
felipepiovezan wrote:
Isn't this going to degrade the performance of _all_ negative queries?
I don't remember off the top of my head what is "m_fallback" when we have an
accelerator table, but this worries me.
https://github.com/llvm/llvm-project/pull/102123
https://github.com/Michael137 edited
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/Michael137 edited
https://github.com/llvm/llvm-project/pull/102708
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -25,6 +26,7 @@ namespace lldb_private {
using StatsClock = std::chrono::high_resolution_clock;
using StatsTimepoint = std::chrono::time_point;
+using Duration = std::chrono::duration;
Michael137 wrote:
Is this used anywhere?
https://github.com/llvm/llvm-p
@@ -615,7 +615,16 @@ bool ValueObject::GetSummaryAsCString(TypeSummaryImpl
*summary_ptr,
m_synthetic_value->UpdateValueIfNeeded(); // the summary might depend on
// the synthetic children being
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
@@ -174,6 +176,83 @@ struct StatisticsOptions {
std::optional m_include_transcript;
};
+/// A class that represents statistics about a TypeSummaryProviders invocations
+class SummaryStatistics {
+public:
+ SummaryStatistics() = default;
+ SummaryStatistics(lldb_private::Co
1 - 100 of 163 matches
Mail list logo
