mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-28 11:18:45 +07:00
ef369904aa
This gets rid of the irq bottom half tasklets and instead performs the work needed in process context. We also convert irq-disabling spinlocks to ordinary spinlocks. This should decrease system latency for other system components, like sound for example but has the potential to increase latency for processes that wait on the GPU. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Sinclair Yeh <syeh@vmware.com>
385 lines
10 KiB
C
385 lines
10 KiB
C
/**************************************************************************
|
|
*
|
|
* Copyright © 2009-2015 VMware, Inc., Palo Alto, CA., USA
|
|
* All Rights Reserved.
|
|
*
|
|
* 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, sub license, 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 NON-INFRINGEMENT. IN NO EVENT SHALL
|
|
* THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS 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 <drm/drmP.h>
|
|
#include "vmwgfx_drv.h"
|
|
|
|
#define VMW_FENCE_WRAP (1 << 24)
|
|
|
|
/**
|
|
* vmw_thread_fn - Deferred (process context) irq handler
|
|
*
|
|
* @irq: irq number
|
|
* @arg: Closure argument. Pointer to a struct drm_device cast to void *
|
|
*
|
|
* This function implements the deferred part of irq processing.
|
|
* The function is guaranteed to run at least once after the
|
|
* vmw_irq_handler has returned with IRQ_WAKE_THREAD.
|
|
*
|
|
*/
|
|
static irqreturn_t vmw_thread_fn(int irq, void *arg)
|
|
{
|
|
struct drm_device *dev = (struct drm_device *)arg;
|
|
struct vmw_private *dev_priv = vmw_priv(dev);
|
|
irqreturn_t ret = IRQ_NONE;
|
|
|
|
if (test_and_clear_bit(VMW_IRQTHREAD_FENCE,
|
|
dev_priv->irqthread_pending)) {
|
|
vmw_fences_update(dev_priv->fman);
|
|
wake_up_all(&dev_priv->fence_queue);
|
|
ret = IRQ_HANDLED;
|
|
}
|
|
|
|
if (test_and_clear_bit(VMW_IRQTHREAD_CMDBUF,
|
|
dev_priv->irqthread_pending)) {
|
|
vmw_cmdbuf_irqthread(dev_priv->cman);
|
|
ret = IRQ_HANDLED;
|
|
}
|
|
|
|
return ret;
|
|
}
|
|
|
|
/**
|
|
* vmw_irq_handler irq handler
|
|
*
|
|
* @irq: irq number
|
|
* @arg: Closure argument. Pointer to a struct drm_device cast to void *
|
|
*
|
|
* This function implements the quick part of irq processing.
|
|
* The function performs fast actions like clearing the device interrupt
|
|
* flags and also reasonably quick actions like waking processes waiting for
|
|
* FIFO space. Other IRQ actions are deferred to the IRQ thread.
|
|
*/
|
|
static irqreturn_t vmw_irq_handler(int irq, void *arg)
|
|
{
|
|
struct drm_device *dev = (struct drm_device *)arg;
|
|
struct vmw_private *dev_priv = vmw_priv(dev);
|
|
uint32_t status, masked_status;
|
|
irqreturn_t ret = IRQ_HANDLED;
|
|
|
|
status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
masked_status = status & READ_ONCE(dev_priv->irq_mask);
|
|
|
|
if (likely(status))
|
|
outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
|
|
if (!status)
|
|
return IRQ_NONE;
|
|
|
|
if (masked_status & SVGA_IRQFLAG_FIFO_PROGRESS)
|
|
wake_up_all(&dev_priv->fifo_queue);
|
|
|
|
if ((masked_status & (SVGA_IRQFLAG_ANY_FENCE |
|
|
SVGA_IRQFLAG_FENCE_GOAL)) &&
|
|
!test_and_set_bit(VMW_IRQTHREAD_FENCE, dev_priv->irqthread_pending))
|
|
ret = IRQ_WAKE_THREAD;
|
|
|
|
if ((masked_status & (SVGA_IRQFLAG_COMMAND_BUFFER |
|
|
SVGA_IRQFLAG_ERROR)) &&
|
|
!test_and_set_bit(VMW_IRQTHREAD_CMDBUF,
|
|
dev_priv->irqthread_pending))
|
|
ret = IRQ_WAKE_THREAD;
|
|
|
|
return ret;
|
|
}
|
|
|
|
static bool vmw_fifo_idle(struct vmw_private *dev_priv, uint32_t seqno)
|
|
{
|
|
|
|
return (vmw_read(dev_priv, SVGA_REG_BUSY) == 0);
|
|
}
|
|
|
|
void vmw_update_seqno(struct vmw_private *dev_priv,
|
|
struct vmw_fifo_state *fifo_state)
|
|
{
|
|
u32 *fifo_mem = dev_priv->mmio_virt;
|
|
uint32_t seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
|
|
|
|
if (dev_priv->last_read_seqno != seqno) {
|
|
dev_priv->last_read_seqno = seqno;
|
|
vmw_marker_pull(&fifo_state->marker_queue, seqno);
|
|
vmw_fences_update(dev_priv->fman);
|
|
}
|
|
}
|
|
|
|
bool vmw_seqno_passed(struct vmw_private *dev_priv,
|
|
uint32_t seqno)
|
|
{
|
|
struct vmw_fifo_state *fifo_state;
|
|
bool ret;
|
|
|
|
if (likely(dev_priv->last_read_seqno - seqno < VMW_FENCE_WRAP))
|
|
return true;
|
|
|
|
fifo_state = &dev_priv->fifo;
|
|
vmw_update_seqno(dev_priv, fifo_state);
|
|
if (likely(dev_priv->last_read_seqno - seqno < VMW_FENCE_WRAP))
|
|
return true;
|
|
|
|
if (!(fifo_state->capabilities & SVGA_FIFO_CAP_FENCE) &&
|
|
vmw_fifo_idle(dev_priv, seqno))
|
|
return true;
|
|
|
|
/**
|
|
* Then check if the seqno is higher than what we've actually
|
|
* emitted. Then the fence is stale and signaled.
|
|
*/
|
|
|
|
ret = ((atomic_read(&dev_priv->marker_seq) - seqno)
|
|
> VMW_FENCE_WRAP);
|
|
|
|
return ret;
|
|
}
|
|
|
|
int vmw_fallback_wait(struct vmw_private *dev_priv,
|
|
bool lazy,
|
|
bool fifo_idle,
|
|
uint32_t seqno,
|
|
bool interruptible,
|
|
unsigned long timeout)
|
|
{
|
|
struct vmw_fifo_state *fifo_state = &dev_priv->fifo;
|
|
|
|
uint32_t count = 0;
|
|
uint32_t signal_seq;
|
|
int ret;
|
|
unsigned long end_jiffies = jiffies + timeout;
|
|
bool (*wait_condition)(struct vmw_private *, uint32_t);
|
|
DEFINE_WAIT(__wait);
|
|
|
|
wait_condition = (fifo_idle) ? &vmw_fifo_idle :
|
|
&vmw_seqno_passed;
|
|
|
|
/**
|
|
* Block command submission while waiting for idle.
|
|
*/
|
|
|
|
if (fifo_idle) {
|
|
down_read(&fifo_state->rwsem);
|
|
if (dev_priv->cman) {
|
|
ret = vmw_cmdbuf_idle(dev_priv->cman, interruptible,
|
|
10*HZ);
|
|
if (ret)
|
|
goto out_err;
|
|
}
|
|
}
|
|
|
|
signal_seq = atomic_read(&dev_priv->marker_seq);
|
|
ret = 0;
|
|
|
|
for (;;) {
|
|
prepare_to_wait(&dev_priv->fence_queue, &__wait,
|
|
(interruptible) ?
|
|
TASK_INTERRUPTIBLE : TASK_UNINTERRUPTIBLE);
|
|
if (wait_condition(dev_priv, seqno))
|
|
break;
|
|
if (time_after_eq(jiffies, end_jiffies)) {
|
|
DRM_ERROR("SVGA device lockup.\n");
|
|
break;
|
|
}
|
|
if (lazy)
|
|
schedule_timeout(1);
|
|
else if ((++count & 0x0F) == 0) {
|
|
/**
|
|
* FIXME: Use schedule_hr_timeout here for
|
|
* newer kernels and lower CPU utilization.
|
|
*/
|
|
|
|
__set_current_state(TASK_RUNNING);
|
|
schedule();
|
|
__set_current_state((interruptible) ?
|
|
TASK_INTERRUPTIBLE :
|
|
TASK_UNINTERRUPTIBLE);
|
|
}
|
|
if (interruptible && signal_pending(current)) {
|
|
ret = -ERESTARTSYS;
|
|
break;
|
|
}
|
|
}
|
|
finish_wait(&dev_priv->fence_queue, &__wait);
|
|
if (ret == 0 && fifo_idle) {
|
|
u32 *fifo_mem = dev_priv->mmio_virt;
|
|
|
|
vmw_mmio_write(signal_seq, fifo_mem + SVGA_FIFO_FENCE);
|
|
}
|
|
wake_up_all(&dev_priv->fence_queue);
|
|
out_err:
|
|
if (fifo_idle)
|
|
up_read(&fifo_state->rwsem);
|
|
|
|
return ret;
|
|
}
|
|
|
|
void vmw_generic_waiter_add(struct vmw_private *dev_priv,
|
|
u32 flag, int *waiter_count)
|
|
{
|
|
spin_lock_bh(&dev_priv->waiter_lock);
|
|
if ((*waiter_count)++ == 0) {
|
|
outl(flag, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
dev_priv->irq_mask |= flag;
|
|
vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
|
|
}
|
|
spin_unlock_bh(&dev_priv->waiter_lock);
|
|
}
|
|
|
|
void vmw_generic_waiter_remove(struct vmw_private *dev_priv,
|
|
u32 flag, int *waiter_count)
|
|
{
|
|
spin_lock_bh(&dev_priv->waiter_lock);
|
|
if (--(*waiter_count) == 0) {
|
|
dev_priv->irq_mask &= ~flag;
|
|
vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
|
|
}
|
|
spin_unlock_bh(&dev_priv->waiter_lock);
|
|
}
|
|
|
|
void vmw_seqno_waiter_add(struct vmw_private *dev_priv)
|
|
{
|
|
vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_ANY_FENCE,
|
|
&dev_priv->fence_queue_waiters);
|
|
}
|
|
|
|
void vmw_seqno_waiter_remove(struct vmw_private *dev_priv)
|
|
{
|
|
vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_ANY_FENCE,
|
|
&dev_priv->fence_queue_waiters);
|
|
}
|
|
|
|
void vmw_goal_waiter_add(struct vmw_private *dev_priv)
|
|
{
|
|
vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_FENCE_GOAL,
|
|
&dev_priv->goal_queue_waiters);
|
|
}
|
|
|
|
void vmw_goal_waiter_remove(struct vmw_private *dev_priv)
|
|
{
|
|
vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_FENCE_GOAL,
|
|
&dev_priv->goal_queue_waiters);
|
|
}
|
|
|
|
int vmw_wait_seqno(struct vmw_private *dev_priv,
|
|
bool lazy, uint32_t seqno,
|
|
bool interruptible, unsigned long timeout)
|
|
{
|
|
long ret;
|
|
struct vmw_fifo_state *fifo = &dev_priv->fifo;
|
|
|
|
if (likely(dev_priv->last_read_seqno - seqno < VMW_FENCE_WRAP))
|
|
return 0;
|
|
|
|
if (likely(vmw_seqno_passed(dev_priv, seqno)))
|
|
return 0;
|
|
|
|
vmw_fifo_ping_host(dev_priv, SVGA_SYNC_GENERIC);
|
|
|
|
if (!(fifo->capabilities & SVGA_FIFO_CAP_FENCE))
|
|
return vmw_fallback_wait(dev_priv, lazy, true, seqno,
|
|
interruptible, timeout);
|
|
|
|
if (!(dev_priv->capabilities & SVGA_CAP_IRQMASK))
|
|
return vmw_fallback_wait(dev_priv, lazy, false, seqno,
|
|
interruptible, timeout);
|
|
|
|
vmw_seqno_waiter_add(dev_priv);
|
|
|
|
if (interruptible)
|
|
ret = wait_event_interruptible_timeout
|
|
(dev_priv->fence_queue,
|
|
vmw_seqno_passed(dev_priv, seqno),
|
|
timeout);
|
|
else
|
|
ret = wait_event_timeout
|
|
(dev_priv->fence_queue,
|
|
vmw_seqno_passed(dev_priv, seqno),
|
|
timeout);
|
|
|
|
vmw_seqno_waiter_remove(dev_priv);
|
|
|
|
if (unlikely(ret == 0))
|
|
ret = -EBUSY;
|
|
else if (likely(ret > 0))
|
|
ret = 0;
|
|
|
|
return ret;
|
|
}
|
|
|
|
static void vmw_irq_preinstall(struct drm_device *dev)
|
|
{
|
|
struct vmw_private *dev_priv = vmw_priv(dev);
|
|
uint32_t status;
|
|
|
|
status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
}
|
|
|
|
void vmw_irq_uninstall(struct drm_device *dev)
|
|
{
|
|
struct vmw_private *dev_priv = vmw_priv(dev);
|
|
uint32_t status;
|
|
|
|
if (!(dev_priv->capabilities & SVGA_CAP_IRQMASK))
|
|
return;
|
|
|
|
if (!dev->irq_enabled)
|
|
return;
|
|
|
|
vmw_write(dev_priv, SVGA_REG_IRQMASK, 0);
|
|
|
|
status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
|
|
|
|
dev->irq_enabled = false;
|
|
free_irq(dev->irq, dev);
|
|
}
|
|
|
|
/**
|
|
* vmw_irq_install - Install the irq handlers
|
|
*
|
|
* @dev: Pointer to the drm device.
|
|
* @irq: The irq number.
|
|
* Return: Zero if successful. Negative number otherwise.
|
|
*/
|
|
int vmw_irq_install(struct drm_device *dev, int irq)
|
|
{
|
|
int ret;
|
|
|
|
if (dev->irq_enabled)
|
|
return -EBUSY;
|
|
|
|
vmw_irq_preinstall(dev);
|
|
|
|
ret = request_threaded_irq(irq, vmw_irq_handler, vmw_thread_fn,
|
|
IRQF_SHARED, VMWGFX_DRIVER_NAME, dev);
|
|
if (ret < 0)
|
|
return ret;
|
|
|
|
dev->irq_enabled = true;
|
|
dev->irq = irq;
|
|
|
|
return ret;
|
|
}
|