blob: 0382ed09eb9872a4148d95088532cfaa7598d93f [file] [log] [blame]
/*
* Copyright (c) 2016-2017, The Linux Foundation. All rights reserved.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 and
* only version 2 as published by the Free Software Foundation.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
*/
#define pr_fmt(fmt) "[drm:%s] " fmt, __func__
#include "sde_kms.h"
#include "sde_hw_lm.h"
#include "sde_hw_ctl.h"
#include "sde_hw_cdm.h"
#include "sde_hw_dspp.h"
#include "sde_hw_pingpong.h"
#include "sde_hw_intf.h"
#include "sde_hw_wb.h"
#include "sde_encoder.h"
#include "sde_connector.h"
#include "sde_hw_dsc.h"
#include "sde_hw_rot.h"
#define RESERVED_BY_OTHER(h, r) \
((h)->rsvp && ((h)->rsvp->enc_id != (r)->enc_id))
#define RM_RQ_LOCK(r) ((r)->top_ctrl & BIT(SDE_RM_TOPCTL_RESERVE_LOCK))
#define RM_RQ_CLEAR(r) ((r)->top_ctrl & BIT(SDE_RM_TOPCTL_RESERVE_CLEAR))
#define RM_RQ_DSPP(r) ((r)->top_ctrl & BIT(SDE_RM_TOPCTL_DSPP))
#define RM_IS_TOPOLOGY_MATCH(t, r) ((t).num_lm == (r).num_lm && \
(t).num_comp_enc == (r).num_enc && \
(t).num_intf == (r).num_intf)
struct sde_rm_topology_def {
enum sde_rm_topology_name top_name;
int num_lm;
int num_comp_enc;
int num_intf;
int num_ctl;
int needs_split_display;
};
static const struct sde_rm_topology_def g_top_table[] = {
{ SDE_RM_TOPOLOGY_NONE, 0, 0, 0, 0, false },
{ SDE_RM_TOPOLOGY_SINGLEPIPE, 1, 0, 1, 1, false },
{ SDE_RM_TOPOLOGY_SINGLEPIPE_DSC, 1, 1, 1, 1, false },
{ SDE_RM_TOPOLOGY_DUALPIPE, 2, 0, 2, 2, true },
{ SDE_RM_TOPOLOGY_DUALPIPE_DSC, 2, 2, 2, 2, true },
{ SDE_RM_TOPOLOGY_DUALPIPE_3DMERGE, 2, 0, 1, 1, false },
{ SDE_RM_TOPOLOGY_DUALPIPE_3DMERGE_DSC, 2, 1, 1, 1, false },
{ SDE_RM_TOPOLOGY_DUALPIPE_DSCMERGE, 2, 2, 1, 1, false },
{ SDE_RM_TOPOLOGY_PPSPLIT, 1, 0, 2, 1, true },
};
/**
* struct sde_rm_requirements - Reservation requirements parameter bundle
* @top_ctrl: topology control preference from kernel client
* @top: selected topology for the display
* @hw_res: Hardware resources required as reported by the encoders
*/
struct sde_rm_requirements {
uint64_t top_ctrl;
const struct sde_rm_topology_def *topology;
struct sde_encoder_hw_resources hw_res;
};
/**
* struct sde_rm_rsvp - Use Case Reservation tagging structure
* Used to tag HW blocks as reserved by a CRTC->Encoder->Connector chain
* By using as a tag, rather than lists of pointers to HW blocks used
* we can avoid some list management since we don't know how many blocks
* of each type a given use case may require.
* @list: List head for list of all reservations
* @seq: Global RSVP sequence number for debugging, especially for
* differentiating differenct allocations for same encoder.
* @enc_id: Reservations are tracked by Encoder DRM object ID.
* CRTCs may be connected to multiple Encoders.
* An encoder or connector id identifies the display path.
* @topology DRM<->HW topology use case
*/
struct sde_rm_rsvp {
struct list_head list;
uint32_t seq;
uint32_t enc_id;
enum sde_rm_topology_name topology;
};
/**
* struct sde_rm_hw_blk - hardware block tracking list member
* @list: List head for list of all hardware blocks tracking items
* @rsvp: Pointer to use case reservation if reserved by a client
* @rsvp_nxt: Temporary pointer used during reservation to the incoming
* request. Will be swapped into rsvp if proposal is accepted
* @type: Type of hardware block this structure tracks
* @id: Hardware ID number, within it's own space, ie. LM_X
* @catalog: Pointer to the hardware catalog entry for this block
* @hw: Pointer to the hardware register access object for this block
*/
struct sde_rm_hw_blk {
struct list_head list;
struct sde_rm_rsvp *rsvp;
struct sde_rm_rsvp *rsvp_nxt;
enum sde_hw_blk_type type;
uint32_t id;
struct sde_hw_blk *hw;
};
/**
* sde_rm_dbg_rsvp_stage - enum of steps in making reservation for event logging
*/
enum sde_rm_dbg_rsvp_stage {
SDE_RM_STAGE_BEGIN,
SDE_RM_STAGE_AFTER_CLEAR,
SDE_RM_STAGE_AFTER_RSVPNEXT,
SDE_RM_STAGE_FINAL
};
static void _sde_rm_print_rsvps(
struct sde_rm *rm,
enum sde_rm_dbg_rsvp_stage stage)
{
struct sde_rm_rsvp *rsvp;
struct sde_rm_hw_blk *blk;
enum sde_hw_blk_type type;
SDE_DEBUG("%d\n", stage);
list_for_each_entry(rsvp, &rm->rsvps, list) {
SDE_DEBUG("%d rsvp[s%ue%u] topology %d\n", stage, rsvp->seq,
rsvp->enc_id, rsvp->topology);
SDE_EVT32(stage, rsvp->seq, rsvp->enc_id, rsvp->topology);
}
for (type = 0; type < SDE_HW_BLK_MAX; type++) {
list_for_each_entry(blk, &rm->hw_blks[type], list) {
if (!blk->rsvp && !blk->rsvp_nxt)
continue;
SDE_DEBUG("%d rsvp[s%ue%u->s%ue%u] %d %d\n", stage,
(blk->rsvp) ? blk->rsvp->seq : 0,
(blk->rsvp) ? blk->rsvp->enc_id : 0,
(blk->rsvp_nxt) ? blk->rsvp_nxt->seq : 0,
(blk->rsvp_nxt) ? blk->rsvp_nxt->enc_id : 0,
blk->type, blk->id);
SDE_EVT32(stage,
(blk->rsvp) ? blk->rsvp->seq : 0,
(blk->rsvp) ? blk->rsvp->enc_id : 0,
(blk->rsvp_nxt) ? blk->rsvp_nxt->seq : 0,
(blk->rsvp_nxt) ? blk->rsvp_nxt->enc_id : 0,
blk->type, blk->id);
}
}
}
struct sde_hw_mdp *sde_rm_get_mdp(struct sde_rm *rm)
{
return rm->hw_mdp;
}
void sde_rm_init_hw_iter(
struct sde_rm_hw_iter *iter,
uint32_t enc_id,
enum sde_hw_blk_type type)
{
memset(iter, 0, sizeof(*iter));
iter->enc_id = enc_id;
iter->type = type;
}
static bool _sde_rm_get_hw_locked(struct sde_rm *rm, struct sde_rm_hw_iter *i)
{
struct list_head *blk_list;
if (!rm || !i || i->type >= SDE_HW_BLK_MAX) {
SDE_ERROR("invalid rm\n");
return false;
}
i->hw = NULL;
blk_list = &rm->hw_blks[i->type];
if (i->blk && (&i->blk->list == blk_list)) {
SDE_DEBUG("attempt resume iteration past last\n");
return false;
}
i->blk = list_prepare_entry(i->blk, blk_list, list);
list_for_each_entry_continue(i->blk, blk_list, list) {
struct sde_rm_rsvp *rsvp = i->blk->rsvp;
if (i->blk->type != i->type) {
SDE_ERROR("found incorrect block type %d on %d list\n",
i->blk->type, i->type);
return false;
}
if ((i->enc_id == 0) || (rsvp && rsvp->enc_id == i->enc_id)) {
i->hw = i->blk->hw;
SDE_DEBUG("found type %d id %d for enc %d\n",
i->type, i->blk->id, i->enc_id);
return true;
}
}
SDE_DEBUG("no match, type %d for enc %d\n", i->type, i->enc_id);
return false;
}
bool sde_rm_get_hw(struct sde_rm *rm, struct sde_rm_hw_iter *i)
{
bool ret;
mutex_lock(&rm->rm_lock);
ret = _sde_rm_get_hw_locked(rm, i);
mutex_unlock(&rm->rm_lock);
return ret;
}
static void _sde_rm_hw_destroy(enum sde_hw_blk_type type, void *hw)
{
switch (type) {
case SDE_HW_BLK_LM:
sde_hw_lm_destroy(hw);
break;
case SDE_HW_BLK_DSPP:
sde_hw_dspp_destroy(hw);
break;
case SDE_HW_BLK_CTL:
sde_hw_ctl_destroy(hw);
break;
case SDE_HW_BLK_CDM:
sde_hw_cdm_destroy(hw);
break;
case SDE_HW_BLK_PINGPONG:
sde_hw_pingpong_destroy(hw);
break;
case SDE_HW_BLK_INTF:
sde_hw_intf_destroy(hw);
break;
case SDE_HW_BLK_WB:
sde_hw_wb_destroy(hw);
break;
case SDE_HW_BLK_DSC:
sde_hw_dsc_destroy(hw);
break;
case SDE_HW_BLK_ROT:
sde_hw_rot_destroy(hw);
break;
case SDE_HW_BLK_SSPP:
/* SSPPs are not managed by the resource manager */
case SDE_HW_BLK_TOP:
/* Top is a singleton, not managed in hw_blks list */
case SDE_HW_BLK_MAX:
default:
SDE_ERROR("unsupported block type %d\n", type);
break;
}
}
int sde_rm_destroy(struct sde_rm *rm)
{
struct sde_rm_rsvp *rsvp_cur, *rsvp_nxt;
struct sde_rm_hw_blk *hw_cur, *hw_nxt;
enum sde_hw_blk_type type;
if (!rm) {
SDE_ERROR("invalid rm\n");
return -EINVAL;
}
list_for_each_entry_safe(rsvp_cur, rsvp_nxt, &rm->rsvps, list) {
list_del(&rsvp_cur->list);
kfree(rsvp_cur);
}
for (type = 0; type < SDE_HW_BLK_MAX; type++) {
list_for_each_entry_safe(hw_cur, hw_nxt, &rm->hw_blks[type],
list) {
list_del(&hw_cur->list);
_sde_rm_hw_destroy(hw_cur->type, hw_cur->hw);
kfree(hw_cur);
}
}
sde_hw_mdp_destroy(rm->hw_mdp);
rm->hw_mdp = NULL;
mutex_destroy(&rm->rm_lock);
return 0;
}
static int _sde_rm_hw_blk_create(
struct sde_rm *rm,
struct sde_mdss_cfg *cat,
void *mmio,
enum sde_hw_blk_type type,
uint32_t id,
void *hw_catalog_info)
{
struct sde_rm_hw_blk *blk;
struct sde_hw_mdp *hw_mdp;
void *hw;
hw_mdp = rm->hw_mdp;
switch (type) {
case SDE_HW_BLK_LM:
hw = sde_hw_lm_init(id, mmio, cat);
break;
case SDE_HW_BLK_DSPP:
hw = sde_hw_dspp_init(id, mmio, cat);
break;
case SDE_HW_BLK_CTL:
hw = sde_hw_ctl_init(id, mmio, cat);
break;
case SDE_HW_BLK_CDM:
hw = sde_hw_cdm_init(id, mmio, cat, hw_mdp);
break;
case SDE_HW_BLK_PINGPONG:
hw = sde_hw_pingpong_init(id, mmio, cat);
break;
case SDE_HW_BLK_INTF:
hw = sde_hw_intf_init(id, mmio, cat);
break;
case SDE_HW_BLK_WB:
hw = sde_hw_wb_init(id, mmio, cat, hw_mdp);
break;
case SDE_HW_BLK_DSC:
hw = sde_hw_dsc_init(id, mmio, cat);
break;
case SDE_HW_BLK_ROT:
hw = sde_hw_rot_init(id, mmio, cat);
break;
case SDE_HW_BLK_SSPP:
/* SSPPs are not managed by the resource manager */
case SDE_HW_BLK_TOP:
/* Top is a singleton, not managed in hw_blks list */
case SDE_HW_BLK_MAX:
default:
SDE_ERROR("unsupported block type %d\n", type);
return -EINVAL;
}
if (IS_ERR_OR_NULL(hw)) {
SDE_ERROR("failed hw object creation: type %d, err %ld\n",
type, PTR_ERR(hw));
return -EFAULT;
}
blk = kzalloc(sizeof(*blk), GFP_KERNEL);
if (!blk) {
_sde_rm_hw_destroy(type, hw);
return -ENOMEM;
}
blk->type = type;
blk->id = id;
blk->hw = hw;
list_add_tail(&blk->list, &rm->hw_blks[type]);
return 0;
}
int sde_rm_init(struct sde_rm *rm,
struct sde_mdss_cfg *cat,
void *mmio,
struct drm_device *dev)
{
int rc, i;
enum sde_hw_blk_type type;
if (!rm || !cat || !mmio || !dev) {
SDE_ERROR("invalid kms\n");
return -EINVAL;
}
/* Clear, setup lists */
memset(rm, 0, sizeof(*rm));
mutex_init(&rm->rm_lock);
INIT_LIST_HEAD(&rm->rsvps);
for (type = 0; type < SDE_HW_BLK_MAX; type++)
INIT_LIST_HEAD(&rm->hw_blks[type]);
rm->dev = dev;
/* Some of the sub-blocks require an mdptop to be created */
rm->hw_mdp = sde_hw_mdptop_init(MDP_TOP, mmio, cat);
if (IS_ERR_OR_NULL(rm->hw_mdp)) {
rc = PTR_ERR(rm->hw_mdp);
rm->hw_mdp = NULL;
SDE_ERROR("failed: mdp hw not available\n");
goto fail;
}
/* Interrogate HW catalog and create tracking items for hw blocks */
for (i = 0; i < cat->mixer_count; i++) {
struct sde_lm_cfg *lm = &cat->mixer[i];
if (lm->pingpong == PINGPONG_MAX) {
SDE_DEBUG("skip mixer %d without pingpong\n", lm->id);
continue;
}
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_LM,
cat->mixer[i].id, &cat->mixer[i]);
if (rc) {
SDE_ERROR("failed: lm hw not available\n");
goto fail;
}
if (!rm->lm_max_width) {
rm->lm_max_width = lm->sblk->maxwidth;
} else if (rm->lm_max_width != lm->sblk->maxwidth) {
/*
* Don't expect to have hw where lm max widths differ.
* If found, take the min.
*/
SDE_ERROR("unsupported: lm maxwidth differs\n");
if (rm->lm_max_width > lm->sblk->maxwidth)
rm->lm_max_width = lm->sblk->maxwidth;
}
}
for (i = 0; i < cat->dspp_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_DSPP,
cat->dspp[i].id, &cat->dspp[i]);
if (rc) {
SDE_ERROR("failed: dspp hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->pingpong_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_PINGPONG,
cat->pingpong[i].id, &cat->pingpong[i]);
if (rc) {
SDE_ERROR("failed: pp hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->dsc_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_DSC,
cat->dsc[i].id, &cat->dsc[i]);
if (rc) {
SDE_ERROR("failed: dsc hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->intf_count; i++) {
if (cat->intf[i].type == INTF_NONE) {
SDE_DEBUG("skip intf %d with type none\n", i);
continue;
}
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_INTF,
cat->intf[i].id, &cat->intf[i]);
if (rc) {
SDE_ERROR("failed: intf hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->wb_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_WB,
cat->wb[i].id, &cat->wb[i]);
if (rc) {
SDE_ERROR("failed: wb hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->rot_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_ROT,
cat->rot[i].id, &cat->rot[i]);
if (rc) {
SDE_ERROR("failed: rot hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->ctl_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_CTL,
cat->ctl[i].id, &cat->ctl[i]);
if (rc) {
SDE_ERROR("failed: ctl hw not available\n");
goto fail;
}
}
for (i = 0; i < cat->cdm_count; i++) {
rc = _sde_rm_hw_blk_create(rm, cat, mmio, SDE_HW_BLK_CDM,
cat->cdm[i].id, &cat->cdm[i]);
if (rc) {
SDE_ERROR("failed: cdm hw not available\n");
goto fail;
}
}
return 0;
fail:
sde_rm_destroy(rm);
return rc;
}
/**
* _sde_rm_check_lm_and_get_connected_blks - check if proposed layer mixer meets
* proposed use case requirements, incl. hardwired dependent blocks like
* pingpong, and dspp.
* @rm: sde resource manager handle
* @rsvp: reservation currently being created
* @reqs: proposed use case requirements
* @lm: proposed layer mixer, function checks if lm, and all other hardwired
* blocks connected to the lm (pp, dspp) are available and appropriate
* @dspp: output parameter, dspp block attached to the layer mixer.
* NULL if dspp was not available, or not matching requirements.
* @pp: output parameter, pingpong block attached to the layer mixer.
* NULL if dspp was not available, or not matching requirements.
* @primary_lm: if non-null, this function check if lm is compatible primary_lm
* as well as satisfying all other requirements
* @Return: true if lm matches all requirements, false otherwise
*/
static bool _sde_rm_check_lm_and_get_connected_blks(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
struct sde_rm_requirements *reqs,
struct sde_rm_hw_blk *lm,
struct sde_rm_hw_blk **dspp,
struct sde_rm_hw_blk **pp,
struct sde_rm_hw_blk *primary_lm)
{
const struct sde_lm_cfg *lm_cfg = to_sde_hw_mixer(lm->hw)->cap;
const struct sde_pingpong_cfg *pp_cfg;
struct sde_rm_hw_iter iter;
*dspp = NULL;
*pp = NULL;
SDE_DEBUG("check lm %d: dspp %d pp %d\n", lm_cfg->id, lm_cfg->dspp,
lm_cfg->pingpong);
/* Check if this layer mixer is a peer of the proposed primary LM */
if (primary_lm) {
const struct sde_lm_cfg *prim_lm_cfg =
to_sde_hw_mixer(primary_lm->hw)->cap;
if (!test_bit(lm_cfg->id, &prim_lm_cfg->lm_pair_mask)) {
SDE_DEBUG("lm %d not peer of lm %d\n", lm_cfg->id,
prim_lm_cfg->id);
return false;
}
}
/* Matches user requirements? */
if ((RM_RQ_DSPP(reqs) && lm_cfg->dspp == DSPP_MAX) ||
(!RM_RQ_DSPP(reqs) && lm_cfg->dspp != DSPP_MAX)) {
SDE_DEBUG("dspp req mismatch lm %d reqdspp %d, lm->dspp %d\n",
lm_cfg->id, (bool)(RM_RQ_DSPP(reqs)),
lm_cfg->dspp);
return false;
}
/* Already reserved? */
if (RESERVED_BY_OTHER(lm, rsvp)) {
SDE_DEBUG("lm %d already reserved\n", lm_cfg->id);
return false;
}
if (lm_cfg->dspp != DSPP_MAX) {
sde_rm_init_hw_iter(&iter, 0, SDE_HW_BLK_DSPP);
while (_sde_rm_get_hw_locked(rm, &iter)) {
if (iter.blk->id == lm_cfg->dspp) {
*dspp = iter.blk;
break;
}
}
if (!*dspp) {
SDE_DEBUG("lm %d failed to retrieve dspp %d\n", lm->id,
lm_cfg->dspp);
return false;
}
if (RESERVED_BY_OTHER(*dspp, rsvp)) {
SDE_DEBUG("lm %d dspp %d already reserved\n",
lm->id, (*dspp)->id);
return false;
}
}
sde_rm_init_hw_iter(&iter, 0, SDE_HW_BLK_PINGPONG);
while (_sde_rm_get_hw_locked(rm, &iter)) {
if (iter.blk->id == lm_cfg->pingpong) {
*pp = iter.blk;
break;
}
}
if (!*pp) {
SDE_ERROR("failed to get pp on lm %d\n", lm_cfg->pingpong);
return false;
}
if (RESERVED_BY_OTHER(*pp, rsvp)) {
SDE_DEBUG("lm %d pp %d already reserved\n", lm->id,
(*pp)->id);
*dspp = NULL;
return false;
}
pp_cfg = to_sde_hw_pingpong((*pp)->hw)->caps;
if ((reqs->topology->top_name == SDE_RM_TOPOLOGY_PPSPLIT) &&
!(test_bit(SDE_PINGPONG_SPLIT, &pp_cfg->features))) {
SDE_DEBUG("pp %d doesn't support ppsplit\n", pp_cfg->id);
*dspp = NULL;
return false;
}
return true;
}
static int _sde_rm_reserve_lms(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
struct sde_rm_requirements *reqs)
{
struct sde_rm_hw_blk *lm[MAX_BLOCKS];
struct sde_rm_hw_blk *dspp[MAX_BLOCKS];
struct sde_rm_hw_blk *pp[MAX_BLOCKS];
struct sde_rm_hw_iter iter_i, iter_j;
int lm_count = 0;
int i, rc = 0;
if (!reqs->topology->num_lm) {
SDE_ERROR("invalid number of lm: %d\n", reqs->topology->num_lm);
return -EINVAL;
}
/* Find a primary mixer */
sde_rm_init_hw_iter(&iter_i, 0, SDE_HW_BLK_LM);
while (lm_count != reqs->topology->num_lm &&
_sde_rm_get_hw_locked(rm, &iter_i)) {
memset(&lm, 0, sizeof(lm));
memset(&dspp, 0, sizeof(dspp));
memset(&pp, 0, sizeof(pp));
lm_count = 0;
lm[lm_count] = iter_i.blk;
if (!_sde_rm_check_lm_and_get_connected_blks(rm, rsvp, reqs,
lm[lm_count], &dspp[lm_count], &pp[lm_count],
NULL))
continue;
++lm_count;
/* Valid primary mixer found, find matching peers */
sde_rm_init_hw_iter(&iter_j, 0, SDE_HW_BLK_LM);
while (lm_count != reqs->topology->num_lm &&
_sde_rm_get_hw_locked(rm, &iter_j)) {
if (iter_i.blk == iter_j.blk)
continue;
if (!_sde_rm_check_lm_and_get_connected_blks(rm, rsvp,
reqs, iter_j.blk, &dspp[lm_count],
&pp[lm_count], iter_i.blk))
continue;
lm[lm_count] = iter_j.blk;
++lm_count;
}
}
if (lm_count != reqs->topology->num_lm) {
SDE_DEBUG("unable to find appropriate mixers\n");
return -ENAVAIL;
}
for (i = 0; i < ARRAY_SIZE(lm); i++) {
if (!lm[i])
break;
lm[i]->rsvp_nxt = rsvp;
pp[i]->rsvp_nxt = rsvp;
if (dspp[i])
dspp[i]->rsvp_nxt = rsvp;
SDE_EVT32(lm[i]->type, rsvp->enc_id, lm[i]->id, pp[i]->id,
dspp[i] ? dspp[i]->id : 0);
}
if (reqs->topology->top_name == SDE_RM_TOPOLOGY_PPSPLIT) {
/* reserve a free PINGPONG_SLAVE block */
rc = -ENAVAIL;
sde_rm_init_hw_iter(&iter_i, 0, SDE_HW_BLK_PINGPONG);
while (_sde_rm_get_hw_locked(rm, &iter_i)) {
const struct sde_hw_pingpong *pp =
to_sde_hw_pingpong(iter_i.blk->hw);
const struct sde_pingpong_cfg *pp_cfg = pp->caps;
if (!(test_bit(SDE_PINGPONG_SLAVE, &pp_cfg->features)))
continue;
if (RESERVED_BY_OTHER(iter_i.blk, rsvp))
continue;
iter_i.blk->rsvp_nxt = rsvp;
rc = 0;
break;
}
}
return rc;
}
static int _sde_rm_reserve_ctls(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
const struct sde_rm_topology_def *top)
{
struct sde_rm_hw_blk *ctls[MAX_BLOCKS];
struct sde_rm_hw_iter iter;
int i = 0;
memset(&ctls, 0, sizeof(ctls));
sde_rm_init_hw_iter(&iter, 0, SDE_HW_BLK_CTL);
while (_sde_rm_get_hw_locked(rm, &iter)) {
const struct sde_hw_ctl *ctl = to_sde_hw_ctl(iter.blk->hw);
unsigned long features = ctl->caps->features;
bool has_split_display, has_ppsplit;
if (RESERVED_BY_OTHER(iter.blk, rsvp))
continue;
has_split_display = BIT(SDE_CTL_SPLIT_DISPLAY) & features;
has_ppsplit = BIT(SDE_CTL_PINGPONG_SPLIT) & features;
SDE_DEBUG("ctl %d caps 0x%lX\n", iter.blk->id, features);
if (top->needs_split_display != has_split_display)
continue;
if (top->top_name == SDE_RM_TOPOLOGY_PPSPLIT && !has_ppsplit)
continue;
ctls[i] = iter.blk;
SDE_DEBUG("ctl %d match\n", iter.blk->id);
if (++i == top->num_ctl)
break;
}
if (i != top->num_ctl)
return -ENAVAIL;
for (i = 0; i < ARRAY_SIZE(ctls) && i < top->num_ctl; i++) {
ctls[i]->rsvp_nxt = rsvp;
SDE_EVT32(ctls[i]->type, rsvp->enc_id, ctls[i]->id);
}
return 0;
}
static int _sde_rm_reserve_dsc(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
const struct sde_rm_topology_def *top)
{
struct sde_rm_hw_iter iter;
int alloc_count = 0;
int num_dsc_enc = top->num_lm;
if (!top->num_comp_enc)
return 0;
sde_rm_init_hw_iter(&iter, 0, SDE_HW_BLK_DSC);
while (_sde_rm_get_hw_locked(rm, &iter)) {
if (RESERVED_BY_OTHER(iter.blk, rsvp))
continue;
iter.blk->rsvp_nxt = rsvp;
SDE_EVT32(iter.blk->type, rsvp->enc_id, iter.blk->id);
if (++alloc_count == num_dsc_enc)
return 0;
}
SDE_ERROR("couldn't reserve %d dsc blocks for enc id %d\n",
num_dsc_enc, rsvp->enc_id);
return -ENAVAIL;
}
static int _sde_rm_reserve_cdm(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
uint32_t id,
enum sde_hw_blk_type type)
{
struct sde_rm_hw_iter iter;
sde_rm_init_hw_iter(&iter, 0, SDE_HW_BLK_CDM);
while (_sde_rm_get_hw_locked(rm, &iter)) {
const struct sde_hw_cdm *cdm = to_sde_hw_cdm(iter.blk->hw);
const struct sde_cdm_cfg *caps = cdm->caps;
bool match = false;
if (RESERVED_BY_OTHER(iter.blk, rsvp))
continue;
if (type == SDE_HW_BLK_INTF && id != INTF_MAX)
match = test_bit(id, &caps->intf_connect);
else if (type == SDE_HW_BLK_WB && id != WB_MAX)
match = test_bit(id, &caps->wb_connect);
SDE_DEBUG("type %d id %d, cdm intfs %lu wbs %lu match %d\n",
type, id, caps->intf_connect, caps->wb_connect,
match);
if (!match)
continue;
iter.blk->rsvp_nxt = rsvp;
SDE_EVT32(iter.blk->type, rsvp->enc_id, iter.blk->id);
break;
}
if (!iter.hw) {
SDE_ERROR("couldn't reserve cdm for type %d id %d\n", type, id);
return -ENAVAIL;
}
return 0;
}
static int _sde_rm_reserve_intf_or_wb(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
uint32_t id,
enum sde_hw_blk_type type,
bool needs_cdm)
{
struct sde_rm_hw_iter iter;
int ret = 0;
/* Find the block entry in the rm, and note the reservation */
sde_rm_init_hw_iter(&iter, 0, type);
while (_sde_rm_get_hw_locked(rm, &iter)) {
if (iter.blk->id != id)
continue;
if (RESERVED_BY_OTHER(iter.blk, rsvp)) {
SDE_ERROR("type %d id %d already reserved\n", type, id);
return -ENAVAIL;
}
iter.blk->rsvp_nxt = rsvp;
SDE_EVT32(iter.blk->type, rsvp->enc_id, iter.blk->id);
break;
}
/* Shouldn't happen since wbs / intfs are fixed at probe */
if (!iter.hw) {
SDE_ERROR("couldn't find type %d id %d\n", type, id);
return -EINVAL;
}
/* Expected only one intf or wb will request cdm */
if (needs_cdm)
ret = _sde_rm_reserve_cdm(rm, rsvp, id, type);
return ret;
}
static int _sde_rm_reserve_intf_related_hw(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
struct sde_encoder_hw_resources *hw_res)
{
int i, ret = 0;
u32 id;
for (i = 0; i < ARRAY_SIZE(hw_res->intfs); i++) {
if (hw_res->intfs[i] == INTF_MODE_NONE)
continue;
id = i + INTF_0;
ret = _sde_rm_reserve_intf_or_wb(rm, rsvp, id,
SDE_HW_BLK_INTF, hw_res->needs_cdm);
if (ret)
return ret;
}
for (i = 0; i < ARRAY_SIZE(hw_res->wbs); i++) {
if (hw_res->wbs[i] == INTF_MODE_NONE)
continue;
id = i + WB_0;
ret = _sde_rm_reserve_intf_or_wb(rm, rsvp, id,
SDE_HW_BLK_WB, hw_res->needs_cdm);
if (ret)
return ret;
}
return ret;
}
static int _sde_rm_make_next_rsvp(
struct sde_rm *rm,
struct drm_encoder *enc,
struct drm_crtc_state *crtc_state,
struct drm_connector_state *conn_state,
struct sde_rm_rsvp *rsvp,
struct sde_rm_requirements *reqs)
{
int ret;
struct sde_rm_topology_def topology;
/* Create reservation info, tag reserved blocks with it as we go */
rsvp->seq = ++rm->rsvp_next_seq;
rsvp->enc_id = enc->base.id;
rsvp->topology = reqs->topology->top_name;
list_add_tail(&rsvp->list, &rm->rsvps);
/*
* Assign LMs and blocks whose usage is tied to them: DSPP & Pingpong.
* Do assignment preferring to give away low-resource mixers first:
* - Check mixers without DSPPs
* - Only then allow to grab from mixers with DSPP capability
*/
ret = _sde_rm_reserve_lms(rm, rsvp, reqs);
if (ret && !RM_RQ_DSPP(reqs)) {
reqs->top_ctrl |= BIT(SDE_RM_TOPCTL_DSPP);
ret = _sde_rm_reserve_lms(rm, rsvp, reqs);
}
if (ret) {
SDE_ERROR("unable to find appropriate mixers\n");
return ret;
}
/*
* Do assignment preferring to give away low-resource CTLs first:
* - Check mixers without Split Display
* - Only then allow to grab from CTLs with split display capability
*/
_sde_rm_reserve_ctls(rm, rsvp, reqs->topology);
if (ret && !reqs->topology->needs_split_display) {
memcpy(&topology, reqs->topology, sizeof(topology));
topology.needs_split_display = true;
_sde_rm_reserve_ctls(rm, rsvp, &topology);
}
if (ret) {
SDE_ERROR("unable to find appropriate CTL\n");
return ret;
}
/* Assign INTFs, WBs, and blks whose usage is tied to them: CTL & CDM */
ret = _sde_rm_reserve_intf_related_hw(rm, rsvp, &reqs->hw_res);
if (ret)
return ret;
ret = _sde_rm_reserve_dsc(rm, rsvp, reqs->topology);
if (ret)
return ret;
return ret;
}
static int _sde_rm_populate_requirements(
struct sde_rm *rm,
struct drm_encoder *enc,
struct drm_crtc_state *crtc_state,
struct drm_connector_state *conn_state,
struct sde_rm_requirements *reqs)
{
const struct drm_display_mode *mode = &crtc_state->mode;
int i;
memset(reqs, 0, sizeof(*reqs));
reqs->top_ctrl = sde_connector_get_property(conn_state,
CONNECTOR_PROP_TOPOLOGY_CONTROL);
sde_encoder_get_hw_resources(enc, &reqs->hw_res, conn_state);
for (i = 0; i < SDE_RM_TOPOLOGY_MAX; i++) {
if (RM_IS_TOPOLOGY_MATCH(g_top_table[i],
reqs->hw_res.topology)) {
reqs->topology = &g_top_table[i];
break;
}
}
if (!reqs->topology) {
SDE_ERROR("invalid topology for the display\n");
return -EINVAL;
}
/* DSC blocks are hardwired for control path 0 and 1 */
if (reqs->topology->num_comp_enc)
reqs->top_ctrl |= BIT(SDE_RM_TOPCTL_DSPP);
SDE_DEBUG("top_ctrl: 0x%llX num_h_tiles: %d\n", reqs->top_ctrl,
reqs->hw_res.display_num_of_h_tiles);
SDE_DEBUG("num_lm: %d num_ctl: %d topology: %d split_display: %d\n",
reqs->topology->num_lm, reqs->topology->num_ctl,
reqs->topology->top_name,
reqs->topology->needs_split_display);
SDE_EVT32(mode->hdisplay, rm->lm_max_width, reqs->topology->num_lm,
reqs->top_ctrl, reqs->topology->top_name,
reqs->topology->num_ctl);
return 0;
}
static struct sde_rm_rsvp *_sde_rm_get_rsvp(
struct sde_rm *rm,
struct drm_encoder *enc)
{
struct sde_rm_rsvp *i;
if (!rm || !enc) {
SDE_ERROR("invalid params\n");
return NULL;
}
if (list_empty(&rm->rsvps))
return NULL;
list_for_each_entry(i, &rm->rsvps, list)
if (i->enc_id == enc->base.id)
return i;
return NULL;
}
static struct drm_connector *_sde_rm_get_connector(
struct drm_encoder *enc)
{
struct drm_connector *conn = NULL;
struct list_head *connector_list =
&enc->dev->mode_config.connector_list;
list_for_each_entry(conn, connector_list, head)
if (conn->encoder == enc)
return conn;
return NULL;
}
/**
* _sde_rm_release_rsvp - release resources and release a reservation
* @rm: KMS handle
* @rsvp: RSVP pointer to release and release resources for
*/
static void _sde_rm_release_rsvp(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
struct drm_connector *conn)
{
struct sde_rm_rsvp *rsvp_c, *rsvp_n;
struct sde_rm_hw_blk *blk;
enum sde_hw_blk_type type;
if (!rsvp)
return;
SDE_DEBUG("rel rsvp %d enc %d\n", rsvp->seq, rsvp->enc_id);
list_for_each_entry_safe(rsvp_c, rsvp_n, &rm->rsvps, list) {
if (rsvp == rsvp_c) {
list_del(&rsvp_c->list);
break;
}
}
for (type = 0; type < SDE_HW_BLK_MAX; type++) {
list_for_each_entry(blk, &rm->hw_blks[type], list) {
if (blk->rsvp == rsvp) {
blk->rsvp = NULL;
SDE_DEBUG("rel rsvp %d enc %d %d %d\n",
rsvp->seq, rsvp->enc_id,
blk->type, blk->id);
}
if (blk->rsvp_nxt == rsvp) {
blk->rsvp_nxt = NULL;
SDE_DEBUG("rel rsvp_nxt %d enc %d %d %d\n",
rsvp->seq, rsvp->enc_id,
blk->type, blk->id);
}
}
}
kfree(rsvp);
}
void sde_rm_release(struct sde_rm *rm, struct drm_encoder *enc)
{
struct sde_rm_rsvp *rsvp;
struct drm_connector *conn;
uint64_t top_ctrl;
if (!rm || !enc) {
SDE_ERROR("invalid params\n");
return;
}
mutex_lock(&rm->rm_lock);
rsvp = _sde_rm_get_rsvp(rm, enc);
if (!rsvp) {
SDE_ERROR("failed to find rsvp for enc %d\n", enc->base.id);
goto end;
}
conn = _sde_rm_get_connector(enc);
if (!conn) {
SDE_ERROR("failed to get connector for enc %d\n", enc->base.id);
goto end;
}
top_ctrl = sde_connector_get_property(conn->state,
CONNECTOR_PROP_TOPOLOGY_CONTROL);
if (top_ctrl & BIT(SDE_RM_TOPCTL_RESERVE_LOCK)) {
SDE_DEBUG("rsvp[s%de%d] not releasing locked resources\n",
rsvp->seq, rsvp->enc_id);
} else {
SDE_DEBUG("release rsvp[s%de%d]\n", rsvp->seq,
rsvp->enc_id);
_sde_rm_release_rsvp(rm, rsvp, conn);
(void) msm_property_set_property(
sde_connector_get_propinfo(conn),
sde_connector_get_property_values(conn->state),
CONNECTOR_PROP_TOPOLOGY_NAME,
SDE_RM_TOPOLOGY_NONE);
}
end:
mutex_unlock(&rm->rm_lock);
}
static int _sde_rm_commit_rsvp(
struct sde_rm *rm,
struct sde_rm_rsvp *rsvp,
struct drm_connector_state *conn_state)
{
struct sde_rm_hw_blk *blk;
enum sde_hw_blk_type type;
int ret = 0;
ret = msm_property_set_property(
sde_connector_get_propinfo(conn_state->connector),
sde_connector_get_property_values(conn_state),
CONNECTOR_PROP_TOPOLOGY_NAME,
rsvp->topology);
if (ret) {
SDE_ERROR("failed to set topology name property, ret %d\n",
ret);
_sde_rm_release_rsvp(rm, rsvp, conn_state->connector);
return ret;
}
/* Swap next rsvp to be the active */
for (type = 0; type < SDE_HW_BLK_MAX; type++) {
list_for_each_entry(blk, &rm->hw_blks[type], list) {
if (blk->rsvp_nxt) {
blk->rsvp = blk->rsvp_nxt;
blk->rsvp_nxt = NULL;
}
}
}
if (!ret) {
SDE_DEBUG("rsrv enc %d topology %d\n", rsvp->enc_id,
rsvp->topology);
SDE_EVT32(rsvp->enc_id, rsvp->topology);
}
return ret;
}
int sde_rm_reserve(
struct sde_rm *rm,
struct drm_encoder *enc,
struct drm_crtc_state *crtc_state,
struct drm_connector_state *conn_state,
bool test_only)
{
struct sde_rm_rsvp *rsvp_cur, *rsvp_nxt;
struct sde_rm_requirements reqs;
int ret;
if (!rm || !enc || !crtc_state || !conn_state) {
SDE_ERROR("invalid arguments\n");
return -EINVAL;
}
/* Check if this is just a page-flip */
if (!drm_atomic_crtc_needs_modeset(crtc_state))
return 0;
SDE_DEBUG("reserving hw for conn %d enc %d crtc %d test_only %d\n",
conn_state->connector->base.id, enc->base.id,
crtc_state->crtc->base.id, test_only);
SDE_EVT32(enc->base.id, conn_state->connector->base.id);
mutex_lock(&rm->rm_lock);
_sde_rm_print_rsvps(rm, SDE_RM_STAGE_BEGIN);
ret = _sde_rm_populate_requirements(rm, enc, crtc_state,
conn_state, &reqs);
if (ret) {
SDE_ERROR("failed to populate hw requirements\n");
goto end;
}
/*
* We only support one active reservation per-hw-block. But to implement
* transactional semantics for test-only, and for allowing failure while
* modifying your existing reservation, over the course of this
* function we can have two reservations:
* Current: Existing reservation
* Next: Proposed reservation. The proposed reservation may fail, or may
* be discarded if in test-only mode.
* If reservation is successful, and we're not in test-only, then we
* replace the current with the next.
*/
rsvp_nxt = kzalloc(sizeof(*rsvp_nxt), GFP_KERNEL);
if (!rsvp_nxt) {
ret = -ENOMEM;
goto end;
}
rsvp_cur = _sde_rm_get_rsvp(rm, enc);
/*
* User can request that we clear out any reservation during the
* atomic_check phase by using this CLEAR bit
*/
if (rsvp_cur && test_only && RM_RQ_CLEAR(&reqs)) {
SDE_DEBUG("test_only & CLEAR: clear rsvp[s%de%d]\n",
rsvp_cur->seq, rsvp_cur->enc_id);
_sde_rm_release_rsvp(rm, rsvp_cur, conn_state->connector);
rsvp_cur = NULL;
_sde_rm_print_rsvps(rm, SDE_RM_STAGE_AFTER_CLEAR);
(void) msm_property_set_property(
sde_connector_get_propinfo(
conn_state->connector),
sde_connector_get_property_values(conn_state),
CONNECTOR_PROP_TOPOLOGY_NAME,
SDE_RM_TOPOLOGY_NONE);
}
/* Check the proposed reservation, store it in hw's "next" field */
ret = _sde_rm_make_next_rsvp(rm, enc, crtc_state, conn_state,
rsvp_nxt, &reqs);
_sde_rm_print_rsvps(rm, SDE_RM_STAGE_AFTER_RSVPNEXT);
if (ret) {
SDE_ERROR("failed to reserve hw resources: %d\n", ret);
_sde_rm_release_rsvp(rm, rsvp_nxt, conn_state->connector);
} else if (test_only && !RM_RQ_LOCK(&reqs)) {
/*
* Normally, if test_only, test the reservation and then undo
* However, if the user requests LOCK, then keep the reservation
* made during the atomic_check phase.
*/
SDE_DEBUG("test_only: discard test rsvp[s%de%d]\n",
rsvp_nxt->seq, rsvp_nxt->enc_id);
_sde_rm_release_rsvp(rm, rsvp_nxt, conn_state->connector);
} else {
if (test_only && RM_RQ_LOCK(&reqs))
SDE_DEBUG("test_only & LOCK: lock rsvp[s%de%d]\n",
rsvp_nxt->seq, rsvp_nxt->enc_id);
_sde_rm_release_rsvp(rm, rsvp_cur, conn_state->connector);
ret = _sde_rm_commit_rsvp(rm, rsvp_nxt, conn_state);
}
_sde_rm_print_rsvps(rm, SDE_RM_STAGE_FINAL);
end:
mutex_unlock(&rm->rm_lock);
return ret;
}