2009-12-11 17:24:15 +08:00
|
|
|
/*
|
|
|
|
* Copyright (C) 2007 Ben Skeggs.
|
|
|
|
* 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, 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 (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 NONINFRINGEMENT.
|
|
|
|
* IN NO EVENT SHALL THE COPYRIGHT OWNER(S) 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.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
|
2016-05-20 07:22:55 +08:00
|
|
|
#include "nouveau_drv.h"
|
2012-07-20 06:17:34 +08:00
|
|
|
#include "nouveau_dma.h"
|
2010-01-08 08:57:39 +08:00
|
|
|
|
2009-12-11 17:24:15 +08:00
|
|
|
void
|
|
|
|
OUT_RINGp(struct nouveau_channel *chan, const void *data, unsigned nr_dwords)
|
|
|
|
{
|
|
|
|
bool is_iomem;
|
2012-07-20 06:17:34 +08:00
|
|
|
u32 *mem = ttm_kmap_obj_virtual(&chan->push.buffer->kmap, &is_iomem);
|
2009-12-11 17:24:15 +08:00
|
|
|
mem = &mem[chan->dma.cur];
|
|
|
|
if (is_iomem)
|
|
|
|
memcpy_toio((void __force __iomem *)mem, data, nr_dwords * 4);
|
|
|
|
else
|
|
|
|
memcpy(mem, data, nr_dwords * 4);
|
|
|
|
chan->dma.cur += nr_dwords;
|
|
|
|
}
|
|
|
|
|
2010-01-15 10:08:57 +08:00
|
|
|
/* Fetch and adjust GPU GET pointer
|
|
|
|
*
|
|
|
|
* Returns:
|
|
|
|
* value >= 0, the adjusted GET pointer
|
|
|
|
* -EINVAL if GET pointer currently outside main push buffer
|
|
|
|
* -EBUSY if timeout exceeded
|
|
|
|
*/
|
|
|
|
static inline int
|
2011-11-19 18:57:52 +08:00
|
|
|
READ_GET(struct nouveau_channel *chan, uint64_t *prev_get, int *timeout)
|
2009-12-11 17:24:15 +08:00
|
|
|
{
|
2011-11-19 18:57:52 +08:00
|
|
|
uint64_t val;
|
2009-12-11 17:24:15 +08:00
|
|
|
|
2015-08-20 12:54:15 +08:00
|
|
|
val = nvif_rd32(&chan->user, chan->user_get);
|
2011-11-19 18:57:52 +08:00
|
|
|
if (chan->user_get_hi)
|
2015-08-20 12:54:15 +08:00
|
|
|
val |= (uint64_t)nvif_rd32(&chan->user, chan->user_get_hi) << 32;
|
2010-01-15 10:08:57 +08:00
|
|
|
|
|
|
|
/* reset counter as long as GET is still advancing, this is
|
|
|
|
* to avoid misdetecting a GPU lockup if the GPU happens to
|
|
|
|
* just be processing an operation that takes a long time
|
|
|
|
*/
|
|
|
|
if (val != *prev_get) {
|
|
|
|
*prev_get = val;
|
|
|
|
*timeout = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((++*timeout & 0xff) == 0) {
|
2012-07-20 06:17:34 +08:00
|
|
|
udelay(1);
|
2010-01-15 10:08:57 +08:00
|
|
|
if (*timeout > 100000)
|
|
|
|
return -EBUSY;
|
2009-12-11 17:24:15 +08:00
|
|
|
}
|
|
|
|
|
2012-07-20 06:17:34 +08:00
|
|
|
if (val < chan->push.vma.offset ||
|
|
|
|
val > chan->push.vma.offset + (chan->dma.max << 2))
|
2010-01-15 10:08:57 +08:00
|
|
|
return -EINVAL;
|
|
|
|
|
2012-07-20 06:17:34 +08:00
|
|
|
return (val - chan->push.vma.offset) >> 2;
|
2009-12-11 17:24:15 +08:00
|
|
|
}
|
|
|
|
|
2010-02-11 14:37:26 +08:00
|
|
|
void
|
|
|
|
nv50_dma_push(struct nouveau_channel *chan, struct nouveau_bo *bo,
|
2010-02-12 08:27:35 +08:00
|
|
|
int delta, int length)
|
2010-02-11 14:37:26 +08:00
|
|
|
{
|
2015-08-20 12:54:15 +08:00
|
|
|
struct nouveau_cli *cli = (void *)chan->user.client;
|
2012-07-20 06:17:34 +08:00
|
|
|
struct nouveau_bo *pb = chan->push.buffer;
|
2015-01-14 13:36:34 +08:00
|
|
|
struct nvkm_vma *vma;
|
2010-02-11 14:37:26 +08:00
|
|
|
int ip = (chan->dma.ib_put * 2) + chan->dma.ib_base;
|
2011-06-07 11:23:47 +08:00
|
|
|
u64 offset;
|
|
|
|
|
2014-08-10 02:10:22 +08:00
|
|
|
vma = nouveau_bo_vma_find(bo, cli->vm);
|
2011-06-07 11:23:47 +08:00
|
|
|
BUG_ON(!vma);
|
|
|
|
offset = vma->offset + delta;
|
2010-02-11 14:37:26 +08:00
|
|
|
|
|
|
|
BUG_ON(chan->dma.ib_free < 1);
|
2012-07-20 06:17:34 +08:00
|
|
|
|
2010-02-12 09:11:54 +08:00
|
|
|
nouveau_bo_wr32(pb, ip++, lower_32_bits(offset));
|
2010-02-12 08:27:35 +08:00
|
|
|
nouveau_bo_wr32(pb, ip++, upper_32_bits(offset) | length << 8);
|
2010-02-11 14:37:26 +08:00
|
|
|
|
|
|
|
chan->dma.ib_put = (chan->dma.ib_put + 1) & chan->dma.ib_max;
|
2010-02-26 03:00:38 +08:00
|
|
|
|
2013-12-11 18:34:45 +08:00
|
|
|
mb();
|
2010-02-26 03:00:38 +08:00
|
|
|
/* Flush writes. */
|
|
|
|
nouveau_bo_rd32(pb, 0);
|
|
|
|
|
2015-08-20 12:54:15 +08:00
|
|
|
nvif_wr32(&chan->user, 0x8c, chan->dma.ib_put);
|
2010-02-11 14:37:26 +08:00
|
|
|
chan->dma.ib_free--;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
nv50_dma_push_wait(struct nouveau_channel *chan, int count)
|
|
|
|
{
|
|
|
|
uint32_t cnt = 0, prev_get = 0;
|
|
|
|
|
|
|
|
while (chan->dma.ib_free < count) {
|
2015-08-20 12:54:15 +08:00
|
|
|
uint32_t get = nvif_rd32(&chan->user, 0x88);
|
2010-02-11 14:37:26 +08:00
|
|
|
if (get != prev_get) {
|
|
|
|
prev_get = get;
|
|
|
|
cnt = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((++cnt & 0xff) == 0) {
|
|
|
|
DRM_UDELAY(1);
|
|
|
|
if (cnt > 100000)
|
|
|
|
return -EBUSY;
|
|
|
|
}
|
|
|
|
|
|
|
|
chan->dma.ib_free = get - chan->dma.ib_put;
|
|
|
|
if (chan->dma.ib_free <= 0)
|
2010-09-30 07:09:42 +08:00
|
|
|
chan->dma.ib_free += chan->dma.ib_max;
|
2010-02-11 14:37:26 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
nv50_dma_wait(struct nouveau_channel *chan, int slots, int count)
|
|
|
|
{
|
2011-11-19 18:57:52 +08:00
|
|
|
uint64_t prev_get = 0;
|
|
|
|
int ret, cnt = 0;
|
2010-02-11 14:37:26 +08:00
|
|
|
|
|
|
|
ret = nv50_dma_push_wait(chan, slots + 1);
|
|
|
|
if (unlikely(ret))
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
while (chan->dma.free < count) {
|
|
|
|
int get = READ_GET(chan, &prev_get, &cnt);
|
|
|
|
if (unlikely(get < 0)) {
|
|
|
|
if (get == -EINVAL)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
return get;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (get <= chan->dma.cur) {
|
|
|
|
chan->dma.free = chan->dma.max - chan->dma.cur;
|
|
|
|
if (chan->dma.free >= count)
|
|
|
|
break;
|
|
|
|
|
|
|
|
FIRE_RING(chan);
|
|
|
|
do {
|
|
|
|
get = READ_GET(chan, &prev_get, &cnt);
|
|
|
|
if (unlikely(get < 0)) {
|
|
|
|
if (get == -EINVAL)
|
|
|
|
continue;
|
|
|
|
return get;
|
|
|
|
}
|
|
|
|
} while (get == 0);
|
|
|
|
chan->dma.cur = 0;
|
|
|
|
chan->dma.put = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
chan->dma.free = get - chan->dma.cur - 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-12-11 17:24:15 +08:00
|
|
|
int
|
2010-02-11 14:37:26 +08:00
|
|
|
nouveau_dma_wait(struct nouveau_channel *chan, int slots, int size)
|
2009-12-11 17:24:15 +08:00
|
|
|
{
|
2011-11-19 18:57:52 +08:00
|
|
|
uint64_t prev_get = 0;
|
|
|
|
int cnt = 0, get;
|
2009-12-11 17:24:15 +08:00
|
|
|
|
2010-02-11 14:37:26 +08:00
|
|
|
if (chan->dma.ib_max)
|
|
|
|
return nv50_dma_wait(chan, slots, size);
|
|
|
|
|
2009-12-11 17:24:15 +08:00
|
|
|
while (chan->dma.free < size) {
|
2010-01-15 10:08:57 +08:00
|
|
|
get = READ_GET(chan, &prev_get, &cnt);
|
|
|
|
if (unlikely(get == -EBUSY))
|
|
|
|
return -EBUSY;
|
2009-12-11 17:24:15 +08:00
|
|
|
|
|
|
|
/* loop until we have a usable GET pointer. the value
|
|
|
|
* we read from the GPU may be outside the main ring if
|
|
|
|
* PFIFO is processing a buffer called from the main ring,
|
|
|
|
* discard these values until something sensible is seen.
|
|
|
|
*
|
|
|
|
* the other case we discard GET is while the GPU is fetching
|
|
|
|
* from the SKIPS area, so the code below doesn't have to deal
|
|
|
|
* with some fun corner cases.
|
|
|
|
*/
|
2010-01-15 10:08:57 +08:00
|
|
|
if (unlikely(get == -EINVAL) || get < NOUVEAU_DMA_SKIPS)
|
2009-12-11 17:24:15 +08:00
|
|
|
continue;
|
|
|
|
|
|
|
|
if (get <= chan->dma.cur) {
|
|
|
|
/* engine is fetching behind us, or is completely
|
|
|
|
* idle (GET == PUT) so we have free space up until
|
|
|
|
* the end of the push buffer
|
|
|
|
*
|
|
|
|
* we can only hit that path once per call due to
|
|
|
|
* looping back to the beginning of the push buffer,
|
|
|
|
* we'll hit the fetching-ahead-of-us path from that
|
|
|
|
* point on.
|
|
|
|
*
|
|
|
|
* the *one* exception to that rule is if we read
|
|
|
|
* GET==PUT, in which case the below conditional will
|
|
|
|
* always succeed and break us out of the wait loop.
|
|
|
|
*/
|
|
|
|
chan->dma.free = chan->dma.max - chan->dma.cur;
|
|
|
|
if (chan->dma.free >= size)
|
|
|
|
break;
|
|
|
|
|
|
|
|
/* not enough space left at the end of the push buffer,
|
|
|
|
* instruct the GPU to jump back to the start right
|
|
|
|
* after processing the currently pending commands.
|
|
|
|
*/
|
2012-07-20 06:17:34 +08:00
|
|
|
OUT_RING(chan, chan->push.vma.offset | 0x20000000);
|
2010-01-15 10:08:57 +08:00
|
|
|
|
|
|
|
/* wait for GET to depart from the skips area.
|
|
|
|
* prevents writing GET==PUT and causing a race
|
|
|
|
* condition that causes us to think the GPU is
|
|
|
|
* idle when it's not.
|
|
|
|
*/
|
|
|
|
do {
|
|
|
|
get = READ_GET(chan, &prev_get, &cnt);
|
|
|
|
if (unlikely(get == -EBUSY))
|
|
|
|
return -EBUSY;
|
|
|
|
if (unlikely(get == -EINVAL))
|
|
|
|
continue;
|
|
|
|
} while (get <= NOUVEAU_DMA_SKIPS);
|
2009-12-11 17:24:15 +08:00
|
|
|
WRITE_PUT(NOUVEAU_DMA_SKIPS);
|
|
|
|
|
|
|
|
/* we're now submitting commands at the start of
|
|
|
|
* the push buffer.
|
|
|
|
*/
|
|
|
|
chan->dma.cur =
|
|
|
|
chan->dma.put = NOUVEAU_DMA_SKIPS;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* engine fetching ahead of us, we have space up until the
|
|
|
|
* current GET pointer. the "- 1" is to ensure there's
|
|
|
|
* space left to emit a jump back to the beginning of the
|
|
|
|
* push buffer if we require it. we can never get GET == PUT
|
|
|
|
* here, so this is safe.
|
|
|
|
*/
|
|
|
|
chan->dma.free = get - chan->dma.cur - 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|