diff options
author | Simon Marchi <simon.marchi@ericsson.com> | 2018-05-20 21:06:03 -0400 |
---|---|---|
committer | Simon Marchi <simon.marchi@polymtl.ca> | 2018-05-20 21:06:36 -0400 |
commit | 284a0e3cbffa92ee5c94065fcdd5a528482344fc (patch) | |
tree | 8b7ad3a53e77adf130a5f33ed306fe76d4ab0ff8 /gdb/common | |
parent | b1aed5de0cc075ab5061453f88cedbf2d3e97ff6 (diff) | |
download | gdb-284a0e3cbffa92ee5c94065fcdd5a528482344fc.zip gdb-284a0e3cbffa92ee5c94065fcdd5a528482344fc.tar.gz gdb-284a0e3cbffa92ee5c94065fcdd5a528482344fc.tar.bz2 |
Introduce obstack_new, poison other "typed" obstack functions
Since we use obstacks with objects that are not default constructible,
we sometimes need to manually call the constructor by hand using
placement new:
foo *f = obstack_alloc (obstack, sizeof (foo));
f = new (f) foo;
It's possible to use allocate_on_obstack instead, but there are types
that we sometimes want to allocate on an obstack, and sometimes on the
regular heap. This patch introduces a utility to make this pattern
simpler if allocate_on_obstack is not an option:
foo *f = obstack_new<foo> (obstack);
Right now there's only one usage (in tdesc_data_init).
To help catch places where we would forget to call new when allocating
such an object on an obstack, this patch also poisons some other methods
of allocating an instance of a type on an obstack:
- OBSTACK_ZALLOC/OBSTACK_CALLOC
- XOBNEW/XOBNEW
- GDBARCH_OBSTACK_ZALLOC/GDBARCH_OBSTACK_CALLOC
Unfortunately, there's no way to catch wrong usages of obstack_alloc.
By pulling on that string though, it tripped on allocating struct
template_symbol using OBSTACK_ZALLOC. The criterion currently used to
know whether it's safe to "malloc" an instance of a struct is whether it
is a POD. Because it inherits from struct symbol, template_symbol is
not a POD. This criterion is a bit too strict however, it should still
safe to allocate memory for a template_symbol and memset it to 0. We
didn't use is_trivially_constructible as the criterion in the first
place only because it is not available in gcc < 5. So here I considered
two alternatives:
1. Relax that criterion to use std::is_trivially_constructible and add a
bit more glue code to make it work with gcc < 5
2. Continue pulling on the string and change how the symbol structures
are allocated and initialized
I managed to do both, but I decided to go with #1 to keep this patch
simpler and more focused. When building with a compiler that does not
have is_trivially_constructible, the check will just not be enforced.
gdb/ChangeLog:
* common/traits.h (HAVE_IS_TRIVIALLY_COPYABLE): Define if
compiler supports std::is_trivially_constructible.
* common/poison.h: Include obstack.h.
(IsMallocable): Define to is_trivially_constructible if the
compiler supports it, define to true_type otherwise.
(xobnew): New.
(XOBNEW): Redefine.
(xobnewvec): New.
(XOBNEWVEC): Redefine.
* gdb_obstack.h (obstack_zalloc): New.
(OBSTACK_ZALLOC): Redefine.
(obstack_calloc): New.
(OBSTACK_CALLOC): Redefine.
(obstack_new): New.
* gdbarch.sh: Include gdb_obstack in gdbarch.h.
(gdbarch_obstack): New declaration in gdbarch.h, definition in
gdbarch.c.
(GDBARCH_OBSTACK_CALLOC, GDBARCH_OBSTACK_ZALLOC): Use
obstack_calloc/obstack_zalloc.
(gdbarch_obstack_zalloc): Remove.
* target-descriptions.c (tdesc_data_init): Use obstack_new.
Diffstat (limited to 'gdb/common')
-rw-r--r-- | gdb/common/poison.h | 31 | ||||
-rw-r--r-- | gdb/common/traits.h | 8 |
2 files changed, 38 insertions, 1 deletions
diff --git a/gdb/common/poison.h b/gdb/common/poison.h index c98d2b3..ddab2c1 100644 --- a/gdb/common/poison.h +++ b/gdb/common/poison.h @@ -21,6 +21,7 @@ #define COMMON_POISON_H #include "traits.h" +#include "obstack.h" /* Poison memset of non-POD types. The idea is catching invalid initialization of non-POD structs that is easy to be introduced as @@ -88,7 +89,11 @@ void *memmove (D *dest, const S *src, size_t n) = delete; objects that require new/delete. */ template<typename T> -using IsMallocable = std::is_pod<T>; +#if HAVE_IS_TRIVIALLY_CONSTRUCTIBLE +using IsMallocable = std::is_trivially_constructible<T>; +#else +using IsMallocable = std::true_type; +#endif template<typename T> using IsFreeable = gdb::Or<std::is_trivially_destructible<T>, std::is_void<T>>; @@ -216,4 +221,28 @@ non-POD data type."); #undef XRESIZEVAR #define XRESIZEVAR(T, P, S) xresizevar<T> (P, S) +template<typename T> +static T * +xobnew (obstack *ob) +{ + static_assert (IsMallocable<T>::value, "Trying to use XOBNEW with a \ +non-POD data type."); + return XOBNEW (ob, T); +} + +#undef XOBNEW +#define XOBNEW(O, T) xobnew<T> (O) + +template<typename T> +static T * +xobnewvec (obstack *ob, size_t n) +{ + static_assert (IsMallocable<T>::value, "Trying to use XOBNEWVEC with a \ +non-POD data type."); + return XOBNEWVEC (ob, T, n); +} + +#undef XOBNEWVEC +#define XOBNEWVEC(O, T, N) xobnewvec<T> (O, N) + #endif /* COMMON_POISON_H */ diff --git a/gdb/common/traits.h b/gdb/common/traits.h index d9e6839..070ef15 100644 --- a/gdb/common/traits.h +++ b/gdb/common/traits.h @@ -33,6 +33,14 @@ # define HAVE_IS_TRIVIALLY_COPYABLE 1 #endif +/* HAVE_IS_TRIVIALLY_CONSTRUCTIBLE is defined as 1 iff + std::is_trivially_constructible is available. GCC only implemented it + in GCC 5. */ +#if (__has_feature(is_trivially_constructible) \ + || (defined __GNUC__ && __GNUC__ >= 5)) +# define HAVE_IS_TRIVIALLY_COPYABLE 1 +#endif + namespace gdb { /* Pre C++14-safe (CWG 1558) version of C++17's std::void_t. See |