Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

8 files changed, 3014 insertions, 0 deletions

diff --git a/drivers/firmware/tegra/Kconfig b/drivers/firmware/tegra/Kconfig
new file mode 100644
index 000000000..cde1ab8bd
--- /dev/null
+++ b/drivers/firmware/tegra/Kconfig
@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: GPL-2.0-only
+menu "Tegra firmware driver"
+
+config TEGRA_IVC
+	bool "Tegra IVC protocol"
+	depends on ARCH_TEGRA
+	help
+	  IVC (Inter-VM Communication) protocol is part of the IPC
+	  (Inter Processor Communication) framework on Tegra. It maintains the
+	  data and the different communication channels in SysRAM or RAM and
+	  keeps the content is synchronization between host CPU and remote
+	  processors.
+
+config TEGRA_BPMP
+	bool "Tegra BPMP driver"
+	depends on ARCH_TEGRA && TEGRA_HSP_MBOX && TEGRA_IVC
+	depends on !CPU_BIG_ENDIAN
+	help
+	  BPMP (Boot and Power Management Processor) is designed to off-loading
+	  the PM functions which include clock/DVFS/thermal/power from the CPU.
+	  It needs HSP as the HW synchronization and notification module and
+	  IVC module as the message communication protocol.
+
+	  This driver manages the IPC interface between host CPU and the
+	  firmware running on BPMP.
+
+endmenu
diff --git a/drivers/firmware/tegra/Makefile b/drivers/firmware/tegra/Makefile
new file mode 100644
index 000000000..620cf3fdd
--- /dev/null
+++ b/drivers/firmware/tegra/Makefile
@@ -0,0 +1,9 @@
+# SPDX-License-Identifier: GPL-2.0-only
+tegra-bpmp-y			= bpmp.o
+tegra-bpmp-$(CONFIG_ARCH_TEGRA_210_SOC)	+= bpmp-tegra210.o
+tegra-bpmp-$(CONFIG_ARCH_TEGRA_186_SOC)	+= bpmp-tegra186.o
+tegra-bpmp-$(CONFIG_ARCH_TEGRA_194_SOC)	+= bpmp-tegra186.o
+tegra-bpmp-$(CONFIG_ARCH_TEGRA_234_SOC)	+= bpmp-tegra186.o
+tegra-bpmp-$(CONFIG_DEBUG_FS)	+= bpmp-debugfs.o
+obj-$(CONFIG_TEGRA_BPMP)	+= tegra-bpmp.o
+obj-$(CONFIG_TEGRA_IVC)		+= ivc.o
diff --git a/drivers/firmware/tegra/bpmp-debugfs.c b/drivers/firmware/tegra/bpmp-debugfs.c
new file mode 100644
index 000000000..3ca2b5d9e
--- /dev/null
+++ b/drivers/firmware/tegra/bpmp-debugfs.c
@@ -0,0 +1,800 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2017, NVIDIA CORPORATION.  All rights reserved.
+ */
+#include <linux/debugfs.h>
+#include <linux/dma-mapping.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+
+#include <soc/tegra/bpmp.h>
+#include <soc/tegra/bpmp-abi.h>
+
+static DEFINE_MUTEX(bpmp_debug_lock);
+
+struct seqbuf {
+	char *buf;
+	size_t pos;
+	size_t size;
+};
+
+static void seqbuf_init(struct seqbuf *seqbuf, void *buf, size_t size)
+{
+	seqbuf->buf = buf;
+	seqbuf->size = size;
+	seqbuf->pos = 0;
+}
+
+static size_t seqbuf_avail(struct seqbuf *seqbuf)
+{
+	return seqbuf->pos < seqbuf->size ? seqbuf->size - seqbuf->pos : 0;
+}
+
+static size_t seqbuf_status(struct seqbuf *seqbuf)
+{
+	return seqbuf->pos <= seqbuf->size ? 0 : -EOVERFLOW;
+}
+
+static int seqbuf_eof(struct seqbuf *seqbuf)
+{
+	return seqbuf->pos >= seqbuf->size;
+}
+
+static int seqbuf_read(struct seqbuf *seqbuf, void *buf, size_t nbyte)
+{
+	nbyte = min(nbyte, seqbuf_avail(seqbuf));
+	memcpy(buf, seqbuf->buf + seqbuf->pos, nbyte);
+	seqbuf->pos += nbyte;
+	return seqbuf_status(seqbuf);
+}
+
+static int seqbuf_read_u32(struct seqbuf *seqbuf, u32 *v)
+{
+	return seqbuf_read(seqbuf, v, 4);
+}
+
+static int seqbuf_read_str(struct seqbuf *seqbuf, const char **str)
+{
+	*str = seqbuf->buf + seqbuf->pos;
+	seqbuf->pos += strnlen(*str, seqbuf_avail(seqbuf));
+	seqbuf->pos++;
+	return seqbuf_status(seqbuf);
+}
+
+static void seqbuf_seek(struct seqbuf *seqbuf, ssize_t offset)
+{
+	seqbuf->pos += offset;
+}
+
+/* map filename in Linux debugfs to corresponding entry in BPMP */
+static const char *get_filename(struct tegra_bpmp *bpmp,
+				const struct file *file, char *buf, int size)
+{
+	const char *root_path, *filename = NULL;
+	char *root_path_buf;
+	size_t root_len;
+	size_t root_path_buf_len = 512;
+
+	root_path_buf = kzalloc(root_path_buf_len, GFP_KERNEL);
+	if (!root_path_buf)
+		goto out;
+
+	root_path = dentry_path(bpmp->debugfs_mirror, root_path_buf,
+				root_path_buf_len);
+	if (IS_ERR(root_path))
+		goto out;
+
+	root_len = strlen(root_path);
+
+	filename = dentry_path(file->f_path.dentry, buf, size);
+	if (IS_ERR(filename)) {
+		filename = NULL;
+		goto out;
+	}
+
+	if (strlen(filename) < root_len || strncmp(filename, root_path, root_len)) {
+		filename = NULL;
+		goto out;
+	}
+
+	filename += root_len;
+
+out:
+	kfree(root_path_buf);
+	return filename;
+}
+
+static int mrq_debug_open(struct tegra_bpmp *bpmp, const char *name,
+			  u32 *fd, u32 *len, bool write)
+{
+	struct mrq_debug_request req = {
+		.cmd = write ? CMD_DEBUG_OPEN_WO : CMD_DEBUG_OPEN_RO,
+	};
+	struct mrq_debug_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUG,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	ssize_t sz_name;
+	int err = 0;
+
+	sz_name = strscpy(req.fop.name, name, sizeof(req.fop.name));
+	if (sz_name < 0) {
+		pr_err("File name too large: %s\n", name);
+		return -EINVAL;
+	}
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err < 0)
+		return err;
+	else if (msg.rx.ret < 0)
+		return -EINVAL;
+
+	*len = resp.fop.datalen;
+	*fd = resp.fop.fd;
+
+	return 0;
+}
+
+static int mrq_debug_close(struct tegra_bpmp *bpmp, u32 fd)
+{
+	struct mrq_debug_request req = {
+		.cmd = CMD_DEBUG_CLOSE,
+		.frd = {
+			.fd = fd,
+		},
+	};
+	struct mrq_debug_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUG,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	int err = 0;
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err < 0)
+		return err;
+	else if (msg.rx.ret < 0)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int mrq_debug_read(struct tegra_bpmp *bpmp, const char *name,
+			  char *data, size_t sz_data, u32 *nbytes)
+{
+	struct mrq_debug_request req = {
+		.cmd = CMD_DEBUG_READ,
+	};
+	struct mrq_debug_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUG,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	u32 fd = 0, len = 0;
+	int remaining, err;
+
+	mutex_lock(&bpmp_debug_lock);
+	err = mrq_debug_open(bpmp, name, &fd, &len, 0);
+	if (err)
+		goto out;
+
+	if (len > sz_data) {
+		err = -EFBIG;
+		goto close;
+	}
+
+	req.frd.fd = fd;
+	remaining = len;
+
+	while (remaining > 0) {
+		err = tegra_bpmp_transfer(bpmp, &msg);
+		if (err < 0) {
+			goto close;
+		} else if (msg.rx.ret < 0) {
+			err = -EINVAL;
+			goto close;
+		}
+
+		if (resp.frd.readlen > remaining) {
+			pr_err("%s: read data length invalid\n", __func__);
+			err = -EINVAL;
+			goto close;
+		}
+
+		memcpy(data, resp.frd.data, resp.frd.readlen);
+		data += resp.frd.readlen;
+		remaining -= resp.frd.readlen;
+	}
+
+	*nbytes = len;
+
+close:
+	err = mrq_debug_close(bpmp, fd);
+out:
+	mutex_unlock(&bpmp_debug_lock);
+	return err;
+}
+
+static int mrq_debug_write(struct tegra_bpmp *bpmp, const char *name,
+			   uint8_t *data, size_t sz_data)
+{
+	struct mrq_debug_request req = {
+		.cmd = CMD_DEBUG_WRITE
+	};
+	struct mrq_debug_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUG,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	u32 fd = 0, len = 0;
+	size_t remaining;
+	int err;
+
+	mutex_lock(&bpmp_debug_lock);
+	err = mrq_debug_open(bpmp, name, &fd, &len, 1);
+	if (err)
+		goto out;
+
+	if (sz_data > len) {
+		err = -EINVAL;
+		goto close;
+	}
+
+	req.fwr.fd = fd;
+	remaining = sz_data;
+
+	while (remaining > 0) {
+		len = min(remaining, sizeof(req.fwr.data));
+		memcpy(req.fwr.data, data, len);
+		req.fwr.datalen = len;
+
+		err = tegra_bpmp_transfer(bpmp, &msg);
+		if (err < 0) {
+			goto close;
+		} else if (msg.rx.ret < 0) {
+			err = -EINVAL;
+			goto close;
+		}
+
+		data += req.fwr.datalen;
+		remaining -= req.fwr.datalen;
+	}
+
+close:
+	err = mrq_debug_close(bpmp, fd);
+out:
+	mutex_unlock(&bpmp_debug_lock);
+	return err;
+}
+
+static int bpmp_debug_show(struct seq_file *m, void *p)
+{
+	struct file *file = m->private;
+	struct inode *inode = file_inode(file);
+	struct tegra_bpmp *bpmp = inode->i_private;
+	char fnamebuf[256];
+	const char *filename;
+	struct mrq_debug_request req = {
+		.cmd = CMD_DEBUG_READ,
+	};
+	struct mrq_debug_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUG,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	u32 fd = 0, len = 0;
+	int remaining, err;
+
+	filename = get_filename(bpmp, file, fnamebuf, sizeof(fnamebuf));
+	if (!filename)
+		return -ENOENT;
+
+	mutex_lock(&bpmp_debug_lock);
+	err = mrq_debug_open(bpmp, filename, &fd, &len, 0);
+	if (err)
+		goto out;
+
+	req.frd.fd = fd;
+	remaining = len;
+
+	while (remaining > 0) {
+		err = tegra_bpmp_transfer(bpmp, &msg);
+		if (err < 0) {
+			goto close;
+		} else if (msg.rx.ret < 0) {
+			err = -EINVAL;
+			goto close;
+		}
+
+		if (resp.frd.readlen > remaining) {
+			pr_err("%s: read data length invalid\n", __func__);
+			err = -EINVAL;
+			goto close;
+		}
+
+		seq_write(m, resp.frd.data, resp.frd.readlen);
+		remaining -= resp.frd.readlen;
+	}
+
+close:
+	err = mrq_debug_close(bpmp, fd);
+out:
+	mutex_unlock(&bpmp_debug_lock);
+	return err;
+}
+
+static ssize_t bpmp_debug_store(struct file *file, const char __user *buf,
+		size_t count, loff_t *f_pos)
+{
+	struct inode *inode = file_inode(file);
+	struct tegra_bpmp *bpmp = inode->i_private;
+	char *databuf = NULL;
+	char fnamebuf[256];
+	const char *filename;
+	ssize_t err;
+
+	filename = get_filename(bpmp, file, fnamebuf, sizeof(fnamebuf));
+	if (!filename)
+		return -ENOENT;
+
+	databuf = memdup_user(buf, count);
+	if (IS_ERR(databuf))
+		return PTR_ERR(databuf);
+
+	err = mrq_debug_write(bpmp, filename, databuf, count);
+	kfree(databuf);
+
+	return err ?: count;
+}
+
+static int bpmp_debug_open(struct inode *inode, struct file *file)
+{
+	return single_open_size(file, bpmp_debug_show, file, SZ_256K);
+}
+
+static const struct file_operations bpmp_debug_fops = {
+	.open		= bpmp_debug_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.write		= bpmp_debug_store,
+	.release	= single_release,
+};
+
+static int bpmp_populate_debugfs_inband(struct tegra_bpmp *bpmp,
+					struct dentry *parent,
+					char *ppath)
+{
+	const size_t pathlen = SZ_256;
+	const size_t bufsize = SZ_16K;
+	struct dentry *dentry;
+	u32 dsize, attrs = 0;
+	struct seqbuf seqbuf;
+	char *buf, *pathbuf;
+	const char *name;
+	int err = 0;
+
+	if (!bpmp || !parent || !ppath)
+		return -EINVAL;
+
+	buf = kmalloc(bufsize, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	pathbuf = kzalloc(pathlen, GFP_KERNEL);
+	if (!pathbuf) {
+		kfree(buf);
+		return -ENOMEM;
+	}
+
+	err = mrq_debug_read(bpmp, ppath, buf, bufsize, &dsize);
+	if (err)
+		goto out;
+
+	seqbuf_init(&seqbuf, buf, dsize);
+
+	while (!seqbuf_eof(&seqbuf)) {
+		err = seqbuf_read_u32(&seqbuf, &attrs);
+		if (err)
+			goto out;
+
+		err = seqbuf_read_str(&seqbuf, &name);
+		if (err < 0)
+			goto out;
+
+		if (attrs & DEBUGFS_S_ISDIR) {
+			size_t len;
+
+			dentry = debugfs_create_dir(name, parent);
+			if (IS_ERR(dentry)) {
+				err = PTR_ERR(dentry);
+				goto out;
+			}
+
+			len = snprintf(pathbuf, pathlen, "%s%s/", ppath, name);
+			if (len >= pathlen) {
+				err = -EINVAL;
+				goto out;
+			}
+
+			err = bpmp_populate_debugfs_inband(bpmp, dentry,
+							   pathbuf);
+			if (err < 0)
+				goto out;
+		} else {
+			umode_t mode;
+
+			mode = attrs & DEBUGFS_S_IRUSR ? 0400 : 0;
+			mode |= attrs & DEBUGFS_S_IWUSR ? 0200 : 0;
+			dentry = debugfs_create_file(name, mode, parent, bpmp,
+						     &bpmp_debug_fops);
+			if (IS_ERR(dentry)) {
+				err = -ENOMEM;
+				goto out;
+			}
+		}
+	}
+
+out:
+	kfree(pathbuf);
+	kfree(buf);
+
+	return err;
+}
+
+static int mrq_debugfs_read(struct tegra_bpmp *bpmp,
+			    dma_addr_t name, size_t sz_name,
+			    dma_addr_t data, size_t sz_data,
+			    size_t *nbytes)
+{
+	struct mrq_debugfs_request req = {
+		.cmd = CMD_DEBUGFS_READ,
+		.fop = {
+			.fnameaddr = (u32)name,
+			.fnamelen = (u32)sz_name,
+			.dataaddr = (u32)data,
+			.datalen = (u32)sz_data,
+		},
+	};
+	struct mrq_debugfs_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUGFS,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	int err;
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err < 0)
+		return err;
+	else if (msg.rx.ret < 0)
+		return -EINVAL;
+
+	*nbytes = (size_t)resp.fop.nbytes;
+
+	return 0;
+}
+
+static int mrq_debugfs_write(struct tegra_bpmp *bpmp,
+			     dma_addr_t name, size_t sz_name,
+			     dma_addr_t data, size_t sz_data)
+{
+	const struct mrq_debugfs_request req = {
+		.cmd = CMD_DEBUGFS_WRITE,
+		.fop = {
+			.fnameaddr = (u32)name,
+			.fnamelen = (u32)sz_name,
+			.dataaddr = (u32)data,
+			.datalen = (u32)sz_data,
+		},
+	};
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUGFS,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+	};
+
+	return tegra_bpmp_transfer(bpmp, &msg);
+}
+
+static int mrq_debugfs_dumpdir(struct tegra_bpmp *bpmp, dma_addr_t addr,
+			       size_t size, size_t *nbytes)
+{
+	const struct mrq_debugfs_request req = {
+		.cmd = CMD_DEBUGFS_DUMPDIR,
+		.dumpdir = {
+			.dataaddr = (u32)addr,
+			.datalen = (u32)size,
+		},
+	};
+	struct mrq_debugfs_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_DEBUGFS,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	int err;
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err < 0)
+		return err;
+	else if (msg.rx.ret < 0)
+		return -EINVAL;
+
+	*nbytes = (size_t)resp.dumpdir.nbytes;
+
+	return 0;
+}
+
+static int debugfs_show(struct seq_file *m, void *p)
+{
+	struct file *file = m->private;
+	struct inode *inode = file_inode(file);
+	struct tegra_bpmp *bpmp = inode->i_private;
+	const size_t datasize = m->size;
+	const size_t namesize = SZ_256;
+	void *datavirt, *namevirt;
+	dma_addr_t dataphys, namephys;
+	char buf[256];
+	const char *filename;
+	size_t len, nbytes;
+	int err;
+
+	filename = get_filename(bpmp, file, buf, sizeof(buf));
+	if (!filename)
+		return -ENOENT;
+
+	namevirt = dma_alloc_coherent(bpmp->dev, namesize, &namephys,
+				      GFP_KERNEL | GFP_DMA32);
+	if (!namevirt)
+		return -ENOMEM;
+
+	datavirt = dma_alloc_coherent(bpmp->dev, datasize, &dataphys,
+				      GFP_KERNEL | GFP_DMA32);
+	if (!datavirt) {
+		err = -ENOMEM;
+		goto free_namebuf;
+	}
+
+	len = strlen(filename);
+	strncpy(namevirt, filename, namesize);
+
+	err = mrq_debugfs_read(bpmp, namephys, len, dataphys, datasize,
+			       &nbytes);
+
+	if (!err)
+		seq_write(m, datavirt, nbytes);
+
+	dma_free_coherent(bpmp->dev, datasize, datavirt, dataphys);
+free_namebuf:
+	dma_free_coherent(bpmp->dev, namesize, namevirt, namephys);
+
+	return err;
+}
+
+static int debugfs_open(struct inode *inode, struct file *file)
+{
+	return single_open_size(file, debugfs_show, file, SZ_128K);
+}
+
+static ssize_t debugfs_store(struct file *file, const char __user *buf,
+		size_t count, loff_t *f_pos)
+{
+	struct inode *inode = file_inode(file);
+	struct tegra_bpmp *bpmp = inode->i_private;
+	const size_t datasize = count;
+	const size_t namesize = SZ_256;
+	void *datavirt, *namevirt;
+	dma_addr_t dataphys, namephys;
+	char fnamebuf[256];
+	const char *filename;
+	size_t len;
+	int err;
+
+	filename = get_filename(bpmp, file, fnamebuf, sizeof(fnamebuf));
+	if (!filename)
+		return -ENOENT;
+
+	namevirt = dma_alloc_coherent(bpmp->dev, namesize, &namephys,
+				      GFP_KERNEL | GFP_DMA32);
+	if (!namevirt)
+		return -ENOMEM;
+
+	datavirt = dma_alloc_coherent(bpmp->dev, datasize, &dataphys,
+				      GFP_KERNEL | GFP_DMA32);
+	if (!datavirt) {
+		err = -ENOMEM;
+		goto free_namebuf;
+	}
+
+	len = strlen(filename);
+	strncpy(namevirt, filename, namesize);
+
+	if (copy_from_user(datavirt, buf, count)) {
+		err = -EFAULT;
+		goto free_databuf;
+	}
+
+	err = mrq_debugfs_write(bpmp, namephys, len, dataphys,
+				count);
+
+free_databuf:
+	dma_free_coherent(bpmp->dev, datasize, datavirt, dataphys);
+free_namebuf:
+	dma_free_coherent(bpmp->dev, namesize, namevirt, namephys);
+
+	return err ?: count;
+}
+
+static const struct file_operations debugfs_fops = {
+	.open		= debugfs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.write		= debugfs_store,
+	.release	= single_release,
+};
+
+static int bpmp_populate_dir(struct tegra_bpmp *bpmp, struct seqbuf *seqbuf,
+			     struct dentry *parent, u32 depth)
+{
+	int err;
+	u32 d, t;
+	const char *name;
+	struct dentry *dentry;
+
+	while (!seqbuf_eof(seqbuf)) {
+		err = seqbuf_read_u32(seqbuf, &d);
+		if (err < 0)
+			return err;
+
+		if (d < depth) {
+			seqbuf_seek(seqbuf, -4);
+			/* go up a level */
+			return 0;
+		} else if (d != depth) {
+			/* malformed data received from BPMP */
+			return -EIO;
+		}
+
+		err = seqbuf_read_u32(seqbuf, &t);
+		if (err < 0)
+			return err;
+		err = seqbuf_read_str(seqbuf, &name);
+		if (err < 0)
+			return err;
+
+		if (t & DEBUGFS_S_ISDIR) {
+			dentry = debugfs_create_dir(name, parent);
+			if (IS_ERR(dentry))
+				return -ENOMEM;
+			err = bpmp_populate_dir(bpmp, seqbuf, dentry, depth+1);
+			if (err < 0)
+				return err;
+		} else {
+			umode_t mode;
+
+			mode = t & DEBUGFS_S_IRUSR ? S_IRUSR : 0;
+			mode |= t & DEBUGFS_S_IWUSR ? S_IWUSR : 0;
+			dentry = debugfs_create_file(name, mode,
+						     parent, bpmp,
+						     &debugfs_fops);
+			if (IS_ERR(dentry))
+				return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+static int bpmp_populate_debugfs_shmem(struct tegra_bpmp *bpmp)
+{
+	struct seqbuf seqbuf;
+	const size_t sz = SZ_512K;
+	dma_addr_t phys;
+	size_t nbytes;
+	void *virt;
+	int err;
+
+	virt = dma_alloc_coherent(bpmp->dev, sz, &phys,
+				  GFP_KERNEL | GFP_DMA32);
+	if (!virt)
+		return -ENOMEM;
+
+	err = mrq_debugfs_dumpdir(bpmp, phys, sz, &nbytes);
+	if (err < 0) {
+		goto free;
+	} else if (nbytes > sz) {
+		err = -EINVAL;
+		goto free;
+	}
+
+	seqbuf_init(&seqbuf, virt, nbytes);
+	err = bpmp_populate_dir(bpmp, &seqbuf, bpmp->debugfs_mirror, 0);
+free:
+	dma_free_coherent(bpmp->dev, sz, virt, phys);
+
+	return err;
+}
+
+int tegra_bpmp_init_debugfs(struct tegra_bpmp *bpmp)
+{
+	struct dentry *root;
+	bool inband;
+	int err;
+
+	inband = tegra_bpmp_mrq_is_supported(bpmp, MRQ_DEBUG);
+
+	if (!inband && !tegra_bpmp_mrq_is_supported(bpmp, MRQ_DEBUGFS))
+		return 0;
+
+	root = debugfs_create_dir("bpmp", NULL);
+	if (IS_ERR(root))
+		return -ENOMEM;
+
+	bpmp->debugfs_mirror = debugfs_create_dir("debug", root);
+	if (IS_ERR(bpmp->debugfs_mirror)) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	if (inband)
+		err = bpmp_populate_debugfs_inband(bpmp, bpmp->debugfs_mirror,
+						   "/");
+	else
+		err = bpmp_populate_debugfs_shmem(bpmp);
+
+out:
+	if (err < 0)
+		debugfs_remove_recursive(root);
+
+	return err;
+}
diff --git a/drivers/firmware/tegra/bpmp-private.h b/drivers/firmware/tegra/bpmp-private.h
new file mode 100644
index 000000000..182bfe396
--- /dev/null
+++ b/drivers/firmware/tegra/bpmp-private.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (c) 2018, NVIDIA CORPORATION.
+ */
+
+#ifndef __FIRMWARE_TEGRA_BPMP_PRIVATE_H
+#define __FIRMWARE_TEGRA_BPMP_PRIVATE_H
+
+#include <soc/tegra/bpmp.h>
+
+struct tegra_bpmp_ops {
+	int (*init)(struct tegra_bpmp *bpmp);
+	void (*deinit)(struct tegra_bpmp *bpmp);
+	bool (*is_response_ready)(struct tegra_bpmp_channel *channel);
+	bool (*is_request_ready)(struct tegra_bpmp_channel *channel);
+	int (*ack_response)(struct tegra_bpmp_channel *channel);
+	int (*ack_request)(struct tegra_bpmp_channel *channel);
+	bool (*is_response_channel_free)(struct tegra_bpmp_channel *channel);
+	bool (*is_request_channel_free)(struct tegra_bpmp_channel *channel);
+	int (*post_response)(struct tegra_bpmp_channel *channel);
+	int (*post_request)(struct tegra_bpmp_channel *channel);
+	int (*ring_doorbell)(struct tegra_bpmp *bpmp);
+	int (*resume)(struct tegra_bpmp *bpmp);
+};
+
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_186_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_234_SOC)
+extern const struct tegra_bpmp_ops tegra186_bpmp_ops;
+#endif
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_210_SOC)
+extern const struct tegra_bpmp_ops tegra210_bpmp_ops;
+#endif
+
+#endif
diff --git a/drivers/firmware/tegra/bpmp-tegra186.c b/drivers/firmware/tegra/bpmp-tegra186.c
new file mode 100644
index 000000000..2e2619904
--- /dev/null
+++ b/drivers/firmware/tegra/bpmp-tegra186.c
@@ -0,0 +1,303 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2018, NVIDIA CORPORATION.
+ */
+
+#include <linux/genalloc.h>
+#include <linux/mailbox_client.h>
+#include <linux/platform_device.h>
+
+#include <soc/tegra/bpmp.h>
+#include <soc/tegra/bpmp-abi.h>
+#include <soc/tegra/ivc.h>
+
+#include "bpmp-private.h"
+
+struct tegra186_bpmp {
+	struct tegra_bpmp *parent;
+
+	struct {
+		struct gen_pool *pool;
+		void __iomem *virt;
+		dma_addr_t phys;
+	} tx, rx;
+
+	struct {
+		struct mbox_client client;
+		struct mbox_chan *channel;
+	} mbox;
+};
+
+static inline struct tegra_bpmp *
+mbox_client_to_bpmp(struct mbox_client *client)
+{
+	struct tegra186_bpmp *priv;
+
+	priv = container_of(client, struct tegra186_bpmp, mbox.client);
+
+	return priv->parent;
+}
+
+static bool tegra186_bpmp_is_message_ready(struct tegra_bpmp_channel *channel)
+{
+	int err;
+
+	err = tegra_ivc_read_get_next_frame(channel->ivc, &channel->ib);
+	if (err) {
+		iosys_map_clear(&channel->ib);
+		return false;
+	}
+
+	return true;
+}
+
+static bool tegra186_bpmp_is_channel_free(struct tegra_bpmp_channel *channel)
+{
+	int err;
+
+	err = tegra_ivc_write_get_next_frame(channel->ivc, &channel->ob);
+	if (err) {
+		iosys_map_clear(&channel->ob);
+		return false;
+	}
+
+	return true;
+}
+
+static int tegra186_bpmp_ack_message(struct tegra_bpmp_channel *channel)
+{
+	return tegra_ivc_read_advance(channel->ivc);
+}
+
+static int tegra186_bpmp_post_message(struct tegra_bpmp_channel *channel)
+{
+	return tegra_ivc_write_advance(channel->ivc);
+}
+
+static int tegra186_bpmp_ring_doorbell(struct tegra_bpmp *bpmp)
+{
+	struct tegra186_bpmp *priv = bpmp->priv;
+	int err;
+
+	err = mbox_send_message(priv->mbox.channel, NULL);
+	if (err < 0)
+		return err;
+
+	mbox_client_txdone(priv->mbox.channel, 0);
+
+	return 0;
+}
+
+static void tegra186_bpmp_ivc_notify(struct tegra_ivc *ivc, void *data)
+{
+	struct tegra_bpmp *bpmp = data;
+	struct tegra186_bpmp *priv = bpmp->priv;
+
+	if (WARN_ON(priv->mbox.channel == NULL))
+		return;
+
+	tegra186_bpmp_ring_doorbell(bpmp);
+}
+
+static int tegra186_bpmp_channel_init(struct tegra_bpmp_channel *channel,
+				      struct tegra_bpmp *bpmp,
+				      unsigned int index)
+{
+	struct tegra186_bpmp *priv = bpmp->priv;
+	size_t message_size, queue_size;
+	struct iosys_map rx, tx;
+	unsigned int offset;
+	int err;
+
+	channel->ivc = devm_kzalloc(bpmp->dev, sizeof(*channel->ivc),
+				    GFP_KERNEL);
+	if (!channel->ivc)
+		return -ENOMEM;
+
+	message_size = tegra_ivc_align(MSG_MIN_SZ);
+	queue_size = tegra_ivc_total_queue_size(message_size);
+	offset = queue_size * index;
+
+	iosys_map_set_vaddr_iomem(&rx, priv->rx.virt + offset);
+	iosys_map_set_vaddr_iomem(&tx, priv->tx.virt + offset);
+
+	err = tegra_ivc_init(channel->ivc, NULL, &rx, priv->rx.phys + offset, &tx,
+			     priv->tx.phys + offset, 1, message_size, tegra186_bpmp_ivc_notify,
+			     bpmp);
+	if (err < 0) {
+		dev_err(bpmp->dev, "failed to setup IVC for channel %u: %d\n",
+			index, err);
+		return err;
+	}
+
+	init_completion(&channel->completion);
+	channel->bpmp = bpmp;
+
+	return 0;
+}
+
+static void tegra186_bpmp_channel_reset(struct tegra_bpmp_channel *channel)
+{
+	/* reset the channel state */
+	tegra_ivc_reset(channel->ivc);
+
+	/* sync the channel state with BPMP */
+	while (tegra_ivc_notified(channel->ivc))
+		;
+}
+
+static void tegra186_bpmp_channel_cleanup(struct tegra_bpmp_channel *channel)
+{
+	tegra_ivc_cleanup(channel->ivc);
+}
+
+static void mbox_handle_rx(struct mbox_client *client, void *data)
+{
+	struct tegra_bpmp *bpmp = mbox_client_to_bpmp(client);
+
+	tegra_bpmp_handle_rx(bpmp);
+}
+
+static int tegra186_bpmp_init(struct tegra_bpmp *bpmp)
+{
+	struct tegra186_bpmp *priv;
+	unsigned int i;
+	int err;
+
+	priv = devm_kzalloc(bpmp->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	bpmp->priv = priv;
+	priv->parent = bpmp;
+
+	priv->tx.pool = of_gen_pool_get(bpmp->dev->of_node, "shmem", 0);
+	if (!priv->tx.pool) {
+		dev_err(bpmp->dev, "TX shmem pool not found\n");
+		return -EPROBE_DEFER;
+	}
+
+	priv->tx.virt = (void __iomem *)gen_pool_dma_alloc(priv->tx.pool, 4096, &priv->tx.phys);
+	if (!priv->tx.virt) {
+		dev_err(bpmp->dev, "failed to allocate from TX pool\n");
+		return -ENOMEM;
+	}
+
+	priv->rx.pool = of_gen_pool_get(bpmp->dev->of_node, "shmem", 1);
+	if (!priv->rx.pool) {
+		dev_err(bpmp->dev, "RX shmem pool not found\n");
+		err = -EPROBE_DEFER;
+		goto free_tx;
+	}
+
+	priv->rx.virt = (void __iomem *)gen_pool_dma_alloc(priv->rx.pool, 4096, &priv->rx.phys);
+	if (!priv->rx.virt) {
+		dev_err(bpmp->dev, "failed to allocate from RX pool\n");
+		err = -ENOMEM;
+		goto free_tx;
+	}
+
+	err = tegra186_bpmp_channel_init(bpmp->tx_channel, bpmp,
+					 bpmp->soc->channels.cpu_tx.offset);
+	if (err < 0)
+		goto free_rx;
+
+	err = tegra186_bpmp_channel_init(bpmp->rx_channel, bpmp,
+					 bpmp->soc->channels.cpu_rx.offset);
+	if (err < 0)
+		goto cleanup_tx_channel;
+
+	for (i = 0; i < bpmp->threaded.count; i++) {
+		unsigned int index = bpmp->soc->channels.thread.offset + i;
+
+		err = tegra186_bpmp_channel_init(&bpmp->threaded_channels[i],
+						 bpmp, index);
+		if (err < 0)
+			goto cleanup_channels;
+	}
+
+	/* mbox registration */
+	priv->mbox.client.dev = bpmp->dev;
+	priv->mbox.client.rx_callback = mbox_handle_rx;
+	priv->mbox.client.tx_block = false;
+	priv->mbox.client.knows_txdone = false;
+
+	priv->mbox.channel = mbox_request_channel(&priv->mbox.client, 0);
+	if (IS_ERR(priv->mbox.channel)) {
+		err = PTR_ERR(priv->mbox.channel);
+		dev_err(bpmp->dev, "failed to get HSP mailbox: %d\n", err);
+		goto cleanup_channels;
+	}
+
+	tegra186_bpmp_channel_reset(bpmp->tx_channel);
+	tegra186_bpmp_channel_reset(bpmp->rx_channel);
+
+	for (i = 0; i < bpmp->threaded.count; i++)
+		tegra186_bpmp_channel_reset(&bpmp->threaded_channels[i]);
+
+	return 0;
+
+cleanup_channels:
+	for (i = 0; i < bpmp->threaded.count; i++) {
+		if (!bpmp->threaded_channels[i].bpmp)
+			continue;
+
+		tegra186_bpmp_channel_cleanup(&bpmp->threaded_channels[i]);
+	}
+
+	tegra186_bpmp_channel_cleanup(bpmp->rx_channel);
+cleanup_tx_channel:
+	tegra186_bpmp_channel_cleanup(bpmp->tx_channel);
+free_rx:
+	gen_pool_free(priv->rx.pool, (unsigned long)priv->rx.virt, 4096);
+free_tx:
+	gen_pool_free(priv->tx.pool, (unsigned long)priv->tx.virt, 4096);
+
+	return err;
+}
+
+static void tegra186_bpmp_deinit(struct tegra_bpmp *bpmp)
+{
+	struct tegra186_bpmp *priv = bpmp->priv;
+	unsigned int i;
+
+	mbox_free_channel(priv->mbox.channel);
+
+	for (i = 0; i < bpmp->threaded.count; i++)
+		tegra186_bpmp_channel_cleanup(&bpmp->threaded_channels[i]);
+
+	tegra186_bpmp_channel_cleanup(bpmp->rx_channel);
+	tegra186_bpmp_channel_cleanup(bpmp->tx_channel);
+
+	gen_pool_free(priv->rx.pool, (unsigned long)priv->rx.virt, 4096);
+	gen_pool_free(priv->tx.pool, (unsigned long)priv->tx.virt, 4096);
+}
+
+static int tegra186_bpmp_resume(struct tegra_bpmp *bpmp)
+{
+	unsigned int i;
+
+	/* reset message channels */
+	tegra186_bpmp_channel_reset(bpmp->tx_channel);
+	tegra186_bpmp_channel_reset(bpmp->rx_channel);
+
+	for (i = 0; i < bpmp->threaded.count; i++)
+		tegra186_bpmp_channel_reset(&bpmp->threaded_channels[i]);
+
+	return 0;
+}
+
+const struct tegra_bpmp_ops tegra186_bpmp_ops = {
+	.init = tegra186_bpmp_init,
+	.deinit = tegra186_bpmp_deinit,
+	.is_response_ready = tegra186_bpmp_is_message_ready,
+	.is_request_ready = tegra186_bpmp_is_message_ready,
+	.ack_response = tegra186_bpmp_ack_message,
+	.ack_request = tegra186_bpmp_ack_message,
+	.is_response_channel_free = tegra186_bpmp_is_channel_free,
+	.is_request_channel_free = tegra186_bpmp_is_channel_free,
+	.post_response = tegra186_bpmp_post_message,
+	.post_request = tegra186_bpmp_post_message,
+	.ring_doorbell = tegra186_bpmp_ring_doorbell,
+	.resume = tegra186_bpmp_resume,
+};
diff --git a/drivers/firmware/tegra/bpmp-tegra210.c b/drivers/firmware/tegra/bpmp-tegra210.c
new file mode 100644
index 000000000..6295f5640
--- /dev/null
+++ b/drivers/firmware/tegra/bpmp-tegra210.c
@@ -0,0 +1,237 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2018, NVIDIA CORPORATION.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#include <soc/tegra/bpmp.h>
+
+#include "bpmp-private.h"
+
+#define TRIGGER_OFFSET		0x000
+#define RESULT_OFFSET(id)	(0xc00 + id * 4)
+#define TRIGGER_ID_SHIFT	16
+#define TRIGGER_CMD_GET		4
+
+#define STA_OFFSET		0
+#define SET_OFFSET		4
+#define CLR_OFFSET		8
+
+#define CH_MASK(ch)	(0x3 << ((ch) * 2))
+#define SL_SIGL(ch)	(0x0 << ((ch) * 2))
+#define SL_QUED(ch)	(0x1 << ((ch) * 2))
+#define MA_FREE(ch)	(0x2 << ((ch) * 2))
+#define MA_ACKD(ch)	(0x3 << ((ch) * 2))
+
+struct tegra210_bpmp {
+	void __iomem *atomics;
+	void __iomem *arb_sema;
+	struct irq_data *tx_irq_data;
+};
+
+static u32 bpmp_channel_status(struct tegra_bpmp *bpmp, unsigned int index)
+{
+	struct tegra210_bpmp *priv = bpmp->priv;
+
+	return __raw_readl(priv->arb_sema + STA_OFFSET) & CH_MASK(index);
+}
+
+static bool tegra210_bpmp_is_response_ready(struct tegra_bpmp_channel *channel)
+{
+	unsigned int index = channel->index;
+
+	return bpmp_channel_status(channel->bpmp, index) == MA_ACKD(index);
+}
+
+static bool tegra210_bpmp_is_request_ready(struct tegra_bpmp_channel *channel)
+{
+	unsigned int index = channel->index;
+
+	return bpmp_channel_status(channel->bpmp, index) == SL_SIGL(index);
+}
+
+static bool
+tegra210_bpmp_is_request_channel_free(struct tegra_bpmp_channel *channel)
+{
+	unsigned int index = channel->index;
+
+	return bpmp_channel_status(channel->bpmp, index) == MA_FREE(index);
+}
+
+static bool
+tegra210_bpmp_is_response_channel_free(struct tegra_bpmp_channel *channel)
+{
+	unsigned int index = channel->index;
+
+	return bpmp_channel_status(channel->bpmp, index) == SL_QUED(index);
+}
+
+static int tegra210_bpmp_post_request(struct tegra_bpmp_channel *channel)
+{
+	struct tegra210_bpmp *priv = channel->bpmp->priv;
+
+	__raw_writel(CH_MASK(channel->index), priv->arb_sema + CLR_OFFSET);
+
+	return 0;
+}
+
+static int tegra210_bpmp_post_response(struct tegra_bpmp_channel *channel)
+{
+	struct tegra210_bpmp *priv = channel->bpmp->priv;
+
+	__raw_writel(MA_ACKD(channel->index), priv->arb_sema + SET_OFFSET);
+
+	return 0;
+}
+
+static int tegra210_bpmp_ack_response(struct tegra_bpmp_channel *channel)
+{
+	struct tegra210_bpmp *priv = channel->bpmp->priv;
+
+	__raw_writel(MA_ACKD(channel->index) ^ MA_FREE(channel->index),
+		     priv->arb_sema + CLR_OFFSET);
+
+	return 0;
+}
+
+static int tegra210_bpmp_ack_request(struct tegra_bpmp_channel *channel)
+{
+	struct tegra210_bpmp *priv = channel->bpmp->priv;
+
+	__raw_writel(SL_QUED(channel->index), priv->arb_sema + SET_OFFSET);
+
+	return 0;
+}
+
+static int tegra210_bpmp_ring_doorbell(struct tegra_bpmp *bpmp)
+{
+	struct tegra210_bpmp *priv = bpmp->priv;
+	struct irq_data *irq_data = priv->tx_irq_data;
+
+	/*
+	 * Tegra Legacy Interrupt Controller (LIC) is used to notify BPMP of
+	 * available messages
+	 */
+	if (irq_data->chip->irq_retrigger)
+		return irq_data->chip->irq_retrigger(irq_data);
+
+	return -EINVAL;
+}
+
+static irqreturn_t rx_irq(int irq, void *data)
+{
+	struct tegra_bpmp *bpmp = data;
+
+	tegra_bpmp_handle_rx(bpmp);
+
+	return IRQ_HANDLED;
+}
+
+static int tegra210_bpmp_channel_init(struct tegra_bpmp_channel *channel,
+				      struct tegra_bpmp *bpmp,
+				      unsigned int index)
+{
+	struct tegra210_bpmp *priv = bpmp->priv;
+	void __iomem *p;
+	u32 address;
+
+	/* Retrieve channel base address from BPMP */
+	writel(index << TRIGGER_ID_SHIFT | TRIGGER_CMD_GET,
+	       priv->atomics + TRIGGER_OFFSET);
+	address = readl(priv->atomics + RESULT_OFFSET(index));
+
+	p = devm_ioremap(bpmp->dev, address, 0x80);
+	if (!p)
+		return -ENOMEM;
+
+	iosys_map_set_vaddr_iomem(&channel->ib, p);
+	iosys_map_set_vaddr_iomem(&channel->ob, p);
+
+	channel->index = index;
+	init_completion(&channel->completion);
+	channel->bpmp = bpmp;
+
+	return 0;
+}
+
+static int tegra210_bpmp_init(struct tegra_bpmp *bpmp)
+{
+	struct platform_device *pdev = to_platform_device(bpmp->dev);
+	struct tegra210_bpmp *priv;
+	unsigned int i;
+	int err;
+
+	priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	bpmp->priv = priv;
+
+	priv->atomics = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(priv->atomics))
+		return PTR_ERR(priv->atomics);
+
+	priv->arb_sema = devm_platform_ioremap_resource(pdev, 1);
+	if (IS_ERR(priv->arb_sema))
+		return PTR_ERR(priv->arb_sema);
+
+	err = tegra210_bpmp_channel_init(bpmp->tx_channel, bpmp,
+					 bpmp->soc->channels.cpu_tx.offset);
+	if (err < 0)
+		return err;
+
+	err = tegra210_bpmp_channel_init(bpmp->rx_channel, bpmp,
+					 bpmp->soc->channels.cpu_rx.offset);
+	if (err < 0)
+		return err;
+
+	for (i = 0; i < bpmp->threaded.count; i++) {
+		unsigned int index = bpmp->soc->channels.thread.offset + i;
+
+		err = tegra210_bpmp_channel_init(&bpmp->threaded_channels[i],
+						 bpmp, index);
+		if (err < 0)
+			return err;
+	}
+
+	err = platform_get_irq_byname(pdev, "tx");
+	if (err < 0)
+		return err;
+
+	priv->tx_irq_data = irq_get_irq_data(err);
+	if (!priv->tx_irq_data) {
+		dev_err(&pdev->dev, "failed to get IRQ data for TX IRQ\n");
+		return -ENOENT;
+	}
+
+	err = platform_get_irq_byname(pdev, "rx");
+	if (err < 0)
+		return err;
+
+	err = devm_request_irq(&pdev->dev, err, rx_irq,
+			       IRQF_NO_SUSPEND, dev_name(&pdev->dev), bpmp);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to request IRQ: %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
+
+const struct tegra_bpmp_ops tegra210_bpmp_ops = {
+	.init = tegra210_bpmp_init,
+	.is_response_ready = tegra210_bpmp_is_response_ready,
+	.is_request_ready = tegra210_bpmp_is_request_ready,
+	.ack_response = tegra210_bpmp_ack_response,
+	.ack_request = tegra210_bpmp_ack_request,
+	.is_response_channel_free = tegra210_bpmp_is_response_channel_free,
+	.is_request_channel_free = tegra210_bpmp_is_request_channel_free,
+	.post_response = tegra210_bpmp_post_response,
+	.post_request = tegra210_bpmp_post_request,
+	.ring_doorbell = tegra210_bpmp_ring_doorbell,
+};
diff --git a/drivers/firmware/tegra/bpmp.c b/drivers/firmware/tegra/bpmp.c
new file mode 100644
index 000000000..042c20439
--- /dev/null
+++ b/drivers/firmware/tegra/bpmp.c
@@ -0,0 +1,882 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION.  All rights reserved.
+ */
+
+#include <linux/clk/tegra.h>
+#include <linux/genalloc.h>
+#include <linux/mailbox_client.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/semaphore.h>
+#include <linux/sched/clock.h>
+
+#include <soc/tegra/bpmp.h>
+#include <soc/tegra/bpmp-abi.h>
+#include <soc/tegra/ivc.h>
+
+#include "bpmp-private.h"
+
+#define MSG_ACK		BIT(0)
+#define MSG_RING	BIT(1)
+#define TAG_SZ		32
+
+static inline struct tegra_bpmp *
+mbox_client_to_bpmp(struct mbox_client *client)
+{
+	return container_of(client, struct tegra_bpmp, mbox.client);
+}
+
+static inline const struct tegra_bpmp_ops *
+channel_to_ops(struct tegra_bpmp_channel *channel)
+{
+	struct tegra_bpmp *bpmp = channel->bpmp;
+
+	return bpmp->soc->ops;
+}
+
+struct tegra_bpmp *tegra_bpmp_get(struct device *dev)
+{
+	struct platform_device *pdev;
+	struct tegra_bpmp *bpmp;
+	struct device_node *np;
+
+	np = of_parse_phandle(dev->of_node, "nvidia,bpmp", 0);
+	if (!np)
+		return ERR_PTR(-ENOENT);
+
+	pdev = of_find_device_by_node(np);
+	if (!pdev) {
+		bpmp = ERR_PTR(-ENODEV);
+		goto put;
+	}
+
+	bpmp = platform_get_drvdata(pdev);
+	if (!bpmp) {
+		bpmp = ERR_PTR(-EPROBE_DEFER);
+		put_device(&pdev->dev);
+		goto put;
+	}
+
+put:
+	of_node_put(np);
+	return bpmp;
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_get);
+
+void tegra_bpmp_put(struct tegra_bpmp *bpmp)
+{
+	if (bpmp)
+		put_device(bpmp->dev);
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_put);
+
+static int
+tegra_bpmp_channel_get_thread_index(struct tegra_bpmp_channel *channel)
+{
+	struct tegra_bpmp *bpmp = channel->bpmp;
+	unsigned int count;
+	int index;
+
+	count = bpmp->soc->channels.thread.count;
+
+	index = channel - channel->bpmp->threaded_channels;
+	if (index < 0 || index >= count)
+		return -EINVAL;
+
+	return index;
+}
+
+static bool tegra_bpmp_message_valid(const struct tegra_bpmp_message *msg)
+{
+	return (msg->tx.size <= MSG_DATA_MIN_SZ) &&
+	       (msg->rx.size <= MSG_DATA_MIN_SZ) &&
+	       (msg->tx.size == 0 || msg->tx.data) &&
+	       (msg->rx.size == 0 || msg->rx.data);
+}
+
+static bool tegra_bpmp_is_response_ready(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->is_response_ready(channel);
+}
+
+static bool tegra_bpmp_is_request_ready(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->is_request_ready(channel);
+}
+
+static int tegra_bpmp_wait_response(struct tegra_bpmp_channel *channel)
+{
+	unsigned long timeout = channel->bpmp->soc->channels.cpu_tx.timeout;
+	ktime_t end;
+
+	end = ktime_add_us(ktime_get(), timeout);
+
+	do {
+		if (tegra_bpmp_is_response_ready(channel))
+			return 0;
+	} while (ktime_before(ktime_get(), end));
+
+	return -ETIMEDOUT;
+}
+
+static int tegra_bpmp_ack_response(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->ack_response(channel);
+}
+
+static int tegra_bpmp_ack_request(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->ack_request(channel);
+}
+
+static bool
+tegra_bpmp_is_request_channel_free(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->is_request_channel_free(channel);
+}
+
+static bool
+tegra_bpmp_is_response_channel_free(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->is_response_channel_free(channel);
+}
+
+static int
+tegra_bpmp_wait_request_channel_free(struct tegra_bpmp_channel *channel)
+{
+	unsigned long timeout = channel->bpmp->soc->channels.cpu_tx.timeout;
+	ktime_t start, now;
+
+	start = ns_to_ktime(local_clock());
+
+	do {
+		if (tegra_bpmp_is_request_channel_free(channel))
+			return 0;
+
+		now = ns_to_ktime(local_clock());
+	} while (ktime_us_delta(now, start) < timeout);
+
+	return -ETIMEDOUT;
+}
+
+static int tegra_bpmp_post_request(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->post_request(channel);
+}
+
+static int tegra_bpmp_post_response(struct tegra_bpmp_channel *channel)
+{
+	const struct tegra_bpmp_ops *ops = channel_to_ops(channel);
+
+	return ops->post_response(channel);
+}
+
+static int tegra_bpmp_ring_doorbell(struct tegra_bpmp *bpmp)
+{
+	return bpmp->soc->ops->ring_doorbell(bpmp);
+}
+
+static ssize_t __tegra_bpmp_channel_read(struct tegra_bpmp_channel *channel,
+					 void *data, size_t size, int *ret)
+{
+	int err;
+
+	if (data && size > 0)
+		tegra_bpmp_mb_read(data, &channel->ib, size);
+
+	err = tegra_bpmp_ack_response(channel);
+	if (err < 0)
+		return err;
+
+	*ret = tegra_bpmp_mb_read_field(&channel->ib, code);
+
+	return 0;
+}
+
+static ssize_t tegra_bpmp_channel_read(struct tegra_bpmp_channel *channel,
+				       void *data, size_t size, int *ret)
+{
+	struct tegra_bpmp *bpmp = channel->bpmp;
+	unsigned long flags;
+	ssize_t err;
+	int index;
+
+	index = tegra_bpmp_channel_get_thread_index(channel);
+	if (index < 0) {
+		err = index;
+		goto unlock;
+	}
+
+	spin_lock_irqsave(&bpmp->lock, flags);
+	err = __tegra_bpmp_channel_read(channel, data, size, ret);
+	clear_bit(index, bpmp->threaded.allocated);
+	spin_unlock_irqrestore(&bpmp->lock, flags);
+
+unlock:
+	up(&bpmp->threaded.lock);
+
+	return err;
+}
+
+static ssize_t __tegra_bpmp_channel_write(struct tegra_bpmp_channel *channel,
+					  unsigned int mrq, unsigned long flags,
+					  const void *data, size_t size)
+{
+	tegra_bpmp_mb_write_field(&channel->ob, code, mrq);
+	tegra_bpmp_mb_write_field(&channel->ob, flags, flags);
+
+	if (data && size > 0)
+		tegra_bpmp_mb_write(&channel->ob, data, size);
+
+	return tegra_bpmp_post_request(channel);
+}
+
+static struct tegra_bpmp_channel *
+tegra_bpmp_write_threaded(struct tegra_bpmp *bpmp, unsigned int mrq,
+			  const void *data, size_t size)
+{
+	unsigned long timeout = bpmp->soc->channels.thread.timeout;
+	unsigned int count = bpmp->soc->channels.thread.count;
+	struct tegra_bpmp_channel *channel;
+	unsigned long flags;
+	unsigned int index;
+	int err;
+
+	err = down_timeout(&bpmp->threaded.lock, usecs_to_jiffies(timeout));
+	if (err < 0)
+		return ERR_PTR(err);
+
+	spin_lock_irqsave(&bpmp->lock, flags);
+
+	index = find_first_zero_bit(bpmp->threaded.allocated, count);
+	if (index == count) {
+		err = -EBUSY;
+		goto unlock;
+	}
+
+	channel = &bpmp->threaded_channels[index];
+
+	if (!tegra_bpmp_is_request_channel_free(channel)) {
+		err = -EBUSY;
+		goto unlock;
+	}
+
+	set_bit(index, bpmp->threaded.allocated);
+
+	err = __tegra_bpmp_channel_write(channel, mrq, MSG_ACK | MSG_RING,
+					 data, size);
+	if (err < 0)
+		goto clear_allocated;
+
+	set_bit(index, bpmp->threaded.busy);
+
+	spin_unlock_irqrestore(&bpmp->lock, flags);
+	return channel;
+
+clear_allocated:
+	clear_bit(index, bpmp->threaded.allocated);
+unlock:
+	spin_unlock_irqrestore(&bpmp->lock, flags);
+	up(&bpmp->threaded.lock);
+
+	return ERR_PTR(err);
+}
+
+static ssize_t tegra_bpmp_channel_write(struct tegra_bpmp_channel *channel,
+					unsigned int mrq, unsigned long flags,
+					const void *data, size_t size)
+{
+	int err;
+
+	err = tegra_bpmp_wait_request_channel_free(channel);
+	if (err < 0)
+		return err;
+
+	return __tegra_bpmp_channel_write(channel, mrq, flags, data, size);
+}
+
+int tegra_bpmp_transfer_atomic(struct tegra_bpmp *bpmp,
+			       struct tegra_bpmp_message *msg)
+{
+	struct tegra_bpmp_channel *channel;
+	int err;
+
+	if (WARN_ON(!irqs_disabled()))
+		return -EPERM;
+
+	if (!tegra_bpmp_message_valid(msg))
+		return -EINVAL;
+
+	channel = bpmp->tx_channel;
+
+	spin_lock(&bpmp->atomic_tx_lock);
+
+	err = tegra_bpmp_channel_write(channel, msg->mrq, MSG_ACK,
+				       msg->tx.data, msg->tx.size);
+	if (err < 0) {
+		spin_unlock(&bpmp->atomic_tx_lock);
+		return err;
+	}
+
+	spin_unlock(&bpmp->atomic_tx_lock);
+
+	err = tegra_bpmp_ring_doorbell(bpmp);
+	if (err < 0)
+		return err;
+
+	err = tegra_bpmp_wait_response(channel);
+	if (err < 0)
+		return err;
+
+	return __tegra_bpmp_channel_read(channel, msg->rx.data, msg->rx.size,
+					 &msg->rx.ret);
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_transfer_atomic);
+
+int tegra_bpmp_transfer(struct tegra_bpmp *bpmp,
+			struct tegra_bpmp_message *msg)
+{
+	struct tegra_bpmp_channel *channel;
+	unsigned long timeout;
+	int err;
+
+	if (WARN_ON(irqs_disabled()))
+		return -EPERM;
+
+	if (!tegra_bpmp_message_valid(msg))
+		return -EINVAL;
+
+	channel = tegra_bpmp_write_threaded(bpmp, msg->mrq, msg->tx.data,
+					    msg->tx.size);
+	if (IS_ERR(channel))
+		return PTR_ERR(channel);
+
+	err = tegra_bpmp_ring_doorbell(bpmp);
+	if (err < 0)
+		return err;
+
+	timeout = usecs_to_jiffies(bpmp->soc->channels.thread.timeout);
+
+	err = wait_for_completion_timeout(&channel->completion, timeout);
+	if (err == 0)
+		return -ETIMEDOUT;
+
+	return tegra_bpmp_channel_read(channel, msg->rx.data, msg->rx.size,
+				       &msg->rx.ret);
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_transfer);
+
+static struct tegra_bpmp_mrq *tegra_bpmp_find_mrq(struct tegra_bpmp *bpmp,
+						  unsigned int mrq)
+{
+	struct tegra_bpmp_mrq *entry;
+
+	list_for_each_entry(entry, &bpmp->mrqs, list)
+		if (entry->mrq == mrq)
+			return entry;
+
+	return NULL;
+}
+
+void tegra_bpmp_mrq_return(struct tegra_bpmp_channel *channel, int code,
+			   const void *data, size_t size)
+{
+	unsigned long flags = tegra_bpmp_mb_read_field(&channel->ib, flags);
+	struct tegra_bpmp *bpmp = channel->bpmp;
+	int err;
+
+	if (WARN_ON(size > MSG_DATA_MIN_SZ))
+		return;
+
+	err = tegra_bpmp_ack_request(channel);
+	if (WARN_ON(err < 0))
+		return;
+
+	if ((flags & MSG_ACK) == 0)
+		return;
+
+	if (WARN_ON(!tegra_bpmp_is_response_channel_free(channel)))
+		return;
+
+	tegra_bpmp_mb_write_field(&channel->ob, code, code);
+
+	if (data && size > 0)
+		tegra_bpmp_mb_write(&channel->ob, data, size);
+
+	err = tegra_bpmp_post_response(channel);
+	if (WARN_ON(err < 0))
+		return;
+
+	if (flags & MSG_RING) {
+		err = tegra_bpmp_ring_doorbell(bpmp);
+		if (WARN_ON(err < 0))
+			return;
+	}
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_mrq_return);
+
+static void tegra_bpmp_handle_mrq(struct tegra_bpmp *bpmp,
+				  unsigned int mrq,
+				  struct tegra_bpmp_channel *channel)
+{
+	struct tegra_bpmp_mrq *entry;
+	u32 zero = 0;
+
+	spin_lock(&bpmp->lock);
+
+	entry = tegra_bpmp_find_mrq(bpmp, mrq);
+	if (!entry) {
+		spin_unlock(&bpmp->lock);
+		tegra_bpmp_mrq_return(channel, -EINVAL, &zero, sizeof(zero));
+		return;
+	}
+
+	entry->handler(mrq, channel, entry->data);
+
+	spin_unlock(&bpmp->lock);
+}
+
+int tegra_bpmp_request_mrq(struct tegra_bpmp *bpmp, unsigned int mrq,
+			   tegra_bpmp_mrq_handler_t handler, void *data)
+{
+	struct tegra_bpmp_mrq *entry;
+	unsigned long flags;
+
+	if (!handler)
+		return -EINVAL;
+
+	entry = devm_kzalloc(bpmp->dev, sizeof(*entry), GFP_KERNEL);
+	if (!entry)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&bpmp->lock, flags);
+
+	entry->mrq = mrq;
+	entry->handler = handler;
+	entry->data = data;
+	list_add(&entry->list, &bpmp->mrqs);
+
+	spin_unlock_irqrestore(&bpmp->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_request_mrq);
+
+void tegra_bpmp_free_mrq(struct tegra_bpmp *bpmp, unsigned int mrq, void *data)
+{
+	struct tegra_bpmp_mrq *entry;
+	unsigned long flags;
+
+	spin_lock_irqsave(&bpmp->lock, flags);
+
+	entry = tegra_bpmp_find_mrq(bpmp, mrq);
+	if (!entry)
+		goto unlock;
+
+	list_del(&entry->list);
+	devm_kfree(bpmp->dev, entry);
+
+unlock:
+	spin_unlock_irqrestore(&bpmp->lock, flags);
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_free_mrq);
+
+bool tegra_bpmp_mrq_is_supported(struct tegra_bpmp *bpmp, unsigned int mrq)
+{
+	struct mrq_query_abi_request req = { .mrq = mrq };
+	struct mrq_query_abi_response resp;
+	struct tegra_bpmp_message msg = {
+		.mrq = MRQ_QUERY_ABI,
+		.tx = {
+			.data = &req,
+			.size = sizeof(req),
+		},
+		.rx = {
+			.data = &resp,
+			.size = sizeof(resp),
+		},
+	};
+	int err;
+
+	err = tegra_bpmp_transfer(bpmp, &msg);
+	if (err || msg.rx.ret)
+		return false;
+
+	return resp.status == 0;
+}
+EXPORT_SYMBOL_GPL(tegra_bpmp_mrq_is_supported);
+
+static void tegra_bpmp_mrq_handle_ping(unsigned int mrq,
+				       struct tegra_bpmp_channel *channel,
+				       void *data)
+{
+	struct mrq_ping_request request;
+	struct mrq_ping_response response;
+
+	tegra_bpmp_mb_read(&request, &channel->ib, sizeof(request));
+
+	memset(&response, 0, sizeof(response));
+	response.reply = request.challenge << 1;
+
+	tegra_bpmp_mrq_return(channel, 0, &response, sizeof(response));
+}
+
+static int tegra_bpmp_ping(struct tegra_bpmp *bpmp)
+{
+	struct mrq_ping_response response;
+	struct mrq_ping_request request;
+	struct tegra_bpmp_message msg;
+	unsigned long flags;
+	ktime_t start, end;
+	int err;
+
+	memset(&request, 0, sizeof(request));
+	request.challenge = 1;
+
+	memset(&response, 0, sizeof(response));
+
+	memset(&msg, 0, sizeof(msg));
+	msg.mrq = MRQ_PING;
+	msg.tx.data = &request;
+	msg.tx.size = sizeof(request);
+	msg.rx.data = &response;
+	msg.rx.size = sizeof(response);
+
+	local_irq_save(flags);
+	start = ktime_get();
+	err = tegra_bpmp_transfer_atomic(bpmp, &msg);
+	end = ktime_get();
+	local_irq_restore(flags);
+
+	if (!err)
+		dev_dbg(bpmp->dev,
+			"ping ok: challenge: %u, response: %u, time: %lld\n",
+			request.challenge, response.reply,
+			ktime_to_us(ktime_sub(end, start)));
+
+	return err;
+}
+
+/* deprecated version of tag query */
+static int tegra_bpmp_get_firmware_tag_old(struct tegra_bpmp *bpmp, char *tag,
+					   size_t size)
+{
+	struct mrq_query_tag_request request;
+	struct tegra_bpmp_message msg;
+	unsigned long flags;
+	dma_addr_t phys;
+	void *virt;
+	int err;
+
+	if (size != TAG_SZ)
+		return -EINVAL;
+
+	virt = dma_alloc_coherent(bpmp->dev, TAG_SZ, &phys,
+				  GFP_KERNEL | GFP_DMA32);
+	if (!virt)
+		return -ENOMEM;
+
+	memset(&request, 0, sizeof(request));
+	request.addr = phys;
+
+	memset(&msg, 0, sizeof(msg));
+	msg.mrq = MRQ_QUERY_TAG;
+	msg.tx.data = &request;
+	msg.tx.size = sizeof(request);
+
+	local_irq_save(flags);
+	err = tegra_bpmp_transfer_atomic(bpmp, &msg);
+	local_irq_restore(flags);
+
+	if (err == 0)
+		memcpy(tag, virt, TAG_SZ);
+
+	dma_free_coherent(bpmp->dev, TAG_SZ, virt, phys);
+
+	return err;
+}
+
+static int tegra_bpmp_get_firmware_tag(struct tegra_bpmp *bpmp, char *tag,
+				       size_t size)
+{
+	if (tegra_bpmp_mrq_is_supported(bpmp, MRQ_QUERY_FW_TAG)) {
+		struct mrq_query_fw_tag_response resp;
+		struct tegra_bpmp_message msg = {
+			.mrq = MRQ_QUERY_FW_TAG,
+			.rx = {
+				.data = &resp,
+				.size = sizeof(resp),
+			},
+		};
+		int err;
+
+		if (size != sizeof(resp.tag))
+			return -EINVAL;
+
+		err = tegra_bpmp_transfer(bpmp, &msg);
+
+		if (err)
+			return err;
+		if (msg.rx.ret < 0)
+			return -EINVAL;
+
+		memcpy(tag, resp.tag, sizeof(resp.tag));
+		return 0;
+	}
+
+	return tegra_bpmp_get_firmware_tag_old(bpmp, tag, size);
+}
+
+static void tegra_bpmp_channel_signal(struct tegra_bpmp_channel *channel)
+{
+	unsigned long flags = tegra_bpmp_mb_read_field(&channel->ob, flags);
+
+	if ((flags & MSG_RING) == 0)
+		return;
+
+	complete(&channel->completion);
+}
+
+void tegra_bpmp_handle_rx(struct tegra_bpmp *bpmp)
+{
+	struct tegra_bpmp_channel *channel;
+	unsigned int i, count;
+	unsigned long *busy;
+
+	channel = bpmp->rx_channel;
+	count = bpmp->soc->channels.thread.count;
+	busy = bpmp->threaded.busy;
+
+	if (tegra_bpmp_is_request_ready(channel)) {
+		unsigned int mrq = tegra_bpmp_mb_read_field(&channel->ib, code);
+
+		tegra_bpmp_handle_mrq(bpmp, mrq, channel);
+	}
+
+	spin_lock(&bpmp->lock);
+
+	for_each_set_bit(i, busy, count) {
+		struct tegra_bpmp_channel *channel;
+
+		channel = &bpmp->threaded_channels[i];
+
+		if (tegra_bpmp_is_response_ready(channel)) {
+			tegra_bpmp_channel_signal(channel);
+			clear_bit(i, busy);
+		}
+	}
+
+	spin_unlock(&bpmp->lock);
+}
+
+static int tegra_bpmp_probe(struct platform_device *pdev)
+{
+	struct tegra_bpmp *bpmp;
+	char tag[TAG_SZ];
+	size_t size;
+	int err;
+
+	bpmp = devm_kzalloc(&pdev->dev, sizeof(*bpmp), GFP_KERNEL);
+	if (!bpmp)
+		return -ENOMEM;
+
+	bpmp->soc = of_device_get_match_data(&pdev->dev);
+	bpmp->dev = &pdev->dev;
+
+	INIT_LIST_HEAD(&bpmp->mrqs);
+	spin_lock_init(&bpmp->lock);
+
+	bpmp->threaded.count = bpmp->soc->channels.thread.count;
+	sema_init(&bpmp->threaded.lock, bpmp->threaded.count);
+
+	size = BITS_TO_LONGS(bpmp->threaded.count) * sizeof(long);
+
+	bpmp->threaded.allocated = devm_kzalloc(&pdev->dev, size, GFP_KERNEL);
+	if (!bpmp->threaded.allocated)
+		return -ENOMEM;
+
+	bpmp->threaded.busy = devm_kzalloc(&pdev->dev, size, GFP_KERNEL);
+	if (!bpmp->threaded.busy)
+		return -ENOMEM;
+
+	spin_lock_init(&bpmp->atomic_tx_lock);
+	bpmp->tx_channel = devm_kzalloc(&pdev->dev, sizeof(*bpmp->tx_channel),
+					GFP_KERNEL);
+	if (!bpmp->tx_channel)
+		return -ENOMEM;
+
+	bpmp->rx_channel = devm_kzalloc(&pdev->dev, sizeof(*bpmp->rx_channel),
+	                                GFP_KERNEL);
+	if (!bpmp->rx_channel)
+		return -ENOMEM;
+
+	bpmp->threaded_channels = devm_kcalloc(&pdev->dev, bpmp->threaded.count,
+					       sizeof(*bpmp->threaded_channels),
+					       GFP_KERNEL);
+	if (!bpmp->threaded_channels)
+		return -ENOMEM;
+
+	err = bpmp->soc->ops->init(bpmp);
+	if (err < 0)
+		return err;
+
+	err = tegra_bpmp_request_mrq(bpmp, MRQ_PING,
+				     tegra_bpmp_mrq_handle_ping, bpmp);
+	if (err < 0)
+		goto deinit;
+
+	err = tegra_bpmp_ping(bpmp);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to ping BPMP: %d\n", err);
+		goto free_mrq;
+	}
+
+	err = tegra_bpmp_get_firmware_tag(bpmp, tag, sizeof(tag));
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to get firmware tag: %d\n", err);
+		goto free_mrq;
+	}
+
+	dev_info(&pdev->dev, "firmware: %.*s\n", (int)sizeof(tag), tag);
+
+	platform_set_drvdata(pdev, bpmp);
+
+	err = of_platform_default_populate(pdev->dev.of_node, NULL, &pdev->dev);
+	if (err < 0)
+		goto free_mrq;
+
+	if (of_find_property(pdev->dev.of_node, "#clock-cells", NULL)) {
+		err = tegra_bpmp_init_clocks(bpmp);
+		if (err < 0)
+			goto free_mrq;
+	}
+
+	if (of_find_property(pdev->dev.of_node, "#reset-cells", NULL)) {
+		err = tegra_bpmp_init_resets(bpmp);
+		if (err < 0)
+			goto free_mrq;
+	}
+
+	if (of_find_property(pdev->dev.of_node, "#power-domain-cells", NULL)) {
+		err = tegra_bpmp_init_powergates(bpmp);
+		if (err < 0)
+			goto free_mrq;
+	}
+
+	err = tegra_bpmp_init_debugfs(bpmp);
+	if (err < 0)
+		dev_err(&pdev->dev, "debugfs initialization failed: %d\n", err);
+
+	return 0;
+
+free_mrq:
+	tegra_bpmp_free_mrq(bpmp, MRQ_PING, bpmp);
+deinit:
+	if (bpmp->soc->ops->deinit)
+		bpmp->soc->ops->deinit(bpmp);
+
+	return err;
+}
+
+static int __maybe_unused tegra_bpmp_resume(struct device *dev)
+{
+	struct tegra_bpmp *bpmp = dev_get_drvdata(dev);
+
+	if (bpmp->soc->ops->resume)
+		return bpmp->soc->ops->resume(bpmp);
+	else
+		return 0;
+}
+
+static const struct dev_pm_ops tegra_bpmp_pm_ops = {
+	.resume_noirq = tegra_bpmp_resume,
+};
+
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_186_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_234_SOC)
+static const struct tegra_bpmp_soc tegra186_soc = {
+	.channels = {
+		.cpu_tx = {
+			.offset = 3,
+			.timeout = 60 * USEC_PER_SEC,
+		},
+		.thread = {
+			.offset = 0,
+			.count = 3,
+			.timeout = 600 * USEC_PER_SEC,
+		},
+		.cpu_rx = {
+			.offset = 13,
+			.timeout = 0,
+		},
+	},
+	.ops = &tegra186_bpmp_ops,
+	.num_resets = 193,
+};
+#endif
+
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_210_SOC)
+static const struct tegra_bpmp_soc tegra210_soc = {
+	.channels = {
+		.cpu_tx = {
+			.offset = 0,
+			.count = 1,
+			.timeout = 60 * USEC_PER_SEC,
+		},
+		.thread = {
+			.offset = 4,
+			.count = 1,
+			.timeout = 600 * USEC_PER_SEC,
+		},
+		.cpu_rx = {
+			.offset = 8,
+			.count = 1,
+			.timeout = 0,
+		},
+	},
+	.ops = &tegra210_bpmp_ops,
+};
+#endif
+
+static const struct of_device_id tegra_bpmp_match[] = {
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_186_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) || \
+    IS_ENABLED(CONFIG_ARCH_TEGRA_234_SOC)
+	{ .compatible = "nvidia,tegra186-bpmp", .data = &tegra186_soc },
+#endif
+#if IS_ENABLED(CONFIG_ARCH_TEGRA_210_SOC)
+	{ .compatible = "nvidia,tegra210-bpmp", .data = &tegra210_soc },
+#endif
+	{ }
+};
+
+static struct platform_driver tegra_bpmp_driver = {
+	.driver = {
+		.name = "tegra-bpmp",
+		.of_match_table = tegra_bpmp_match,
+		.pm = &tegra_bpmp_pm_ops,
+		.suppress_bind_attrs = true,
+	},
+	.probe = tegra_bpmp_probe,
+};
+builtin_platform_driver(tegra_bpmp_driver);
diff --git a/drivers/firmware/tegra/ivc.c b/drivers/firmware/tegra/ivc.c
new file mode 100644
index 000000000..8c9aff980
--- /dev/null
+++ b/drivers/firmware/tegra/ivc.c
@@ -0,0 +1,721 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2014-2016, NVIDIA CORPORATION.  All rights reserved.
+ */
+
+#include <soc/tegra/ivc.h>
+
+#define TEGRA_IVC_ALIGN 64
+
+/*
+ * IVC channel reset protocol.
+ *
+ * Each end uses its tx_channel.state to indicate its synchronization state.
+ */
+enum tegra_ivc_state {
+	/*
+	 * This value is zero for backwards compatibility with services that
+	 * assume channels to be initially zeroed. Such channels are in an
+	 * initially valid state, but cannot be asynchronously reset, and must
+	 * maintain a valid state at all times.
+	 *
+	 * The transmitting end can enter the established state from the sync or
+	 * ack state when it observes the receiving endpoint in the ack or
+	 * established state, indicating that has cleared the counters in our
+	 * rx_channel.
+	 */
+	TEGRA_IVC_STATE_ESTABLISHED = 0,
+
+	/*
+	 * If an endpoint is observed in the sync state, the remote endpoint is
+	 * allowed to clear the counters it owns asynchronously with respect to
+	 * the current endpoint. Therefore, the current endpoint is no longer
+	 * allowed to communicate.
+	 */
+	TEGRA_IVC_STATE_SYNC,
+
+	/*
+	 * When the transmitting end observes the receiving end in the sync
+	 * state, it can clear the w_count and r_count and transition to the ack
+	 * state. If the remote endpoint observes us in the ack state, it can
+	 * return to the established state once it has cleared its counters.
+	 */
+	TEGRA_IVC_STATE_ACK
+};
+
+/*
+ * This structure is divided into two-cache aligned parts, the first is only
+ * written through the tx.channel pointer, while the second is only written
+ * through the rx.channel pointer. This delineates ownership of the cache
+ * lines, which is critical to performance and necessary in non-cache coherent
+ * implementations.
+ */
+struct tegra_ivc_header {
+	union {
+		struct {
+			/* fields owned by the transmitting end */
+			u32 count;
+			u32 state;
+		};
+
+		u8 pad[TEGRA_IVC_ALIGN];
+	} tx;
+
+	union {
+		/* fields owned by the receiving end */
+		u32 count;
+		u8 pad[TEGRA_IVC_ALIGN];
+	} rx;
+};
+
+#define tegra_ivc_header_read_field(hdr, field) \
+	iosys_map_rd_field(hdr, 0, struct tegra_ivc_header, field)
+
+#define tegra_ivc_header_write_field(hdr, field, value) \
+	iosys_map_wr_field(hdr, 0, struct tegra_ivc_header, field, value)
+
+static inline void tegra_ivc_invalidate(struct tegra_ivc *ivc, dma_addr_t phys)
+{
+	if (!ivc->peer)
+		return;
+
+	dma_sync_single_for_cpu(ivc->peer, phys, TEGRA_IVC_ALIGN,
+				DMA_FROM_DEVICE);
+}
+
+static inline void tegra_ivc_flush(struct tegra_ivc *ivc, dma_addr_t phys)
+{
+	if (!ivc->peer)
+		return;
+
+	dma_sync_single_for_device(ivc->peer, phys, TEGRA_IVC_ALIGN,
+				   DMA_TO_DEVICE);
+}
+
+static inline bool tegra_ivc_empty(struct tegra_ivc *ivc, struct iosys_map *map)
+{
+	/*
+	 * This function performs multiple checks on the same values with
+	 * security implications, so create snapshots with READ_ONCE() to
+	 * ensure that these checks use the same values.
+	 */
+	u32 tx = tegra_ivc_header_read_field(map, tx.count);
+	u32 rx = tegra_ivc_header_read_field(map, rx.count);
+
+	/*
+	 * Perform an over-full check to prevent denial of service attacks
+	 * where a server could be easily fooled into believing that there's
+	 * an extremely large number of frames ready, since receivers are not
+	 * expected to check for full or over-full conditions.
+	 *
+	 * Although the channel isn't empty, this is an invalid case caused by
+	 * a potentially malicious peer, so returning empty is safer, because
+	 * it gives the impression that the channel has gone silent.
+	 */
+	if (tx - rx > ivc->num_frames)
+		return true;
+
+	return tx == rx;
+}
+
+static inline bool tegra_ivc_full(struct tegra_ivc *ivc, struct iosys_map *map)
+{
+	u32 tx = tegra_ivc_header_read_field(map, tx.count);
+	u32 rx = tegra_ivc_header_read_field(map, rx.count);
+
+	/*
+	 * Invalid cases where the counters indicate that the queue is over
+	 * capacity also appear full.
+	 */
+	return tx - rx >= ivc->num_frames;
+}
+
+static inline u32 tegra_ivc_available(struct tegra_ivc *ivc, struct iosys_map *map)
+{
+	u32 tx = tegra_ivc_header_read_field(map, tx.count);
+	u32 rx = tegra_ivc_header_read_field(map, rx.count);
+
+	/*
+	 * This function isn't expected to be used in scenarios where an
+	 * over-full situation can lead to denial of service attacks. See the
+	 * comment in tegra_ivc_empty() for an explanation about special
+	 * over-full considerations.
+	 */
+	return tx - rx;
+}
+
+static inline void tegra_ivc_advance_tx(struct tegra_ivc *ivc)
+{
+	unsigned int count = tegra_ivc_header_read_field(&ivc->tx.map, tx.count);
+
+	tegra_ivc_header_write_field(&ivc->tx.map, tx.count, count + 1);
+
+	if (ivc->tx.position == ivc->num_frames - 1)
+		ivc->tx.position = 0;
+	else
+		ivc->tx.position++;
+}
+
+static inline void tegra_ivc_advance_rx(struct tegra_ivc *ivc)
+{
+	unsigned int count = tegra_ivc_header_read_field(&ivc->rx.map, rx.count);
+
+	tegra_ivc_header_write_field(&ivc->rx.map, rx.count, count + 1);
+
+	if (ivc->rx.position == ivc->num_frames - 1)
+		ivc->rx.position = 0;
+	else
+		ivc->rx.position++;
+}
+
+static inline int tegra_ivc_check_read(struct tegra_ivc *ivc)
+{
+	unsigned int offset = offsetof(struct tegra_ivc_header, tx.count);
+	unsigned int state;
+
+	/*
+	 * tx.channel->state is set locally, so it is not synchronized with
+	 * state from the remote peer. The remote peer cannot reset its
+	 * transmit counters until we've acknowledged its synchronization
+	 * request, so no additional synchronization is required because an
+	 * asynchronous transition of rx.channel->state to
+	 * TEGRA_IVC_STATE_ACK is not allowed.
+	 */
+	state = tegra_ivc_header_read_field(&ivc->tx.map, tx.state);
+	if (state != TEGRA_IVC_STATE_ESTABLISHED)
+		return -ECONNRESET;
+
+	/*
+	 * Avoid unnecessary invalidations when performing repeated accesses
+	 * to an IVC channel by checking the old queue pointers first.
+	 *
+	 * Synchronization is only necessary when these pointers indicate
+	 * empty or full.
+	 */
+	if (!tegra_ivc_empty(ivc, &ivc->rx.map))
+		return 0;
+
+	tegra_ivc_invalidate(ivc, ivc->rx.phys + offset);
+
+	if (tegra_ivc_empty(ivc, &ivc->rx.map))
+		return -ENOSPC;
+
+	return 0;
+}
+
+static inline int tegra_ivc_check_write(struct tegra_ivc *ivc)
+{
+	unsigned int offset = offsetof(struct tegra_ivc_header, rx.count);
+	unsigned int state;
+
+	state = tegra_ivc_header_read_field(&ivc->tx.map, tx.state);
+	if (state != TEGRA_IVC_STATE_ESTABLISHED)
+		return -ECONNRESET;
+
+	if (!tegra_ivc_full(ivc, &ivc->tx.map))
+		return 0;
+
+	tegra_ivc_invalidate(ivc, ivc->tx.phys + offset);
+
+	if (tegra_ivc_full(ivc, &ivc->tx.map))
+		return -ENOSPC;
+
+	return 0;
+}
+
+static int tegra_ivc_frame_virt(struct tegra_ivc *ivc, const struct iosys_map *header,
+				unsigned int frame, struct iosys_map *map)
+{
+	size_t offset = sizeof(struct tegra_ivc_header) + ivc->frame_size * frame;
+
+	if (WARN_ON(frame >= ivc->num_frames))
+		return -EINVAL;
+
+	*map = IOSYS_MAP_INIT_OFFSET(header, offset);
+
+	return 0;
+}
+
+static inline dma_addr_t tegra_ivc_frame_phys(struct tegra_ivc *ivc,
+					      dma_addr_t phys,
+					      unsigned int frame)
+{
+	unsigned long offset;
+
+	offset = sizeof(struct tegra_ivc_header) + ivc->frame_size * frame;
+
+	return phys + offset;
+}
+
+static inline void tegra_ivc_invalidate_frame(struct tegra_ivc *ivc,
+					      dma_addr_t phys,
+					      unsigned int frame,
+					      unsigned int offset,
+					      size_t size)
+{
+	if (!ivc->peer || WARN_ON(frame >= ivc->num_frames))
+		return;
+
+	phys = tegra_ivc_frame_phys(ivc, phys, frame) + offset;
+
+	dma_sync_single_for_cpu(ivc->peer, phys, size, DMA_FROM_DEVICE);
+}
+
+static inline void tegra_ivc_flush_frame(struct tegra_ivc *ivc,
+					 dma_addr_t phys,
+					 unsigned int frame,
+					 unsigned int offset,
+					 size_t size)
+{
+	if (!ivc->peer || WARN_ON(frame >= ivc->num_frames))
+		return;
+
+	phys = tegra_ivc_frame_phys(ivc, phys, frame) + offset;
+
+	dma_sync_single_for_device(ivc->peer, phys, size, DMA_TO_DEVICE);
+}
+
+/* directly peek at the next frame rx'ed */
+int tegra_ivc_read_get_next_frame(struct tegra_ivc *ivc, struct iosys_map *map)
+{
+	int err;
+
+	if (WARN_ON(ivc == NULL))
+		return -EINVAL;
+
+	err = tegra_ivc_check_read(ivc);
+	if (err < 0)
+		return err;
+
+	/*
+	 * Order observation of ivc->rx.position potentially indicating new
+	 * data before data read.
+	 */
+	smp_rmb();
+
+	tegra_ivc_invalidate_frame(ivc, ivc->rx.phys, ivc->rx.position, 0,
+				   ivc->frame_size);
+
+	return tegra_ivc_frame_virt(ivc, &ivc->rx.map, ivc->rx.position, map);
+}
+EXPORT_SYMBOL(tegra_ivc_read_get_next_frame);
+
+int tegra_ivc_read_advance(struct tegra_ivc *ivc)
+{
+	unsigned int rx = offsetof(struct tegra_ivc_header, rx.count);
+	unsigned int tx = offsetof(struct tegra_ivc_header, tx.count);
+	int err;
+
+	/*
+	 * No read barriers or synchronization here: the caller is expected to
+	 * have already observed the channel non-empty. This check is just to
+	 * catch programming errors.
+	 */
+	err = tegra_ivc_check_read(ivc);
+	if (err < 0)
+		return err;
+
+	tegra_ivc_advance_rx(ivc);
+
+	tegra_ivc_flush(ivc, ivc->rx.phys + rx);
+
+	/*
+	 * Ensure our write to ivc->rx.position occurs before our read from
+	 * ivc->tx.position.
+	 */
+	smp_mb();
+
+	/*
+	 * Notify only upon transition from full to non-full. The available
+	 * count can only asynchronously increase, so the worst possible
+	 * side-effect will be a spurious notification.
+	 */
+	tegra_ivc_invalidate(ivc, ivc->rx.phys + tx);
+
+	if (tegra_ivc_available(ivc, &ivc->rx.map) == ivc->num_frames - 1)
+		ivc->notify(ivc, ivc->notify_data);
+
+	return 0;
+}
+EXPORT_SYMBOL(tegra_ivc_read_advance);
+
+/* directly poke at the next frame to be tx'ed */
+int tegra_ivc_write_get_next_frame(struct tegra_ivc *ivc, struct iosys_map *map)
+{
+	int err;
+
+	err = tegra_ivc_check_write(ivc);
+	if (err < 0)
+		return err;
+
+	return tegra_ivc_frame_virt(ivc, &ivc->tx.map, ivc->tx.position, map);
+}
+EXPORT_SYMBOL(tegra_ivc_write_get_next_frame);
+
+/* advance the tx buffer */
+int tegra_ivc_write_advance(struct tegra_ivc *ivc)
+{
+	unsigned int tx = offsetof(struct tegra_ivc_header, tx.count);
+	unsigned int rx = offsetof(struct tegra_ivc_header, rx.count);
+	int err;
+
+	err = tegra_ivc_check_write(ivc);
+	if (err < 0)
+		return err;
+
+	tegra_ivc_flush_frame(ivc, ivc->tx.phys, ivc->tx.position, 0,
+			      ivc->frame_size);
+
+	/*
+	 * Order any possible stores to the frame before update of
+	 * ivc->tx.position.
+	 */
+	smp_wmb();
+
+	tegra_ivc_advance_tx(ivc);
+	tegra_ivc_flush(ivc, ivc->tx.phys + tx);
+
+	/*
+	 * Ensure our write to ivc->tx.position occurs before our read from
+	 * ivc->rx.position.
+	 */
+	smp_mb();
+
+	/*
+	 * Notify only upon transition from empty to non-empty. The available
+	 * count can only asynchronously decrease, so the worst possible
+	 * side-effect will be a spurious notification.
+	 */
+	tegra_ivc_invalidate(ivc, ivc->tx.phys + rx);
+
+	if (tegra_ivc_available(ivc, &ivc->tx.map) == 1)
+		ivc->notify(ivc, ivc->notify_data);
+
+	return 0;
+}
+EXPORT_SYMBOL(tegra_ivc_write_advance);
+
+void tegra_ivc_reset(struct tegra_ivc *ivc)
+{
+	unsigned int offset = offsetof(struct tegra_ivc_header, tx.count);
+
+	tegra_ivc_header_write_field(&ivc->tx.map, tx.state, TEGRA_IVC_STATE_SYNC);
+	tegra_ivc_flush(ivc, ivc->tx.phys + offset);
+	ivc->notify(ivc, ivc->notify_data);
+}
+EXPORT_SYMBOL(tegra_ivc_reset);
+
+/*
+ * =======================================================
+ *  IVC State Transition Table - see tegra_ivc_notified()
+ * =======================================================
+ *
+ *	local	remote	action
+ *	-----	------	-----------------------------------
+ *	SYNC	EST	<none>
+ *	SYNC	ACK	reset counters; move to EST; notify
+ *	SYNC	SYNC	reset counters; move to ACK; notify
+ *	ACK	EST	move to EST; notify
+ *	ACK	ACK	move to EST; notify
+ *	ACK	SYNC	reset counters; move to ACK; notify
+ *	EST	EST	<none>
+ *	EST	ACK	<none>
+ *	EST	SYNC	reset counters; move to ACK; notify
+ *
+ * ===============================================================
+ */
+
+int tegra_ivc_notified(struct tegra_ivc *ivc)
+{
+	unsigned int offset = offsetof(struct tegra_ivc_header, tx.count);
+	enum tegra_ivc_state rx_state, tx_state;
+
+	/* Copy the receiver's state out of shared memory. */
+	tegra_ivc_invalidate(ivc, ivc->rx.phys + offset);
+	rx_state = tegra_ivc_header_read_field(&ivc->rx.map, tx.state);
+	tx_state = tegra_ivc_header_read_field(&ivc->tx.map, tx.state);
+
+	if (rx_state == TEGRA_IVC_STATE_SYNC) {
+		offset = offsetof(struct tegra_ivc_header, tx.count);
+
+		/*
+		 * Order observation of TEGRA_IVC_STATE_SYNC before stores
+		 * clearing tx.channel.
+		 */
+		smp_rmb();
+
+		/*
+		 * Reset tx.channel counters. The remote end is in the SYNC
+		 * state and won't make progress until we change our state,
+		 * so the counters are not in use at this time.
+		 */
+		tegra_ivc_header_write_field(&ivc->tx.map, tx.count, 0);
+		tegra_ivc_header_write_field(&ivc->rx.map, rx.count, 0);
+
+		ivc->tx.position = 0;
+		ivc->rx.position = 0;
+
+		/*
+		 * Ensure that counters appear cleared before new state can be
+		 * observed.
+		 */
+		smp_wmb();
+
+		/*
+		 * Move to ACK state. We have just cleared our counters, so it
+		 * is now safe for the remote end to start using these values.
+		 */
+		tegra_ivc_header_write_field(&ivc->tx.map, tx.state, TEGRA_IVC_STATE_ACK);
+		tegra_ivc_flush(ivc, ivc->tx.phys + offset);
+
+		/*
+		 * Notify remote end to observe state transition.
+		 */
+		ivc->notify(ivc, ivc->notify_data);
+
+	} else if (tx_state == TEGRA_IVC_STATE_SYNC &&
+		   rx_state == TEGRA_IVC_STATE_ACK) {
+		offset = offsetof(struct tegra_ivc_header, tx.count);
+
+		/*
+		 * Order observation of ivc_state_sync before stores clearing
+		 * tx_channel.
+		 */
+		smp_rmb();
+
+		/*
+		 * Reset tx.channel counters. The remote end is in the ACK
+		 * state and won't make progress until we change our state,
+		 * so the counters are not in use at this time.
+		 */
+		tegra_ivc_header_write_field(&ivc->tx.map, tx.count, 0);
+		tegra_ivc_header_write_field(&ivc->rx.map, rx.count, 0);
+
+		ivc->tx.position = 0;
+		ivc->rx.position = 0;
+
+		/*
+		 * Ensure that counters appear cleared before new state can be
+		 * observed.
+		 */
+		smp_wmb();
+
+		/*
+		 * Move to ESTABLISHED state. We know that the remote end has
+		 * already cleared its counters, so it is safe to start
+		 * writing/reading on this channel.
+		 */
+		tegra_ivc_header_write_field(&ivc->tx.map, tx.state, TEGRA_IVC_STATE_ESTABLISHED);
+		tegra_ivc_flush(ivc, ivc->tx.phys + offset);
+
+		/*
+		 * Notify remote end to observe state transition.
+		 */
+		ivc->notify(ivc, ivc->notify_data);
+
+	} else if (tx_state == TEGRA_IVC_STATE_ACK) {
+		offset = offsetof(struct tegra_ivc_header, tx.count);
+
+		/*
+		 * At this point, we have observed the peer to be in either
+		 * the ACK or ESTABLISHED state. Next, order observation of
+		 * peer state before storing to tx.channel.
+		 */
+		smp_rmb();
+
+		/*
+		 * Move to ESTABLISHED state. We know that we have previously
+		 * cleared our counters, and we know that the remote end has
+		 * cleared its counters, so it is safe to start writing/reading
+		 * on this channel.
+		 */
+		tegra_ivc_header_write_field(&ivc->tx.map, tx.state, TEGRA_IVC_STATE_ESTABLISHED);
+		tegra_ivc_flush(ivc, ivc->tx.phys + offset);
+
+		/*
+		 * Notify remote end to observe state transition.
+		 */
+		ivc->notify(ivc, ivc->notify_data);
+
+	} else {
+		/*
+		 * There is no need to handle any further action. Either the
+		 * channel is already fully established, or we are waiting for
+		 * the remote end to catch up with our current state. Refer
+		 * to the diagram in "IVC State Transition Table" above.
+		 */
+	}
+
+	if (tx_state != TEGRA_IVC_STATE_ESTABLISHED)
+		return -EAGAIN;
+
+	return 0;
+}
+EXPORT_SYMBOL(tegra_ivc_notified);
+
+size_t tegra_ivc_align(size_t size)
+{
+	return ALIGN(size, TEGRA_IVC_ALIGN);
+}
+EXPORT_SYMBOL(tegra_ivc_align);
+
+unsigned tegra_ivc_total_queue_size(unsigned queue_size)
+{
+	if (!IS_ALIGNED(queue_size, TEGRA_IVC_ALIGN)) {
+		pr_err("%s: queue_size (%u) must be %u-byte aligned\n",
+		       __func__, queue_size, TEGRA_IVC_ALIGN);
+		return 0;
+	}
+
+	return queue_size + sizeof(struct tegra_ivc_header);
+}
+EXPORT_SYMBOL(tegra_ivc_total_queue_size);
+
+static int tegra_ivc_check_params(unsigned long rx, unsigned long tx,
+				  unsigned int num_frames, size_t frame_size)
+{
+	BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct tegra_ivc_header, tx.count),
+				 TEGRA_IVC_ALIGN));
+	BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct tegra_ivc_header, rx.count),
+				 TEGRA_IVC_ALIGN));
+	BUILD_BUG_ON(!IS_ALIGNED(sizeof(struct tegra_ivc_header),
+				 TEGRA_IVC_ALIGN));
+
+	if ((uint64_t)num_frames * (uint64_t)frame_size >= 0x100000000UL) {
+		pr_err("num_frames * frame_size overflows\n");
+		return -EINVAL;
+	}
+
+	if (!IS_ALIGNED(frame_size, TEGRA_IVC_ALIGN)) {
+		pr_err("frame size not adequately aligned: %zu\n", frame_size);
+		return -EINVAL;
+	}
+
+	/*
+	 * The headers must at least be aligned enough for counters
+	 * to be accessed atomically.
+	 */
+	if (!IS_ALIGNED(rx, TEGRA_IVC_ALIGN)) {
+		pr_err("IVC channel start not aligned: %#lx\n", rx);
+		return -EINVAL;
+	}
+
+	if (!IS_ALIGNED(tx, TEGRA_IVC_ALIGN)) {
+		pr_err("IVC channel start not aligned: %#lx\n", tx);
+		return -EINVAL;
+	}
+
+	if (rx < tx) {
+		if (rx + frame_size * num_frames > tx) {
+			pr_err("queue regions overlap: %#lx + %zx > %#lx\n",
+			       rx, frame_size * num_frames, tx);
+			return -EINVAL;
+		}
+	} else {
+		if (tx + frame_size * num_frames > rx) {
+			pr_err("queue regions overlap: %#lx + %zx > %#lx\n",
+			       tx, frame_size * num_frames, rx);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static inline void iosys_map_copy(struct iosys_map *dst, const struct iosys_map *src)
+{
+	*dst = *src;
+}
+
+static inline unsigned long iosys_map_get_address(const struct iosys_map *map)
+{
+	if (map->is_iomem)
+		return (unsigned long)map->vaddr_iomem;
+
+	return (unsigned long)map->vaddr;
+}
+
+static inline void *iosys_map_get_vaddr(const struct iosys_map *map)
+{
+	if (WARN_ON(map->is_iomem))
+		return NULL;
+
+	return map->vaddr;
+}
+
+int tegra_ivc_init(struct tegra_ivc *ivc, struct device *peer, const struct iosys_map *rx,
+		   dma_addr_t rx_phys, const struct iosys_map *tx, dma_addr_t tx_phys,
+		   unsigned int num_frames, size_t frame_size,
+		   void (*notify)(struct tegra_ivc *ivc, void *data),
+		   void *data)
+{
+	size_t queue_size;
+	int err;
+
+	if (WARN_ON(!ivc || !notify))
+		return -EINVAL;
+
+	/*
+	 * All sizes that can be returned by communication functions should
+	 * fit in an int.
+	 */
+	if (frame_size > INT_MAX)
+		return -E2BIG;
+
+	err = tegra_ivc_check_params(iosys_map_get_address(rx), iosys_map_get_address(tx),
+				     num_frames, frame_size);
+	if (err < 0)
+		return err;
+
+	queue_size = tegra_ivc_total_queue_size(num_frames * frame_size);
+
+	if (peer) {
+		ivc->rx.phys = dma_map_single(peer, iosys_map_get_vaddr(rx), queue_size,
+					      DMA_BIDIRECTIONAL);
+		if (dma_mapping_error(peer, ivc->rx.phys))
+			return -ENOMEM;
+
+		ivc->tx.phys = dma_map_single(peer, iosys_map_get_vaddr(tx), queue_size,
+					      DMA_BIDIRECTIONAL);
+		if (dma_mapping_error(peer, ivc->tx.phys)) {
+			dma_unmap_single(peer, ivc->rx.phys, queue_size,
+					 DMA_BIDIRECTIONAL);
+			return -ENOMEM;
+		}
+	} else {
+		ivc->rx.phys = rx_phys;
+		ivc->tx.phys = tx_phys;
+	}
+
+	iosys_map_copy(&ivc->rx.map, rx);
+	iosys_map_copy(&ivc->tx.map, tx);
+	ivc->peer = peer;
+	ivc->notify = notify;
+	ivc->notify_data = data;
+	ivc->frame_size = frame_size;
+	ivc->num_frames = num_frames;
+
+	/*
+	 * These values aren't necessarily correct until the channel has been
+	 * reset.
+	 */
+	ivc->tx.position = 0;
+	ivc->rx.position = 0;
+
+	return 0;
+}
+EXPORT_SYMBOL(tegra_ivc_init);
+
+void tegra_ivc_cleanup(struct tegra_ivc *ivc)
+{
+	if (ivc->peer) {
+		size_t size = tegra_ivc_total_queue_size(ivc->num_frames *
+							 ivc->frame_size);
+
+		dma_unmap_single(ivc->peer, ivc->rx.phys, size,
+				 DMA_BIDIRECTIONAL);
+		dma_unmap_single(ivc->peer, ivc->tx.phys, size,
+				 DMA_BIDIRECTIONAL);
+	}
+}
+EXPORT_SYMBOL(tegra_ivc_cleanup);