| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Introduce Cutil.class_of_attribute to return the class of the given attribute: one among
Attr_type attribute related to types (e.g. "aligned")
Attr_struct attribute related to struct/union/enum types (e.g. "packed")
Attr_function attribute related to function types (e.g. "noreturn")
Attr_name attribute related to variable and function declarations (e.g. "section")
Attr_unknown attribute was not declared
Cutil.declare_attribute is used to associate a class to a custom attribute.
Standard attributes (const, volatile, _Alignas, etc) are Attr_type.
cfronted/C2C.ml: declare the few attributes that CompCert honors currently.
cparser/GCC.ml: a bigger list of attributes taken from GCC, for reference only.
|
|
|
|
| |
Before, we were doing C90, there was no official syntax for such attributes, and we used ours. With C99 we can use "ty [ attributes N ]" to print "array with attributes of N elements of type ty".
|
|
|
|
|
|
|
| |
Owing to the peculiarities of array types in Cutil.change_attributes_type, type-related attributes of the array element type were duplicated on the array type. E.g. elaborating 'const int a[10][5]' produced
"a is an array of 5 const arrays of 10 const ints"
instead of
"a is an array of 5 arrays of 10 const ints"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The treatment of attributes in the current CompCert is often surprising. For example,
attribute(xxx) char * x;
is parsed as "x is a pointer to a (char modified by attribute "xxx")", while for most attributes (e.g. section attributes) the expected meaning is "x, modified by attribute "xxx", has type pointer to char".
CompCert's current treatment comes from the fact that attributes are processed very much like the standard type modifiers `const` and `volatile`, i.e.
const char * x;
is really "x is a pointer to a const char", not "x is a const pointer to char".
This experiment introduces a distinction between type-related attributes (which include the standard modifiers `const` and `volatile`) and other attributes. The other, non-type-related attributes are "floated up" during elaboration so that they apply to the variable or function being declared or defined. In the examples above,
attribute(xxx) char * x; // "attribute(xxx)" applies to "x"
const char * x; // "const" applies to "char"
This may be a step in the right direction but is not the final story. In particular, the `packed` attribute is special-cased when applied to `struct`, like it was before, and future attributes concerning calling conventions would need to be floated up to function types but not higher than that.
|
|
|
|
| |
Follow-up to commit 1df1830
|
| |
|
|
|
|
| |
Those three functions can be useful to implement front-ends for languages other than C.
|
|
|
|
|
|
|
|
|
| |
The new option -fdiagnostics-format allows it to switch between
the three different format version:
-ccomp (default) with file:line:
-vi with file+line:
-msvc with file(line):
Bug 19872
|
| |
|
| |
|
|
|
|
|
|
| |
Controls whether the [-Woption] is printed in the diagnostic
message for mappable warnings/errors.
Bug 19872
|
|
|
|
|
|
|
| |
The option -fmax-errors limits the number of errors that are
reported before the compilation is aborted. The default 0 means no
limit.
Bug 19872
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
GCC treats files with .sx extension in the same way as it treats
files with .S suffix.
|
|
|
|
|
| |
Anonymous members no longer are printed in the debug information.
Fix 20798
|
| |
|
| |
|
|
|
|
|
|
|
| |
Instead of just accepting a string the function is changed to
accept a format string. This removes a lot of artificial sprintfs
in calls to the functions.
Bug 19872
|
|
|
|
|
|
| |
Instead of escaping all newlines etc for the help options use
quoted strings.
Bug 19872
|
|
|
|
|
|
| |
More functions are now documented. Furthermore compcert now prints
"ccomp:" instead of nothing for unknown locations.
Bug 19872
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
If CompCert crashes because of an uncaught exception the exception
is caught toplevel and the backtrace is printed plus an additional
message to include the backtrace in a support request, if buildnr
and tag are available.
Bug 20681.
|
| |
|
| |
|
| |
|
|\
| |
| | |
Next try for support of anonymous structs.
|
| |
| |
| |
| | |
"try ...; true with _ -> false" is dangerous if "..." raises unexpected exceptions such as Out_of_memory or Stack_overflow.
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| | |
The naming of anonymous structs is performed by an additional step
in elab_struct_or_union_info instead of in elab_field_group.
Also the aux functions are renamed to access.
Bug 20003
|
| |
| |
| |
| |
| |
| | |
Instead of using idents the anonymous fileds get names of the
for <anon>_c where c is a counter of all anonymous members.
Bug 20003
|
| |
| |
| |
| |
| |
| |
| |
| | |
Applied to the 92-constructor 'operation' type, 'decide equality' produces a huge transparent term that causes the VM compiler to generate huge code and exceeed a memory limit of Coq on 32-bit platforms. (The limit is OCaml's, really.)
The lib/BoolEqual.v file defines alternative tactics to build decidable equalities where the transparent part of the definition is smaller (O(N^2) instead of O(N^3)). The proof parts are still huge (O(N^3)) but they are opaque.
Fixes #151
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| | |
Since the test for emit constants has moved before the printing of
the instruction the no_fallthrough of the last test should be used.
Bug 20598
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Switch tables were able to screw up the book keeping for emiting
constants in code. Now we estimate the size of an instruction
before printing it by the safe upper bound of 12 for normal
instructions, 1024 for inline assembler and
(2 or 3 + length switch tbl) * 4 for switch tables depending on
thumb etc.
Bug 20598
|
| | |
|
|/ |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
The arm dwarf float registers constants are larger than 2 bytes.
Bug 20489
|
| |
|