intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
/*
|
|
|
|
|
* Copyright © 2017 Google, Inc.
|
|
|
|
|
*
|
|
|
|
|
* Permission is hereby granted, free of charge, to any person obtaining a
|
|
|
|
|
* copy of this software and associated documentation files (the "Software"),
|
|
|
|
|
* to deal in the Software without restriction, including without limitation
|
|
|
|
|
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
|
|
|
|
* and/or sell copies of the Software, and to permit persons to whom the
|
|
|
|
|
* Software is furnished to do so, subject to the following conditions:
|
|
|
|
|
*
|
|
|
|
|
* The above copyright notice and this permission notice (including the next
|
|
|
|
|
* paragraph) shall be included in all copies or substantial portions of the
|
|
|
|
|
* Software.
|
|
|
|
|
*
|
|
|
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
|
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
|
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
|
|
|
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
|
|
|
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
|
|
|
|
|
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
|
|
|
|
|
* IN THE SOFTWARE.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include <stdarg.h>
|
|
|
|
|
#include <stdio.h>
|
2021-06-16 10:20:35 -07:00
|
|
|
#include <stdlib.h>
|
|
|
|
|
#include <string.h>
|
2023-02-17 15:17:26 -08:00
|
|
|
#include "c11/threads.h"
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
#include "util/detect_os.h"
|
|
|
|
|
#include "util/log.h"
|
2021-06-16 10:20:35 -07:00
|
|
|
#include "util/ralloc.h"
|
2023-02-17 15:17:26 -08:00
|
|
|
#include "util/u_debug.h"
|
|
|
|
|
|
2024-06-06 01:19:35 +08:00
|
|
|
#if DETECT_OS_POSIX
|
2023-02-17 16:15:21 -08:00
|
|
|
#include <syslog.h>
|
|
|
|
|
#include "util/u_process.h"
|
|
|
|
|
#endif
|
|
|
|
|
|
2023-02-17 15:17:26 -08:00
|
|
|
#if DETECT_OS_ANDROID
|
|
|
|
|
#include <android/log.h>
|
|
|
|
|
#endif
|
|
|
|
|
|
2023-02-17 17:44:29 -08:00
|
|
|
#if DETECT_OS_WINDOWS
|
|
|
|
|
#include <windows.h>
|
|
|
|
|
#endif
|
|
|
|
|
|
2023-02-17 15:17:26 -08:00
|
|
|
enum mesa_log_control {
|
|
|
|
|
MESA_LOG_CONTROL_NULL = 1 << 0,
|
|
|
|
|
MESA_LOG_CONTROL_FILE = 1 << 1,
|
2023-02-17 16:15:21 -08:00
|
|
|
MESA_LOG_CONTROL_SYSLOG = 1 << 2,
|
|
|
|
|
MESA_LOG_CONTROL_ANDROID = 1 << 3,
|
2023-02-17 17:44:29 -08:00
|
|
|
MESA_LOG_CONTROL_WINDBG = 1 << 4,
|
2023-02-17 15:17:26 -08:00
|
|
|
MESA_LOG_CONTROL_LOGGER_MASK = 0xff,
|
2023-02-17 15:22:17 -08:00
|
|
|
|
|
|
|
|
MESA_LOG_CONTROL_WAIT = 1 << 8,
|
2023-02-17 15:17:26 -08:00
|
|
|
};
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
|
2023-02-17 15:17:26 -08:00
|
|
|
static const struct debug_control mesa_log_control_options[] = {
|
|
|
|
|
/* loggers */
|
|
|
|
|
{ "null", MESA_LOG_CONTROL_NULL },
|
|
|
|
|
{ "file", MESA_LOG_CONTROL_FILE },
|
2023-02-17 16:15:21 -08:00
|
|
|
{ "syslog", MESA_LOG_CONTROL_SYSLOG },
|
2023-02-17 15:17:26 -08:00
|
|
|
{ "android", MESA_LOG_CONTROL_ANDROID },
|
2023-02-17 17:44:29 -08:00
|
|
|
{ "windbg", MESA_LOG_CONTROL_WINDBG },
|
2023-02-17 15:22:17 -08:00
|
|
|
/* flags */
|
|
|
|
|
{ "wait", MESA_LOG_CONTROL_WAIT },
|
2023-02-17 15:17:26 -08:00
|
|
|
{ NULL, 0 },
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
static uint32_t mesa_log_control;
|
2023-02-17 17:37:27 -08:00
|
|
|
static FILE *mesa_log_file;
|
2023-02-17 15:17:26 -08:00
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
mesa_log_init_once(void)
|
|
|
|
|
{
|
|
|
|
|
mesa_log_control = parse_debug_string(os_get_option("MESA_LOG"),
|
|
|
|
|
mesa_log_control_options);
|
|
|
|
|
|
|
|
|
|
if (!(mesa_log_control & MESA_LOG_CONTROL_LOGGER_MASK)) {
|
|
|
|
|
/* pick the default loggers */
|
|
|
|
|
#if DETECT_OS_ANDROID
|
|
|
|
|
mesa_log_control |= MESA_LOG_CONTROL_ANDROID;
|
|
|
|
|
#else
|
|
|
|
|
mesa_log_control |= MESA_LOG_CONTROL_FILE;
|
|
|
|
|
#endif
|
2023-02-17 17:44:29 -08:00
|
|
|
|
|
|
|
|
#if DETECT_OS_WINDOWS
|
|
|
|
|
/* stderr from windows applications without console is not usually
|
|
|
|
|
* visible, so communicate with the debugger instead */
|
|
|
|
|
mesa_log_control |= MESA_LOG_CONTROL_WINDBG;
|
|
|
|
|
#endif
|
2023-02-17 15:17:26 -08:00
|
|
|
}
|
2023-02-17 16:15:21 -08:00
|
|
|
|
2023-02-17 17:37:27 -08:00
|
|
|
mesa_log_file = stderr;
|
|
|
|
|
|
|
|
|
|
#if !DETECT_OS_WINDOWS
|
2024-01-29 12:42:18 +00:00
|
|
|
if (__normal_user()) {
|
2023-02-17 17:37:27 -08:00
|
|
|
const char *log_file = os_get_option("MESA_LOG_FILE");
|
|
|
|
|
if (log_file) {
|
|
|
|
|
FILE *fp = fopen(log_file, "w");
|
|
|
|
|
if (fp) {
|
|
|
|
|
mesa_log_file = fp;
|
|
|
|
|
mesa_log_control |= MESA_LOG_CONTROL_FILE;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
|
2024-06-06 01:19:35 +08:00
|
|
|
#if DETECT_OS_POSIX
|
2023-02-17 16:15:21 -08:00
|
|
|
if (mesa_log_control & MESA_LOG_CONTROL_SYSLOG)
|
|
|
|
|
openlog(util_get_process_name(), LOG_NDELAY | LOG_PID, LOG_USER);
|
|
|
|
|
#endif
|
2023-02-17 15:17:26 -08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
mesa_log_init(void)
|
|
|
|
|
{
|
|
|
|
|
static once_flag once = ONCE_FLAG_INIT;
|
|
|
|
|
call_once(&once, mesa_log_init_once);
|
|
|
|
|
}
|
2023-02-17 14:58:02 -08:00
|
|
|
|
2024-07-03 19:08:10 +10:00
|
|
|
void
|
|
|
|
|
mesa_log_if_debug(enum mesa_log_level level, const char *outputString)
|
|
|
|
|
{
|
|
|
|
|
static int debug = -1;
|
|
|
|
|
|
|
|
|
|
/* Init the local 'debug' var once. */
|
|
|
|
|
if (debug == -1) {
|
|
|
|
|
const char *env = getenv("MESA_DEBUG");
|
|
|
|
|
bool silent = env && strstr(env, "silent") != NULL;
|
|
|
|
|
#ifndef NDEBUG
|
|
|
|
|
/* in debug builds, print messages unless MESA_DEBUG="silent" */
|
|
|
|
|
if (silent)
|
|
|
|
|
debug = 0;
|
|
|
|
|
else
|
|
|
|
|
debug = 1;
|
|
|
|
|
#else
|
|
|
|
|
/* in release builds, print messages if any MESA_DEBUG value other than
|
|
|
|
|
* MESA_DEBUG="silent" is set
|
|
|
|
|
*/
|
|
|
|
|
debug = env && !silent;
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Now only print the string if we're required to do so. */
|
|
|
|
|
if (debug)
|
|
|
|
|
mesa_log(level, "Mesa", "%s", outputString);
|
|
|
|
|
}
|
|
|
|
|
|
2023-02-17 14:58:02 -08:00
|
|
|
static inline const char *
|
|
|
|
|
level_to_str(enum mesa_log_level l)
|
|
|
|
|
{
|
|
|
|
|
switch (l) {
|
|
|
|
|
case MESA_LOG_ERROR: return "error";
|
|
|
|
|
case MESA_LOG_WARN: return "warning";
|
|
|
|
|
case MESA_LOG_INFO: return "info";
|
|
|
|
|
case MESA_LOG_DEBUG: return "debug";
|
|
|
|
|
}
|
|
|
|
|
|
2025-07-23 09:17:35 +02:00
|
|
|
UNREACHABLE("bad mesa_log_level");
|
2023-02-17 14:58:02 -08:00
|
|
|
}
|
|
|
|
|
|
2023-02-21 10:02:35 -08:00
|
|
|
enum logger_vasnprintf_affix {
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_TAG = 1 << 0,
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_LEVEL = 1 << 1,
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_NEWLINE = 1 << 2,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/* Try vsnprintf first and fall back to vasprintf if buf is too small. This
|
|
|
|
|
* function handles all errors and never fails.
|
|
|
|
|
*/
|
|
|
|
|
static char *
|
|
|
|
|
logger_vasnprintf(char *buf,
|
|
|
|
|
int size,
|
|
|
|
|
int affixes,
|
|
|
|
|
enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
2023-04-20 16:12:57 -07:00
|
|
|
va_list in_va)
|
2023-02-21 10:02:35 -08:00
|
|
|
{
|
|
|
|
|
struct {
|
|
|
|
|
char *cur;
|
|
|
|
|
int rem;
|
|
|
|
|
int total;
|
|
|
|
|
bool invalid;
|
|
|
|
|
} state = {
|
|
|
|
|
.cur = buf,
|
|
|
|
|
.rem = size,
|
|
|
|
|
};
|
|
|
|
|
|
2023-04-20 16:12:57 -07:00
|
|
|
va_list va;
|
|
|
|
|
va_copy(va, in_va);
|
|
|
|
|
|
2023-02-21 10:02:35 -08:00
|
|
|
#define APPEND(state, func, ...) \
|
|
|
|
|
do { \
|
|
|
|
|
int ret = func(state.cur, state.rem, __VA_ARGS__); \
|
|
|
|
|
if (ret < 0) { \
|
|
|
|
|
state.invalid = true; \
|
|
|
|
|
} else { \
|
|
|
|
|
state.total += ret; \
|
|
|
|
|
if (ret >= state.rem) \
|
|
|
|
|
ret = state.rem; \
|
|
|
|
|
state.cur += ret; \
|
|
|
|
|
state.rem -= ret; \
|
|
|
|
|
} \
|
|
|
|
|
} while (false)
|
|
|
|
|
|
|
|
|
|
if (affixes & LOGGER_VASNPRINTF_AFFIX_TAG)
|
|
|
|
|
APPEND(state, snprintf, "%s: ", tag);
|
|
|
|
|
if (affixes & LOGGER_VASNPRINTF_AFFIX_LEVEL)
|
|
|
|
|
APPEND(state, snprintf, "%s: ", level_to_str(level));
|
|
|
|
|
|
|
|
|
|
APPEND(state, vsnprintf, format, va);
|
|
|
|
|
|
|
|
|
|
if (affixes & LOGGER_VASNPRINTF_AFFIX_NEWLINE) {
|
|
|
|
|
if (state.cur == buf || state.cur[-1] != '\n')
|
|
|
|
|
APPEND(state, snprintf, "\n");
|
|
|
|
|
}
|
|
|
|
|
#undef APPEND
|
|
|
|
|
|
|
|
|
|
assert(size >= 64);
|
|
|
|
|
if (state.invalid) {
|
|
|
|
|
strncpy(buf, "invalid message format", size);
|
|
|
|
|
} else if (state.total >= size) {
|
|
|
|
|
/* print again into alloc to avoid truncation */
|
|
|
|
|
void *alloc = malloc(state.total + 1);
|
|
|
|
|
if (alloc) {
|
|
|
|
|
buf = logger_vasnprintf(alloc, state.total + 1, affixes, level, tag,
|
2023-04-20 16:12:57 -07:00
|
|
|
format, in_va);
|
2023-02-21 10:02:35 -08:00
|
|
|
assert(buf == alloc);
|
|
|
|
|
} else {
|
|
|
|
|
/* pretty-truncate the message */
|
|
|
|
|
strncpy(buf + size - 4, "...", 4);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2023-04-20 16:12:57 -07:00
|
|
|
va_end(va);
|
|
|
|
|
|
2023-02-21 10:02:35 -08:00
|
|
|
return buf;
|
|
|
|
|
}
|
|
|
|
|
|
2023-02-17 14:58:02 -08:00
|
|
|
static void
|
|
|
|
|
logger_file(enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
|
|
|
|
va_list va)
|
|
|
|
|
{
|
2023-02-17 17:37:27 -08:00
|
|
|
FILE *fp = mesa_log_file;
|
2023-02-21 10:02:35 -08:00
|
|
|
char local_msg[1024];
|
|
|
|
|
char *msg = logger_vasnprintf(local_msg, sizeof(local_msg),
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_TAG |
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_LEVEL |
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_NEWLINE,
|
|
|
|
|
level, tag, format, va);
|
2023-02-17 14:58:02 -08:00
|
|
|
|
2023-02-21 10:02:35 -08:00
|
|
|
fprintf(fp, "%s", msg);
|
2023-02-17 17:37:27 -08:00
|
|
|
fflush(fp);
|
2023-02-17 14:58:02 -08:00
|
|
|
|
2023-02-21 10:02:35 -08:00
|
|
|
if (msg != local_msg)
|
|
|
|
|
free(msg);
|
2023-02-17 14:58:02 -08:00
|
|
|
}
|
|
|
|
|
|
2024-06-06 01:19:35 +08:00
|
|
|
#if DETECT_OS_POSIX
|
2023-02-17 16:15:21 -08:00
|
|
|
|
|
|
|
|
static inline int
|
|
|
|
|
level_to_syslog(enum mesa_log_level l)
|
|
|
|
|
{
|
|
|
|
|
switch (l) {
|
|
|
|
|
case MESA_LOG_ERROR: return LOG_ERR;
|
|
|
|
|
case MESA_LOG_WARN: return LOG_WARNING;
|
|
|
|
|
case MESA_LOG_INFO: return LOG_INFO;
|
|
|
|
|
case MESA_LOG_DEBUG: return LOG_DEBUG;
|
|
|
|
|
}
|
|
|
|
|
|
2025-07-23 09:17:35 +02:00
|
|
|
UNREACHABLE("bad mesa_log_level");
|
2023-02-17 16:15:21 -08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
logger_syslog(enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
|
|
|
|
va_list va)
|
|
|
|
|
{
|
|
|
|
|
char local_msg[1024];
|
|
|
|
|
char *msg = logger_vasnprintf(local_msg, sizeof(local_msg),
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_TAG, level, tag, format, va);
|
|
|
|
|
|
|
|
|
|
syslog(level_to_syslog(level), "%s", msg);
|
|
|
|
|
|
|
|
|
|
if (msg != local_msg)
|
|
|
|
|
free(msg);
|
|
|
|
|
}
|
|
|
|
|
|
2024-06-06 01:19:35 +08:00
|
|
|
#endif /* DETECT_OS_POSIX */
|
2023-02-17 16:15:21 -08:00
|
|
|
|
2023-02-17 15:17:26 -08:00
|
|
|
#if DETECT_OS_ANDROID
|
2023-02-17 14:58:02 -08:00
|
|
|
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
static inline android_LogPriority
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
level_to_android(enum mesa_log_level l)
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
{
|
|
|
|
|
switch (l) {
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
case MESA_LOG_ERROR: return ANDROID_LOG_ERROR;
|
|
|
|
|
case MESA_LOG_WARN: return ANDROID_LOG_WARN;
|
|
|
|
|
case MESA_LOG_INFO: return ANDROID_LOG_INFO;
|
|
|
|
|
case MESA_LOG_DEBUG: return ANDROID_LOG_DEBUG;
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
}
|
|
|
|
|
|
2025-07-23 09:17:35 +02:00
|
|
|
UNREACHABLE("bad mesa_log_level");
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
}
|
|
|
|
|
|
2023-02-17 14:58:02 -08:00
|
|
|
static void
|
|
|
|
|
logger_android(enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
|
|
|
|
va_list va)
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
{
|
2023-02-17 15:22:17 -08:00
|
|
|
/* Android can truncate/drop messages
|
|
|
|
|
*
|
|
|
|
|
* - the internal buffer for vsnprintf has a fixed size (usually 1024)
|
|
|
|
|
* - the socket to logd is non-blocking
|
|
|
|
|
*
|
|
|
|
|
* and provides no way to detect. Try our best.
|
|
|
|
|
*/
|
|
|
|
|
char local_msg[1024];
|
|
|
|
|
char *msg = logger_vasnprintf(local_msg, sizeof(local_msg), 0, level, tag,
|
|
|
|
|
format, va);
|
|
|
|
|
|
|
|
|
|
__android_log_write(level_to_android(level), tag, msg);
|
|
|
|
|
|
|
|
|
|
if (msg != local_msg)
|
|
|
|
|
free(msg);
|
|
|
|
|
|
|
|
|
|
/* increase the chance of logd doing its part */
|
|
|
|
|
if (mesa_log_control & MESA_LOG_CONTROL_WAIT)
|
|
|
|
|
thrd_yield();
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
}
|
2023-02-17 14:58:02 -08:00
|
|
|
|
2023-02-17 15:17:26 -08:00
|
|
|
#endif /* DETECT_OS_ANDROID */
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
|
2023-02-17 17:44:29 -08:00
|
|
|
#if DETECT_OS_WINDOWS
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
logger_windbg(enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
|
|
|
|
va_list va)
|
|
|
|
|
{
|
|
|
|
|
char local_msg[1024];
|
|
|
|
|
char *msg = logger_vasnprintf(local_msg, sizeof(local_msg),
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_TAG |
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_LEVEL |
|
|
|
|
|
LOGGER_VASNPRINTF_AFFIX_NEWLINE,
|
|
|
|
|
level, tag, format, va);
|
|
|
|
|
|
|
|
|
|
OutputDebugStringA(msg);
|
|
|
|
|
|
|
|
|
|
if (msg != local_msg)
|
|
|
|
|
free(msg);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
#endif /* DETECT_OS_WINDOWS */
|
|
|
|
|
|
2023-02-17 17:37:27 -08:00
|
|
|
/* This is for use with debug functions that take a FILE, such as
|
|
|
|
|
* nir_print_shader, although switching to nir_log_shader* is preferred.
|
|
|
|
|
*/
|
|
|
|
|
FILE *
|
|
|
|
|
mesa_log_get_file(void)
|
|
|
|
|
{
|
|
|
|
|
mesa_log_init();
|
|
|
|
|
return mesa_log_file;
|
|
|
|
|
}
|
|
|
|
|
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
void
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
mesa_log(enum mesa_log_level level, const char *tag, const char *format, ...)
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
{
|
|
|
|
|
va_list va;
|
|
|
|
|
|
|
|
|
|
va_start(va, format);
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
mesa_log_v(level, tag, format, va);
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
va_end(va);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
mesa_log_v(enum mesa_log_level level, const char *tag, const char *format,
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
va_list va)
|
|
|
|
|
{
|
2023-02-17 15:17:26 -08:00
|
|
|
static const struct {
|
|
|
|
|
enum mesa_log_control bit;
|
|
|
|
|
void (*log)(enum mesa_log_level level,
|
|
|
|
|
const char *tag,
|
|
|
|
|
const char *format,
|
|
|
|
|
va_list va);
|
|
|
|
|
} loggers[] = {
|
|
|
|
|
{ MESA_LOG_CONTROL_FILE, logger_file },
|
2024-06-06 01:19:35 +08:00
|
|
|
#if DETECT_OS_POSIX
|
2023-02-17 16:15:21 -08:00
|
|
|
{ MESA_LOG_CONTROL_SYSLOG, logger_syslog },
|
|
|
|
|
#endif
|
2023-02-17 15:17:26 -08:00
|
|
|
#if DETECT_OS_ANDROID
|
|
|
|
|
{ MESA_LOG_CONTROL_ANDROID, logger_android },
|
2023-02-17 17:44:29 -08:00
|
|
|
#endif
|
|
|
|
|
#if DETECT_OS_WINDOWS
|
|
|
|
|
{ MESA_LOG_CONTROL_WINDBG, logger_windbg },
|
mesa: Promote Intel's simple logging façade for Android to util/
I'm bringing up freedreno Vulkan on an Android phone, and my pains are
exactly what Chad said when working on Intel's vulkan for Android in
aa716db0f64d ("intel: Add simple logging façade for Android (v2)"):
On Android, stdio goes to /dev/null. On Android, remote gdb is even
more painful than the usual remote gdb. On Android, nothing works like
you expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing
fancy. It follows the time-honored Unix tradition of spewing
everything to stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
Compared to trusty fprintf, _mesa_log[ewi]() is actually usable on
Android. Compared to os_log_message(), this has different error levels
and supports format arguments.
The only code change in the move is wrapping flockfile/funlockfile in
!DETECT_OS_WINDOWS, since mingw32 doesn't have it. Windows likely wants
different logging code, anyway.
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6806>
2020-09-21 12:53:14 -07:00
|
|
|
#endif
|
2023-02-17 15:17:26 -08:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
mesa_log_init();
|
|
|
|
|
|
|
|
|
|
for (uint32_t i = 0; i < ARRAY_SIZE(loggers); i++) {
|
|
|
|
|
if (mesa_log_control & loggers[i].bit) {
|
|
|
|
|
va_list copy;
|
|
|
|
|
va_copy(copy, va);
|
|
|
|
|
loggers[i].log(level, tag, format, copy);
|
|
|
|
|
va_end(copy);
|
|
|
|
|
}
|
|
|
|
|
}
|
intel: Add simple logging façade for Android (v2)
I'm bringing up Vulkan in the Android container of Chrome OS (ARC++).
On Android, stdio goes to /dev/null. On Android, remote gdb is even more
painful than the usual remote gdb. On Android, nothing works like you
expect and debugging is hell. I need logging.
This patch introduces a small, simple logging API that can easily wrap
Android's API. On non-Android platforms, this logger does nothing fancy.
It follows the time-honored Unix tradition of spewing everything to
stderr with minimal fuss.
My goal here is not perfection. My goal is to make a minimal, clean API,
that people hate merely a little instead of a lot, and that's good
enough to let me bring up Android Vulkan. And it needs to be fast,
which means it must be small. No one wants to their game to miss frames
while aiming a flaming bow into the jaws of an angry robot t-rex, and
thus become t-rex breakfast, because some fool had too much fun desiging
a bloated, ideal logging API.
If people like it, perhaps we should quickly promote it to src/util.
The API looks like this:
#define INTEL_LOG_TAG "intel-vulkan"
#define DEBUG
intel_logd("try hard thing with foo=%d", foo);
n = try_foo(...);
if (n < 0) {
intel_loge("%s:%d: foo failed bigtime", __FILE__, __LINE__);
return VK_ERROR_DEVICE_LOST;
}
And produces this on non-Android:
intel-vulkan: debug: try hard thing with foo=93
intel-vulkan: error: anv_device.c:182: foo failed bigtime
v2: Fix meson build. [for dcbaker]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
2017-08-25 14:14:34 -07:00
|
|
|
}
|
2021-06-16 10:20:35 -07:00
|
|
|
|
2024-07-03 19:08:10 +10:00
|
|
|
void
|
|
|
|
|
_mesa_log(const char *fmtString, ...)
|
|
|
|
|
{
|
|
|
|
|
char s[MAX_LOG_MESSAGE_LENGTH];
|
|
|
|
|
va_list args;
|
|
|
|
|
va_start(args, fmtString);
|
|
|
|
|
vsnprintf(s, MAX_LOG_MESSAGE_LENGTH, fmtString, args);
|
|
|
|
|
va_end(args);
|
|
|
|
|
mesa_log_if_debug(MESA_LOG_INFO, s);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
|
|
|
|
_mesa_log_direct(const char *string)
|
|
|
|
|
{
|
|
|
|
|
mesa_log_if_debug(MESA_LOG_INFO, string);
|
|
|
|
|
}
|
|
|
|
|
|
2021-06-16 10:20:35 -07:00
|
|
|
struct log_stream *
|
2023-04-15 09:47:19 -07:00
|
|
|
_mesa_log_stream_create(enum mesa_log_level level, const char *tag)
|
2021-06-16 10:20:35 -07:00
|
|
|
{
|
|
|
|
|
struct log_stream *stream = ralloc(NULL, struct log_stream);
|
|
|
|
|
stream->level = level;
|
|
|
|
|
stream->tag = tag;
|
|
|
|
|
stream->msg = ralloc_strdup(stream, "");
|
|
|
|
|
stream->pos = 0;
|
|
|
|
|
return stream;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
|
|
|
|
mesa_log_stream_destroy(struct log_stream *stream)
|
|
|
|
|
{
|
|
|
|
|
/* If you left trailing stuff in the log stream, flush it out as a line. */
|
|
|
|
|
if (stream->pos != 0)
|
|
|
|
|
mesa_log(stream->level, stream->tag, "%s", stream->msg);
|
|
|
|
|
|
|
|
|
|
ralloc_free(stream);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
mesa_log_stream_flush(struct log_stream *stream, size_t scan_offset)
|
|
|
|
|
{
|
|
|
|
|
char *end;
|
|
|
|
|
char *next = stream->msg;
|
|
|
|
|
while ((end = strchr(stream->msg + scan_offset, '\n'))) {
|
|
|
|
|
*end = 0;
|
|
|
|
|
mesa_log(stream->level, stream->tag, "%s", next);
|
|
|
|
|
next = end + 1;
|
|
|
|
|
scan_offset = next - stream->msg;
|
|
|
|
|
}
|
|
|
|
|
if (next != stream->msg) {
|
|
|
|
|
/* Clear out the lines we printed and move any trailing chars to the start. */
|
|
|
|
|
size_t remaining = stream->msg + stream->pos - next;
|
|
|
|
|
memmove(stream->msg, next, remaining);
|
|
|
|
|
stream->pos = remaining;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void mesa_log_stream_printf(struct log_stream *stream, const char *format, ...)
|
|
|
|
|
{
|
|
|
|
|
size_t old_pos = stream->pos;
|
|
|
|
|
|
|
|
|
|
va_list va;
|
|
|
|
|
va_start(va, format);
|
|
|
|
|
ralloc_vasprintf_rewrite_tail(&stream->msg, &stream->pos, format, va);
|
|
|
|
|
va_end(va);
|
|
|
|
|
|
|
|
|
|
mesa_log_stream_flush(stream, old_pos);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
|
|
|
|
_mesa_log_multiline(enum mesa_log_level level, const char *tag, const char *lines)
|
|
|
|
|
{
|
|
|
|
|
struct log_stream tmp = {
|
|
|
|
|
.level = level,
|
|
|
|
|
.tag = tag,
|
|
|
|
|
.msg = strdup(lines),
|
|
|
|
|
.pos = strlen(lines),
|
|
|
|
|
};
|
|
|
|
|
mesa_log_stream_flush(&tmp, 0);
|
|
|
|
|
free(tmp.msg);
|
|
|
|
|
}
|