| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
| |
Like, for example the clang, CompCert now prints a more detailed
descriptions of the debug information in the assembler file. For
each abbreviation and debug entry the dwarf attributes and their encodings
are added.
Bug 17392.
|
|
|
|
|
|
| |
Instead of pushing strings around use the actual section. However
the string is still used in the Hashtbl.
Bug 17392.
|
| |
|
|
|
|
|
|
|
|
| |
Instead of using a string they now take an optional string, which
should be none if the backend is not the diab backend and the
corresponding section is the text section and Some s with s being
the custom section name else.
Bug 17392.
|
|
|
|
|
|
|
|
| |
GCC prints all string larger than 3 characters in the debug_str
section which reduces the size of the debug information since entries
containing the same string now map to the same string in the
debug_str sections.
Bug 17392.
|
|
|
|
|
|
|
| |
The dwarf 2 standard allows more attributes for certain debuggint
entries than used by gcc or diab data. Since they are also not
set by compcert they can be removed.
Bug 17392.
|
|
|
|
|
| |
The function exists_type is not really used so we can remove it.
Bug 17392.
|
|
|
|
|
|
|
| |
Instead of making each filed mutuable we use a reference to a record
of type implem. Now only the default implementation and the default
debug information need to be upated to add a new function.
Bug 17392.
|
|
|
|
|
|
| |
Instead of using a tuple we now use a record with descriptive names
for the different entries.
Bug 17392
|
|
|
|
|
|
| |
Since the strip functions might be useful in other context and is
more general then the debug information.
Bug 17392.
|
|
|
|
|
|
| |
The atom to global variable debug id mapping is never used so we
do not need to insert global variables into it.
Bug 17392.
|
|
|
|
|
|
|
| |
Instead of defining two functions for adding the mapping from atom
to debug id we use one function which then sets the corresponding
values.
Bug 17392.
|
|
|
|
|
|
| |
The name_of_section function no returns the correct name for the
debug sections, the prologue and epilogue directives are added and
the labels for the live ranges are introduced in the Asmexpand pass.
|
| |
|
| |
|
|
|
|
| |
information.
|
| |
|
|\
| |
| |
| |
| | |
Conflicts:
debug/DebugInformation.ml
|
| | |
|
| | |
|
| | |
|
|/
|
|
|
| |
Since files without function definition contain no function in the
Section_text the filenum would be empty otherwise.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
in the Debuging information.
|
| |
|
|
|
|
| |
Producer as well as target system.
|
|
|
|
| |
the code more robust and added indentation for convertCompositeDef
|
|
|
|
|
|
| |
In order to avoid the problem that the stamp is not correct for
implict declarations I insert all possible stamps of a function
into my mapping and assign them one debug id.
|
|
|
|
|
|
|
| |
Changed the sizeof function to take into account the bytes needed
for the sleb128/uleb128 encoding of the DW_OP_* arguments and
changed the end_live_range function to only close functions where
the live range is currently open.
|
| |
|
|
|
|
|
|
| |
If a user uses the #pragma use_section for functions the diab linker
requires a separate debug_info section for each entry. This commit
adds functionality to emulate this behavior.
|
|
|
|
| |
scopes before the last statement.
|
| |
|
|
|
|
|
|
|
| |
In order to remove unnecessary dependecies the implemenation type is made
and the DebugInit file initializes the fields in the record. This allows
it to move more functions behind the Debug interface without introducing
circular dependencies.
|
|
|
|
|
|
|
|
| |
Debuging code.
The old functions to store the symbol for the Global variables and retrive this
is no longer needed since the atom is stored in DebugInformation. Also the
Debug.Abbrev module is no longer needed.
|
| |
|
|
|
|
|
| |
This commit adds furher support for location information for local
variables and starts with the implementation of the debug_loc section.
|
|
|
|
|
| |
In order to avoid the usage of too many labels we replace the
debug statements during the Asmexpand phase.
|
|
|
|
| |
atom + scope id Hashtable.
|
|
|
|
|
|
| |
Local variables are now added with bogus lexical scopes to reflect
the actually lexical scopes. Also this commit fixes assembler problems
of the das when a user section with the name ".text" is defined.
|
|
|
|
|
|
| |
Instead of creating separate annotations for the local variables
we call the Debug.add_lvar_scope and we construct a mapping from
function id + scope id to scope information.
|
|\
| |
| |
| |
| |
| | |
Conflicts:
debug/CtoDwarf.ml
debug/DwarfPrinter.ml
|
| |
| |
| |
| |
| |
| | |
The debug printer now uses udata for member sizes since structs can
have a size larger than 255 bytes. Also functions that are inlined
no longer have an high and low pc in their abbreviation entry.
|
| |
| |
| |
| |
| | |
In the case of a structure definition we do not need to set the
structure declaration field set to false.
|
| |
| |
| |
| |
| | |
Instead of reimplementing the whole scope handling in the debug information
use the existing functionality and fill the scopes explicitly in the functions.
|