path: root/drivers/gpu/drm/xe/xe_guc_ct.c

// SPDX-License-Identifier: MIT
/*
 * Copyright © 2022 Intel Corporation
 */

#include "xe_guc_ct.h"

#include <linux/bitfield.h>
#include <linux/circ_buf.h>
#include <linux/delay.h>

#include <kunit/static_stub.h>

#include <drm/drm_managed.h>

#include "abi/guc_actions_abi.h"
#include "abi/guc_actions_sriov_abi.h"
#include "abi/guc_klvs_abi.h"
#include "xe_bo.h"
#include "xe_device.h"
#include "xe_gt.h"
#include "xe_gt_pagefault.h"
#include "xe_gt_printk.h"
#include "xe_gt_sriov_pf_control.h"
#include "xe_gt_sriov_pf_monitor.h"
#include "xe_gt_tlb_invalidation.h"
#include "xe_guc.h"
#include "xe_guc_relay.h"
#include "xe_guc_submit.h"
#include "xe_map.h"
#include "xe_pm.h"
#include "xe_trace_guc.h"

/* Used when a CT send wants to block and / or receive data */
struct g2h_fence {
	u32 *response_buffer;
	u32 seqno;
	u32 response_data;
	u16 response_len;
	u16 error;
	u16 hint;
	u16 reason;
	bool retry;
	bool fail;
	bool done;
};

static void g2h_fence_init(struct g2h_fence *g2h_fence, u32 *response_buffer)
{
	g2h_fence->response_buffer = response_buffer;
	g2h_fence->response_data = 0;
	g2h_fence->response_len = 0;
	g2h_fence->fail = false;
	g2h_fence->retry = false;
	g2h_fence->done = false;
	g2h_fence->seqno = ~0x0;
}

static bool g2h_fence_needs_alloc(struct g2h_fence *g2h_fence)
{
	return g2h_fence->seqno == ~0x0;
}

static struct xe_guc *
ct_to_guc(struct xe_guc_ct *ct)
{
	return container_of(ct, struct xe_guc, ct);
}

static struct xe_gt *
ct_to_gt(struct xe_guc_ct *ct)
{
	return container_of(ct, struct xe_gt, uc.guc.ct);
}

static struct xe_device *
ct_to_xe(struct xe_guc_ct *ct)
{
	return gt_to_xe(ct_to_gt(ct));
}

/**
 * DOC: GuC CTB Blob
 *
 * We allocate single blob to hold both CTB descriptors and buffers:
 *
 *      +--------+-----------------------------------------------+------+
 *      | offset | contents                                      | size |
 *      +========+===============================================+======+
 *      | 0x0000 | H2G CTB Descriptor (send)                     |      |
 *      +--------+-----------------------------------------------+  4K  |
 *      | 0x0800 | G2H CTB Descriptor (g2h)                      |      |
 *      +--------+-----------------------------------------------+------+
 *      | 0x1000 | H2G CT Buffer (send)                          | n*4K |
 *      |        |                                               |      |
 *      +--------+-----------------------------------------------+------+
 *      | 0x1000 | G2H CT Buffer (g2h)                           | m*4K |
 *      | + n*4K |                                               |      |
 *      +--------+-----------------------------------------------+------+
 *
 * Size of each ``CT Buffer`` must be multiple of 4K.
 * We don't expect too many messages in flight at any time, unless we are
 * using the GuC submission. In that case each request requires a minimum
 * 2 dwords which gives us a maximum 256 queue'd requests. Hopefully this
 * enough space to avoid backpressure on the driver. We increase the size
 * of the receive buffer (relative to the send) to ensure a G2H response
 * CTB has a landing spot.
 *
 * In addition to submissions, the G2H buffer needs to be able to hold
 * enough space for recoverable page fault notifications. The number of
 * page faults is interrupt driven and can be as much as the number of
 * compute resources available. However, most of the actual work for these
 * is in a separate page fault worker thread. Therefore we only need to
 * make sure the queue has enough space to handle all of the submissions
 * and responses and an extra buffer for incoming page faults.
 */

#define CTB_DESC_SIZE		ALIGN(sizeof(struct guc_ct_buffer_desc), SZ_2K)
#define CTB_H2G_BUFFER_SIZE	(SZ_4K)
#define CTB_G2H_BUFFER_SIZE	(SZ_128K)
#define G2H_ROOM_BUFFER_SIZE	(CTB_G2H_BUFFER_SIZE / 2)

/**
 * xe_guc_ct_queue_proc_time_jiffies - Return maximum time to process a full
 * CT command queue
 * @ct: the &xe_guc_ct. Unused at this moment but will be used in the future.
 *
 * Observation is that a 4KiB buffer full of commands takes a little over a
 * second to process. Use that to calculate maximum time to process a full CT
 * command queue.
 *
 * Return: Maximum time to process a full CT queue in jiffies.
 */
long xe_guc_ct_queue_proc_time_jiffies(struct xe_guc_ct *ct)
{
	BUILD_BUG_ON(!IS_ALIGNED(CTB_H2G_BUFFER_SIZE, SZ_4));
	return (CTB_H2G_BUFFER_SIZE / SZ_4K) * HZ;
}

static size_t guc_ct_size(void)
{
	return 2 * CTB_DESC_SIZE + CTB_H2G_BUFFER_SIZE +
		CTB_G2H_BUFFER_SIZE;
}

static void guc_ct_fini(struct drm_device *drm, void *arg)
{
	struct xe_guc_ct *ct = arg;

	destroy_workqueue(ct->g2h_wq);
	xa_destroy(&ct->fence_lookup);
}

static void receive_g2h(struct xe_guc_ct *ct);
static void g2h_worker_func(struct work_struct *w);
static void safe_mode_worker_func(struct work_struct *w);

static void primelockdep(struct xe_guc_ct *ct)
{
	if (!IS_ENABLED(CONFIG_LOCKDEP))
		return;

	fs_reclaim_acquire(GFP_KERNEL);
	might_lock(&ct->lock);
	fs_reclaim_release(GFP_KERNEL);
}

int xe_guc_