kernel/drivers/gpu/drm/nouveau/nvkm/falcon/cmdq.c

215 lines
5.5 KiB
C
Raw Normal View History

2024-07-22 17:22:30 +08:00
/*
* Copyright (c) 2017, NVIDIA CORPORATION. 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 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 "qmgr.h"
static bool
nvkm_falcon_cmdq_has_room(struct nvkm_falcon_cmdq *cmdq, u32 size, bool *rewind)
{
u32 head = nvkm_falcon_rd32(cmdq->qmgr->falcon, cmdq->head_reg);
u32 tail = nvkm_falcon_rd32(cmdq->qmgr->falcon, cmdq->tail_reg);
u32 free;
size = ALIGN(size, QUEUE_ALIGNMENT);
if (head >= tail) {
free = cmdq->offset + cmdq->size - head;
free -= HDR_SIZE;
if (size > free) {
*rewind = true;
head = cmdq->offset;
}
}
if (head < tail)
free = tail - head - 1;
return size <= free;
}
static void
nvkm_falcon_cmdq_push(struct nvkm_falcon_cmdq *cmdq, void *data, u32 size)
{
struct nvkm_falcon *falcon = cmdq->qmgr->falcon;
nvkm_falcon_load_dmem(falcon, data, cmdq->position, size, 0);
cmdq->position += ALIGN(size, QUEUE_ALIGNMENT);
}
static void
nvkm_falcon_cmdq_rewind(struct nvkm_falcon_cmdq *cmdq)
{
struct nvfw_falcon_cmd cmd;
cmd.unit_id = NV_FALCON_CMD_UNIT_ID_REWIND;
cmd.size = sizeof(cmd);
nvkm_falcon_cmdq_push(cmdq, &cmd, cmd.size);
cmdq->position = cmdq->offset;
}
static int
nvkm_falcon_cmdq_open(struct nvkm_falcon_cmdq *cmdq, u32 size)
{
struct nvkm_falcon *falcon = cmdq->qmgr->falcon;
bool rewind = false;
mutex_lock(&cmdq->mutex);
if (!nvkm_falcon_cmdq_has_room(cmdq, size, &rewind)) {
FLCNQ_DBG(cmdq, "queue full");
mutex_unlock(&cmdq->mutex);
return -EAGAIN;
}
cmdq->position = nvkm_falcon_rd32(falcon, cmdq->head_reg);
if (rewind)
nvkm_falcon_cmdq_rewind(cmdq);
return 0;
}
static void
nvkm_falcon_cmdq_close(struct nvkm_falcon_cmdq *cmdq)
{
nvkm_falcon_wr32(cmdq->qmgr->falcon, cmdq->head_reg, cmdq->position);
mutex_unlock(&cmdq->mutex);
}
static int
nvkm_falcon_cmdq_write(struct nvkm_falcon_cmdq *cmdq, struct nvfw_falcon_cmd *cmd)
{
static unsigned timeout = 2000;
unsigned long end_jiffies = jiffies + msecs_to_jiffies(timeout);
int ret = -EAGAIN;
while (ret == -EAGAIN && time_before(jiffies, end_jiffies))
ret = nvkm_falcon_cmdq_open(cmdq, cmd->size);
if (ret) {
FLCNQ_ERR(cmdq, "timeout waiting for queue space");
return ret;
}
nvkm_falcon_cmdq_push(cmdq, cmd, cmd->size);
nvkm_falcon_cmdq_close(cmdq);
return ret;
}
/* specifies that we want to know the command status in the answer message */
#define CMD_FLAGS_STATUS BIT(0)
/* specifies that we want an interrupt when the answer message is queued */
#define CMD_FLAGS_INTR BIT(1)
int
nvkm_falcon_cmdq_send(struct nvkm_falcon_cmdq *cmdq, struct nvfw_falcon_cmd *cmd,
nvkm_falcon_qmgr_callback cb, void *priv,
unsigned long timeout)
{
struct nvkm_falcon_qmgr_seq *seq;
int ret;
if (!wait_for_completion_timeout(&cmdq->ready,
msecs_to_jiffies(1000))) {
FLCNQ_ERR(cmdq, "timeout waiting for queue ready");
return -ETIMEDOUT;
}
seq = nvkm_falcon_qmgr_seq_acquire(cmdq->qmgr);
if (IS_ERR(seq))
return PTR_ERR(seq);
cmd->seq_id = seq->id;
cmd->ctrl_flags = CMD_FLAGS_STATUS | CMD_FLAGS_INTR;
seq->state = SEQ_STATE_USED;
seq->async = !timeout;
seq->callback = cb;
seq->priv = priv;
ret = nvkm_falcon_cmdq_write(cmdq, cmd);
if (ret) {
seq->state = SEQ_STATE_PENDING;
nvkm_falcon_qmgr_seq_release(cmdq->qmgr, seq);
return ret;
}
if (!seq->async) {
if (!wait_for_completion_timeout(&seq->done, timeout)) {
FLCNQ_ERR(cmdq, "timeout waiting for reply");
return -ETIMEDOUT;
}
ret = seq->result;
nvkm_falcon_qmgr_seq_release(cmdq->qmgr, seq);
}
return ret;
}
void
nvkm_falcon_cmdq_fini(struct nvkm_falcon_cmdq *cmdq)
{
reinit_completion(&cmdq->ready);
}
void
nvkm_falcon_cmdq_init(struct nvkm_falcon_cmdq *cmdq,
u32 index, u32 offset, u32 size)
{
const struct nvkm_falcon_func *func = cmdq->qmgr->falcon->func;
cmdq->head_reg = func->cmdq.head + index * func->cmdq.stride;
cmdq->tail_reg = func->cmdq.tail + index * func->cmdq.stride;
cmdq->offset = offset;
cmdq->size = size;
complete_all(&cmdq->ready);
FLCNQ_DBG(cmdq, "initialised @ index %d offset 0x%08x size 0x%08x",
index, cmdq->offset, cmdq->size);
}
void
nvkm_falcon_cmdq_del(struct nvkm_falcon_cmdq **pcmdq)
{
struct nvkm_falcon_cmdq *cmdq = *pcmdq;
if (cmdq) {
kfree(*pcmdq);
*pcmdq = NULL;
}
}
int
nvkm_falcon_cmdq_new(struct nvkm_falcon_qmgr *qmgr, const char *name,
struct nvkm_falcon_cmdq **pcmdq)
{
struct nvkm_falcon_cmdq *cmdq = *pcmdq;
if (!(cmdq = *pcmdq = kzalloc(sizeof(*cmdq), GFP_KERNEL)))
return -ENOMEM;
cmdq->qmgr = qmgr;
cmdq->name = name;
mutex_init(&cmdq->mutex);
init_completion(&cmdq->ready);
return 0;
}