date:20231115

[Lldb-commits] [lldb] Add support for arm64 registers in minidump core file saving. (PR #72315)

2023-11-15 Thread antoine moynault via lldb-commits


antmox wrote:

Thanks @clayborg, the buildbot is now green again : 
https://lab.llvm.org/buildbot/#/builders/219/builds/6947


https://github.com/llvm/llvm-project/pull/72315
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 

DavidSpickett wrote:

> This is new for me. Thank you!

New to me too actually. I only went looking for it because I was going to 
suggest a two stage format like `%%%ds` and I got that feeling that someone 
must have solved this problem before.

> In Stream.h, we tried to keep up with the file standard. Hence, the comments:

Brilliant, yes it was on my list to check for these comments in fact. Keep them 
for sure.

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -231,6 +231,18 @@ class Stream {
   /// The string to be output to the stream.
   size_t PutCString(llvm::StringRef cstr);
 
+  /// Output a C string to the stream with color highlighting.
+  ///
+  /// Print a C string \a text to the stream, applying color highlighting to
+  /// the specified \a pattern within the string.
+  ///
+  /// \param[in] text
+  /// The string to be output to the stream.
+  ///
+  /// \param[in] pattern
+  /// The portion of the \a text string to be colorized for highlighting.

DavidSpickett wrote:

It should be noted that this is:
* A regex pattern that is matched as many times as possible throughout the 
string.
* If it is nullptr, then no highlighting is done.
* The highlight colour is red.

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -163,7 +166,10 @@ bool SymbolContext::DumpStopContext(Stream *s, 
ExecutionContextScope *exe_scope,
   dumped_something = true;
   if (symbol->GetType() == eSymbolTypeTrampoline)
 s->PutCString("symbol stub for: ");
-  symbol->GetName().Dump(s);
+  if (pattern)
+Address::DumpName(s, symbol->GetName().GetStringRef(), pattern);
+  else
+symbol->GetName().Dump(s);

DavidSpickett wrote:

This one still needs fixing, can be one call to PutCString... I think.

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -231,6 +231,18 @@ class Stream {
   /// The string to be output to the stream.
   size_t PutCString(llvm::StringRef cstr);
 
+  /// Output a C string to the stream with color highlighting.
+  ///
+  /// Print a C string \a text to the stream, applying color highlighting to
+  /// the specified \a pattern within the string.

DavidSpickett wrote:

Maybe:
Print a C string \a text to the stream, applying color highlighting to the 
portions of the string matched by the regex pattern \a pattern.

Just to emphasise the regex nature of this.

(someone may argue to call this parameter `re_pattern` as well, but don't 
change that now, only if later reviewers feel it's needed)

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -70,6 +72,32 @@ size_t Stream::PutCString(llvm::StringRef str) {
   return bytes_written;
 }
 
+void Stream::PutCStringColorHighlighted(llvm::StringRef text,
+const char *pattern) {
+  if (!pattern) {
+PutCString(text.data());
+return;
+  }
+
+  // If pattern is not nullptr, we should use color
+  llvm::Regex reg_pattern(pattern);
+  llvm::SmallVector matches;
+  llvm::StringRef remaining = text;
+  std::string format_str = lldb_private::ansi::FormatAnsiTerminalCodes(
+  "${ansi.fg.red}%s${ansi.normal}");
+  size_t last_pos = 0;
+  while (reg_pattern.match(remaining, &matches)) {
+llvm::StringRef match = matches[0];
+size_t match_start_pos = match.data() - remaining.data();
+Write(remaining.data(), match_start_pos);
+Printf(format_str.c_str(), match.str().c_str());

DavidSpickett wrote:

Oh clever, by using the match instead of the text itself you don't need the 
`.*s` dance. Good catch!

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?Jos=C3=A9?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -70,6 +72,32 @@ size_t Stream::PutCString(llvm::StringRef str) {
   return bytes_written;
 }
 
+void Stream::PutCStringColorHighlighted(llvm::StringRef text,
+const char *pattern) {
+  if (!pattern) {
+PutCString(text.data());
+return;
+  }
+
+  // If pattern is not nullptr, we should use color
+  llvm::Regex reg_pattern(pattern);
+  llvm::SmallVector matches;
+  llvm::StringRef remaining = text;
+  std::string format_str = lldb_private::ansi::FormatAnsiTerminalCodes(
+  "${ansi.fg.red}%s${ansi.normal}");
+  size_t last_pos = 0;
+  while (reg_pattern.match(remaining, &matches)) {
+llvm::StringRef match = matches[0];
+size_t match_start_pos = match.data() - remaining.data();
+Write(remaining.data(), match_start_pos);
+Printf(format_str.c_str(), match.str().c_str());
+last_pos = match_start_pos + match.size();
+remaining = remaining.drop_front(last_pos);
+  }
+  if (remaining.size())
+PutCString(remaining.data());

DavidSpickett wrote:

Same here, don't need .data()

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior 
Message-ID:
In-Reply-To: 



@@ -70,6 +72,32 @@ size_t Stream::PutCString(llvm::StringRef str) {
   return bytes_written;
 }
 
+void Stream::PutCStringColorHighlighted(llvm::StringRef text,
+const char *pattern) {
+  if (!pattern) {
+PutCString(text.data());

DavidSpickett wrote:

`size_t PutCString(llvm::StringRef cstr);` so you don't need the .data();

What likely happens is that .data gives you const char* and StringRef has a 
constructor from const char *, so it ends up back as a string ref.

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] 35b10ac - [lldb][DWARFASTParserClang] DWARFv5: support DW_TAG_variable static data members declarations (#72236)

2023-11-15 Thread via lldb-commits


Author: Michael Buch
Date: 2023-11-15T11:01:52Z
New Revision: 35b10acf201ab873fab4423a37ae6dbe52591f6d

URL: 
https://github.com/llvm/llvm-project/commit/35b10acf201ab873fab4423a37ae6dbe52591f6d
DIFF: 
https://github.com/llvm/llvm-project/commit/35b10acf201ab873fab4423a37ae6dbe52591f6d.diff

LOG: [lldb][DWARFASTParserClang] DWARFv5: support DW_TAG_variable static data 
members declarations (#72236)

The accepted DWARFv5 issue 161118.1: "DW_TAG for C++ static data
members" specifies that static data member declaration be described by
DW_TAG_variable. Make sure we recognize such members.

Depends on:
* https://github.com/llvm/llvm-project/pull/72234
* https://github.com/llvm/llvm-project/pull/72235

Added: 


Modified: 
lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp

Removed: 




diff  --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp 
b/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
index f5628b2753da2a7..37efe70461977ad 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
@@ -144,7 +144,7 @@ static bool ShouldIgnoreArtificialField(llvm::StringRef 
FieldName) {
 
 std::optional
 DWARFASTParserClang::FindConstantOnVariableDefinition(DWARFDIE die) {
-  assert(die.Tag() == llvm::dwarf::DW_TAG_member);
+  assert(die.Tag() == DW_TAG_member || die.Tag() == DW_TAG_variable);
 
   auto *dwarf = die.GetDWARF();
   if (!dwarf)
@@ -2889,7 +2889,7 @@ void DWARFASTParserClang::CreateStaticMemberVariable(
 const DWARFDIE &die, const MemberAttributes &attrs,
 const lldb_private::CompilerType &class_clang_type) {
   Log *log = GetLog(DWARFLog::TypeCompletion | DWARFLog::Lookups);
-  assert(die.Tag() == DW_TAG_member);
+  assert(die.Tag() == DW_TAG_member || die.Tag() == DW_TAG_variable);
 
   Type *var_type = die.ResolveTypeUID(attrs.encoding_form.Reference());
 
@@ -2965,6 +2965,7 @@ void DWARFASTParserClang::ParseSingleMember(
   // data members is DW_AT_declaration, so we check it instead.
   // FIXME: Since DWARFv5, static data members are marked DW_AT_variable so we
   // can consistently detect them on both GCC and Clang without below 
heuristic.
+  // Remove this block if we ever drop DWARFv4 support.
   if (attrs.member_byte_offset == UINT32_MAX &&
   attrs.data_bit_offset == UINT64_MAX && attrs.is_declaration) {
 CreateStaticMemberVariable(die, attrs, class_clang_type);
@@ -3195,6 +3196,10 @@ bool DWARFASTParserClang::ParseChildMembers(
   }
   break;
 
+case DW_TAG_variable: {
+  const MemberAttributes attrs(die, parent_die, module_sp);
+  CreateStaticMemberVariable(die, attrs, class_clang_type);
+} break;
 case DW_TAG_member:
   ParseSingleMember(die, parent_die, class_clang_type,
 default_accessibility, layout_info, last_field_info);



___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] [lldb][DWARFASTParserClang] DWARFv5: support DW_TAG_variable static data members declarations (PR #72236)

2023-11-15 Thread Michael Buch via lldb-commits


https://github.com/Michael137 closed 
https://github.com/llvm/llvm-project/pull/72236
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

[Lldb-commits] [lldb] Colorize output when searching for symbols in lldb (PR #69422)

2023-11-15 Thread David Spickett via lldb-commits

=?utf-8?q?José?= L. Junior
Message-ID:
In-Reply-To:

DavidSpickett wrote:

Guess I'm gonna take the 100th comment here :)

This is almost ready for "proper" review (people that aren't me). So first
thing is to address the few minor comments I have from this latest review round.

Then, since there's so much discussion here already, you need to make it very
clear that this is ready for review. So update the PR title and description
with what you wish to be shown when the change is landed.

The title of the first commit `[lldb] ` is fine (and thank you for
including the tag!), just paste that into the PR title. For the body, describe
how you've implemented the change and, if you can find it, note the feature
request issue that originally started all this (if that's been lost it's fine,
this is a good feature regardless).

The description should also make it clear who worked on the patch. Given that
@taalhaataahir0102 opened the PR, it will be landed with their address (unless
changed), so you should add a co-author tag so that everyone gets the proper
credit. This also goes into the PR description.

See
https://docs.github.com/en/pull-requests/committing-changes-to-your-project/creating-and-editing-commits/creating-a-commit-with-multiple-authors.

Then take the opportunity to squash your commits and rebase them onto the
latest main branch. You will do fixup commits on top of this when other
reviewers request changes, then the final set of commits will be squashed and
the PR title/message is used as the commit message.

Once you've done that, take the PR out of draft mode and as soon as I see
that's done I'll add a comment with some context for reviewers and make sure it
gets the right labeling and such.

(I did consider making you open a new PR, but it is useful for reviewers to see
what we have already discussed so let's just stay with this one, if people
complain you can make a new one and blame me :) )

Thank you for sticking with this for as long as you have. Other reviewers will
have more comments I'm sure, so don't get disheartened by that, you've already
achieved a lot to get to this point.

https://github.com/llvm/llvm-project/pull/69422
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits

98 matches

Mail list logo