208 lines
5.4 KiB
C
208 lines
5.4 KiB
C
/*
|
|
* Copyright (C) 2010-2011 ARM Limited. All rights reserved.
|
|
*
|
|
* This program is free software and is provided to you under the terms of the GNU General Public License version 2
|
|
* as published by the Free Software Foundation, and any use by you of this program is subject to the terms of such GNU licence.
|
|
*
|
|
* A copy of the licence is included with the program, and can also be obtained from Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
|
*/
|
|
|
|
#include "mali_kernel_utilization.h"
|
|
#include "mali_osk.h"
|
|
#include "mali_platform.h"
|
|
|
|
/* Define how often to calculate and report GPU utilization, in milliseconds */
|
|
#define MALI_GPU_UTILIZATION_TIMEOUT 1000
|
|
|
|
static _mali_osk_lock_t *time_data_lock;
|
|
|
|
static _mali_osk_atomic_t num_running_cores;
|
|
|
|
static u64 period_start_time = 0;
|
|
static u64 work_start_time = 0;
|
|
static u64 accumulated_work_time = 0;
|
|
|
|
static _mali_osk_timer_t *utilization_timer = NULL;
|
|
static mali_bool timer_running = MALI_FALSE;
|
|
|
|
|
|
static void calculate_gpu_utilization(void* arg)
|
|
{
|
|
u64 time_now;
|
|
u64 time_period;
|
|
u32 leading_zeroes;
|
|
u32 shift_val;
|
|
u32 work_normalized;
|
|
u32 period_normalized;
|
|
u32 utilization;
|
|
|
|
_mali_osk_lock_wait(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
if (accumulated_work_time == 0 && work_start_time == 0)
|
|
{
|
|
/* Don't reschedule timer, this will be started if new work arrives */
|
|
timer_running = MALI_FALSE;
|
|
|
|
_mali_osk_lock_signal(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
/* No work done for this period, report zero usage */
|
|
mali_gpu_utilization_handler(0);
|
|
|
|
return;
|
|
}
|
|
|
|
time_now = _mali_osk_time_get_ns();
|
|
time_period = time_now - period_start_time;
|
|
|
|
/* If we are currently busy, update working period up to now */
|
|
if (work_start_time != 0)
|
|
{
|
|
accumulated_work_time += (time_now - work_start_time);
|
|
work_start_time = time_now;
|
|
}
|
|
|
|
/*
|
|
* We have two 64-bit values, a dividend and a divisor.
|
|
* To avoid dependencies to a 64-bit divider, we shift down the two values
|
|
* equally first.
|
|
* We shift the dividend up and possibly the divisor down, making the result X in 256.
|
|
*/
|
|
|
|
/* Shift the 64-bit values down so they fit inside a 32-bit integer */
|
|
leading_zeroes = _mali_osk_clz((u32)(time_period >> 32));
|
|
shift_val = 32 - leading_zeroes;
|
|
work_normalized = (u32)(accumulated_work_time >> shift_val);
|
|
period_normalized = (u32)(time_period >> shift_val);
|
|
|
|
/*
|
|
* Now, we should report the usage in parts of 256
|
|
* this means we must shift up the dividend or down the divisor by 8
|
|
* (we could do a combination, but we just use one for simplicity,
|
|
* but the end result should be good enough anyway)
|
|
*/
|
|
if (period_normalized > 0x00FFFFFF)
|
|
{
|
|
/* The divisor is so big that it is safe to shift it down */
|
|
period_normalized >>= 8;
|
|
}
|
|
else
|
|
{
|
|
/*
|
|
* The divisor is so small that we can shift up the dividend, without loosing any data.
|
|
* (dividend is always smaller than the divisor)
|
|
*/
|
|
work_normalized <<= 8;
|
|
}
|
|
|
|
utilization = work_normalized / period_normalized;
|
|
|
|
accumulated_work_time = 0;
|
|
period_start_time = time_now; /* starting a new period */
|
|
|
|
_mali_osk_lock_signal(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
_mali_osk_timer_add(utilization_timer, _mali_osk_time_mstoticks(MALI_GPU_UTILIZATION_TIMEOUT));
|
|
|
|
mali_gpu_utilization_handler(utilization);
|
|
}
|
|
|
|
|
|
|
|
_mali_osk_errcode_t mali_utilization_init(void)
|
|
{
|
|
time_data_lock = _mali_osk_lock_init( _MALI_OSK_LOCKFLAG_SPINLOCK_IRQ|_MALI_OSK_LOCKFLAG_NONINTERRUPTABLE, 0, 0 );
|
|
if (NULL == time_data_lock)
|
|
{
|
|
return _MALI_OSK_ERR_FAULT;
|
|
}
|
|
|
|
_mali_osk_atomic_init(&num_running_cores, 0);
|
|
|
|
utilization_timer = _mali_osk_timer_init();
|
|
if (NULL == utilization_timer)
|
|
{
|
|
_mali_osk_lock_term(time_data_lock);
|
|
return _MALI_OSK_ERR_FAULT;
|
|
}
|
|
_mali_osk_timer_setcallback(utilization_timer, calculate_gpu_utilization, NULL);
|
|
|
|
return _MALI_OSK_ERR_OK;
|
|
}
|
|
|
|
void mali_utilization_suspend(void)
|
|
{
|
|
if (NULL != utilization_timer)
|
|
{
|
|
_mali_osk_timer_del(utilization_timer);
|
|
timer_running = MALI_FALSE;
|
|
}
|
|
}
|
|
|
|
void mali_utilization_term(void)
|
|
{
|
|
if (NULL != utilization_timer)
|
|
{
|
|
_mali_osk_timer_del(utilization_timer);
|
|
timer_running = MALI_FALSE;
|
|
_mali_osk_timer_term(utilization_timer);
|
|
utilization_timer = NULL;
|
|
}
|
|
|
|
_mali_osk_atomic_term(&num_running_cores);
|
|
|
|
_mali_osk_lock_term(time_data_lock);
|
|
}
|
|
|
|
|
|
|
|
void mali_utilization_core_start(void)
|
|
{
|
|
if (_mali_osk_atomic_inc_return(&num_running_cores) == 1)
|
|
{
|
|
/*
|
|
* We went from zero cores working, to one core working,
|
|
* we now consider the entire GPU for being busy
|
|
*/
|
|
|
|
_mali_osk_lock_wait(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
work_start_time = _mali_osk_time_get_ns();
|
|
|
|
if (timer_running != MALI_TRUE)
|
|
{
|
|
timer_running = MALI_TRUE;
|
|
period_start_time = work_start_time; /* starting a new period */
|
|
|
|
_mali_osk_lock_signal(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
_mali_osk_timer_add(utilization_timer, _mali_osk_time_mstoticks(MALI_GPU_UTILIZATION_TIMEOUT));
|
|
}
|
|
else
|
|
{
|
|
_mali_osk_lock_signal(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
|
|
void mali_utilization_core_end(void)
|
|
{
|
|
if (_mali_osk_atomic_dec_return(&num_running_cores) == 0)
|
|
{
|
|
/*
|
|
* No more cores are working, so accumulate the time we was busy.
|
|
*/
|
|
u64 time_now;
|
|
|
|
_mali_osk_lock_wait(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
|
|
time_now = _mali_osk_time_get_ns();
|
|
accumulated_work_time += (time_now - work_start_time);
|
|
work_start_time = 0;
|
|
|
|
_mali_osk_lock_signal(time_data_lock, _MALI_OSK_LOCKMODE_RW);
|
|
}
|
|
}
|