search for: safe_memdup

...@@ -5107,7 +5109,7 @@ and generate_client_actions () = #define error guestfs_error //#define perrorf guestfs_perrorf -//#define safe_malloc guestfs_safe_malloc +#define safe_malloc guestfs_safe_malloc #define safe_realloc guestfs_safe_realloc //#define safe_strdup guestfs_safe_strdup #define safe_memdup guestfs_safe_memdup @@ -5396,8 +5398,20 @@ check_state (guestfs_h *g, const char *caller) pr " /* caller will free this */\n"; pr " return safe_memdup (g, &ret.%s, sizeof (ret.%s));\n" n n | RBufferOut n -> - pr " *size_r...

...deletions(-) diff --git a/src/generator.ml b/src/generator.ml index 2317541..c261ea2 100644 --- a/src/generator.ml +++ b/src/generator.ml @@ -5398,20 +5398,20 @@ check_state (guestfs_h *g, const char *caller) pr " /* caller will free this */\n"; pr " return safe_memdup (g, &ret.%s, sizeof (ret.%s));\n" n n | RBufferOut n -> - pr " /* RBufferOut is tricky: If the buffer is zero-length, then\n"; - pr " * _val might be NULL here. To make the API saner for\n"; - pr " * callers, we turn this case into a uniqu...

OK, I'll try to fix the epoch thing for all three. Nikos On 17 November 2014 13:49, Richard W.M. Jones <rjones@redhat.com> wrote: > On Mon, Nov 17, 2014 at 01:41:28PM +0200, Nikos Skalkotos wrote: >> OK, I'll make the suggested changes and I'll try to come up with a new >> patch by tomorrow or the day after tomorrow. I just noticed that a >> space is

The safe_malloc (etc) functions call g->abort_fn on failure. That's not appropriate for language bindings, and we never intended that these internal functions be used from language bindings, that was just a historical accident. This patch series removes any external use of the safe_* functions. Rich.

...free { + struct deferred_free *next; + void (*freefn) (void *); + void *data; +}; +#endif + /* Per-filesystem data stored for inspect_os. */ enum inspect_os_format { OS_FORMAT_UNKNOWN = 0, @@ -474,6 +499,14 @@ extern char *guestfs___safe_asprintf (guestfs_h *g, const char *fs, ...) #define safe_memdup guestfs___safe_memdup #define safe_asprintf guestfs___safe_asprintf +/* These functions are used internally by the CLEANUP_* macros. + * Don't call them directly. + */ +extern void guestfs___cleanup_free (void *ptr); +#ifndef HAVE_ATTRIBUTE_CLEANUP +extern void guestfs___defer_free (guestfs_...

This rather complex set of patches allow non-optargs functions to gain optional arguments, while preserving source and binary backwards compatibility. The problem is that we cannot add an optional argument to an existing function. For example, we might want to add flags to the 'lvresize' API which currently has no optional arguments.

This patch series changes a small part of the library to use __attribute__((cleanup(...))) to automatically free memory when pointers go out of the current scope. In general terms this seems to be a small win although you do have to use it carefully. For functions where you can completely get rid of the "exit code paths", it can simplify things. For a good example, see the

Inspired by libvirt's virCommand* internal mini-library, this adds some internal APIs for running commands. The first patch contains the new APIs. The subsequent patches change various parts of the library over to use it. Rich.

...et.%s.%s_len + 1));\n" + n n; + pr " ctx.ret.%s.%s_val[ctx.ret.%s.%s_len] = NULL;\n" n n n n; + pr " return ctx.ret.%s.%s_val;\n" n n | RStruct (n, _) -> - pr " /* caller will free this */\n"; - pr " return safe_memdup (g, &ctx.ret.%s, sizeof (ctx.ret.%s));\n" n n + pr " /* caller will free this */\n"; + pr " return safe_memdup (g, &ctx.ret.%s, sizeof (ctx.ret.%s));\n" n n | RStructList (n, _) -> - pr " /* caller will free this */\n";...

libguestfs has used double and triple underscores in identifiers. These aren't valid for global names in C++. (http://stackoverflow.com/a/228797) These large but completely mechanical patches change the illegal identifiers to legal ones. Rich.