luc14n0/glib
1
0
Fork 0
mirror of https://gitlab.gnome.org/GNOME/glib.git synced 2024-11-06 01:16:17 +01:00
Code Issues Packages Projects Releases Wiki Activity
78b63da382
glib/gio/gmarshal-internal.c

2634 lines
93 KiB
C
Raw Normal View History

gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
/* GObject - GLib Type, Object, Parameter and Signal Library
gio: Add SPDX license headers automatically Add SPDX license (but not copyright) headers to all files which follow a certain pattern in their existing non-machine-readable header comment. This commit was entirely generated using the command: ``` git ls-files gio/*.[ch] | xargs perl -0777 -pi -e 's/\n \*\n \* This library is free software; you can redistribute it and\/or\n \* modify it under the terms of the GNU Lesser General Public/\n \*\n \* SPDX-License-Identifier: LGPL-2.1-or-later\n \*\n \* This library is free software; you can redistribute it and\/or\n \* modify it under the terms of the GNU Lesser General Public/igs' ``` Signed-off-by: Philip Withnall <pwithnall@endlessos.org> Helps: #1415
2022-05-18 10:12:45 +02:00
*
* SPDX-License-Identifier: LGPL-2.1-or-later
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General
* Public License along with this library; if not, see <http://www.gnu.org/licenses/>.
*/
#include "config.h"
#include <glib-object.h>
#include "gmarshal-internal.h"
#ifdef G_ENABLE_DEBUG
#define g_marshal_value_peek_boolean(v) g_value_get_boolean (v)
#define g_marshal_value_peek_char(v) g_value_get_schar (v)
#define g_marshal_value_peek_uchar(v) g_value_get_uchar (v)
#define g_marshal_value_peek_int(v) g_value_get_int (v)
#define g_marshal_value_peek_uint(v) g_value_get_uint (v)
#define g_marshal_value_peek_long(v) g_value_get_long (v)
#define g_marshal_value_peek_ulong(v) g_value_get_ulong (v)
#define g_marshal_value_peek_int64(v) g_value_get_int64 (v)
#define g_marshal_value_peek_uint64(v) g_value_get_uint64 (v)
#define g_marshal_value_peek_enum(v) g_value_get_enum (v)
#define g_marshal_value_peek_flags(v) g_value_get_flags (v)
#define g_marshal_value_peek_float(v) g_value_get_float (v)
#define g_marshal_value_peek_double(v) g_value_get_double (v)
#define g_marshal_value_peek_string(v) (char*) g_value_get_string (v)
#define g_marshal_value_peek_param(v) g_value_get_param (v)
#define g_marshal_value_peek_boxed(v) g_value_get_boxed (v)
#define g_marshal_value_peek_pointer(v) g_value_get_pointer (v)
#define g_marshal_value_peek_object(v) g_value_get_object (v)
#define g_marshal_value_peek_variant(v) g_value_get_variant (v)
#else /* !G_ENABLE_DEBUG */
/* WARNING: This code accesses GValues directly, which is UNSUPPORTED API.
* Do not access GValues directly in your code. Instead, use the
* g_value_get_*() functions
*/
#define g_marshal_value_peek_boolean(v) (v)->data[0].v_int
#define g_marshal_value_peek_char(v) (v)->data[0].v_int
#define g_marshal_value_peek_uchar(v) (v)->data[0].v_uint
#define g_marshal_value_peek_int(v) (v)->data[0].v_int
#define g_marshal_value_peek_uint(v) (v)->data[0].v_uint
#define g_marshal_value_peek_long(v) (v)->data[0].v_long
#define g_marshal_value_peek_ulong(v) (v)->data[0].v_ulong
#define g_marshal_value_peek_int64(v) (v)->data[0].v_int64
#define g_marshal_value_peek_uint64(v) (v)->data[0].v_uint64
#define g_marshal_value_peek_enum(v) (v)->data[0].v_long
#define g_marshal_value_peek_flags(v) (v)->data[0].v_ulong
#define g_marshal_value_peek_float(v) (v)->data[0].v_float
#define g_marshal_value_peek_double(v) (v)->data[0].v_double
#define g_marshal_value_peek_string(v) (v)->data[0].v_pointer
#define g_marshal_value_peek_param(v) (v)->data[0].v_pointer
#define g_marshal_value_peek_boxed(v) (v)->data[0].v_pointer
#define g_marshal_value_peek_pointer(v) (v)->data[0].v_pointer
#define g_marshal_value_peek_object(v) (v)->data[0].v_pointer
#define g_marshal_value_peek_variant(v) (v)->data[0].v_pointer
#endif /* !G_ENABLE_DEBUG */
/* BOOLEAN:OBJECT */
void
_g_cclosure_marshal_BOOLEAN__OBJECT (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 2);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_object (param_values + 1),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__OBJECTv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT callback;
gboolean v_return;
gpointer arg0;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:OBJECT,FLAGS */
void
_g_cclosure_marshal_BOOLEAN__OBJECT_FLAGS (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT_FLAGS) (gpointer data1,
gpointer arg1,
guint arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT_FLAGS callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT_FLAGS) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_flags (param_values + 2),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__OBJECT_FLAGSv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT_FLAGS) (gpointer data1,
gpointer arg1,
guint arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT_FLAGS callback;
gboolean v_return;
gpointer arg0;
guint arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (guint) va_arg (args_copy, guint);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT_FLAGS) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
arg1,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:OBJECT,OBJECT */
void
_g_cclosure_marshal_BOOLEAN__OBJECT_OBJECT (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT_OBJECT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT_OBJECT callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__OBJECT_OBJECTv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__OBJECT_OBJECT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__OBJECT_OBJECT callback;
gboolean v_return;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
arg1,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if (arg1 != NULL)
g_object_unref (arg1);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:POINTER,INT */
void
_g_cclosure_marshal_BOOLEAN__POINTER_INT (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__POINTER_INT) (gpointer data1,
gpointer arg1,
gint arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__POINTER_INT callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__POINTER_INT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_pointer (param_values + 1),
g_marshal_value_peek_int (param_values + 2),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__POINTER_INTv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__POINTER_INT) (gpointer data1,
gpointer arg1,
gint arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__POINTER_INT callback;
gboolean v_return;
gpointer arg0;
gint arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
arg1 = (gint) va_arg (args_copy, gint);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__POINTER_INT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
arg1,
data2);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:STRING */
void
_g_cclosure_marshal_BOOLEAN__STRING (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__STRING) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__STRING callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 2);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__STRING) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_string (param_values + 1),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__STRINGv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__STRING) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__STRING callback;
gboolean v_return;
gpointer arg0;
va_list args_copy;
gmarshal-internal: Fix string leakage due to misplaced check
2021-10-11 12:59:05 +02:00
g_return_if_fail (return_value != NULL);
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__STRING) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:UINT */
void
_g_cclosure_marshal_BOOLEAN__UINT (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__UINT) (gpointer data1,
guint arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__UINT callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 2);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__UINT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_uint (param_values + 1),
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__UINTv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__UINT) (gpointer data1,
guint arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__UINT callback;
gboolean v_return;
guint arg0;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (guint) va_arg (args_copy, guint);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__UINT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
data2);
g_value_set_boolean (return_value, v_return);
}
/* BOOLEAN:VOID */
void
_g_cclosure_marshal_BOOLEAN__VOID (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__VOID) (gpointer data1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__VOID callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 1);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__VOID) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
data2);
g_value_set_boolean (return_value, v_return);
}
void
_g_cclosure_marshal_BOOLEAN__VOIDv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gboolean (*GMarshalFunc_BOOLEAN__VOID) (gpointer data1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_BOOLEAN__VOID callback;
gboolean v_return;
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_BOOLEAN__VOID) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
data2);
g_value_set_boolean (return_value, v_return);
}
/* INT:BOXED */
void
_g_cclosure_marshal_INT__BOXED (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gint (*GMarshalFunc_INT__BOXED) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_INT__BOXED callback;
gint v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 2);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_INT__BOXED) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_boxed (param_values + 1),
data2);
g_value_set_int (return_value, v_return);
}
void
_g_cclosure_marshal_INT__BOXEDv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gint (*GMarshalFunc_INT__BOXED) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_INT__BOXED callback;
gint v_return;
gpointer arg0;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_boxed_copy (param_types[0] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg0);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_INT__BOXED) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_boxed_free (param_types[0] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg0);
g_value_set_int (return_value, v_return);
}
/* INT:OBJECT */
void
_g_cclosure_marshal_INT__OBJECT (GClosure *closure,
GValue *return_value,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef gint (*GMarshalFunc_INT__OBJECT) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_INT__OBJECT callback;
gint v_return;
g_return_if_fail (return_value != NULL);
g_return_if_fail (n_param_values == 2);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_INT__OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
g_marshal_value_peek_object (param_values + 1),
data2);
g_value_set_int (return_value, v_return);
}
void
_g_cclosure_marshal_INT__OBJECTv (GClosure *closure,
GValue *return_value,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef gint (*GMarshalFunc_INT__OBJECT) (gpointer data1,
gpointer arg1,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_INT__OBJECT callback;
gint v_return;
gpointer arg0;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
va_end (args_copy);
g_return_if_fail (return_value != NULL);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_INT__OBJECT) (marshal_data ? marshal_data : cc->callback);
v_return = callback (data1,
arg0,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
g_value_set_int (return_value, v_return);
}
/* VOID:BOOLEAN,BOXED */
void
_g_cclosure_marshal_VOID__BOOLEAN_BOXED (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__BOOLEAN_BOXED) (gpointer data1,
gboolean arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__BOOLEAN_BOXED callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__BOOLEAN_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_boolean (param_values + 1),
g_marshal_value_peek_boxed (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__BOOLEAN_BOXEDv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__BOOLEAN_BOXED) (gpointer data1,
gboolean arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__BOOLEAN_BOXED callback;
gboolean arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gboolean) va_arg (args_copy, gboolean);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_boxed_copy (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__BOOLEAN_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_boxed_free (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
}
/* VOID:ENUM,OBJECT */
void
_g_cclosure_marshal_VOID__ENUM_OBJECT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__ENUM_OBJECT) (gpointer data1,
gint arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__ENUM_OBJECT callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__ENUM_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_enum (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__ENUM_OBJECTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__ENUM_OBJECT) (gpointer data1,
gint arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__ENUM_OBJECT callback;
gint arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gint) va_arg (args_copy, gint);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__ENUM_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if (arg1 != NULL)
g_object_unref (arg1);
}
/* VOID:ENUM,OBJECT,OBJECT */
void
_g_cclosure_marshal_VOID__ENUM_OBJECT_OBJECT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__ENUM_OBJECT_OBJECT) (gpointer data1,
gint arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__ENUM_OBJECT_OBJECT callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__ENUM_OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_enum (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
g_marshal_value_peek_object (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__ENUM_OBJECT_OBJECTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__ENUM_OBJECT_OBJECT) (gpointer data1,
gint arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__ENUM_OBJECT_OBJECT callback;
gint arg0;
gpointer arg1;
gpointer arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gint) va_arg (args_copy, gint);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if (arg2 != NULL)
arg2 = g_object_ref (arg2);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__ENUM_OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if (arg1 != NULL)
g_object_unref (arg1);
if (arg2 != NULL)
g_object_unref (arg2);
}
/* VOID:INT,INT,INT */
void
_g_cclosure_marshal_VOID__INT_INT_INT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__INT_INT_INT) (gpointer data1,
gint arg1,
gint arg2,
gint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__INT_INT_INT callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__INT_INT_INT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_int (param_values + 1),
g_marshal_value_peek_int (param_values + 2),
g_marshal_value_peek_int (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__INT_INT_INTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__INT_INT_INT) (gpointer data1,
gint arg1,
gint arg2,
gint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__INT_INT_INT callback;
gint arg0;
gint arg1;
gint arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gint) va_arg (args_copy, gint);
arg1 = (gint) va_arg (args_copy, gint);
arg2 = (gint) va_arg (args_copy, gint);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__INT_INT_INT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
}
/* VOID:OBJECT,OBJECT */
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__OBJECT_OBJECTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT callback;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if (arg1 != NULL)
g_object_unref (arg1);
}
/* VOID:OBJECT,OBJECT,ENUM */
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_ENUM (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_ENUM) (gpointer data1,
gpointer arg1,
gpointer arg2,
gint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_ENUM callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_ENUM) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
g_marshal_value_peek_enum (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_ENUMv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_ENUM) (gpointer data1,
gpointer arg1,
gpointer arg2,
gint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_ENUM callback;
gpointer arg0;
gpointer arg1;
gint arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
arg2 = (gint) va_arg (args_copy, gint);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_ENUM) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if (arg1 != NULL)
g_object_unref (arg1);
}
/* VOID:OBJECT,OBJECT,STRING,STRING,VARIANT */
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer arg4,
gpointer arg5,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT callback;
g_return_if_fail (n_param_values == 6);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
g_marshal_value_peek_string (param_values + 3),
g_marshal_value_peek_string (param_values + 4),
g_marshal_value_peek_variant (param_values + 5),
data2);
}
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_STRING_STRING_VARIANTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer arg4,
gpointer arg5,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT callback;
gpointer arg0;
gpointer arg1;
gpointer arg2;
gpointer arg3;
gpointer arg4;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_strdup (arg2);
arg3 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[3] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg3 != NULL)
arg3 = g_strdup (arg3);
arg4 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[4] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg4 != NULL)
arg4 = g_variant_ref_sink (arg4);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_STRING_STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
arg3,
arg4,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if (arg1 != NULL)
g_object_unref (arg1);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_free (arg2);
if ((param_types[3] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg3 != NULL)
g_free (arg3);
if ((param_types[4] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg4 != NULL)
g_variant_unref (arg4);
}
/* VOID:OBJECT,OBJECT,VARIANT,BOXED */
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_VARIANT_BOXED (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer arg4,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED callback;
g_return_if_fail (n_param_values == 5);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_object (param_values + 2),
g_marshal_value_peek_variant (param_values + 3),
g_marshal_value_peek_boxed (param_values + 4),
data2);
}
void
_g_cclosure_marshal_VOID__OBJECT_OBJECT_VARIANT_BOXEDv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer arg4,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED callback;
gpointer arg0;
gpointer arg1;
gpointer arg2;
gpointer arg3;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if (arg1 != NULL)
arg1 = g_object_ref (arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_variant_ref_sink (arg2);
arg3 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[3] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg3 != NULL)
arg3 = g_boxed_copy (param_types[3] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg3);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_OBJECT_VARIANT_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
arg3,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if (arg1 != NULL)
g_object_unref (arg1);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_variant_unref (arg2);
if ((param_types[3] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg3 != NULL)
g_boxed_free (param_types[3] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg3);
}
/* VOID:OBJECT,VARIANT */
void
_g_cclosure_marshal_VOID__OBJECT_VARIANT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__OBJECT_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_VARIANT callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_object (param_values + 1),
g_marshal_value_peek_variant (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__OBJECT_VARIANTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__OBJECT_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__OBJECT_VARIANT callback;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if (arg0 != NULL)
arg0 = g_object_ref (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_variant_ref_sink (arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__OBJECT_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if (arg0 != NULL)
g_object_unref (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_variant_unref (arg1);
}
/* VOID:POINTER,INT,STRING */
void
_g_cclosure_marshal_VOID__POINTER_INT_STRING (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__POINTER_INT_STRING) (gpointer data1,
gpointer arg1,
gint arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__POINTER_INT_STRING callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__POINTER_INT_STRING) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_pointer (param_values + 1),
g_marshal_value_peek_int (param_values + 2),
g_marshal_value_peek_string (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__POINTER_INT_STRINGv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__POINTER_INT_STRING) (gpointer data1,
gpointer arg1,
gint arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__POINTER_INT_STRING callback;
gpointer arg0;
gint arg1;
gpointer arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
arg1 = (gint) va_arg (args_copy, gint);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_strdup (arg2);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__POINTER_INT_STRING) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_free (arg2);
}
/* VOID:STRING,BOOLEAN */
void
_g_cclosure_marshal_VOID__STRING_BOOLEAN (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_BOOLEAN) (gpointer data1,
gpointer arg1,
gboolean arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOOLEAN callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOOLEAN) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_boolean (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_BOOLEANv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_BOOLEAN) (gpointer data1,
gpointer arg1,
gboolean arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOOLEAN callback;
gpointer arg0;
gboolean arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gboolean) va_arg (args_copy, gboolean);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOOLEAN) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
}
/* VOID:STRING,BOXED */
void
_g_cclosure_marshal_VOID__STRING_BOXED (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOXED callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_boxed (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_BOXEDv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOXED callback;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_boxed_copy (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_boxed_free (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
}
/* VOID:STRING,BOXED,BOXED */
void
_g_cclosure_marshal_VOID__STRING_BOXED_BOXED (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_BOXED_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOXED_BOXED callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOXED_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_boxed (param_values + 2),
g_marshal_value_peek_boxed (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_BOXED_BOXEDv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_BOXED_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_BOXED_BOXED callback;
gpointer arg0;
gpointer arg1;
gpointer arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_boxed_copy (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_boxed_copy (param_types[2] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg2);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_BOXED_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_boxed_free (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_boxed_free (param_types[2] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg2);
}
/* VOID:STRING,INT64,INT64 */
void
_g_cclosure_marshal_VOID__STRING_INT64_INT64 (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_INT64_INT64) (gpointer data1,
gpointer arg1,
gint64 arg2,
gint64 arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_INT64_INT64 callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_INT64_INT64) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_int64 (param_values + 2),
g_marshal_value_peek_int64 (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_INT64_INT64v (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_INT64_INT64) (gpointer data1,
gpointer arg1,
gint64 arg2,
gint64 arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_INT64_INT64 callback;
gpointer arg0;
gint64 arg1;
gint64 arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gint64) va_arg (args_copy, gint64);
arg2 = (gint64) va_arg (args_copy, gint64);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_INT64_INT64) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
}
/* VOID:STRING,STRING,STRING,FLAGS */
void
_g_cclosure_marshal_VOID__STRING_STRING_STRING_FLAGS (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
guint arg4,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS callback;
g_return_if_fail (n_param_values == 5);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_string (param_values + 2),
g_marshal_value_peek_string (param_values + 3),
g_marshal_value_peek_flags (param_values + 4),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_STRING_STRING_FLAGSv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
guint arg4,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS callback;
gpointer arg0;
gpointer arg1;
gpointer arg2;
guint arg3;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_strdup (arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_strdup (arg2);
arg3 = (guint) va_arg (args_copy, guint);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_STRING_STRING_FLAGS) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
arg3,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_free (arg1);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_free (arg2);
}
/* VOID:STRING,STRING,VARIANT */
void
_g_cclosure_marshal_VOID__STRING_STRING_VARIANT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_STRING_VARIANT callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_string (param_values + 2),
g_marshal_value_peek_variant (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_STRING_VARIANTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_STRING_VARIANT callback;
gpointer arg0;
gpointer arg1;
gpointer arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_strdup (arg1);
arg2 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
arg2 = g_variant_ref_sink (arg2);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_free (arg1);
if ((param_types[2] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg2 != NULL)
g_variant_unref (arg2);
}
/* VOID:STRING,VARIANT */
void
_g_cclosure_marshal_VOID__STRING_VARIANT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_VARIANT callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_string (param_values + 1),
g_marshal_value_peek_variant (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__STRING_VARIANTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__STRING_VARIANT) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__STRING_VARIANT callback;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_strdup (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_variant_ref_sink (arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__STRING_VARIANT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_free (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_variant_unref (arg1);
}
/* VOID:UINT,UINT,UINT */
void
_g_cclosure_marshal_VOID__UINT_UINT_UINT (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__UINT_UINT_UINT) (gpointer data1,
guint arg1,
guint arg2,
guint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__UINT_UINT_UINT callback;
g_return_if_fail (n_param_values == 4);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__UINT_UINT_UINT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_uint (param_values + 1),
g_marshal_value_peek_uint (param_values + 2),
g_marshal_value_peek_uint (param_values + 3),
data2);
}
void
_g_cclosure_marshal_VOID__UINT_UINT_UINTv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__UINT_UINT_UINT) (gpointer data1,
guint arg1,
guint arg2,
guint arg3,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__UINT_UINT_UINT callback;
guint arg0;
guint arg1;
guint arg2;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (guint) va_arg (args_copy, guint);
arg1 = (guint) va_arg (args_copy, guint);
arg2 = (guint) va_arg (args_copy, guint);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__UINT_UINT_UINT) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
arg2,
data2);
}
/* VOID:VARIANT,BOXED */
void
_g_cclosure_marshal_VOID__VARIANT_BOXED (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
guint n_param_values,
const GValue *param_values,
gpointer invocation_hint G_GNUC_UNUSED,
gpointer marshal_data)
{
typedef void (*GMarshalFunc_VOID__VARIANT_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__VARIANT_BOXED callback;
g_return_if_fail (n_param_values == 3);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = g_value_peek_pointer (param_values + 0);
}
else
{
data1 = g_value_peek_pointer (param_values + 0);
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__VARIANT_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
g_marshal_value_peek_variant (param_values + 1),
g_marshal_value_peek_boxed (param_values + 2),
data2);
}
void
_g_cclosure_marshal_VOID__VARIANT_BOXEDv (GClosure *closure,
GValue *return_value G_GNUC_UNUSED,
gpointer instance,
va_list args,
gpointer marshal_data,
int n_params,
GType *param_types)
{
typedef void (*GMarshalFunc_VOID__VARIANT_BOXED) (gpointer data1,
gpointer arg1,
gpointer arg2,
gpointer data2);
GCClosure *cc = (GCClosure *) closure;
gpointer data1, data2;
GMarshalFunc_VOID__VARIANT_BOXED callback;
gpointer arg0;
gpointer arg1;
va_list args_copy;
Replace all uses of G_VA_COPY with Standard C va_copy Signed-off-by: Simon McVittie <smcv@collabora.com>
2022-10-21 15:48:47 +02:00
va_copy (args_copy, args);
gio: remove use of generic marshaller from GIO objects Using the generic marshaller has drawbacks beyond performance. One such drawback is that it breaks the stack unwinding from the Linux kernel due to having unsufficient data to walk past ffi_call_unixt64. That means that performance profiling by application developers looks grouped among seemingly unrelated code paths. While we can't fix the kernel unwinding here, we can provide proper c_marshallers and va_marshallers for objects within Gio so that performance profiling of applications is more reliable. Related to GNOME/Initiatives#10
2019-05-31 04:29:18 +02:00
arg0 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
arg0 = g_variant_ref_sink (arg0);
arg1 = (gpointer) va_arg (args_copy, gpointer);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
arg1 = g_boxed_copy (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
va_end (args_copy);
if (G_CCLOSURE_SWAP_DATA (closure))
{
data1 = closure->data;
data2 = instance;
}
else
{
data1 = instance;
data2 = closure->data;
}
callback = (GMarshalFunc_VOID__VARIANT_BOXED) (marshal_data ? marshal_data : cc->callback);
callback (data1,
arg0,
arg1,
data2);
if ((param_types[0] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg0 != NULL)
g_variant_unref (arg0);
if ((param_types[1] & G_SIGNAL_TYPE_STATIC_SCOPE) == 0 && arg1 != NULL)
g_boxed_free (param_types[1] & ~G_SIGNAL_TYPE_STATIC_SCOPE, arg1);
}
Reference in New Issue Copy Permalink