1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
|
/*
* Copyright 2014 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.
*/
/*
* KFD Interrupts.
*
* AMD GPUs deliver interrupts by pushing an interrupt description onto the
* interrupt ring and then sending an interrupt. KGD receives the interrupt
* in ISR and sends us a pointer to each new entry on the interrupt ring.
*
* We generally can't process interrupt-signaled events from ISR, so we call
* out to each interrupt client module (currently only the scheduler) to ask if
* each interrupt is interesting. If they return true, then it requires further
* processing so we copy it to an internal interrupt ring and call each
* interrupt client again from a work-queue.
*
* There's no acknowledgment for the interrupts we use. The hardware simply
* queues a new interrupt each time without waiting.
*
* The fixed-size internal queue means that it's possible for us to lose
* interrupts because we have no back-pressure to the hardware.
*/
#include <linux/slab.h>
#include <linux/device.h>
#include "kfd_priv.h"
#define KFD_INTERRUPT_RING_SIZE 1024
static void interrupt_wq(struct work_struct *);
int kfd_interrupt_init(struct kfd_dev *kfd)
{
void *interrupt_ring = kmalloc_array(KFD_INTERRUPT_RING_SIZE,
kfd->device_info->ih_ring_entry_size,
GFP_KERNEL);
if (!interrupt_ring)
return -ENOMEM;
kfd->interrupt_ring = interrupt_ring;
kfd->interrupt_ring_size =
KFD_INTERRUPT_RING_SIZE * kfd->device_info->ih_ring_entry_size;
atomic_set(&kfd->interrupt_ring_wptr, 0);
atomic_set(&kfd->interrupt_ring_rptr, 0);
spin_lock_init(&kfd->interrupt_lock);
INIT_WORK(&kfd->interrupt_work, interrupt_wq);
kfd->interrupts_active = true;
/*
* After this function returns, the interrupt will be enabled. This
* barrier ensures that the interrupt running on a different processor
* sees all the above writes.
*/
smp_wmb();
return 0;
}
void kfd_interrupt_exit(struct kfd_dev *kfd)
{
/*
* Stop the interrupt handler from writing to the ring and scheduling
* workqueue items. The spinlock ensures that any interrupt running
* after we have unlocked sees interrupts_active = false.
*/
unsigned long flags;
spin_lock_irqsave(&kfd->interrupt_lock, flags);
kfd->interrupts_active = false;
spin_unlock_irqrestore(&kfd->interrupt_lock, flags);
/*
* Flush_scheduled_work ensures that there are no outstanding
* work-queue items that will access interrupt_ring. New work items
* can't be created because we stopped interrupt handling above.
*/
flush_scheduled_work();
kfree(kfd->interrupt_ring);
}
/*
* This assumes that it can't be called concurrently with itself
* but only with dequeue_ih_ring_entry.
*/
bool enqueue_ih_ring_entry(struct kfd_dev *kfd, const void *ih_ring_entry)
{
unsigned int rptr = atomic_read(&kfd->interrupt_ring_rptr);
unsigned int wptr = atomic_read(&kfd->interrupt_ring_wptr);
if ((rptr - wptr) % kfd->interrupt_ring_size ==
kfd->device_info->ih_ring_entry_size) {
/* This is very bad, the system is likely to hang. */
dev_err_ratelimited(kfd_chardev(),
"Interrupt ring overflow, dropping interrupt.\n");
return false;
}
memcpy(kfd->interrupt_ring + wptr, ih_ring_entry,
kfd->device_info->ih_ring_entry_size);
wptr = (wptr + kfd->device_info->ih_ring_entry_size) %
kfd->interrupt_ring_size;
smp_wmb(); /* Ensure memcpy'd data is visible before wptr update. */
atomic_set(&kfd->interrupt_ring_wptr, wptr);
return true;
}
/*
* This assumes that it can't be called concurrently with itself
* but only with enqueue_ih_ring_entry.
*/
static bool dequeue_ih_ring_entry(struct kfd_dev *kfd, void *ih_ring_entry)
{
/*
* Assume that wait queues have an implicit barrier, i.e. anything that
* happened in the ISR before it queued work is visible.
*/
unsigned int wptr = atomic_read(&kfd->interrupt_ring_wptr);
unsigned int rptr = atomic_read(&kfd->interrupt_ring_rptr);
if (rptr == wptr)
return false;
memcpy(ih_ring_entry, kfd->interrupt_ring + rptr,
kfd->device_info->ih_ring_entry_size);
rptr = (rptr + kfd->device_info->ih_ring_entry_size) %
kfd->interrupt_ring_size;
/*
* Ensure the rptr write update is not visible until
* memcpy has finished reading.
*/
smp_mb();
atomic_set(&kfd->interrupt_ring_rptr, rptr);
return true;
}
static void interrupt_wq(struct work_struct *work)
{
struct kfd_dev *dev = container_of(work, struct kfd_dev,
interrupt_work);
uint32_t ih_ring_entry[DIV_ROUND_UP(
dev->device_info->ih_ring_entry_size,
sizeof(uint32_t))];
while (dequeue_ih_ring_entry(dev, ih_ring_entry))
dev->device_info->event_interrupt_class->interrupt_wq(dev,
ih_ring_entry);
}
bool interrupt_is_wanted(struct kfd_dev *dev, const uint32_t *ih_ring_entry)
{
/* integer and bitwise OR so there is no boolean short-circuiting */
unsigned int wanted = 0;
wanted |= dev->device_info->event_interrupt_class->interrupt_isr(dev,
ih_ring_entry);
return wanted != 0;
}
|
