mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-28 11:18:45 +07:00
13f1bfd3b3
As our allocations are not device specific, we can move our slab caches to a global scope. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190228102035.5857-2-chris@chris-wilson.co.uk
141 lines
2.9 KiB
C
141 lines
2.9 KiB
C
/*
|
|
* SPDX-License-Identifier: MIT
|
|
*
|
|
* Copyright © 2019 Intel Corporation
|
|
*/
|
|
|
|
#include <linux/slab.h>
|
|
#include <linux/workqueue.h>
|
|
|
|
#include "i915_active.h"
|
|
#include "i915_gem_context.h"
|
|
#include "i915_gem_object.h"
|
|
#include "i915_globals.h"
|
|
#include "i915_request.h"
|
|
#include "i915_scheduler.h"
|
|
#include "i915_vma.h"
|
|
|
|
int __init i915_globals_init(void)
|
|
{
|
|
int err;
|
|
|
|
err = i915_global_active_init();
|
|
if (err)
|
|
return err;
|
|
|
|
err = i915_global_context_init();
|
|
if (err)
|
|
goto err_active;
|
|
|
|
err = i915_global_objects_init();
|
|
if (err)
|
|
goto err_context;
|
|
|
|
err = i915_global_request_init();
|
|
if (err)
|
|
goto err_objects;
|
|
|
|
err = i915_global_scheduler_init();
|
|
if (err)
|
|
goto err_request;
|
|
|
|
err = i915_global_vma_init();
|
|
if (err)
|
|
goto err_scheduler;
|
|
|
|
return 0;
|
|
|
|
err_scheduler:
|
|
i915_global_scheduler_exit();
|
|
err_request:
|
|
i915_global_request_exit();
|
|
err_objects:
|
|
i915_global_objects_exit();
|
|
err_context:
|
|
i915_global_context_exit();
|
|
err_active:
|
|
i915_global_active_exit();
|
|
return err;
|
|
}
|
|
|
|
static void i915_globals_shrink(void)
|
|
{
|
|
/*
|
|
* kmem_cache_shrink() discards empty slabs and reorders partially
|
|
* filled slabs to prioritise allocating from the mostly full slabs,
|
|
* with the aim of reducing fragmentation.
|
|
*/
|
|
i915_global_active_shrink();
|
|
i915_global_context_shrink();
|
|
i915_global_objects_shrink();
|
|
i915_global_request_shrink();
|
|
i915_global_scheduler_shrink();
|
|
i915_global_vma_shrink();
|
|
}
|
|
|
|
static atomic_t active;
|
|
static atomic_t epoch;
|
|
struct park_work {
|
|
struct rcu_work work;
|
|
int epoch;
|
|
};
|
|
|
|
static void __i915_globals_park(struct work_struct *work)
|
|
{
|
|
struct park_work *wrk = container_of(work, typeof(*wrk), work.work);
|
|
|
|
/* Confirm nothing woke up in the last grace period */
|
|
if (wrk->epoch == atomic_read(&epoch))
|
|
i915_globals_shrink();
|
|
|
|
kfree(wrk);
|
|
}
|
|
|
|
void i915_globals_park(void)
|
|
{
|
|
struct park_work *wrk;
|
|
|
|
/*
|
|
* Defer shrinking the global slab caches (and other work) until
|
|
* after a RCU grace period has completed with no activity. This
|
|
* is to try and reduce the latency impact on the consumers caused
|
|
* by us shrinking the caches the same time as they are trying to
|
|
* allocate, with the assumption being that if we idle long enough
|
|
* for an RCU grace period to elapse since the last use, it is likely
|
|
* to be longer until we need the caches again.
|
|
*/
|
|
if (!atomic_dec_and_test(&active))
|
|
return;
|
|
|
|
wrk = kmalloc(sizeof(*wrk), GFP_KERNEL);
|
|
if (!wrk)
|
|
return;
|
|
|
|
wrk->epoch = atomic_inc_return(&epoch);
|
|
INIT_RCU_WORK(&wrk->work, __i915_globals_park);
|
|
queue_rcu_work(system_wq, &wrk->work);
|
|
}
|
|
|
|
void i915_globals_unpark(void)
|
|
{
|
|
atomic_inc(&epoch);
|
|
atomic_inc(&active);
|
|
}
|
|
|
|
void __exit i915_globals_exit(void)
|
|
{
|
|
/* Flush any residual park_work */
|
|
rcu_barrier();
|
|
flush_scheduled_work();
|
|
|
|
i915_global_vma_exit();
|
|
i915_global_scheduler_exit();
|
|
i915_global_request_exit();
|
|
i915_global_objects_exit();
|
|
i915_global_context_exit();
|
|
i915_global_active_exit();
|
|
|
|
/* And ensure that our DESTROY_BY_RCU slabs are truly destroyed */
|
|
rcu_barrier();
|
|
}
|