linux-zen-desktop/drivers/gpu/drm/amd/amdkfd/kfd_events.c

1410 lines
36 KiB
C
Raw Normal View History

2023-08-30 17:31:07 +02:00
// SPDX-License-Identifier: GPL-2.0 OR MIT
/*
* Copyright 2014-2022 Advanced Micro Devices, 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 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 COPYRIGHT HOLDER(S) OR AUTHOR(S) 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 <linux/mm_types.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <linux/sched/signal.h>
#include <linux/sched/mm.h>
#include <linux/uaccess.h>
#include <linux/mman.h>
#include <linux/memory.h>
#include "kfd_priv.h"
#include "kfd_events.h"
#include "kfd_iommu.h"
#include <linux/device.h>
/*
* Wrapper around wait_queue_entry_t
*/
struct kfd_event_waiter {
wait_queue_entry_t wait;
struct kfd_event *event; /* Event to wait for */
bool activated; /* Becomes true when event is signaled */
2023-10-24 12:59:35 +02:00
bool event_age_enabled; /* set to true when last_event_age is non-zero */
2023-08-30 17:31:07 +02:00
};
/*
* Each signal event needs a 64-bit signal slot where the signaler will write
* a 1 before sending an interrupt. (This is needed because some interrupts
* do not contain enough spare data bits to identify an event.)
* We get whole pages and map them to the process VA.
* Individual signal events use their event_id as slot index.
*/
struct kfd_signal_page {
uint64_t *kernel_address;
uint64_t __user *user_address;
bool need_to_free_pages;
};
static uint64_t *page_slots(struct kfd_signal_page *page)
{
return page->kernel_address;
}
static struct kfd_signal_page *allocate_signal_page(struct kfd_process *p)
{
void *backing_store;
struct kfd_signal_page *page;
page = kzalloc(sizeof(*page), GFP_KERNEL);
if (!page)
return NULL;
backing_store = (void *) __get_free_pages(GFP_KERNEL,
get_order(KFD_SIGNAL_EVENT_LIMIT * 8));
if (!backing_store)
goto fail_alloc_signal_store;
/* Initialize all events to unsignaled */
memset(backing_store, (uint8_t) UNSIGNALED_EVENT_SLOT,
KFD_SIGNAL_EVENT_LIMIT * 8);
page->kernel_address = backing_store;
page->need_to_free_pages = true;
pr_debug("Allocated new event signal page at %p, for process %p\n",
page, p);
return page;
fail_alloc_signal_store:
kfree(page);
return NULL;
}
static int allocate_event_notification_slot(struct kfd_process *p,
struct kfd_event *ev,
const int *restore_id)
{
int id;
if (!p->signal_page) {
p->signal_page = allocate_signal_page(p);
if (!p->signal_page)
return -ENOMEM;
/* Oldest user mode expects 256 event slots */
p->signal_mapped_size = 256*8;
}
if (restore_id) {
id = idr_alloc(&p->event_idr, ev, *restore_id, *restore_id + 1,
GFP_KERNEL);
} else {
/*
* Compatibility with old user mode: Only use signal slots
* user mode has mapped, may be less than
* KFD_SIGNAL_EVENT_LIMIT. This also allows future increase
* of the event limit without breaking user mode.
*/
id = idr_alloc(&p->event_idr, ev, 0, p->signal_mapped_size / 8,
GFP_KERNEL);
}
if (id < 0)
return id;
ev->event_id = id;
page_slots(p->signal_page)[id] = UNSIGNALED_EVENT_SLOT;
return 0;
}
/*
* Assumes that p->event_mutex or rcu_readlock is held and of course that p is
* not going away.
*/
static struct kfd_event *lookup_event_by_id(struct kfd_process *p, uint32_t id)
{
return idr_find(&p->event_idr, id);
}
/**
* lookup_signaled_event_by_partial_id - Lookup signaled event from partial ID
* @p: Pointer to struct kfd_process
* @id: ID to look up
* @bits: Number of valid bits in @id
*
* Finds the first signaled event with a matching partial ID. If no
* matching signaled event is found, returns NULL. In that case the
* caller should assume that the partial ID is invalid and do an
* exhaustive search of all siglaned events.
*
* If multiple events with the same partial ID signal at the same
* time, they will be found one interrupt at a time, not necessarily
* in the same order the interrupts occurred. As long as the number of
* interrupts is correct, all signaled events will be seen by the
* driver.
*/
static struct kfd_event *lookup_signaled_event_by_partial_id(
struct kfd_process *p, uint32_t id, uint32_t bits)
{
struct kfd_event *ev;
if (!p->signal_page || id >= KFD_SIGNAL_EVENT_LIMIT)
return NULL;
/* Fast path for the common case that @id is not a partial ID
* and we only need a single lookup.
*/
if (bits > 31 || (1U << bits) >= KFD_SIGNAL_EVENT_LIMIT) {
if (page_slots(p->signal_page)[id] == UNSIGNALED_EVENT_SLOT)
return NULL;
return idr_find(&p->event_idr, id);
}
/* General case for partial IDs: Iterate over all matching IDs
* and find the first one that has signaled.
*/
for (ev = NULL; id < KFD_SIGNAL_EVENT_LIMIT && !ev; id += 1U << bits) {
if (page_slots(p->signal_page)[id] == UNSIGNALED_EVENT_SLOT)
continue;
ev = idr_find(&p->event_idr, id);
}
return ev;
}
static int create_signal_event(struct file *devkfd, struct kfd_process *p,
struct kfd_event *ev, const int *restore_id)
{
int ret;
if (p->signal_mapped_size &&
p->signal_event_count == p->signal_mapped_size / 8) {
if (!p->signal_event_limit_reached) {
pr_debug("Signal event wasn't created because limit was reached\n");
p->signal_event_limit_reached = true;
}
return -ENOSPC;
}
ret = allocate_event_notification_slot(p, ev, restore_id);
if (ret) {
pr_warn("Signal event wasn't created because out of kernel memory\n");
return ret;
}
p->signal_event_count++;
ev->user_signal_address = &p->signal_page->user_address[ev->event_id];
pr_debug("Signal event number %zu created with id %d, address %p\n",
p->signal_event_count, ev->event_id,
ev->user_signal_address);
return 0;
}
static int create_other_event(struct kfd_process *p, struct kfd_event *ev, const int *restore_id)
{
int id;
if (restore_id)
id = idr_alloc(&p->event_idr, ev, *restore_id, *restore_id + 1,
GFP_KERNEL);
else
/* Cast KFD_LAST_NONSIGNAL_EVENT to uint32_t. This allows an
* intentional integer overflow to -1 without a compiler
* warning. idr_alloc treats a negative value as "maximum
* signed integer".
*/
id = idr_alloc(&p->event_idr, ev, KFD_FIRST_NONSIGNAL_EVENT_ID,
(uint32_t)KFD_LAST_NONSIGNAL_EVENT_ID + 1,
GFP_KERNEL);
if (id < 0)
return id;
ev->event_id = id;
return 0;
}
int kfd_event_init_process(struct kfd_process *p)
{
int id;
mutex_init(&p->event_mutex);
idr_init(&p->event_idr);
p->signal_page = NULL;
p->signal_event_count = 1;
/* Allocate event ID 0. It is used for a fast path to ignore bogus events
* that are sent by the CP without a context ID
*/
id = idr_alloc(&p->event_idr, NULL, 0, 1, GFP_KERNEL);
if (id < 0) {
idr_destroy(&p->event_idr);
mutex_destroy(&p->event_mutex);
return id;
}
return 0;
}
static void destroy_event(struct kfd_process *p, struct kfd_event *ev)
{
struct kfd_event_waiter *waiter;
/* Wake up pending waiters. They will return failure */
spin_lock(&ev->lock);
list_for_each_entry(waiter, &ev->wq.head, wait.entry)
WRITE_ONCE(waiter->event, NULL);
wake_up_all(&ev->wq);
spin_unlock(&ev->lock);
if (ev->type == KFD_EVENT_TYPE_SIGNAL ||
ev->type == KFD_EVENT_TYPE_DEBUG)
p->signal_event_count--;
idr_remove(&p->event_idr, ev->event_id);
kfree_rcu(ev, rcu);
}
static void destroy_events(struct kfd_process *p)
{
struct kfd_event *ev;
uint32_t id;
idr_for_each_entry(&p->event_idr, ev, id)
if (ev)
destroy_event(p, ev);
idr_destroy(&p->event_idr);
mutex_destroy(&p->event_mutex);
}
/*
* We assume that the process is being destroyed and there is no need to
* unmap the pages or keep bookkeeping data in order.
*/
static void shutdown_signal_page(struct kfd_process *p)
{
struct kfd_signal_page *page = p->signal_page;
if (page) {
if (page->need_to_free_pages)
free_pages((unsigned long)page->kernel_address,
get_order(KFD_SIGNAL_EVENT_LIMIT * 8));
kfree(page);
}
}
void kfd_event_free_process(struct kfd_process *p)
{
destroy_events(p);
shutdown_signal_page(p);
}
static bool event_can_be_gpu_signaled(const struct kfd_event *ev)
{
return ev->type == KFD_EVENT_TYPE_SIGNAL ||
ev->type == KFD_EVENT_TYPE_DEBUG;
}
static bool event_can_be_cpu_signaled(const struct kfd_event *ev)
{
return ev->type == KFD_EVENT_TYPE_SIGNAL;
}
static int kfd_event_page_set(struct kfd_process *p, void *kernel_address,
uint64_t size, uint64_t user_handle)
{
struct kfd_signal_page *page;
if (p->signal_page)
return -EBUSY;
page = kzalloc(sizeof(*page), GFP_KERNEL);
if (!page)
return -ENOMEM;
/* Initialize all events to unsignaled */
memset(kernel_address, (uint8_t) UNSIGNALED_EVENT_SLOT,
KFD_SIGNAL_EVENT_LIMIT * 8);
page->kernel_address = kernel_address;
p->signal_page = page;
p->signal_mapped_size = size;
p->signal_handle = user_handle;
return 0;
}
int kfd_kmap_event_page(struct kfd_process *p, uint64_t event_page_offset)
{
2023-10-24 12:59:35 +02:00
struct kfd_node *kfd;
2023-08-30 17:31:07 +02:00
struct kfd_process_device *pdd;
void *mem, *kern_addr;
uint64_t size;
int err = 0;
if (p->signal_page) {
pr_err("Event page is already set\n");
return -EINVAL;
}
pdd = kfd_process_device_data_by_id(p, GET_GPU_ID(event_page_offset));
if (!pdd) {
pr_err("Getting device by id failed in %s\n", __func__);
return -EINVAL;
}
kfd = pdd->dev;
pdd = kfd_bind_process_to_device(kfd, p);
if (IS_ERR(pdd))
return PTR_ERR(pdd);
mem = kfd_process_device_translate_handle(pdd,
GET_IDR_HANDLE(event_page_offset));
if (!mem) {
pr_err("Can't find BO, offset is 0x%llx\n", event_page_offset);
return -EINVAL;
}
err = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(mem, &kern_addr, &size);
if (err) {
pr_err("Failed to map event page to kernel\n");
return err;
}
err = kfd_event_page_set(p, kern_addr, size, event_page_offset);
if (err) {
pr_err("Failed to set event page\n");
amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(mem);
return err;
}
return err;
}
int kfd_event_create(struct file *devkfd, struct kfd_process *p,
uint32_t event_type, bool auto_reset, uint32_t node_id,
uint32_t *event_id, uint32_t *event_trigger_data,
uint64_t *event_page_offset, uint32_t *event_slot_index)
{
int ret = 0;
struct kfd_event *ev = kzalloc(sizeof(*ev), GFP_KERNEL);
if (!ev)
return -ENOMEM;
ev->type = event_type;
ev->auto_reset = auto_reset;
ev->signaled = false;
spin_lock_init(&ev->lock);
init_waitqueue_head(&ev->wq);
*event_page_offset = 0;
mutex_lock(&p->event_mutex);
switch (event_type) {
case KFD_EVENT_TYPE_SIGNAL:
case KFD_EVENT_TYPE_DEBUG:
ret = create_signal_event(devkfd, p, ev, NULL);
if (!ret) {
*event_page_offset = KFD_MMAP_TYPE_EVENTS;
*event_slot_index = ev->event_id;
}
break;
default:
ret = create_other_event(p, ev, NULL);
break;
}
if (!ret) {
*event_id = ev->event_id;
*event_trigger_data = ev->event_id;
2023-10-24 12:59:35 +02:00
ev->event_age = 1;
2023-08-30 17:31:07 +02:00
} else {
kfree(ev);
}
mutex_unlock(&p->event_mutex);
return ret;
}
int kfd_criu_restore_event(struct file *devkfd,
struct kfd_process *p,
uint8_t __user *user_priv_ptr,
uint64_t *priv_data_offset,
uint64_t max_priv_data_size)
{
struct kfd_criu_event_priv_data *ev_priv;
struct kfd_event *ev = NULL;
int ret = 0;
ev_priv = kmalloc(sizeof(*ev_priv), GFP_KERNEL);
if (!ev_priv)
return -ENOMEM;
ev = kzalloc(sizeof(*ev), GFP_KERNEL);
if (!ev) {
ret = -ENOMEM;
goto exit;
}
if (*priv_data_offset + sizeof(*ev_priv) > max_priv_data_size) {
ret = -EINVAL;
goto exit;
}
ret = copy_from_user(ev_priv, user_priv_ptr + *priv_data_offset, sizeof(*ev_priv));
if (ret) {
ret = -EFAULT;
goto exit;
}
*priv_data_offset += sizeof(*ev_priv);
if (ev_priv->user_handle) {
ret = kfd_kmap_event_page(p, ev_priv->user_handle);
if (ret)
goto exit;
}
ev->type = ev_priv->type;
ev->auto_reset = ev_priv->auto_reset;
ev->signaled = ev_priv->signaled;
spin_lock_init(&ev->lock);
init_waitqueue_head(&ev->wq);
mutex_lock(&p->event_mutex);
switch (ev->type) {
case KFD_EVENT_TYPE_SIGNAL:
case KFD_EVENT_TYPE_DEBUG:
ret = create_signal_event(devkfd, p, ev, &ev_priv->event_id);
break;
case KFD_EVENT_TYPE_MEMORY:
memcpy(&ev->memory_exception_data,
&ev_priv->memory_exception_data,
sizeof(struct kfd_hsa_memory_exception_data));
ret = create_other_event(p, ev, &ev_priv->event_id);
break;
case KFD_EVENT_TYPE_HW_EXCEPTION:
memcpy(&ev->hw_exception_data,
&ev_priv->hw_exception_data,
sizeof(struct kfd_hsa_hw_exception_data));
ret = create_other_event(p, ev, &ev_priv->event_id);
break;
}
mutex_unlock(&p->event_mutex);
exit:
if (ret)
kfree(ev);
kfree(ev_priv);
return ret;
}
int kfd_criu_checkpoint_events(struct kfd_process *p,
uint8_t __user *user_priv_data,
uint64_t *priv_data_offset)
{
struct kfd_criu_event_priv_data *ev_privs;
int i = 0;
int ret = 0;
struct kfd_event *ev;
uint32_t ev_id;
uint32_t num_events = kfd_get_num_events(p);
if (!num_events)
return 0;
ev_privs = kvzalloc(num_events * sizeof(*ev_privs), GFP_KERNEL);
if (!ev_privs)
return -ENOMEM;
idr_for_each_entry(&p->event_idr, ev, ev_id) {
struct kfd_criu_event_priv_data *ev_priv;
/*
* Currently, all events have same size of private_data, but the current ioctl's
* and CRIU plugin supports private_data of variable sizes
*/
ev_priv = &ev_privs[i];
ev_priv->object_type = KFD_CRIU_OBJECT_TYPE_EVENT;
/* We store the user_handle with the first event */
if (i == 0 && p->signal_page)
ev_priv->user_handle = p->signal_handle;
ev_priv->event_id = ev->event_id;
ev_priv->auto_reset = ev->auto_reset;
ev_priv->type = ev->type;
ev_priv->signaled = ev->signaled;
if (ev_priv->type == KFD_EVENT_TYPE_MEMORY)
memcpy(&ev_priv->memory_exception_data,
&ev->memory_exception_data,
sizeof(struct kfd_hsa_memory_exception_data));
else if (ev_priv->type == KFD_EVENT_TYPE_HW_EXCEPTION)
memcpy(&ev_priv->hw_exception_data,
&ev->hw_exception_data,
sizeof(struct kfd_hsa_hw_exception_data));
pr_debug("Checkpointed event[%d] id = 0x%08x auto_reset = %x type = %x signaled = %x\n",
i,
ev_priv->event_id,
ev_priv->auto_reset,
ev_priv->type,
ev_priv->signaled);
i++;
}
ret = copy_to_user(user_priv_data + *priv_data_offset,
ev_privs, num_events * sizeof(*ev_privs));
if (ret) {
pr_err("Failed to copy events priv to user\n");
ret = -EFAULT;
}
*priv_data_offset += num_events * sizeof(*ev_privs);
kvfree(ev_privs);
return ret;
}
int kfd_get_num_events(struct kfd_process *p)
{
struct kfd_event *ev;
uint32_t id;
u32 num_events = 0;
idr_for_each_entry(&p->event_idr, ev, id)
num_events++;
return num_events;
}
/* Assumes that p is current. */
int kfd_event_destroy(struct kfd_process *p, uint32_t event_id)
{
struct kfd_event *ev;
int ret = 0;
mutex_lock(&p->event_mutex);
ev = lookup_event_by_id(p, event_id);
if (ev)
destroy_event(p, ev);
else
ret = -EINVAL;
mutex_unlock(&p->event_mutex);
return ret;
}
static void set_event(struct kfd_event *ev)
{
struct kfd_event_waiter *waiter;
/* Auto reset if the list is non-empty and we're waking
* someone. waitqueue_active is safe here because we're
* protected by the ev->lock, which is also held when
* updating the wait queues in kfd_wait_on_events.
*/
ev->signaled = !ev->auto_reset || !waitqueue_active(&ev->wq);
2023-10-24 12:59:35 +02:00
if (!(++ev->event_age)) {
/* Never wrap back to reserved/default event age 0/1 */
ev->event_age = 2;
WARN_ONCE(1, "event_age wrap back!");
}
2023-08-30 17:31:07 +02:00
list_for_each_entry(waiter, &ev->wq.head, wait.entry)
WRITE_ONCE(waiter->activated, true);
wake_up_all(&ev->wq);
}
/* Assumes that p is current. */
int kfd_set_event(struct kfd_process *p, uint32_t event_id)
{
int ret = 0;
struct kfd_event *ev;
rcu_read_lock();
ev = lookup_event_by_id(p, event_id);
if (!ev) {
ret = -EINVAL;
goto unlock_rcu;
}
spin_lock(&ev->lock);
if (event_can_be_cpu_signaled(ev))
set_event(ev);
else
ret = -EINVAL;
spin_unlock(&ev->lock);
unlock_rcu:
rcu_read_unlock();
return ret;
}
static void reset_event(struct kfd_event *ev)
{
ev->signaled = false;
}
/* Assumes that p is current. */
int kfd_reset_event(struct kfd_process *p, uint32_t event_id)
{
int ret = 0;
struct kfd_event *ev;
rcu_read_lock();
ev = lookup_event_by_id(p, event_id);
if (!ev) {
ret = -EINVAL;
goto unlock_rcu;
}
spin_lock(&ev->lock);
if (event_can_be_cpu_signaled(ev))
reset_event(ev);
else
ret = -EINVAL;
spin_unlock(&ev->lock);
unlock_rcu:
rcu_read_unlock();
return ret;
}
static void acknowledge_signal(struct kfd_process *p, struct kfd_event *ev)
{
WRITE_ONCE(page_slots(p->signal_page)[ev->event_id], UNSIGNALED_EVENT_SLOT);
}
static void set_event_from_interrupt(struct kfd_process *p,
struct kfd_event *ev)
{
if (ev && event_can_be_gpu_signaled(ev)) {
acknowledge_signal(p, ev);
spin_lock(&ev->lock);
set_event(ev);
spin_unlock(&ev->lock);
}
}
void kfd_signal_event_interrupt(u32 pasid, uint32_t partial_id,
uint32_t valid_id_bits)
{
struct kfd_event *ev = NULL;
/*
* Because we are called from arbitrary context (workqueue) as opposed
* to process context, kfd_process could attempt to exit while we are
* running so the lookup function increments the process ref count.
*/
struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
if (!p)
return; /* Presumably process exited. */
rcu_read_lock();
if (valid_id_bits)
ev = lookup_signaled_event_by_partial_id(p, partial_id,
valid_id_bits);
if (ev) {
set_event_from_interrupt(p, ev);
} else if (p->signal_page) {
/*
* Partial ID lookup failed. Assume that the event ID
* in the interrupt payload was invalid and do an
* exhaustive search of signaled events.
*/
uint64_t *slots = page_slots(p->signal_page);
uint32_t id;
if (valid_id_bits)
pr_debug_ratelimited("Partial ID invalid: %u (%u valid bits)\n",
partial_id, valid_id_bits);
if (p->signal_event_count < KFD_SIGNAL_EVENT_LIMIT / 64) {
/* With relatively few events, it's faster to
* iterate over the event IDR
*/
idr_for_each_entry(&p->event_idr, ev, id) {
if (id >= KFD_SIGNAL_EVENT_LIMIT)
break;
if (READ_ONCE(slots[id]) != UNSIGNALED_EVENT_SLOT)
set_event_from_interrupt(p, ev);
}
} else {
/* With relatively many events, it's faster to
* iterate over the signal slots and lookup
* only signaled events from the IDR.
*/
for (id = 1; id < KFD_SIGNAL_EVENT_LIMIT; id++)
if (READ_ONCE(slots[id]) != UNSIGNALED_EVENT_SLOT) {
ev = lookup_event_by_id(p, id);
set_event_from_interrupt(p, ev);
}
}
}
rcu_read_unlock();
kfd_unref_process(p);
}
static struct kfd_event_waiter *alloc_event_waiters(uint32_t num_events)
{
struct kfd_event_waiter *event_waiters;
uint32_t i;
event_waiters = kcalloc(num_events, sizeof(struct kfd_event_waiter),
GFP_KERNEL);
if (!event_waiters)
return NULL;
for (i = 0; i < num_events; i++)
init_wait(&event_waiters[i].wait);
return event_waiters;
}
static int init_event_waiter(struct kfd_process *p,
struct kfd_event_waiter *waiter,
2023-10-24 12:59:35 +02:00
struct kfd_event_data *event_data)
2023-08-30 17:31:07 +02:00
{
2023-10-24 12:59:35 +02:00
struct kfd_event *ev = lookup_event_by_id(p, event_data->event_id);
2023-08-30 17:31:07 +02:00
if (!ev)
return -EINVAL;
spin_lock(&ev->lock);
waiter->event = ev;
waiter->activated = ev->signaled;
ev->signaled = ev->signaled && !ev->auto_reset;
2023-10-24 12:59:35 +02:00
/* last_event_age = 0 reserved for backward compatible */
if (waiter->event->type == KFD_EVENT_TYPE_SIGNAL &&
event_data->signal_event_data.last_event_age) {
waiter->event_age_enabled = true;
if (ev->event_age != event_data->signal_event_data.last_event_age)
waiter->activated = true;
}
2023-08-30 17:31:07 +02:00
if (!waiter->activated)
add_wait_queue(&ev->wq, &waiter->wait);
spin_unlock(&ev->lock);
return 0;
}
/* test_event_condition - Test condition of events being waited for
* @all: Return completion only if all events have signaled
* @num_events: Number of events to wait for
* @event_waiters: Array of event waiters, one per event
*
* Returns KFD_IOC_WAIT_RESULT_COMPLETE if all (or one) event(s) have
* signaled. Returns KFD_IOC_WAIT_RESULT_TIMEOUT if no (or not all)
* events have signaled. Returns KFD_IOC_WAIT_RESULT_FAIL if any of
* the events have been destroyed.
*/
static uint32_t test_event_condition(bool all, uint32_t num_events,
struct kfd_event_waiter *event_waiters)
{
uint32_t i;
uint32_t activated_count = 0;
for (i = 0; i < num_events; i++) {
if (!READ_ONCE(event_waiters[i].event))
return KFD_IOC_WAIT_RESULT_FAIL;
if (READ_ONCE(event_waiters[i].activated)) {
if (!all)
return KFD_IOC_WAIT_RESULT_COMPLETE;
activated_count++;
}
}
return activated_count == num_events ?
KFD_IOC_WAIT_RESULT_COMPLETE : KFD_IOC_WAIT_RESULT_TIMEOUT;
}
/*
* Copy event specific data, if defined.
* Currently only memory exception events have additional data to copy to user
*/
static int copy_signaled_event_data(uint32_t num_events,
struct kfd_event_waiter *event_waiters,
struct kfd_event_data __user *data)
{
2023-10-24 12:59:35 +02:00
void *src;
void __user *dst;
2023-08-30 17:31:07 +02:00
struct kfd_event_waiter *waiter;
struct kfd_event *event;
2023-10-24 12:59:35 +02:00
uint32_t i, size = 0;
2023-08-30 17:31:07 +02:00
for (i = 0; i < num_events; i++) {
waiter = &event_waiters[i];
event = waiter->event;
if (!event)
return -EINVAL; /* event was destroyed */
2023-10-24 12:59:35 +02:00
if (waiter->activated) {
if (event->type == KFD_EVENT_TYPE_MEMORY) {
dst = &data[i].memory_exception_data;
src = &event->memory_exception_data;
size = sizeof(struct kfd_hsa_memory_exception_data);
} else if (event->type == KFD_EVENT_TYPE_SIGNAL &&
waiter->event_age_enabled) {
dst = &data[i].signal_event_data.last_event_age;
src = &event->event_age;
size = sizeof(u64);
}
if (size && copy_to_user(dst, src, size))
2023-08-30 17:31:07 +02:00
return -EFAULT;
}
}
return 0;
}
static long user_timeout_to_jiffies(uint32_t user_timeout_ms)
{
if (user_timeout_ms == KFD_EVENT_TIMEOUT_IMMEDIATE)
return 0;
if (user_timeout_ms == KFD_EVENT_TIMEOUT_INFINITE)
return MAX_SCHEDULE_TIMEOUT;
/*
* msecs_to_jiffies interprets all values above 2^31-1 as infinite,
* but we consider them finite.
* This hack is wrong, but nobody is likely to notice.
*/
user_timeout_ms = min_t(uint32_t, user_timeout_ms, 0x7FFFFFFF);
return msecs_to_jiffies(user_timeout_ms) + 1;
}
static void free_waiters(uint32_t num_events, struct kfd_event_waiter *waiters,
bool undo_auto_reset)
{
uint32_t i;
for (i = 0; i < num_events; i++)
if (waiters[i].event) {
spin_lock(&waiters[i].event->lock);
remove_wait_queue(&waiters[i].event->wq,
&waiters[i].wait);
if (undo_auto_reset && waiters[i].activated &&
waiters[i].event && waiters[i].event->auto_reset)
set_event(waiters[i].event);
spin_unlock(&waiters[i].event->lock);
}
kfree(waiters);
}
int kfd_wait_on_events(struct kfd_process *p,
uint32_t num_events, void __user *data,
bool all, uint32_t *user_timeout_ms,
uint32_t *wait_result)
{
struct kfd_event_data __user *events =
(struct kfd_event_data __user *) data;
uint32_t i;
int ret = 0;
struct kfd_event_waiter *event_waiters = NULL;
long timeout = user_timeout_to_jiffies(*user_timeout_ms);
event_waiters = alloc_event_waiters(num_events);
if (!event_waiters) {
ret = -ENOMEM;
goto out;
}
/* Use p->event_mutex here to protect against concurrent creation and
* destruction of events while we initialize event_waiters.
*/
mutex_lock(&p->event_mutex);
for (i = 0; i < num_events; i++) {
struct kfd_event_data event_data;
if (copy_from_user(&event_data, &events[i],
sizeof(struct kfd_event_data))) {
ret = -EFAULT;
goto out_unlock;
}
2023-10-24 12:59:35 +02:00
ret = init_event_waiter(p, &event_waiters[i], &event_data);
2023-08-30 17:31:07 +02:00
if (ret)
goto out_unlock;
}
/* Check condition once. */
*wait_result = test_event_condition(all, num_events, event_waiters);
if (*wait_result == KFD_IOC_WAIT_RESULT_COMPLETE) {
ret = copy_signaled_event_data(num_events,
event_waiters, events);
goto out_unlock;
} else if (WARN_ON(*wait_result == KFD_IOC_WAIT_RESULT_FAIL)) {
/* This should not happen. Events shouldn't be
* destroyed while we're holding the event_mutex
*/
goto out_unlock;
}
mutex_unlock(&p->event_mutex);
while (true) {
if (fatal_signal_pending(current)) {
ret = -EINTR;
break;
}
if (signal_pending(current)) {
ret = -ERESTARTSYS;
if (*user_timeout_ms != KFD_EVENT_TIMEOUT_IMMEDIATE &&
*user_timeout_ms != KFD_EVENT_TIMEOUT_INFINITE)
*user_timeout_ms = jiffies_to_msecs(
max(0l, timeout-1));
break;
}
/* Set task state to interruptible sleep before
* checking wake-up conditions. A concurrent wake-up
* will put the task back into runnable state. In that
* case schedule_timeout will not put the task to
* sleep and we'll get a chance to re-check the
* updated conditions almost immediately. Otherwise,
* this race condition would lead to a soft hang or a
* very long sleep.
*/
set_current_state(TASK_INTERRUPTIBLE);
*wait_result = test_event_condition(all, num_events,
event_waiters);
if (*wait_result != KFD_IOC_WAIT_RESULT_TIMEOUT)
break;
if (timeout <= 0)
break;
timeout = schedule_timeout(timeout);
}
__set_current_state(TASK_RUNNING);
mutex_lock(&p->event_mutex);
/* copy_signaled_event_data may sleep. So this has to happen
* after the task state is set back to RUNNING.
*
* The event may also have been destroyed after signaling. So
* copy_signaled_event_data also must confirm that the event
* still exists. Therefore this must be under the p->event_mutex
* which is also held when events are destroyed.
*/
if (!ret && *wait_result == KFD_IOC_WAIT_RESULT_COMPLETE)
ret = copy_signaled_event_data(num_events,
event_waiters, events);
out_unlock:
free_waiters(num_events, event_waiters, ret == -ERESTARTSYS);
mutex_unlock(&p->event_mutex);
out:
if (ret)
*wait_result = KFD_IOC_WAIT_RESULT_FAIL;
else if (*wait_result == KFD_IOC_WAIT_RESULT_FAIL)
ret = -EIO;
return ret;
}
int kfd_event_mmap(struct kfd_process *p, struct vm_area_struct *vma)
{
unsigned long pfn;
struct kfd_signal_page *page;
int ret;
/* check required size doesn't exceed the allocated size */
if (get_order(KFD_SIGNAL_EVENT_LIMIT * 8) <
get_order(vma->vm_end - vma->vm_start)) {
pr_err("Event page mmap requested illegal size\n");
return -EINVAL;
}
page = p->signal_page;
if (!page) {
/* Probably KFD bug, but mmap is user-accessible. */
pr_debug("Signal page could not be found\n");
return -EINVAL;
}
pfn = __pa(page->kernel_address);
pfn >>= PAGE_SHIFT;
vm_flags_set(vma, VM_IO | VM_DONTCOPY | VM_DONTEXPAND | VM_NORESERVE
| VM_DONTDUMP | VM_PFNMAP);
pr_debug("Mapping signal page\n");
pr_debug(" start user address == 0x%08lx\n", vma->vm_start);
pr_debug(" end user address == 0x%08lx\n", vma->vm_end);
pr_debug(" pfn == 0x%016lX\n", pfn);
pr_debug(" vm_flags == 0x%08lX\n", vma->vm_flags);
pr_debug(" size == 0x%08lX\n",
vma->vm_end - vma->vm_start);
page->user_address = (uint64_t __user *)vma->vm_start;
/* mapping the page to user process */
ret = remap_pfn_range(vma, vma->vm_start, pfn,
vma->vm_end - vma->vm_start, vma->vm_page_prot);
if (!ret)
p->signal_mapped_size = vma->vm_end - vma->vm_start;
return ret;
}
/*
* Assumes that p is not going away.
*/
static void lookup_events_by_type_and_signal(struct kfd_process *p,
int type, void *event_data)
{
struct kfd_hsa_memory_exception_data *ev_data;
struct kfd_event *ev;
uint32_t id;
bool send_signal = true;
ev_data = (struct kfd_hsa_memory_exception_data *) event_data;
rcu_read_lock();
id = KFD_FIRST_NONSIGNAL_EVENT_ID;
idr_for_each_entry_continue(&p->event_idr, ev, id)
if (ev->type == type) {
send_signal = false;
dev_dbg(kfd_device,
"Event found: id %X type %d",
ev->event_id, ev->type);
spin_lock(&ev->lock);
set_event(ev);
if (ev->type == KFD_EVENT_TYPE_MEMORY && ev_data)
ev->memory_exception_data = *ev_data;
spin_unlock(&ev->lock);
}
if (type == KFD_EVENT_TYPE_MEMORY) {
dev_warn(kfd_device,
"Sending SIGSEGV to process %d (pasid 0x%x)",
p->lead_thread->pid, p->pasid);
send_sig(SIGSEGV, p->lead_thread, 0);
}
/* Send SIGTERM no event of type "type" has been found*/
if (send_signal) {
if (send_sigterm) {
dev_warn(kfd_device,
"Sending SIGTERM to process %d (pasid 0x%x)",
p->lead_thread->pid, p->pasid);
send_sig(SIGTERM, p->lead_thread, 0);
} else {
dev_err(kfd_device,
"Process %d (pasid 0x%x) got unhandled exception",
p->lead_thread->pid, p->pasid);
}
}
rcu_read_unlock();
}
#ifdef KFD_SUPPORT_IOMMU_V2
2023-10-24 12:59:35 +02:00
void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid,
2023-08-30 17:31:07 +02:00
unsigned long address, bool is_write_requested,
bool is_execute_requested)
{
struct kfd_hsa_memory_exception_data memory_exception_data;
struct vm_area_struct *vma;
int user_gpu_id;
/*
* Because we are called from arbitrary context (workqueue) as opposed
* to process context, kfd_process could attempt to exit while we are
* running so the lookup function increments the process ref count.
*/
struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
struct mm_struct *mm;
if (!p)
return; /* Presumably process exited. */
/* Take a safe reference to the mm_struct, which may otherwise
* disappear even while the kfd_process is still referenced.
*/
mm = get_task_mm(p->lead_thread);
if (!mm) {
kfd_unref_process(p);
return; /* Process is exiting */
}
user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
if (unlikely(user_gpu_id == -EINVAL)) {
WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id);
return;
}
memset(&memory_exception_data, 0, sizeof(memory_exception_data));
mmap_read_lock(mm);
vma = find_vma(mm, address);
memory_exception_data.gpu_id = user_gpu_id;
memory_exception_data.va = address;
/* Set failure reason */
memory_exception_data.failure.NotPresent = 1;
memory_exception_data.failure.NoExecute = 0;
memory_exception_data.failure.ReadOnly = 0;
if (vma && address >= vma->vm_start) {
memory_exception_data.failure.NotPresent = 0;
if (is_write_requested && !(vma->vm_flags & VM_WRITE))
memory_exception_data.failure.ReadOnly = 1;
else
memory_exception_data.failure.ReadOnly = 0;
if (is_execute_requested && !(vma->vm_flags & VM_EXEC))
memory_exception_data.failure.NoExecute = 1;
else
memory_exception_data.failure.NoExecute = 0;
}
mmap_read_unlock(mm);
mmput(mm);
pr_debug("notpresent %d, noexecute %d, readonly %d\n",
memory_exception_data.failure.NotPresent,
memory_exception_data.failure.NoExecute,
memory_exception_data.failure.ReadOnly);
/* Workaround on Raven to not kill the process when memory is freed
* before IOMMU is able to finish processing all the excessive PPRs
*/
if (KFD_GC_VERSION(dev) != IP_VERSION(9, 1, 0) &&
KFD_GC_VERSION(dev) != IP_VERSION(9, 2, 2) &&
KFD_GC_VERSION(dev) != IP_VERSION(9, 3, 0))
lookup_events_by_type_and_signal(p, KFD_EVENT_TYPE_MEMORY,
&memory_exception_data);
kfd_unref_process(p);
}
#endif /* KFD_SUPPORT_IOMMU_V2 */
void kfd_signal_hw_exception_event(u32 pasid)
{
/*
* Because we are called from arbitrary context (workqueue) as opposed
* to process context, kfd_process could attempt to exit while we are
* running so the lookup function increments the process ref count.
*/
struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
if (!p)
return; /* Presumably process exited. */
lookup_events_by_type_and_signal(p, KFD_EVENT_TYPE_HW_EXCEPTION, NULL);
kfd_unref_process(p);
}
2023-10-24 12:59:35 +02:00
void kfd_signal_vm_fault_event(struct kfd_node *dev, u32 pasid,
struct kfd_vm_fault_info *info,
struct kfd_hsa_memory_exception_data *data)
2023-08-30 17:31:07 +02:00
{
struct kfd_event *ev;
uint32_t id;
struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
struct kfd_hsa_memory_exception_data memory_exception_data;
int user_gpu_id;
if (!p)
return; /* Presumably process exited. */
user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
if (unlikely(user_gpu_id == -EINVAL)) {
WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id);
return;
}
2023-10-24 12:59:35 +02:00
/* SoC15 chips and onwards will pass in data from now on. */
if (!data) {
memset(&memory_exception_data, 0, sizeof(memory_exception_data));
memory_exception_data.gpu_id = user_gpu_id;
memory_exception_data.failure.imprecise = true;
/* Set failure reason */
if (info) {
memory_exception_data.va = (info->page_addr) <<
PAGE_SHIFT;
memory_exception_data.failure.NotPresent =
info->prot_valid ? 1 : 0;
memory_exception_data.failure.NoExecute =
info->prot_exec ? 1 : 0;
memory_exception_data.failure.ReadOnly =
info->prot_write ? 1 : 0;
memory_exception_data.failure.imprecise = 0;
}
2023-08-30 17:31:07 +02:00
}
rcu_read_lock();
id = KFD_FIRST_NONSIGNAL_EVENT_ID;
idr_for_each_entry_continue(&p->event_idr, ev, id)
if (ev->type == KFD_EVENT_TYPE_MEMORY) {
spin_lock(&ev->lock);
2023-10-24 12:59:35 +02:00
ev->memory_exception_data = data ? *data :
memory_exception_data;
2023-08-30 17:31:07 +02:00
set_event(ev);
spin_unlock(&ev->lock);
}
rcu_read_unlock();
kfd_unref_process(p);
}
2023-10-24 12:59:35 +02:00
void kfd_signal_reset_event(struct kfd_node *dev)
2023-08-30 17:31:07 +02:00
{
struct kfd_hsa_hw_exception_data hw_exception_data;
struct kfd_hsa_memory_exception_data memory_exception_data;
struct kfd_process *p;
struct kfd_event *ev;
unsigned int temp;
uint32_t id, idx;
int reset_cause = atomic_read(&dev->sram_ecc_flag) ?
KFD_HW_EXCEPTION_ECC :
KFD_HW_EXCEPTION_GPU_HANG;
/* Whole gpu reset caused by GPU hang and memory is lost */
memset(&hw_exception_data, 0, sizeof(hw_exception_data));
hw_exception_data.memory_lost = 1;
hw_exception_data.reset_cause = reset_cause;
memset(&memory_exception_data, 0, sizeof(memory_exception_data));
memory_exception_data.ErrorType = KFD_MEM_ERR_SRAM_ECC;
memory_exception_data.failure.imprecise = true;
idx = srcu_read_lock(&kfd_processes_srcu);
hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) {
int user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
if (unlikely(user_gpu_id == -EINVAL)) {
WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id);
continue;
}
rcu_read_lock();
id = KFD_FIRST_NONSIGNAL_EVENT_ID;
idr_for_each_entry_continue(&p->event_idr, ev, id) {
if (ev->type == KFD_EVENT_TYPE_HW_EXCEPTION) {
spin_lock(&ev->lock);
ev->hw_exception_data = hw_exception_data;
ev->hw_exception_data.gpu_id = user_gpu_id;
set_event(ev);
spin_unlock(&ev->lock);
}
if (ev->type == KFD_EVENT_TYPE_MEMORY &&
reset_cause == KFD_HW_EXCEPTION_ECC) {
spin_lock(&ev->lock);
ev->memory_exception_data = memory_exception_data;
ev->memory_exception_data.gpu_id = user_gpu_id;
set_event(ev);
spin_unlock(&ev->lock);
}
}
rcu_read_unlock();
}
srcu_read_unlock(&kfd_processes_srcu, idx);
}
2023-10-24 12:59:35 +02:00
void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid)
2023-08-30 17:31:07 +02:00
{
struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
struct kfd_hsa_memory_exception_data memory_exception_data;
struct kfd_hsa_hw_exception_data hw_exception_data;
struct kfd_event *ev;
uint32_t id = KFD_FIRST_NONSIGNAL_EVENT_ID;
int user_gpu_id;
if (!p)
return; /* Presumably process exited. */
user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id);
if (unlikely(user_gpu_id == -EINVAL)) {
WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id);
return;
}
memset(&hw_exception_data, 0, sizeof(hw_exception_data));
hw_exception_data.gpu_id = user_gpu_id;
hw_exception_data.memory_lost = 1;
hw_exception_data.reset_cause = KFD_HW_EXCEPTION_ECC;
memset(&memory_exception_data, 0, sizeof(memory_exception_data));
memory_exception_data.ErrorType = KFD_MEM_ERR_POISON_CONSUMED;
memory_exception_data.gpu_id = user_gpu_id;
memory_exception_data.failure.imprecise = true;
rcu_read_lock();
idr_for_each_entry_continue(&p->event_idr, ev, id) {
if (ev->type == KFD_EVENT_TYPE_HW_EXCEPTION) {
spin_lock(&ev->lock);
ev->hw_exception_data = hw_exception_data;
set_event(ev);
spin_unlock(&ev->lock);
}
if (ev->type == KFD_EVENT_TYPE_MEMORY) {
spin_lock(&ev->lock);
ev->memory_exception_data = memory_exception_data;
set_event(ev);
spin_unlock(&ev->lock);
}
}
rcu_read_unlock();
/* user application will handle SIGBUS signal */
send_sig(SIGBUS, p->lead_thread, 0);
kfd_unref_process(p);
}