Fixed an issue where our new accelerator tables could cause a crash
when we got a full 32 bit hash match, yet a C string mismatch.
We had a member variable in DWARFMappedHash::Prologue named
"min_hash_data_byte_size" the would compute the byte size of HashData
so we could skip hash data efficiently. It started out with a byte size
value of 4. When we read the table in from disk, we would clear the
atom array and read it from disk, and the byte size would still be set
to 4. We would then, as we read each atom from disk, increment this count.
So the byte size of the HashData was off, which means when we get a lookup
whose 32 bit hash does matches, but the C string does NOT match (which is
very very rare), then we try and skip the data for that hash and we would
add an incorrect offset and get off in our parsing of the hash data and
cause this crash.
To fix this I added a few safeguards:
1 - I now correctly clear the hash data size when we reset the atom array using the new DWARFMappedHash::Prologue::ClearAtoms() function.
2 - I now correctly always let the AppendAtom() calculate the byte size of the hash (before we were doing things manually some times, which was correct, but not good)
3 - I also track if the size of each HashData is a fixed byte size or not, and "do the right thing" when we need to skip the data.
4 - If we do get off in the weeds, then I make sure to return an error and stop any further parsing from happening.
llvm-svn: 147334
LLDB (python bindings) Crashing in lldb::SBDebugger::DeleteTarget(lldb::SBTarget&)
Need to check the validity of (SBTarget&)target passed to SBDebugger::DeleteTarget()
before calling target->Destroy().
llvm-svn: 147213
Switch from GetReturnValue, which was hardly ever used, to GetReturnValueObject
which is much more convenient.
Return the "return value object" as a persistent variable if requested.
llvm-svn: 147157
parser has hitherto been an implementation waiting
for a use. I have now tied the '-o' option for
the expression command -- which indicates that the
result is an Objective-C object and needs to be
printed -- to the ExpressionParser, which
communicates the desired type to Clang.
Now, if the result of an expression is determined
by an Objective-C method call for which there is
no type information, that result is implicitly
cast to id if and only if the -o option is passed
to the expression command. (Otherwise if there
is no explicit cast Clang will issue an error.
This behavior is identical to what happened before
r146756.)
Also added a testcase for -o enabled and disabled.
llvm-svn: 147099
with incomplete definition data were being converted.
Now Clang attempts to complete RecordDecls before
converting them, avoiding a nasty crash.
llvm-svn: 147029
types that have been imported multiple times.
The discussion below uses this diagram:
ASTContext A B C
Decl Da Db Dc
ASTImporter \-Iab-/\-Iac-/
\-----Iac----/
When a Decl D is imported from ASTContext A to
ASTContext B, the ASTImporter Iab records the
pair <Da, Db> in a DenseMap. That way, if Iab
ever encounters Da again (for example, as the
DeclContext for another Decl), it can use the
imported version. This is not an optimization,
it is critical: if I import the field "st_dev"
as part of importing "struct stat," the field
must have DeclContext equal to the parent
structure or we end up with multiple different
Decls containing different parts of "struct
stat." "struct stat" is imported once and
recorded in the DenseMap; then the ASTImporter
finds that same version when looking for the
DeclContext of "st_dev."
The bug arises when Db is imported into another
ASTContext C and ASTContext B goes away. This
often occurs when LLDB produces result variables
for expressions. Ibc is aware of the transport
of Db to Dc, but a brand new ASTImporter, Iac,
is responsible for completing Dc from its source
upon request. That ASTImporter has no mappings,
so it will produce a clone of Dc when attempting
to import its children. That means that type
completion operations on Dc will fail.
The solution is to create Iac as soon as Ibc
imports D from B to C, and inform Iac of the
mapping between Da and Dc. This allows type
completion to happen correctly.
llvm-svn: 147016
rdar://problem/10577182
Audit lldb API impl for places where we need to perform a NULL check
Add a NULL check for SBValue.CreateValueFromExpression().
llvm-svn: 146954
rdar://problem/10577182
Audit lldb API impl for places where we need to perform a NULL check
Add a NULL check for SBTarget.AttachToProcessWithName() so it will not hang.
llvm-svn: 146948
rdar://problem/10577182
Audit lldb API impl for places where we need to perform a NULL check
Add NULL checks for SBCommandReturnObject.AppendMessage().
llvm-svn: 146911
rdar://problem/10577182
Audit lldb API impl for places where we need to perform a NULL check
Add NULL checks for SBCommandInterpreter APIs.
llvm-svn: 146909
rdar://problem/10577182
Audit lldb API impl for places where we need to perform a NULL check
Add NULL checks for SBModule and SBSection APIs.
llvm-svn: 146899
"id" from being found by the parser as an
externally-defined type. Before, "id" would
sometimes make it through if it was defined in
a namespace, but this sometimes caused
confusion, for example when it conflicted with
std::locale::id.
llvm-svn: 146891
as part of the thread format output.
Currently this is only done for the ThreadPlanStepOut.
Add a convenience API ABI::GetReturnValueObject.
Change the ValueObject::EvaluationPoint to BE an ExecutionContextScope, rather than
trying to hand out one of its subsidiary object's pointers. That way this will always
be good.
llvm-svn: 146806
enhanced disassembler classify disassemblers by
target triple (as a string) rather than just by
the architecture component of the triple. This
fixes a problem where different variants of the
Thumb instruction set were lumped into the same
hash bucket.
llvm-svn: 146799
valobj.AddressOf() returns None when an address is expected in a SyntheticChildrenProvider
Patch from Enrico Granata:
The problem was that the frozen object created by the expression parser was a copy of the contents of the StgClosure, rather than a pointer to it. Thus, the expression parser was correctly computing the result of the arithmetic&cast operation along with its address, but only saving it in the live object. This meant that the frozen copy acted as an address-less variable, hence the problem.
The fix attached to this email lets the expression parser store the "live address" in the frozen copy of the address when the object is built without a valid address of its own.
Doing so, along with delegating ValueObjectConstResult to calculate its own address when necessary, solves the issue. I have also added a new test case to check for regressions in this area, and checked that existing test cases pass correctly.
llvm-svn: 146768
we handle Objective-C method calls. Currently,
LLDB treats the result of an Objective-C method
as unknown if the type information doesn't have
the method's signature. Now Clang can cast the
result to id if it isn't explicitly cast.
I also added a test case for this, as well as a
fix for a type import problem that this feature
exposed.
llvm-svn: 146756
1 -- an address pointing off into non-executable memory -- don't
abort the unwind. We'll use the ABI's default UnwindPlan to try
to get out of frame 1 and on many platforms with a standard frame
chain stack layout we can get back on track and get a valid frame
2. This preserves the lldb behavior to-date; the change last week
to require the memory region to be executable broke it.
I'd like to mark this frame specially when displayed to the user;
I tried to override the places where the frame's pc value is returned
to change it to a sentinel value (e.g. LLDB_INVALID_ADDRESS) but
couldn't get that to work cleanly so I backed that part out for
now. When this happens we'll often miss one of the user's actual
frames, the one that's of most interest to the user, so I'd like
to make this visually distinctive.
Note that a frame in non-executable memory region is only allowed
for frame 1. After that we should be solid on the unwind and any
pc address in non-executable memory indicates a failure and we
should stop unwinding.
llvm-svn: 146723
Instead of getting the location of the variable and converting the hex string to an int, just use
val.AddressOf().GetValueAsUnsigned() to compute the address of the memory region to read from.
llvm-svn: 146719
Added a static memory pressure function in SBDebugger:
void SBDebugger::MemoryPressureDetected ()
This can be called by applications that detect memory pressure to cause LLDB to release cached information.
llvm-svn: 146640
size_t
SBProcess::ReadCStringFromMemory (addr_t addr, void *buf, size_t size, lldb::SBError &error);
uint64_t
SBProcess::ReadUnsignedFromMemory (addr_t addr, uint32_t byte_size, lldb::SBError &error);
lldb::addr_t
SBProcess::ReadPointerFromMemory (addr_t addr, lldb::SBError &error);
These ReadCStringFromMemory() has some SWIG type magic that makes it return the
python string directly and the "buf" is not needed:
error = SBError()
max_cstr_len = 256
cstr = lldb.process.ReadCStringFromMemory (0x1000, max_cstr_len, error)
if error.Success():
....
The other two functions behave as expteced. This will make it easier to get integer values
from the inferior process that are correctly byte swapped. Also for pointers, the correct
pointer byte size will be used.
Also cleaned up a few printf style warnings for the 32 bit lldb build on darwin.
llvm-svn: 146636
lldb::SBValue::CreateValueFromAddress does not verify SBType::GetPointerType succeeds
SBValue::CreateValueFromAddress() should check the validity of type and its derived pointer type
before using it. Add a test case.
llvm-svn: 146629
clients to disassemble a series of raw bytes as
demonstrated by a new testcase.
In the future, this API will also allow clients
to provide a callback that adds comments for
addresses in the disassembly.
I also modified the SWIG harness to ensure that
Python ByteArrays work as well as strings as
sources of raw data.
llvm-svn: 146611
LLDBSwigPythonCallCommand crashes when a command script returns an object
Add more robustness to LLDBSwigPythonCallCommand. It should check whether the returned Python object
is a string, and only assign it as the error msg when the check holds.
Also add a regression test.
llvm-svn: 146584
dispatch functions that are implemented in hand-written assembly.
There is also hand-written eh_frame instructions for unwinding
from these functions.
Normally we don't use eh_frame instructions for the currently
executing function, prefering the assembly instruction profiling
method. But in these hand-written dispatch functions, the
profiling is doomed and we should use the eh_frame instructions.
Unfortunately there's no easy way to flag/extend the eh_frame/debug_frame
sections to annotate if the unwind instructions are accurate at
all addresses ("asynchronous") or if they are only accurate at locations
that can throw an exception ("synchronous" and the normal case for
gcc/clang generated eh_frame/debug_frame CFI).
<rdar://problem/10508134>
llvm-svn: 146551
the expression parser to locate instances where
dyn_cast<>() and isa<>() are used on types, and
replace them with getAs<>() as appropriate.
The difference is that dyn_cast<>() and isa<>()
are essentially LLVM/Clang's equivalent of RTTI
-- that is, they try to downcast the object and
return NULL if they cannot -- but getAs<>() can
traverse typedefs to perform a semantic cast.
llvm-svn: 146537
Check that the pc value for frames up the stack is in a
mapped+executable region of memory.
Check that the stack pointer for frames up the stack is
in a mapped+readable region of memory.
If the unwinder ever makes a mistake walking the stack,
these checks will help to keep it from going too far into
the weeds.
These aren't fixing any bugs that I know of, but they
add extra robustness to a complicated task.
llvm-svn: 146478
if this is a mapped/executable region of memory. If it isn't, we've jumped
through a bad pointer and we know how to unwind the stack correctly based
on the ABI.
Previously I had 0x0 special cased but if you jumped to 0x2 on x86_64 one
frame would be skipped because the unwinder would try using the x86_64
ArchDefaultUnwindPlan which relied on the rbp.
Fixes <rdar://problem/10508291>
llvm-svn: 146477
validates the "self," "this," and "_cmd" pointers
that get passed into expressions. It used to check
them aggressively for validity before allowing the
expression to run as an object method; now, this
functionality is gated by a bool and off by default.
Now the default is that when LLDB is stopped in a
method of a class, code entered using "expr" will
always masquerade as an instance method. If for
some reason "self," "this," or "_cmd" is unavailable
it will be reported as NULL. This may cause the
expression to crash if it relies on those pointers,
but for example getting the addresses of ivars will
now work as the user would expect.
llvm-svn: 146465
Modified the Xcode project to not strip liblldb-core.a for BuildAndIntegration builds
and to correctly strip only debug symbols from the command line binaries.
llvm-svn: 146462
There were two problems associated with this radar:
1. "settings show target.source-map" failed to show the source-map after, for example,
"settings set target.source-map /Volumes/data/lldb/svn/trunk/test/source-manager /Volumes/data/lldb/svn/trunk/test/source-manager/hidden"
has been executed to set the source-map.
2. "list -n main" failed to display the source of the main() function after we properly set the source-map.
The first was fixed by adding the missing functionality to TargetInstanceSettings::GetInstanceSettingsValue (Target.cpp)
and updating the support files PathMappingList.h/.cpp; the second by modifying SourceManager.cpp to fix several places
with incorrect logic.
Also added a test case test_move_and_then_display_source() to TestSourceManager.py, which moves main.c to hidden/main.c,
sets target.source-map to perform the directory mapping, and then verifies that "list -n main" can still show the main()
function.
llvm-svn: 146422
We will return a valid range when possible and omit the "permissions" key
when the memory is not readable, writeable or executeable. This will help us
know the difference between an error back from this packet and unsupported,
from just "this address isn't in a valid region".
llvm-svn: 146394
<rdar://problem/10561406>
Stopped the SymbolFileDWARF::FindFunctions (...) from always calculating
the line table entry for all functions that were found. This can slow down
the expression parser if it ends up finding a bunch of matches. Fixed the
places that were relying on the line table entry being filled in.
Discovered a recursive stack blowout that happened when "main" didn't have
line info for it and there was no line information for "main"
llvm-svn: 146330
- Even if a frame isn't present, we always try
to use FindGlobalVariable to find variables.
Instead of using frame->TrackGlobalVariable()
to promote the VariableSP into a ValueObject,
we now simply use ValueObjectVariable.
- When requesting the value of a variable, we
allow returning of the "live version" of the
variable -- that is, the variable in the
target instead of a pointer to its freeze
dried version in LLDB -- even if there is no
process present.
llvm-svn: 146315
An assertion was firing when parsing types due to trying to complete parent
class decl contenxt types too often.
Also, relax where "dsymutil" binary can come from in the Makefile.rules.
llvm-svn: 146310
hard to ensure it doesn't get invalidated out from under us. Instead look it up from the ThreadID
and StackID when asked for it.
<rdar://problem/10554409>
llvm-svn: 146309
that if we prefer the current compile unit, followed by any compile units that
already had their DIEs parsed, followed by the rest of the matches, that we
might save some memory. This turned out not to help much. The code is commented
out, but I want to check it in so I don't lose the code in case it could help
later.
Added the ability to efficiently find the objective C class implementation
when using the new .apple_types acclerator tables with the type flags. If the
type flags are not available, we default back to what we were doing before.
llvm-svn: 146250
PlatformDarwin.cpp -- call it from both PlatformRemoteiOS.cpp
and the native process PlatformDarwin.cpp when running on an arm
system.
Bump lldb version number to 94.
llvm-svn: 146249
that got hit, and first check the condition, and if that location's condition says we should stop, then we
run the callback. In the end if any location's condition and callback say we should stop, then we stop.
llvm-svn: 146242
translation unit has a interface for a class "Bar" that contains hidden ivars
in the implementation and we make sure we can see these hidden ivars. We also
test the case where we stop in translation unit that contains the
implementation first. So the test runs two tests:
1 - run and stop where we have an interface, run to main and print and make
sure we find the hidden ivar
2 - run and stop where we have an implementation, run to main and print and make
sure we find the hidden ivar
llvm-svn: 146216
in the context in which it was originally found, the
expression parser now goes hunting for it in all modules
(in the appropriate namespace, if applicable). This means
that forward-declared types that exist in another shared
library will now be resolved correctly.
Added a test case to cover this. The test case also tests
"frame variable," which does not have this functionality
yet.
llvm-svn: 146204
pointer to make the result of an expression. LLDB now
dumps the ivars of the Objective-C object and all of
its parents. This just required fixing a bug where we
didn't distinguish between Objective-C object pointers
and regular C-style pointers.
Also added a testcase to verify that this continues to
work.
llvm-svn: 146164
that is in a class from the expression parser, and it was causing an
assertion. There is now a function that will correctly resolve a type
even if it is in a class.
llvm-svn: 146141
take a SymbolFile reference and a lldb::user_id_t and be used in objects
which represent things in debug symbols that have types where we don't need
to know the true type yet, such as in lldb_private::Variable objects. This
allows us to defer resolving the type until something is used. More specifically
this allows us to get 1000 local variables from the current function, and if
the user types "frame variable argc", we end up _only_ resolving the type for
"argc" and not for the 999 other local variables. We can expand the use of this
as needed in the future.
Modified the DWARFMappedHash class to be able to read the HashData that has
more than just the DIE offset. It currently will read the atoms in the header
definition and read the data correctly. Currently only the DIE offset and
type flags are supported. This is needed for adding type flags to the
.apple_types hash accelerator tables.
Fixed a assertion crash that would happen if we have a variable that had a
DW_AT_const_value instead of a location where "location.LocationContains_DW_OP_addr()"
would end up asserting when it tried to parse the variable location as a
DWARF opcode list.
Decreased the amount of memory that LLDB would use when evaluating an expression
by 3x - 4x for clang. There was a place in the namespace lookup code that was
parsing all namespaces with a certain name in a DWARF file instead of stopping
when it found the first match. This was causing all of the compile units with
a matching namespace to get parsed into memory and causing unnecessary memory
bloat.
Improved "Target::EvaluateExpression(...)" to not try and find a variable
when the expression contains characters that would certainly cause an expression
to need to be evaluated by the debugger.
llvm-svn: 146130
for now to fix testcases. Once we have a valid use
for the function information (i.e., once properties
returning UnknownAnyTy are allowed, once we read
return type information from the runtime, among
other uses) I will re-enable this.
llvm-svn: 146129
symbols. Now we find the correct method.
Unfortunately we don't get the superclass from the
runtime yet so the method doesn't import correctly
(and I added a check to make sure that doesn't hurt
us) but once we get that information right we will
report methods correctly to the parser as well.
Getting superclass information requires a common AST
context for all Objective-C runtime information,
meaning that the superclass and the subclass are in
the same AST context in all cases. That is the next
thing that needs to be done here.
llvm-svn: 146089
for use in the benchmark against lldb's disassembly speed. Note that the lldb
executable path can already be specified using the LLDB_EXEC env variable.
rdar://problem/7511194
llvm-svn: 146050
from symbols more accessible, I have added a second
map to the ClangASTImporter: the ObjCInterfaceMetaMap.
This map keeps track of all type definitions found for
a particular Objective-C interface, allowing the
ClangASTSource to refer to all possible sources when
looking for method definitions.
There is a bug in lookup that I still need to figure out,
but after that we should be able to report full method
information for Objective-C classes shown in symbols.
Also fixed some errors I ran into when enabling the maps
for the persistent type store. The persistent type store
previously did not use the ClangASTImporter to import
types, instead using ASTImporters that got allocated each
time a type needed copying. To support the requirements
of the persistent type store -- namely, that types must be
copied, completed, and then completely severed from their
origin in the parser's AST context (which will go away) --
I added a new function called DeportType which severs all
these connections.
llvm-svn: 145914
methods. The Clang dump is now much more verbose,
but when somebody types "target modules lookup -t"
that is typically what they're looking for.
llvm-svn: 145892
and CompleteTagDeclarationDefinition() on Objective-C
interfaces populated by SymbolFileSymtab::FindTypes(),
we should mark the interface as forward-declared when
we create it.
llvm-svn: 145825
and fixes we did. Now that objective C classes are represented by symbols with
their own type, there were a few more places in the objective C code that needed
to be fixed when searching for dynamic types.
Cleaned up the objective C runtime plug-in a bit to not keep having to create
constant strings and make one less memory access when we find an "isa" in the
objective C cache.
llvm-svn: 145799
add them to a fast lookup map. lldb_private::Symtab now export the following
public typedefs:
namespace lldb_private {
class Symtab {
typedef std::vector<uint32_t> IndexCollection;
typedef UniqueCStringMap<uint32_t> NameToIndexMap;
};
}
Clients can then find symbols by name and or type and end up with a
Symtab::IndexCollection that is filled with indexes. These indexes can then
be put into a name to index lookup map and control if the mangled and
demangled names get added to the map:
bool add_demangled = true;
bool add_mangled = true;
Symtab::NameToIndexMap name_to_index;
symtab->AppendSymbolNamesToMap (indexes, add_demangled, add_mangled, name_to_index).
This can be repeated as many times as needed to get a lookup table that
you are happy with, and then this can be sorted:
name_to_index.Sort();
Now name lookups can be done using a subset of the symbols you extracted from
the symbol table. This is currently being used to extract objective C types
from object files when there is no debug info in SymbolFileSymtab.
Cleaned up how the objective C types were being vended to be more efficient
and fixed some errors in the regular expression that was being used.
llvm-svn: 145777
class. The thing with Objective C classes is the debug info might have a
definition that isn't just a forward decl, but it is incomplete. So we need to
look and see if we can find the complete definition and avoid recursing a lot
due to the fact that our accelerator tables will have many versions of the
type, but only one complete one. We might not also have the complete type
and we need to deal with this correctly.
llvm-svn: 145759
Objective-C, making symbol lookups for various raw
Objective-C symbols work correctly. The IR interpreter
makes these lookups because Clang has emitted raw
symbol references for ivars and classes.
Also improved performance in SymbolFiles, caching the
result of asking for SymbolFile abilities.
llvm-svn: 145758
for all our external AST sources that lets us associate
arbitrary flags with the types we put into the AST
contexts. Also added an API on ClangASTContext that
allows access to these flags given only an ASTContext
and a type.
Because we don't have access to RTTI, and because at
some point in the future we might encounter external
AST sources that we didn't make (so they don't subclass
ClangExternalASTSourceCommon) I added a magic number
that we check before doing anything else, so that we
can catch that problem as soon as it appears.
llvm-svn: 145748
object file can correctly make these symbols which will abstract us from the
file format and ABI and we can then ask for the objective C class symbol for
a class and find out which object file it was defined in.
llvm-svn: 145744
the function it is being asked to step through, so that even if we get the trampoline
target wrong (for instance) we will still not lose control.
The other fix here is to tighten up the handling of the case where the current plan
doesn't explain the stop, but a plan above us does. In that case, if the plan that
does explain the stop says it is done, we need to clean up the plans below it and
continue on with our processing.
llvm-svn: 145740
Fixed an issue where if we have the DWARF equivalent of:
struct foo;
class foo { ... };
Or vice versa, we wouldn't be able to find the complete type. Since many
compilers allow forward declarations to have struct and definitions to have
class, we need to be able to deal with both cases. This commit fixes this in
the DWARF parser.
llvm-svn: 145733
to use ConstStrings. The const char*s were assumed to
be from ConstStrings before, but since storing a full-on
ConstString is no more expensive than storing a const
char* it makes better sense to enforce uniqueness with
the type checker.
llvm-svn: 145688
Fixed templates with NonTypeTemplateParmDecl objects. For example:
template <unsigned N>
....
This fixes SmallVector and all of the other classes that have template params
that are non types.
llvm-svn: 145667
will allow us to represent a process/thread ID using a pointer for the OS
plug-ins where they might want to represent the process or thread ID using
the address of the process or thread structure.
llvm-svn: 145644
in the face of failures to import types, since blithely
passing on NULL types can sometimes lead to trouble.
Also eliminated a use of getAs and replaced it with
dyn_cast, which is more robust.
llvm-svn: 145628
enhancements. With these enhancements, the return values
of Objective-C methods with unknown return types can be
implicitly cast to id for the purpose of making method
calls.
So what would have required this:
(int)[(id)[ClassWithNoDebugInfo methodReturningObject] methodReturningInt]
can now be written as:
(int)[[ClassWithNoDebugInfo methodReturningObject] methodReturningInt]
llvm-svn: 145567
robust:
- Now a client can specify what kind of symbols
are needed; notably, this allows looking up
Objective-C class symbols specifically.
- In the class of symbols being looked up, if
one is non-NULL and others are NULL, LLDB now
prefers the non-NULL one.
llvm-svn: 145554
ClangASTSource::~ClangASTSource() was calling
ClangASTContext *scratch_clang_ast_context = m_target->GetScratchClangASTContext();
which had the side effect of deleting this very ClangASTSource instance. Not good.
Change it to
// We are in the process of destruction, don't create clang ast context on demand
// by passing false to Target::GetScratchClangASTContext(create_on_demand).
ClangASTContext *scratch_clang_ast_context = m_target->GetScratchClangASTContext(false);
The Target::GetScratchClangASTContext(bool create_on_demand=true) has a new signature.
llvm-svn: 145537
to find Objective-C class types by looking in the
symbol tables for the individual object files.
I did this as follows:
- I added code to SymbolFileSymtab that vends
Clang types for symbols matching the pattern
"_OBJC_CLASS_$_NSMyClassName," making them
appear as Objective-C classes. This only occurs
in modules that do not have debug information,
since otherwise SymbolFileDWARF would be in
charge of looking up types.
- I made a new SymbolVendor subclass for the
Apple Objective-C runtime that is in charge of
making global lookups of Objective-C types. It
currently just sends out type lookup requests to
the appropriate SymbolFiles, but in the future we
will probably extend it to query the runtime more
completely.
I also modified a testcase whose behavior is changed
by the fact that we now actually return an Objective-C
type for __NSCFString.
llvm-svn: 145526
Fix wrong test logic in test_modules_search_paths(). Add additional exercising of 'target modules search-paths list/query".
There is a reproducible crash if 'target modules search-paths clear' is exercised during test teardown.
So we currently comment out the stmt as follows:
# Add teardown hook to clear image-search-paths after the test.
# rdar://problem/10501020
# Uncomment the following to reproduce 10501020.
#self.addTearDownHook(lambda: self.runCmd("target modules search-paths clear"))
llvm-svn: 145466
Fixed an issue where if we are debugging on a remote platform and set a
platform path for our executable, it was not being honored by the new
launch functions that used the ProcessLaunchInfo.
llvm-svn: 145371
as well as attached a new priority description as to why and when you would
want to implement each packet.
Also documented the additions we have made to the stop reply packet and why
the extra information is necessary.
llvm-svn: 145357
management of what allocations remain after an
expression finishes executing. This saves around
2.5KiB per expression for simple expressions.
llvm-svn: 145342
so that we can do Python scripting like this:
target = self.dbg.CreateTarget(self.exe)
self.dbg.SetAsync(True)
process = target.LaunchSimple(None, None, os.getcwd())
process.PutSTDIN("Line 1 Entered.\n")
process.PutSTDIN("Line 2 Entered.\n")
process.PutSTDIN("Line 3 Entered.\n")
Add TestProcessIO.py to exercise the process IO API: PutSTDIN()/GetSTDOUT()/GetSTDERR().
llvm-svn: 145282
to launch a process for debugging. Since this isn't supported on all platforms,
we need to do what we used to do if this isn't supported. I added:
bool
Platform::CanDebugProcess ();
This will get checked before trying to launch a process for debugging and then
fall back to launching the process through the current host debugger. This
should solve the issue for linux and keep the platform code clean.
Centralized logging code for logging errors, warnings and logs when reporting
things for modules or symbol files. Both lldb_private::Module and
lldb_private::SymbolFile now have the following member functions:
void
LogMessage (Log *log, const char *format, ...);
void
ReportWarning (const char *format, ...);
void
ReportError (const char *format, ...);
These will all output the module name and object (if any) such as:
"error: lldb.so ...."
"warning: my_archive.a(foo.o) ...."
This will keep the output consistent and stop a lot of logging calls from
having to try and output all of the information that uniquely identifies
a module or symbol file. Many places in the code were grabbing the path to the
object file manually and if the module represented a .o file in an archive, we
would see log messages like:
error: foo.a - some error happened
llvm-svn: 145219
to 30% of memory. The size doubling was killing us and we ended up with up to
just under 50% of empty capacity. Cleaning this up saves us a ton of memory.
llvm-svn: 145086
Fixed an issue with the options for memory read where --count couldn't be used
with the --binary option when writing data to a file.
Also removed the GDB format option from the --binary version of memory read.
llvm-svn: 145067
something like "display/4i $pc" (or something like this). With LLDB we already
were showing 3 lines of source before and 3 lines of source after the current
source line when showing a stop context. We now improve this by allowing the
user to control the number of lines with the new "stop-line-count-before" and
"stop-line-count-after" settings. Also, there is a new setting for how many
disassembly lines to show: "stop-disassembly-count". This will control how many
source lines are shown when there is no source or when we have no source line
info.
settings set stop-line-count-before 3
settings set stop-line-count-after 3
settings set stop-disassembly-count 4
settings set stop-disassembly-display no-source
The default values are set as shown above and allow 3 lines of source before
and after (what we used to do) the current stop location, and will display 4
lines of disassembly if the source is not available or if we have no debug
info. If both "stop-source-context-before" and "stop-source-context-after" are
set to zero, this will disable showing any source when stopped. The
"stop-disassembly-display" setting is an enumeration that allows you to control
when to display disassembly. It has 3 possible values:
"never" - never show disassembly no matter what
"no-source" - only show disassembly when there is no source line info or the source files are missing
"always" - always show disassembly.
llvm-svn: 145050
several patches. These patches fix a problem
where templated types were not being completed the
first time they were used, and fix a variety of
minor issues I discovered while fixing that problem.
One of the previous local patches was resolved in
the most recent Clang, so I removed it. The others
will be removed in due course.
llvm-svn: 144984
1 - the DIE collections no longer have the NULL tags which saves up to 25%
of the memory on typical C++ code
2 - faster parsing by not having to run the SetDIERelations() function anymore
it is done when parsing the DWARF very efficiently.
llvm-svn: 144983
the name of the PLT entry. This solution assumes a naming convention agreed upon by us and the system folks,
and isn't general. The general solution requires actually finding & calling the resolver function if it
hasn't been called yet. That's more tricky.
llvm-svn: 144981
from a process and hooked it up to the new packet that was recently added
to our GDB remote executable named debugserver. Now Process has the following
new calls:
virtual Error
Process::GetMemoryRegionInfo (lldb::addr_t load_addr, MemoryRegionInfo &range_info);
virtual uint32_t
GetLoadAddressPermissions (lldb::addr_t load_addr);
Only the first one needs to be implemented by subclasses that can add this
support.
Cleaned up the way the new packet was implemented in debugserver to be more
useful as an API inside debugserver. Also found an error where finding a region
for an address actually will pick up the next region that follows the address
in the query so we also need ot make sure that the address we requested the
region for falls into the region that gets returned.
llvm-svn: 144976
we say that the vectors of DWARFDebugInfoEntry objects were the highest on the
the list.
With these changes we cut our memory usage by 40%!!! I did this by reducing
the size of the DWARFDebugInfoEntry from a previous:
uint32_t offset
uint32_t parent_idx
uint32_t sibling_idx
Abbrev * abbrev_ptr
which was 20 bytes, but rounded up to 24 bytes due to alignment. Now we have:
uint32_t offset
uint32_t parent_idx
uint32_t sibling_idx
uint32_t abbr_idx:15, // 32767 possible abbreviation codes
has_children:1, // 0 = no children, 1 = has children
tag:16; // DW_TAG_XXX value
This gets us down to 16 bytes per DIE. I tested some VERY large DWARF files
(900MB) and found there were only ~700 unique abbreviations, so 32767 should
be enough for any sane compiler. If it isn't there are built in assertions
that will fire off and tell us.
llvm-svn: 144975
to allow variables in the persistent variable store to know
how to complete themselves from debug information. That
fixes a variety of bugs during dematerialization of
expression results and also makes persistent variable and
result variables ($foo, $4, ...) more useful.
I have also added logging improvements that make it much
easier to figure out how types are moving from place to
place, and made some checking a little more aggressive.
The commit includes patches to Clang which are currently being
integrated into Clang proper; once these fixes are in Clang
top-of-tree, these patches will be removed. The patches don't
fix API; rather, they fix some internal bugs in Clang's
ASTImporter that were exposed when LLDB was moving types from
place to place multiple times.
llvm-svn: 144969
Use this option with care as you would need to build the inferior(s) by hand
and build the executable(s) with the correct name(s). This option can be used
with '-# n' to stress test certain test cases for n number of times.
An example:
[11:55:11] johnny:/Volumes/data/lldb/svn/trunk/test/python_api/value $ ls
Makefile TestValueAPI.pyc linked_list
TestValueAPI.py change_values main.c
[11:55:14] johnny:/Volumes/data/lldb/svn/trunk/test/python_api/value $ make EXE=test_with_dsym
clang -gdwarf-2 -O0 -arch x86_64 -c -o main.o main.c
clang -gdwarf-2 -O0 -arch x86_64 main.o -o "test_with_dsym"
/usr/bin/dsymutil -o "test_with_dsym.dSYM" "test_with_dsym"
[11:55:20] johnny:/Volumes/data/lldb/svn/trunk/test/python_api/value $ cd ../..
[11:55:24] johnny:/Volumes/data/lldb/svn/trunk/test $ ./dotest.py -v -# 10 -S -f ValueAPITestCase.test_with_dsym
LLDB build dir: /Volumes/data/lldb/svn/trunk/build/Debug
LLDB-89
Path: /Volumes/data/lldb/svn/trunk
URL: https://johnny@llvm.org/svn/llvm-project/lldb/trunk
Repository Root: https://johnny@llvm.org/svn/llvm-project
Repository UUID: 91177308-0d34-0410-b5e6-96231b3b80d8
Revision: 144914
Node Kind: directory
Schedule: normal
Last Changed Author: gclayton
Last Changed Rev: 144911
Last Changed Date: 2011-11-17 09:22:31 -0800 (Thu, 17 Nov 2011)
Session logs for test failures/errors/unexpected successes will go into directory '2011-11-17-11_55_29'
Command invoked: python ./dotest.py -v -# 10 -S -f ValueAPITestCase.test_with_dsym
----------------------------------------------------------------------
Collected 1 test
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 1.163s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.200s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.198s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.199s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.239s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 1.215s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.105s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.098s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 0.195s
OK
1: test_with_dsym (TestValueAPI.ValueAPITestCase)
Exercise some SBValue APIs. ... ok
----------------------------------------------------------------------
Ran 1 test in 1.197s
OK
[11:55:34] johnny:/Volumes/data/lldb/svn/trunk/test $
llvm-svn: 144919
turned out to be unitialized data in the ProcessLaunchInfo default constructor.
Turning on MallocScribble in the environment helped track this down.
When we launch and attach using the host layer, we now inform the process that
it shouldn't detach when by calling an accessor.
llvm-svn: 144882
the thread specific data and were destroying the thread specfic data more
than once.
Also added the ability to ask a lldb::StateType if it is stopped with an
additional paramter of "must_exist" which means that the state must be a
stopped state for a process that still exists. This means that eStateExited
and eStateUnloaded will no longer return true if "must_exist" is set to true.
llvm-svn: 144875
completion information between different AST
contexts. It works like this:
- If a Decl is imported from a context that
has completion metadata, then that Decl
is associated with the same completion
information (possibly none) as the Decl
it was imported from.
- If a Decl is imported from a context that
does not have completion metadata, then it
is marked as completable by consulting the
Decl and context it was imported from.
llvm-svn: 144838
for each AST context it knows about in a single
object. This makes it faster to look up the
appropriate ASTImpoter for a given ASTContext
pair and also makes it much easier to delete all
metadata for a given AST context.
In the future, this fix will allow the
ClangASTImporter to propagate completion
information between the metadata for different
AST contexts as its minions move AST objects
around.
llvm-svn: 144835
rather than individually on behalf of each
ASTContext. This allows the ASTImporter to know
about all containers of types, which will let it
be smarter about forwarding information about
type origins. That means that the following
sequence of steps will be possible (after a few
more changes):
- Import a type from a Module's ASTContext into
an expression parser ASTContext, tracking its
origin information -- this works now.
- Because the result of the expression uses that
type, import it from the expression parser
ASTContext into the Target's scratch AST
context, forwarding the origin information --
this needs to be added.
- For a later expression that uses the result,
import the type from the Target's scratch AST
context, still forwarding origin information
-- this also needs to be added.
- Use the intact origin information to complete
the type as needed -- this works now if the
origin information is present.
To this end, I made the following changes:
- ASTImporter top-level copy functions now
require both a source and a destination AST
context parameter.
- The ASTImporter now knows how to purge
records related to an ASTContext that is
going away.
- The Target now owns and creates the ASTImporter
whenever the main executable changes or (in the
absence of a main executable) on demand.
llvm-svn: 144802
After recent changes we weren't reaping child processes resulting in many
zombie processes.
This was fixed by adding more settings to the ProcessLaunchOptions class
that allow clients to specify a callback function and baton to be notified
when their process dies. If one is not supplied a default callback will be
used that "does the right thing".
Cleaned up a race condition in the ProcessGDBRemote class that would attempt
to monitor when debugserver died.
Added an extra boolean to the process monitor callbacks that indicate if a
process exited or not. If your process exited with a zero exit status and no
signal, both items could be zero.
Modified the process monitor functions to not require a callback function
in order to reap the child process.
llvm-svn: 144780
NULL-terminated C string to store the contents
of the expression prefix file. This meant that
expressions, when printing the contents of the
prefix into the expression's text, would
invariably put in bad data after the end of the
expression.
Now, instead, we store the prefix contents in a
std::string, which handles null-termination
correctly.
llvm-svn: 144760
types. First, I added handling for the memset intrinsic
in the IR, which is used to zero out the returned struct.
Second, I fixed the object-checking instrumentation
to objc_msgSend_stret, and generally tightened up how
the object-checking functions get inserted.
llvm-svn: 144741
info for us to attach by pid, or by name and will also allow us to eventually
do a lot more powerful attaches. If you look at the options for the "platform
process list" command, there are many options which we should be able to
specify. This will allow us to do things like "attach to a process named 'tcsh'
that has a parent process ID of 123", or "attach to a process named 'x' which
has an effective user ID of 345".
I finished up the --shell implementation so that it can be used without the
--tty option in "process launch". The "--shell" option now can take an
optional argument which is the path to the shell to use (or a partial name
like "sh" which we will find using the current PATH environment variable).
Modified the Process::Attach to use the new ProcessAttachInfo as the sole
argument and centralized a lot of code that was in the "process attach"
Execute function so that everyone can take advantage of the powerful new
attach functionality.
llvm-svn: 144615
of problems with Objective-C object completion. To go
along with the LLVM/Clang-side fixes, we have a variety
of Objective-C improvements.
Fixes include:
- It is now possible to run expressions when stopped in
an Objective-C class method and have "self" act just
like "self" would act in the class method itself (i.e.,
[self classMethod] works without casting the return
type if debug info is present). To accomplish this,
the expression masquerades as a class method added by
a category.
- Objective-C objects can now provide methods and
properties and methods to Clang on demand (i.e., the
ASTImporter sets hasExternalVisibleDecls on Objective-C
interface objects).
- Objective-C built-in types, which had long been a bone
of contention (should we be using "id"? "id*"?), are
now fetched correctly using accessor functions on
ClangASTContext. We inhibit searches for them in the
debug information.
There are also a variety of logging fixes, and I made two
changes to the test suite:
- Enabled a test case for Objective-C properties in the
current translation unit.
- Added a test case for calling Objective-C class methods
when stopped in a class method.
llvm-svn: 144607
Fixed an issues with the SBType and SBTypeMember classes:
- Fixed SBType to be able to dump itself from python
- Fixed SBType::GetNumberOfFields() to return the correct value for objective C interfaces
- Fixed SBTypeMember to be able to dump itself from python
- Fixed the SBTypeMember ability to get a field offset in bytes (the value
being returned was wrong)
- Added the SBTypeMember ability to get a field offset in bits
Cleaned up a lot of the Stream usage in the SB API files.
llvm-svn: 144493
This is the actual fix for the above radar where global variables that weren't
initialized were not being shown correctly when leaving the DWARF in the .o
files. Global variables that aren't intialized have symbols in the .o files
that specify they are undefined and external to the .o file, yet document the
size of the variable. This allows the compiler to emit a single copy, but makes
it harder for our DWARF in .o files with the executable having a debug map
because the symbol for the global in the .o file doesn't exist in a section
that we can assign a fixed up linked address to, and also the DWARF contains
an invalid address in the "DW_OP_addr" location (always zero). This means that
the DWARF is incorrect and actually maps all such global varaibles to the
first file address in the .o file which is usually the first function. So we
can fix this in either of two ways: make a new fake section in the .o file
so that we have a file address in the .o file that we can relink, or fix the
the variable as it is created in the .o file DWARF parser and actually give it
the file address from the executable. Each variable contains a
SymbolContextScope, or a single pointer that helps us to recreate where the
variables came from (which module, file, function, etc). This context helps
us to resolve any file addresses that might be in the location description of
the variable by pointing us to which file the file address comes from, so we
can just replace the SymbolContextScope and also fix up the location, which we
would have had to do for the other case as well, and update the file address.
Now globals display correctly.
The above changes made it possible to determine if a variable is a global
or static variable when parsing DWARF. The DWARF emits a DW_TAG_variable tag
for each variable (local, global, or static), yet DWARF provides no way for
us to classify these variables into these categories. We can now detect when
a variable has a simple address expressions as its location and this will help
us classify these correctly.
While making the above changes I also noticed that we had two symbol types:
eSymbolTypeExtern and eSymbolTypeUndefined which mean essentially the same
thing: the symbol is not defined in the current object file. Symbol objects
also have a bit that specifies if a symbol is externally visible, so I got
rid of the eSymbolTypeExtern symbol type and moved all code locations that
used it to use the eSymbolTypeUndefined type.
llvm-svn: 144489
the --tty option. So you can now get shell expansion and file redirection:
(lldb) process launch --tty --shell -- *.jpg < in.txt > out.txt
Again, the "--tty" is mandatory for now until we hook this up to other
functions. The shell is also currently hard coded to "/bin/bash" and not the
"SHELL" variable. "/bin/tcsh" was causing problems which I need to dig into.
llvm-svn: 144443
interfaces. This allows us to pull in Objective-C
method types on demand, which is also now implemented.
Also added a minor fix to prevent multiple-definition
errors for "Class" and "id".
llvm-svn: 144405
problem is that we had a bitfield that kept
track of what had been found so far, and inhibited
searches when the local variable bit was set.
This bitfield was not being initialized correctly.
llvm-svn: 144400
the argument description in the command name could cause a command
alias to crash, e.g.
command alias zzz target stop-hook delete 1
because the "name" is used to re-fetch the exact CommandObject when
adding the final arg.
<rdar://problem/10423753>
llvm-svn: 144330
Greg Clayton