search for: s_buf

...= malloc (sizeof (value));\n" n; + pr " if (%s_user_data == NULL) caml_raise_out_of_memory ();\n" n; + pr " *%s_user_data = %sv;\n" n n; + pr " caml_register_generational_global_root (%s_user_data);\n" n; pr " struct nbd_buffer *%s_buf = NBD_buffer_val (%sv);\n" n n; pr " const void *%s = %s_buf->data;\n" n n; - pr " size_t %s = %s_buf->len;\n" count n - | BytesOut (n, count) -> - pr " void *%s = Bytes_val (%sv);\n" n n; - pr " size_t %s = caml_stri...

...well with anothers applications when I force them to use jack as input/output. -> probably not a jack configuration problem. 3) If I kill jackd and I use chan_alsa.so with the real soundcard, it works. -> probably not a network or sip configuration problem. 4) If I replace "f_buf[i] = s_buf[i] * (1.0 / SHRT_MAX);" with "f_buf[i] = 0.5 * sin(0.3454 * ((float) i));" in app_jack.c and I retry the test 2, I get test sound. It looks like no sound was read in channel... Do you have any idea ? Fabien

...pr " char *%s = NULL;\n" n; pr " Py_ssize_t %s;\n" count | BytesPersistIn (n, _) | BytesPersistOut (n, _) -> @@ -279,11 +279,10 @@ let print_python_binding name { args; optargs; ret; may_set_error } = n; pr " struct py_aio_buffer *%s_buf;\n" n | Closure { cbname } -> - pr " struct user_data *%s_user_data = alloc_user_data ();\n" cbname; - pr " if (%s_user_data == NULL) goto out;\n" cbname; + pr " struct user_data *%s_user_data = NULL;\n" cbname; + pr " PyOb...

...- pr " PyObject *py_ret;\n"; + pr " PyObject *py_ret = NULL;\n"; List.iter ( function | Bool n -> pr " int %s;\n" n @@ -279,7 +280,7 @@ let print_python_binding name { args; optargs; ret; may_set_error } = pr " struct py_aio_buffer *%s_buf;\n" n | Closure { cbname } -> pr " struct user_data *%s_user_data = alloc_user_data ();\n" cbname; - pr " if (%s_user_data == NULL) return NULL;\n" cbname; + pr " if (%s_user_data == NULL) goto out;\n" cbname; pr " nbd_...

...+ List.iter (fun cb -> print_callback cb) cbs | _ -> () ) args; @@ -3901,10 +3907,6 @@ let print_python_binding name { args; ret } = pr " PyObject *%s; /* PyCapsule pointing to struct py_aio_buffer */\n" n; pr " struct py_aio_buffer *%s_buf;\n" n - | Callback (n, _) -> - pr " struct %s_%s_data _%s_data, *%s_data = &_%s_data;\n" name n n n n - | CallbackPersist (n, _) -> - pr " struct %s_%s_data *%s_data;\n" name n n | Flags n -> pr " uint32_t %s_u32;\n"...

...n 0;\n"; pr "}\n"; pr "\n" | _ -> () @@ -3966,7 +3952,7 @@ let print_python_binding name { args; ret } = pr " PyObject *%s; /* PyCapsule pointing to struct py_aio_buffer */\n" n; pr " struct py_aio_buffer *%s_buf;\n" n - | Closure (_, { cbname }) -> + | Closure { cbname } -> pr " PyObject *%s_user_data;\n" cbname | Flags n -> pr " uint32_t %s_u32;\n" n; @@ -4032,7 +4018,7 @@ let print_python_binding name { args; ret } = | BytesIn (n, _) | B...

Posting now that I got something to compile (at the expense of breaking OCaml bindings), but I'm open to ideas on how to improve it. Eric Blake (2): generator: Tweak print_c_arg_list to take alternate first arg RFC: generator: Handle shared callbacks in Python generator/generator | 556 ++++++++++++++++++++++---------------------- 1 file changed, 280 insertions(+), 276 deletions(-) --

v1 was here: https://www.redhat.com/archives/libguestfs/2019-August/msg00103.html In v2 I've made several changes: - Fix Python callbacks so if they don't return something which is int-like, we assume they mean to return 0. - Add nbd.Buffer free() method. Read commit message in patch 2 to see what this is about. - Fixed the asynch copy test to deal with the unbelievably

...may_set_error } = (* Functions with a Closure parameter are special because we * have to generate wrapper functions which translate the * callbacks back to Python. @@ -4157,9 +4179,6 @@ let print_python_binding name { args; ret; may_set_error } = pr " struct py_aio_buffer *%s_buf;\n" n | Closure { cbname } -> pr " PyObject *%s_user_data;\n" cbname - | Flags n -> - pr " uint32_t %s_u32;\n" n; - pr " unsigned int %s; /* really uint32_t */\n" n | Int n -> pr " int %s;\n" n | Int64 n...

...| BytesPersistIn _ | BytesPersistOut _ | Callback _ | Flags _ | Int _ | Int64 _ | Path _ | SockAddrAndLen _ | StringList _ | UInt _ | UInt32 _ -> () ) args; @@ -3054,6 +3072,9 @@ let print_python_binding name { args; ret } = pr " struct py_aio_buffer *%s_buf;\n" n | Callback (n, _) -> pr " struct %s_%s_data callback_data;\n" name n + | Flags n -> + pr " uint32_t %s_u32;\n" n; + pr " unsigned int %s = 0; /* really uint32_t */\n" n | Int n -> pr " int %s;\n" n...

...pr " return ret;\n"; pr "}\n"; pr "\n" @@ -3966,7 +3951,7 @@ let print_python_binding name { args; ret } = pr " PyObject *%s; /* PyCapsule pointing to struct py_aio_buffer */\n" n; pr " struct py_aio_buffer *%s_buf;\n" n - | Closure (_, { cbname }) -> + | Closure { cbname } -> pr " PyObject *%s_user_data;\n" cbname | Flags n -> pr " uint32_t %s_u32;\n" n; @@ -4032,7 +4017,7 @@ let print_python_binding name { args; ret } = | BytesIn (n, _) | B...

...+ pr "}\n"; + pr "\n" | _ -> () ) args; @@ -3996,11 +3966,8 @@ let print_python_binding name { args; ret } = pr " PyObject *%s; /* PyCapsule pointing to struct py_aio_buffer */\n" n; pr " struct py_aio_buffer *%s_buf;\n" n - | Closure (false, cls) -> - pr " struct %s_user_data _user_data;\n" name; - pr " struct %s_user_data *user_data = &_user_data;\n" name - | Closure (true, cls) -> - pr " struct %s_user_data *user_data;\n" name + | Clo...

...pr " return ret;\n"; pr "}\n"; pr "\n" @@ -3887,7 +3872,7 @@ let print_python_binding name { args; ret } = pr " PyObject *%s; /* PyCapsule pointing to struct py_aio_buffer */\n" n; pr " struct py_aio_buffer *%s_buf;\n" n - | Closure (_, { cbname }) -> + | Closure { cbname } -> pr " PyObject *%s_user_data;\n" cbname | Flags n -> pr " uint32_t %s_u32;\n" n; @@ -3953,7 +3938,7 @@ let print_python_binding name { args; ret } = | BytesIn (n, _) | B...

As proposed here: https://www.redhat.com/archives/libguestfs/2019-August/msg00130.html I didn't actually read Eric's replies to that yet because I've been concentrating on writing these patches all day. Anyway here they are and I'll look at what Eric said about the proposal next. Rich.

This implements most of what I wrote here: https://www.redhat.com/archives/libguestfs/2019-July/msg00213.html

This applies on top of patches 1 & 2 here (instead of patch 3): https://www.redhat.com/archives/libguestfs/2019-May/msg00206.html https://www.redhat.com/archives/libguestfs/2019-May/msg00207.html Rich.

v1 was here: https://www.redhat.com/archives/libguestfs/2019-July/thread.html#00231 The changes address everything that Eric picked up in his review of the first two patches. I have also added two more patches (4 and 5) which respectively fix docs and change int status -> unsigned status, as discussed. Passes make, check, check-valgrind. Rich.

...pr "PyObject *\n"; pr "nbd_internal_py_%s (PyObject *self, PyObject *args)\n" name; pr "{\n"; @@ -4390,7 +4386,7 @@ let print_python_binding name { args; optargs; ret; may_set_error } = | BytesPersistIn (n, _) | BytesPersistOut (n, _) -> pr ", %s_buf->data, %s_buf->len" n n | Closure { cbname } -> - pr ", %s_%s_wrapper" name cbname; + pr ", %s_wrapper" cbname; pr ", %s_user_data" cbname | Enum (n, _) -> pr ", %s" n | Flags (n, _) -> pr ", %s_u32&...

...er_data;\n" cbname; | OFlags (n, { flag_prefix }) -> pr " uint32_t %s;\n" n; pr " if (%sv != Val_int (0)) /* Some [ list of %s.t ] */\n" @@ -5120,15 +5121,16 @@ let print_ocaml_binding (name, { args; optargs; ret }) = pr " void *%s = %s_buf->data;\n" n n; pr " size_t %s = %s_buf->len;\n" count n | Closure { cbname } -> + pr " nbd_%s_callback %s_callback;\n" cbname cbname; + pr " struct user_data *%s_user_data = alloc_user_data ();\n" cbname; pr " /*...

I think I've addressed everything that was raised in review. Some of the highlights: - Callbacks should be freed reliably along all exit paths. - There's a simple test of closure lifetimes. - I've tried to use VALID|FREE in all the places where I'm confident that it's safe and correct to do. There may be more places. Note this is an optimization and shouldn't