Merge tag 'asoc-v5.19' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Updates for v5.19

This is quite a big update, partly due to the addition of some larger
drivers (more of which is to follow since at least the AVS driver is
still a work in progress) and partly due to Charles' work sorting out
our handling of endianness.  As has been the case recently it's much
more about drivers than the core.

 - Overhaul of endianness specification for data formats, avoiding
   needless restrictions due to CODECs.
 - Initial stages of Intel AVS driver merge.
 - Introduction of v4 IPC mechanism for SOF.
 - TDM mode support for AK4613.
 - Support for Analog Devices ADAU1361, Cirrus Logic CS35L45, Maxim
   MAX98396, MediaTek MT8186, NXP i.MX8 micfil and SAI interfaces,
   nVidia Tegra186 ASRC, and Texas Instruments TAS2764 and TAS2780

1 files changed, 250 insertions, 0 deletions

diff --git a/sound/soc/intel/avs/apl.c b/sound/soc/intel/avs/apl.c
new file mode 100644
index 000000000000..b8e2b23c9f64
--- /dev/null
+++ b/sound/soc/intel/avs/apl.c
@@ -0,0 +1,250 @@
+// SPDX-License-Identifier: GPL-2.0-only
+//
+// Copyright(c) 2021-2022 Intel Corporation. All rights reserved.
+//
+// Authors: Cezary Rojewski <cezary.rojewski@intel.com>
+//          Amadeusz Slawinski <amadeuszx.slawinski@linux.intel.com>
+//
+
+#include <linux/devcoredump.h>
+#include <linux/slab.h>
+#include "avs.h"
+#include "messages.h"
+#include "path.h"
+#include "topology.h"
+
+static int apl_enable_logs(struct avs_dev *adev, enum avs_log_enable enable, u32 aging_period,
+			   u32 fifo_full_period, unsigned long resource_mask, u32 *priorities)
+{
+	struct apl_log_state_info *info;
+	u32 size, num_cores = adev->hw_cfg.dsp_cores;
+	int ret, i;
+
+	if (fls_long(resource_mask) > num_cores)
+		return -EINVAL;
+	size = struct_size(info, logs_core, num_cores);
+	info = kzalloc(size, GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+
+	info->aging_timer_period = aging_period;
+	info->fifo_full_timer_period = fifo_full_period;
+	info->core_mask = resource_mask;
+	if (enable)
+		for_each_set_bit(i, &resource_mask, num_cores) {
+			info->logs_core[i].enable = enable;
+			info->logs_core[i].min_priority = *priorities++;
+		}
+	else
+		for_each_set_bit(i, &resource_mask, num_cores)
+			info->logs_core[i].enable = enable;
+
+	ret = avs_ipc_set_enable_logs(adev, (u8 *)info, size);
+	kfree(info);
+	if (ret)
+		return AVS_IPC_RET(ret);
+
+	return 0;
+}
+
+static int apl_log_buffer_status(struct avs_dev *adev, union avs_notify_msg *msg)
+{
+	struct apl_log_buffer_layout layout;
+	unsigned long flags;
+	void __iomem *addr, *buf;
+
+	addr = avs_log_buffer_addr(adev, msg->log.core);
+	if (!addr)
+		return -ENXIO;
+
+	memcpy_fromio(&layout, addr, sizeof(layout));
+
+	spin_lock_irqsave(&adev->dbg.trace_lock, flags);
+	if (!kfifo_initialized(&adev->dbg.trace_fifo))
+		/* consume the logs regardless of consumer presence */
+		goto update_read_ptr;
+
+	buf = apl_log_payload_addr(addr);
+
+	if (layout.read_ptr > layout.write_ptr) {
+		__kfifo_fromio_locked(&adev->dbg.trace_fifo, buf + layout.read_ptr,
+				      apl_log_payload_size(adev) - layout.read_ptr,
+				      &adev->dbg.fifo_lock);
+		layout.read_ptr = 0;
+	}
+	__kfifo_fromio_locked(&adev->dbg.trace_fifo, buf + layout.read_ptr,
+			      layout.write_ptr - layout.read_ptr, &adev->dbg.fifo_lock);
+
+	wake_up(&adev->dbg.trace_waitq);
+
+update_read_ptr:
+	spin_unlock_irqrestore(&adev->dbg.trace_lock, flags);
+	writel(layout.write_ptr, addr);
+	return 0;
+}
+
+static int apl_wait_log_entry(struct avs_dev *adev, u32 core, struct apl_log_buffer_layout *layout)
+{
+	unsigned long timeout;
+	void __iomem *addr;
+
+	addr = avs_log_buffer_addr(adev, core);
+	if (!addr)
+		return -ENXIO;
+
+	timeout = jiffies + msecs_to_jiffies(10);
+
+	do {
+		memcpy_fromio(layout, addr, sizeof(*layout));
+		if (layout->read_ptr != layout->write_ptr)
+			return 0;
+		usleep_range(500, 1000);
+	} while (!time_after(jiffies, timeout));
+
+	return -ETIMEDOUT;
+}
+
+/* reads log header and tests its type */
+#define apl_is_entry_stackdump(addr) ((readl(addr) >> 30) & 0x1)
+
+static int apl_coredump(struct avs_dev *adev, union avs_notify_msg *msg)
+{
+	struct apl_log_buffer_layout layout;
+	void __iomem *addr, *buf;
+	size_t dump_size;
+	u16 offset = 0;
+	u8 *dump, *pos;
+
+	dump_size = AVS_FW_REGS_SIZE + msg->ext.coredump.stack_dump_size;
+	dump = vzalloc(dump_size);
+	if (!dump)
+		return -ENOMEM;
+
+	memcpy_fromio(dump, avs_sram_addr(adev, AVS_FW_REGS_WINDOW), AVS_FW_REGS_SIZE);
+
+	if (!msg->ext.coredump.stack_dump_size)
+		goto exit;
+
+	/* Dump the registers even if an external error prevents gathering the stack. */
+	addr = avs_log_buffer_addr(adev, msg->ext.coredump.core_id);
+	if (!addr)
+		goto exit;
+
+	buf = apl_log_payload_addr(addr);
+	memcpy_fromio(&layout, addr, sizeof(layout));
+	if (!apl_is_entry_stackdump(buf + layout.read_ptr)) {
+		/*
+		 * DSP awaits the remaining logs to be
+		 * gathered before dumping stack
+		 */
+		msg->log.core = msg->ext.coredump.core_id;
+		avs_dsp_op(adev, log_buffer_status, msg);
+	}
+
+	pos = dump + AVS_FW_REGS_SIZE;
+	/* gather the stack */
+	do {
+		u32 count;
+
+		if (apl_wait_log_entry(adev, msg->ext.coredump.core_id, &layout))
+			break;
+
+		if (layout.read_ptr > layout.write_ptr) {
+			count = apl_log_payload_size(adev) - layout.read_ptr;
+			memcpy_fromio(pos + offset, buf + layout.read_ptr, count);
+			layout.read_ptr = 0;
+			offset += count;
+		}
+		count = layout.write_ptr - layout.read_ptr;
+		memcpy_fromio(pos + offset, buf + layout.read_ptr, count);
+		offset += count;
+
+		/* update read pointer */
+		writel(layout.write_ptr, addr);
+	} while (offset < msg->ext.coredump.stack_dump_size);
+
+exit:
+	dev_coredumpv(adev->dev, dump, dump_size, GFP_KERNEL);
+
+	return 0;
+}
+
+static bool apl_lp_streaming(struct avs_dev *adev)
+{
+	struct avs_path *path;
+
+	/* Any gateway without buffer allocated in LP area disqualifies D0IX. */
+	list_for_each_entry(path, &adev->path_list, node) {
+		struct avs_path_pipeline *ppl;
+
+		list_for_each_entry(ppl, &path->ppl_list, node) {
+			struct avs_path_module *mod;
+
+			list_for_each_entry(mod, &ppl->mod_list, node) {
+				struct avs_tplg_modcfg_ext *cfg;
+
+				cfg = mod->template->cfg_ext;
+
+				/* only copiers have gateway attributes */
+				if (!guid_equal(&cfg->type, &AVS_COPIER_MOD_UUID))
+					continue;
+				/* non-gateway copiers do not prevent PG */
+				if (cfg->copier.dma_type == INVALID_OBJECT_ID)
+					continue;
+
+				if (!mod->gtw_attrs.lp_buffer_alloc)
+					return false;
+			}
+		}
+	}
+
+	return true;
+}
+
+static bool apl_d0ix_toggle(struct avs_dev *adev, struct avs_ipc_msg *tx, bool wake)
+{
+	/* wake in all cases */
+	if (wake)
+		return true;
+
+	/*
+	 * If no pipelines are running, allow for d0ix schedule.
+	 * If all gateways have lp=1, allow for d0ix schedule.
+	 * If any gateway with lp=0 is allocated, abort scheduling d0ix.
+	 *
+	 * Note: for cAVS 1.5+ and 1.8, D0IX is LP-firmware transition,
+	 * not the power-gating mechanism known from cAVS 2.0.
+	 */
+	return apl_lp_streaming(adev);
+}
+
+static int apl_set_d0ix(struct avs_dev *adev, bool enable)
+{
+	bool streaming = false;
+	int ret;
+
+	if (enable)
+		/* Either idle or all gateways with lp=1. */
+		streaming = !list_empty(&adev->path_list);
+
+	ret = avs_ipc_set_d0ix(adev, enable, streaming);
+	return AVS_IPC_RET(ret);
+}
+
+const struct avs_dsp_ops apl_dsp_ops = {
+	.power = avs_dsp_core_power,
+	.reset = avs_dsp_core_reset,
+	.stall = avs_dsp_core_stall,
+	.irq_handler = avs_dsp_irq_handler,
+	.irq_thread = avs_dsp_irq_thread,
+	.int_control = avs_dsp_interrupt_control,
+	.load_basefw = avs_hda_load_basefw,
+	.load_lib = avs_hda_load_library,
+	.transfer_mods = avs_hda_transfer_modules,
+	.enable_logs = apl_enable_logs,
+	.log_buffer_offset = skl_log_buffer_offset,
+	.log_buffer_status = apl_log_buffer_status,
+	.coredump = apl_coredump,
+	.d0ix_toggle = apl_d0ix_toggle,
+	.set_d0ix = apl_set_d0ix,
+};