Resguardo inicial

linux-7.2.3/arch/um/drivers/Kconfig

@@ -0,0 +1,190 @@
+# SPDX-License-Identifier: GPL-2.0
+
+menu "UML Character Devices"
+
+config STDERR_CONSOLE
+	bool "stderr console"
+	default y
+	help
+	  console driver which dumps all printk messages to stderr.
+
+config SSL
+	bool "Virtual serial line"
+	help
+	  The User-Mode Linux environment allows you to create virtual serial
+	  lines on the UML that are usually made to show up on the host as
+	  ttys or ptys.
+
+	  See <http://user-mode-linux.sourceforge.net/old/input.html> for more
+	  information and command line examples of how to use this facility.
+
+	  Unless you have a specific reason for disabling this, say Y.
+
+config NULL_CHAN
+	bool "null channel support"
+	help
+	  This option enables support for attaching UML consoles and serial
+	  lines to a device similar to /dev/null.  Data written to it disappears
+	  and there is never any data to be read.
+
+config PORT_CHAN
+	bool "port channel support"
+	help
+	  This option enables support for attaching UML consoles and serial
+	  lines to host portals.  They may be accessed with 'telnet <host>
+	  <port number>'.  Any number of consoles and serial lines may be
+	  attached to a single portal, although what UML device you get when
+	  you telnet to that portal will be unpredictable.
+	  It is safe to say 'Y' here.
+
+config PTY_CHAN
+	bool "pty channel support"
+	help
+	  This option enables support for attaching UML consoles and serial
+	  lines to host pseudo-terminals.  Access to both traditional
+	  pseudo-terminals (/dev/pty*) and pts pseudo-terminals are controlled
+	  with this option.  The assignment of UML devices to host devices
+	  will be announced in the kernel message log.
+	  It is safe to say 'Y' here.
+
+config TTY_CHAN
+	bool "tty channel support"
+	help
+	  This option enables support for attaching UML consoles and serial
+	  lines to host terminals.  Access to both virtual consoles
+	  (/dev/tty*) and the slave side of pseudo-terminals (/dev/ttyp* and
+	  /dev/pts/*) are controlled by this option.
+	  It is safe to say 'Y' here.
+
+config XTERM_CHAN
+	bool "xterm channel support"
+	help
+	  This option enables support for attaching UML consoles and serial
+	  lines to xterms.  Each UML device so assigned will be brought up in
+	  its own xterm.
+	  It is safe to say 'Y' here.
+
+config XTERM_CHAN_DEFAULT_EMULATOR
+	string "xterm channel default terminal emulator"
+	depends on XTERM_CHAN
+	default "xterm"
+	help
+	  This option allows changing the default terminal emulator.
+
+config NOCONFIG_CHAN
+	bool
+	default !(XTERM_CHAN && TTY_CHAN && PTY_CHAN && PORT_CHAN && NULL_CHAN)
+
+config CON_ZERO_CHAN
+	string "Default main console channel initialization"
+	default "fd:0,fd:1"
+	help
+	  This is the string describing the channel to which the main console
+	  will be attached by default.  This value can be overridden from the
+	  command line.  The default value is "fd:0,fd:1", which attaches the
+	  main console to stdin and stdout.
+	  It is safe to leave this unchanged.
+
+config CON_CHAN
+	string "Default console channel initialization"
+	default "xterm"
+	help
+	  This is the string describing the channel to which all consoles
+	  except the main console will be attached by default.  This value can
+	  be overridden from the command line.  The default value is "xterm",
+	  which brings them up in xterms.
+	  It is safe to leave this unchanged, although you may wish to change
+	  this if you expect the UML that you build to be run in environments
+	  which don't have X or xterm available.
+
+config SSL_CHAN
+	string "Default serial line channel initialization"
+	default "pty"
+	help
+	  This is the string describing the channel to which the serial lines
+	  will be attached by default.  This value can be overridden from the
+	  command line.  The default value is "pty", which attaches them to
+	  traditional pseudo-terminals.
+	  It is safe to leave this unchanged, although you may wish to change
+	  this if you expect the UML that you build to be run in environments
+	  which don't have a set of /dev/pty* devices.
+
+config UML_SOUND
+	tristate "Sound support"
+	depends on SOUND
+	select SOUND_OSS_CORE
+	help
+	  This option enables UML sound support.  If enabled, it will pull in
+	  the UML hostaudio relay, which acts as a intermediary
+	  between the host's dsp and mixer devices and the UML sound system.
+	  It is safe to say 'Y' here.
+
+endmenu
+
+menu "UML Network Devices"
+	depends on NET
+
+config UML_NET_VECTOR
+	bool "Vector I/O high performance network devices"
+	select MAY_HAVE_RUNTIME_DEPS
+	help
+	  This User-Mode Linux network driver uses multi-message send
+	  and receive functions. The host running the UML guest must have
+	  a linux kernel version above 3.0 and a libc version > 2.13.
+	  This driver provides tap, raw, gre and l2tpv3 network transports.
+
+	  For more information, including explanations of the networking
+	  and sample configurations, see
+	  <file:Documentation/virt/uml/user_mode_linux_howto_v2.rst>.
+
+endmenu
+
+config VIRTIO_UML
+	bool "UML driver for virtio devices"
+	select VIRTIO
+	help
+	  This driver provides support for virtio based paravirtual device
+	  drivers over vhost-user sockets.
+
+config UML_RTC
+	bool "UML RTC driver"
+	depends on RTC_CLASS
+	# there's no use in this if PM_SLEEP isn't enabled ...
+	depends on PM_SLEEP
+	help
+	  When PM_SLEEP is configured, it may be desirable to wake up using
+	  rtcwake, especially in time-travel mode. This driver enables that
+	  by providing a fake RTC clock that causes a wakeup at the right
+	  time.
+
+config UML_PCI
+	bool
+	select FORCE_PCI
+	select IRQ_MSI_LIB
+	select UML_IOMEM_EMULATION
+	select UML_DMA_EMULATION
+	select PCI_MSI
+	select PCI_LOCKLESS_CONFIG
+
+config UML_PCI_OVER_VIRTIO
+	bool "Enable PCI over VIRTIO device simulation"
+	# in theory, just VIRTIO is enough, but that causes recursion
+	depends on VIRTIO_UML
+	select UML_PCI
+
+config UML_PCI_OVER_VIRTIO_DEVICE_ID
+	int "set the virtio device ID for PCI emulation"
+	default -1
+	depends on UML_PCI_OVER_VIRTIO
+	help
+	  There's no official device ID assigned (yet), set the one you
+	  wish to use for experimentation here. The default of -1 is
+	  not valid and will cause the driver to fail at probe.
+
+config UML_PCI_OVER_VFIO
+	bool "Enable VFIO-based PCI passthrough"
+	select UML_PCI
+	help
+	  This driver provides support for VFIO-based PCI passthrough.
+	  Currently, only MSI-X capable devices are supported, and it
+	  is assumed that drivers will use MSI-X.

linux-7.2.3/arch/um/drivers/Makefile

@@ -0,0 +1,55 @@
+# SPDX-License-Identifier: GPL-2.0
+# 
+# Copyright (C) 2000, 2002, 2003 Jeff Dike (jdike@karaya.com)
+#
+
+# pcap is broken in 2.5 because kbuild doesn't allow pcap.a to be linked
+# in to pcap.o
+
+vector-objs := vector_kern.o vector_user.o vector_transports.o
+mconsole-objs := mconsole_kern.o mconsole_user.o
+hostaudio-objs := hostaudio_kern.o
+ubd-objs := ubd_kern.o ubd_user.o
+port-objs := port_kern.o port_user.o
+harddog-objs := harddog_kern.o
+harddog-builtin-$(CONFIG_UML_WATCHDOG) := harddog_user.o harddog_user_exp.o
+rtc-objs := rtc_kern.o rtc_user.o
+vfio_uml-objs := vfio_kern.o vfio_user.o
+
+#XXX: The call below does not work because the flags are added before the
+# object name, so nothing from the library gets linked.
+#$(call if_changed,ld)
+
+# When the above is fixed, don't forget to add this too!
+#targets += $(obj)/pcap.o
+
+obj-y := stdio_console.o fd.o chan_kern.o chan_user.o line.o
+obj-$(CONFIG_SSL) += ssl.o
+obj-$(CONFIG_STDERR_CONSOLE) += stderr_console.o
+
+obj-$(CONFIG_UML_NET_VECTOR) += vector.o
+obj-$(CONFIG_MCONSOLE) += mconsole.o
+obj-$(CONFIG_BLK_DEV_UBD) += ubd.o 
+obj-$(CONFIG_UML_SOUND) += hostaudio.o
+obj-$(CONFIG_NULL_CHAN) += null.o 
+obj-$(CONFIG_PORT_CHAN) += port.o
+obj-$(CONFIG_PTY_CHAN) += pty.o
+obj-$(CONFIG_TTY_CHAN) += tty.o 
+obj-$(CONFIG_XTERM_CHAN) += xterm.o xterm_kern.o
+obj-$(CONFIG_UML_WATCHDOG) += harddog.o
+obj-y += $(harddog-builtin-y) $(harddog-builtin-m)
+obj-$(CONFIG_BLK_DEV_COW_COMMON) += cow_user.o
+obj-$(CONFIG_UML_RANDOM) += random.o
+obj-$(CONFIG_VIRTIO_UML) += virtio_uml.o
+obj-$(CONFIG_UML_RTC) += rtc.o
+obj-$(CONFIG_UML_PCI) += virt-pci.o
+obj-$(CONFIG_UML_PCI_OVER_VIRTIO) += virtio_pcidev.o
+obj-$(CONFIG_UML_PCI_OVER_VFIO) += vfio_uml.o
+
+# pcap_user.o must be added explicitly.
+USER_OBJS := fd.o null.o pty.o tty.o xterm.o vector_user.o
+CFLAGS_null.o = -DDEV_NULL=$(DEV_NULL_PATH)
+
+CFLAGS_xterm.o += '-DCONFIG_XTERM_CHAN_DEFAULT_EMULATOR="$(CONFIG_XTERM_CHAN_DEFAULT_EMULATOR)"'
+
+include $(srctree)/arch/um/scripts/Makefile.rules

linux-7.2.3/arch/um/drivers/chan.h

@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __CHAN_KERN_H__
+#define __CHAN_KERN_H__
+
+#include <linux/tty.h>
+#include <linux/list.h>
+#include <linux/console.h>
+#include "chan_user.h"
+#include "line.h"
+
+struct chan {
+	struct list_head list;
+	struct list_head free_list;
+	struct line *line;
+	char *dev;
+	unsigned int primary:1;
+	unsigned int input:1;
+	unsigned int output:1;
+	unsigned int opened:1;
+	unsigned int enabled:1;
+	int fd_in;
+	int fd_out; /* only different to fd_in if blocking output is needed */
+	const struct chan_ops *ops;
+	void *data;
+};
+
+extern void chan_interrupt(struct line *line, int irq);
+extern int parse_chan_pair(char *str, struct line *line, int device,
+			   const struct chan_opts *opts, char **error_out);
+extern int write_chan(struct chan *chan, const u8 *buf, size_t len,
+			     int write_irq);
+extern int console_write_chan(struct chan *chan, const char *buf, 
+			      int len);
+extern int console_open_chan(struct line *line, struct console *co);
+extern void deactivate_chan(struct chan *chan, int irq);
+extern void chan_enable_winch(struct chan *chan, struct tty_port *port);
+extern int enable_chan(struct line *line);
+extern void close_chan(struct line *line);
+extern int chan_window_size(struct line *line, 
+			     unsigned short *rows_out, 
+			     unsigned short *cols_out);
+extern int chan_config_string(struct line *line, char *str, int size,
+			      char **error_out);
+
+#endif

linux-7.2.3/arch/um/drivers/chan_kern.c

@@ -0,0 +1,610 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <linux/slab.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include "chan.h"
+#include <os.h>
+#include <irq_kern.h>
+
+#ifdef CONFIG_NOCONFIG_CHAN
+static void *not_configged_init(char *str, int device,
+				const struct chan_opts *opts)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return NULL;
+}
+
+static int not_configged_open(int input, int output, int primary, void *data,
+			      char **dev_out)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return -ENODEV;
+}
+
+static void not_configged_close(int fd, void *data)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+}
+
+static int not_configged_read(int fd, u8 *c_out, void *data)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return -EIO;
+}
+
+static int not_configged_write(int fd, const u8 *buf, size_t len, void *data)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return -EIO;
+}
+
+static int not_configged_console_write(int fd, const char *buf, int len)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return -EIO;
+}
+
+static int not_configged_window_size(int fd, void *data, unsigned short *rows,
+				     unsigned short *cols)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+	return -ENODEV;
+}
+
+static void not_configged_free(void *data)
+{
+	printk(KERN_ERR "Using a channel type which is configured out of "
+	       "UML\n");
+}
+
+static const struct chan_ops not_configged_ops = {
+	.init		= not_configged_init,
+	.open		= not_configged_open,
+	.close		= not_configged_close,
+	.read		= not_configged_read,
+	.write		= not_configged_write,
+	.console_write	= not_configged_console_write,
+	.window_size	= not_configged_window_size,
+	.free		= not_configged_free,
+	.winch		= 0,
+};
+#endif /* CONFIG_NOCONFIG_CHAN */
+
+static inline bool need_output_blocking(void)
+{
+	return time_travel_mode == TT_MODE_INFCPU ||
+	       time_travel_mode == TT_MODE_EXTERNAL;
+}
+
+static int open_one_chan(struct chan *chan)
+{
+	int fd, err;
+
+	if (chan->opened)
+		return 0;
+
+	if (chan->ops->open == NULL)
+		fd = 0;
+	else fd = (*chan->ops->open)(chan->input, chan->output, chan->primary,
+				     chan->data, &chan->dev);
+	if (fd < 0)
+		return fd;
+
+	err = os_set_fd_block(fd, 0);
+	if (err)
+		goto out_close;
+
+	chan->fd_in = fd;
+	chan->fd_out = fd;
+
+	/*
+	 * In time-travel modes infinite-CPU and external we need to guarantee
+	 * that any writes to the output succeed immdiately from the point of
+	 * the VM. The best way to do this is to put the FD in blocking mode
+	 * and simply wait/retry until everything is written.
+	 * As every write is guaranteed to complete, we also do not need to
+	 * request an IRQ for the output.
+	 *
+	 * Note that input cannot happen in a time synchronized way. We permit
+	 * it, but time passes very quickly if anything waits for a read.
+	 */
+	if (chan->output && need_output_blocking()) {
+		err = os_dup_file(chan->fd_out);
+		if (err < 0)
+			goto out_close;
+
+		chan->fd_out = err;
+
+		err = os_set_fd_block(chan->fd_out, 1);
+		if (err) {
+			os_close_file(chan->fd_out);
+			goto out_close;
+		}
+	}
+
+	chan->opened = 1;
+	return 0;
+
+out_close:
+	(*chan->ops->close)(fd, chan->data);
+	return err;
+}
+
+static int open_chan(struct list_head *chans)
+{
+	struct list_head *ele;
+	struct chan *chan;
+	int ret, err = 0;
+
+	list_for_each(ele, chans) {
+		chan = list_entry(ele, struct chan, list);
+		ret = open_one_chan(chan);
+		if (chan->primary)
+			err = ret;
+	}
+	return err;
+}
+
+void chan_enable_winch(struct chan *chan, struct tty_port *port)
+{
+	if (chan && chan->primary && chan->ops->winch)
+		register_winch(chan->fd_in, port);
+}
+
+static void line_timer_cb(struct work_struct *work)
+{
+	struct line *line = container_of(work, struct line, task.work);
+
+	if (!line->throttled)
+		chan_interrupt(line, line->read_irq);
+}
+
+int enable_chan(struct line *line)
+{
+	struct list_head *ele;
+	struct chan *chan;
+	int err;
+
+	INIT_DELAYED_WORK(&line->task, line_timer_cb);
+
+	list_for_each(ele, &line->chan_list) {
+		chan = list_entry(ele, struct chan, list);
+		err = open_one_chan(chan);
+		if (err) {
+			if (chan->primary)
+				goto out_close;
+
+			continue;
+		}
+
+		if (chan->enabled)
+			continue;
+		err = line_setup_irq(chan->fd_in, chan->input,
+				     chan->output && !need_output_blocking(),
+				     line, chan);
+		if (err)
+			goto out_close;
+
+		chan->enabled = 1;
+	}
+
+	return 0;
+
+ out_close:
+	close_chan(line);
+	return err;
+}
+
+/* Items are added in IRQ context, when free_irq can't be called, and
+ * removed in process context, when it can.
+ * This handles interrupt sources which disappear, and which need to
+ * be permanently disabled.  This is discovered in IRQ context, but
+ * the freeing of the IRQ must be done later.
+ */
+static DEFINE_RAW_SPINLOCK(irqs_to_free_lock);
+static LIST_HEAD(irqs_to_free);
+
+void free_irqs(void)
+{
+	struct chan *chan;
+	LIST_HEAD(list);
+	struct list_head *ele;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&irqs_to_free_lock, flags);
+	list_splice_init(&irqs_to_free, &list);
+	raw_spin_unlock_irqrestore(&irqs_to_free_lock, flags);
+
+	list_for_each(ele, &list) {
+		chan = list_entry(ele, struct chan, free_list);
+
+		if (chan->input && chan->enabled)
+			um_free_irq(chan->line->read_irq, chan);
+		if (chan->output && chan->enabled &&
+		    !need_output_blocking())
+			um_free_irq(chan->line->write_irq, chan);
+		chan->enabled = 0;
+	}
+}
+
+static void close_one_chan(struct chan *chan, int delay_free_irq)
+{
+	unsigned long flags;
+
+	if (!chan->opened)
+		return;
+
+	if (delay_free_irq) {
+		raw_spin_lock_irqsave(&irqs_to_free_lock, flags);
+		list_add(&chan->free_list, &irqs_to_free);
+		raw_spin_unlock_irqrestore(&irqs_to_free_lock, flags);
+	} else {
+		if (chan->input && chan->enabled)
+			um_free_irq(chan->line->read_irq, chan);
+		if (chan->output && chan->enabled &&
+		    !need_output_blocking())
+			um_free_irq(chan->line->write_irq, chan);
+		chan->enabled = 0;
+	}
+	if (chan->fd_out != chan->fd_in)
+		os_close_file(chan->fd_out);
+	if (chan->ops->close != NULL)
+		(*chan->ops->close)(chan->fd_in, chan->data);
+
+	chan->opened = 0;
+	chan->fd_in = -1;
+	chan->fd_out = -1;
+}
+
+void close_chan(struct line *line)
+{
+	struct chan *chan;
+
+	/* Close in reverse order as open in case more than one of them
+	 * refers to the same device and they save and restore that device's
+	 * state.  Then, the first one opened will have the original state,
+	 * so it must be the last closed.
+	 */
+	list_for_each_entry_reverse(chan, &line->chan_list, list) {
+		close_one_chan(chan, 0);
+	}
+}
+
+void deactivate_chan(struct chan *chan, int irq)
+{
+	if (chan && chan->enabled)
+		deactivate_fd(chan->fd_in, irq);
+}
+
+int write_chan(struct chan *chan, const u8 *buf, size_t len, int write_irq)
+{
+	int n, ret = 0;
+
+	if (len == 0 || !chan || !chan->ops->write)
+		return 0;
+
+	n = chan->ops->write(chan->fd_out, buf, len, chan->data);
+	if (chan->primary) {
+		ret = n;
+	}
+	return ret;
+}
+
+int console_write_chan(struct chan *chan, const char *buf, int len)
+{
+	int n, ret = 0;
+
+	if (!chan || !chan->ops->console_write)
+		return 0;
+
+	n = chan->ops->console_write(chan->fd_out, buf, len);
+	if (chan->primary)
+		ret = n;
+	return ret;
+}
+
+int console_open_chan(struct line *line, struct console *co)
+{
+	int err;
+
+	err = open_chan(&line->chan_list);
+	if (err)
+		return err;
+
+	printk(KERN_INFO "Console initialized on /dev/%s%d\n", co->name,
+	       co->index);
+	return 0;
+}
+
+int chan_window_size(struct line *line, unsigned short *rows_out,
+		      unsigned short *cols_out)
+{
+	struct chan *chan;
+
+	chan = line->chan_in;
+	if (chan && chan->primary) {
+		if (chan->ops->window_size == NULL)
+			return 0;
+		return chan->ops->window_size(chan->fd_in, chan->data,
+					      rows_out, cols_out);
+	}
+	chan = line->chan_out;
+	if (chan && chan->primary) {
+		if (chan->ops->window_size == NULL)
+			return 0;
+		return chan->ops->window_size(chan->fd_in, chan->data,
+					      rows_out, cols_out);
+	}
+	return 0;
+}
+
+static void free_one_chan(struct chan *chan)
+{
+	list_del(&chan->list);
+
+	close_one_chan(chan, 0);
+
+	if (chan->ops->free != NULL)
+		(*chan->ops->free)(chan->data);
+
+	if (chan->primary && chan->output)
+		ignore_sigio_fd(chan->fd_in);
+	kfree(chan);
+}
+
+static void free_chan(struct list_head *chans)
+{
+	struct list_head *ele, *next;
+	struct chan *chan;
+
+	list_for_each_safe(ele, next, chans) {
+		chan = list_entry(ele, struct chan, list);
+		free_one_chan(chan);
+	}
+}
+
+static int one_chan_config_string(struct chan *chan, char *str, int size,
+				  char **error_out)
+{
+	int n = 0;
+
+	if (chan == NULL) {
+		CONFIG_CHUNK(str, size, n, "none", 1);
+		return n;
+	}
+
+	CONFIG_CHUNK(str, size, n, chan->ops->type, 0);
+
+	if (chan->dev == NULL) {
+		CONFIG_CHUNK(str, size, n, "", 1);
+		return n;
+	}
+
+	CONFIG_CHUNK(str, size, n, ":", 0);
+	CONFIG_CHUNK(str, size, n, chan->dev, 0);
+
+	return n;
+}
+
+static int chan_pair_config_string(struct chan *in, struct chan *out,
+				   char *str, int size, char **error_out)
+{
+	int n;
+
+	n = one_chan_config_string(in, str, size, error_out);
+	str += n;
+	size -= n;
+
+	if (in == out) {
+		CONFIG_CHUNK(str, size, n, "", 1);
+		return n;
+	}
+
+	CONFIG_CHUNK(str, size, n, ",", 1);
+	n = one_chan_config_string(out, str, size, error_out);
+	str += n;
+	size -= n;
+	CONFIG_CHUNK(str, size, n, "", 1);
+
+	return n;
+}
+
+int chan_config_string(struct line *line, char *str, int size,
+		       char **error_out)
+{
+	struct chan *in = line->chan_in, *out = line->chan_out;
+
+	if (in && !in->primary)
+		in = NULL;
+	if (out && !out->primary)
+		out = NULL;
+
+	return chan_pair_config_string(in, out, str, size, error_out);
+}
+
+struct chan_type {
+	char *key;
+	const struct chan_ops *ops;
+};
+
+static const struct chan_type chan_table[] = {
+	{ "fd", &fd_ops },
+
+#ifdef CONFIG_NULL_CHAN
+	{ "null", &null_ops },
+#else
+	{ "null", &not_configged_ops },
+#endif
+
+#ifdef CONFIG_PORT_CHAN
+	{ "port", &port_ops },
+#else
+	{ "port", &not_configged_ops },
+#endif
+
+#ifdef CONFIG_PTY_CHAN
+	{ "pty", &pty_ops },
+	{ "pts", &pts_ops },
+#else
+	{ "pty", &not_configged_ops },
+	{ "pts", &not_configged_ops },
+#endif
+
+#ifdef CONFIG_TTY_CHAN
+	{ "tty", &tty_ops },
+#else
+	{ "tty", &not_configged_ops },
+#endif
+
+#ifdef CONFIG_XTERM_CHAN
+	{ "xterm", &xterm_ops },
+#else
+	{ "xterm", &not_configged_ops },
+#endif
+};
+
+static struct chan *parse_chan(struct line *line, char *str, int device,
+			       const struct chan_opts *opts, char **error_out)
+{
+	const struct chan_type *entry;
+	const struct chan_ops *ops;
+	struct chan *chan;
+	void *data;
+	int i;
+
+	ops = NULL;
+	data = NULL;
+	for(i = 0; i < ARRAY_SIZE(chan_table); i++) {
+		entry = &chan_table[i];
+		if (!strncmp(str, entry->key, strlen(entry->key))) {
+			ops = entry->ops;
+			str += strlen(entry->key);
+			break;
+		}
+	}
+	if (ops == NULL) {
+		*error_out = "No match for configured backends";
+		return NULL;
+	}
+
+	data = (*ops->init)(str, device, opts);
+	if (data == NULL) {
+		*error_out = "Configuration failed";
+		return NULL;
+	}
+
+	chan = kmalloc_obj(*chan, GFP_ATOMIC);
+	if (chan == NULL) {
+		*error_out = "Memory allocation failed";
+		return NULL;
+	}
+	*chan = ((struct chan) { .list	 	= LIST_HEAD_INIT(chan->list),
+				 .free_list 	=
+				 	LIST_HEAD_INIT(chan->free_list),
+				 .line		= line,
+				 .primary	= 1,
+				 .input		= 0,
+				 .output 	= 0,
+				 .opened  	= 0,
+				 .enabled  	= 0,
+				 .fd_in		= -1,
+				 .fd_out	= -1,
+				 .ops 		= ops,
+				 .data 		= data });
+	return chan;
+}
+
+int parse_chan_pair(char *str, struct line *line, int device,
+		    const struct chan_opts *opts, char **error_out)
+{
+	struct list_head *chans = &line->chan_list;
+	struct chan *new;
+	char *in, *out;
+
+	if (!list_empty(chans)) {
+		line->chan_in = line->chan_out = NULL;
+		free_chan(chans);
+		INIT_LIST_HEAD(chans);
+	}
+
+	if (!str)
+		return 0;
+
+	out = strchr(str, ',');
+	if (out != NULL) {
+		in = str;
+		*out = '\0';
+		out++;
+		new = parse_chan(line, in, device, opts, error_out);
+		if (new == NULL)
+			return -1;
+
+		new->input = 1;
+		list_add(&new->list, chans);
+		line->chan_in = new;
+
+		new = parse_chan(line, out, device, opts, error_out);
+		if (new == NULL)
+			return -1;
+
+		list_add(&new->list, chans);
+		new->output = 1;
+		line->chan_out = new;
+	}
+	else {
+		new = parse_chan(line, str, device, opts, error_out);
+		if (new == NULL)
+			return -1;
+
+		list_add(&new->list, chans);
+		new->input = 1;
+		new->output = 1;
+		line->chan_in = line->chan_out = new;
+	}
+	return 0;
+}
+
+void chan_interrupt(struct line *line, int irq)
+{
+	struct tty_port *port = &line->port;
+	struct chan *chan = line->chan_in;
+	int err;
+	u8 c;
+
+	if (!chan || !chan->ops->read)
+		goto out;
+
+	do {
+		if (!tty_buffer_request_room(port, 1)) {
+			schedule_delayed_work(&line->task, 1);
+			goto out;
+		}
+		err = chan->ops->read(chan->fd_in, &c, chan->data);
+		if (err > 0)
+			tty_insert_flip_char(port, c, TTY_NORMAL);
+	} while (err > 0);
+
+	if (err == -EIO) {
+		if (chan->primary) {
+			tty_port_tty_hangup(&line->port, false);
+			if (line->chan_out != chan)
+				close_one_chan(line->chan_out, 1);
+		}
+		close_one_chan(chan, 1);
+		if (chan->primary)
+			return;
+	}
+ out:
+	tty_flip_buffer_push(port);
+}

linux-7.2.3/arch/um/drivers/chan_user.c

@@ -0,0 +1,323 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <sched.h>
+#include <signal.h>
+#include <termios.h>
+#include <sys/ioctl.h>
+#include "chan_user.h"
+#include <os.h>
+#include <um_malloc.h>
+
+void generic_close(int fd, void *unused)
+{
+	close(fd);
+}
+
+int generic_read(int fd, __u8 *c_out, void *unused)
+{
+	int n;
+
+	CATCH_EINTR(n = read(fd, c_out, sizeof(*c_out)));
+	if (n > 0)
+		return n;
+	else if (n == 0)
+		return -EIO;
+	else if (errno == EAGAIN)
+		return 0;
+	return -errno;
+}
+
+/* XXX Trivial wrapper around write */
+
+int generic_write(int fd, const __u8 *buf, size_t n, void *unused)
+{
+	int written = 0;
+	int err;
+
+	/* The FD may be in blocking mode, as such, need to retry short writes,
+	 * they may have been interrupted by a signal.
+	 */
+	do {
+		errno = 0;
+		err = write(fd, buf + written, n - written);
+		if (err > 0) {
+			written += err;
+			continue;
+		}
+	} while (err < 0 && errno == EINTR);
+
+	if (written > 0)
+		return written;
+	else if (errno == EAGAIN)
+		return 0;
+	else if (err == 0)
+		return -EIO;
+	return -errno;
+}
+
+int generic_window_size(int fd, void *unused, unsigned short *rows_out,
+			unsigned short *cols_out)
+{
+	struct winsize size;
+	int ret;
+
+	if (ioctl(fd, TIOCGWINSZ, &size) < 0)
+		return -errno;
+
+	ret = ((*rows_out != size.ws_row) || (*cols_out != size.ws_col));
+
+	*rows_out = size.ws_row;
+	*cols_out = size.ws_col;
+
+	return ret;
+}
+
+void generic_free(void *data)
+{
+	kfree(data);
+}
+
+int generic_console_write(int fd, const char *buf, int n)
+{
+	sigset_t old, no_sigio;
+	struct termios save, new;
+	int err;
+
+	if (isatty(fd)) {
+		sigemptyset(&no_sigio);
+		sigaddset(&no_sigio, SIGIO);
+		if (sigprocmask(SIG_BLOCK, &no_sigio, &old))
+			goto error;
+
+		CATCH_EINTR(err = tcgetattr(fd, &save));
+		if (err)
+			goto error;
+		new = save;
+		/*
+		 * The terminal becomes a bit less raw, to handle \n also as
+		 * "Carriage Return", not only as "New Line". Otherwise, the new
+		 * line won't start at the first column.
+		 */
+		new.c_oflag |= OPOST;
+		CATCH_EINTR(err = tcsetattr(fd, TCSAFLUSH, &new));
+		if (err)
+			goto error;
+	}
+	err = generic_write(fd, buf, n, NULL);
+	/*
+	 * Restore raw mode, in any case; we *must* ignore any error apart
+	 * EINTR, except for debug.
+	 */
+	if (isatty(fd)) {
+		CATCH_EINTR(tcsetattr(fd, TCSAFLUSH, &save));
+		sigprocmask(SIG_SETMASK, &old, NULL);
+	}
+
+	return err;
+error:
+	return -errno;
+}
+
+/*
+ * UML SIGWINCH handling
+ *
+ * The point of this is to handle SIGWINCH on consoles which have host
+ * ttys and relay them inside UML to whatever might be running on the
+ * console and cares about the window size (since SIGWINCH notifies
+ * about terminal size changes).
+ *
+ * So, we have a separate thread for each host tty attached to a UML
+ * device (side-issue - I'm annoyed that one thread can't have
+ * multiple controlling ttys for the purpose of handling SIGWINCH, but
+ * I imagine there are other reasons that doesn't make any sense).
+ *
+ * SIGWINCH can't be received synchronously, so you have to set up to
+ * receive it as a signal.  That being the case, if you are going to
+ * wait for it, it is convenient to sit in sigsuspend() and wait for
+ * the signal to bounce you out of it (see below for how we make sure
+ * to exit only on SIGWINCH).
+ */
+
+static void winch_handler(int sig)
+{
+}
+
+struct winch_data {
+	int pty_fd;
+	int pipe_fd;
+};
+
+static __noreturn int winch_thread(void *arg)
+{
+	struct winch_data *data = arg;
+	sigset_t sigs;
+	int pty_fd, pipe_fd;
+	int count;
+	char c = 1;
+
+	os_set_pdeathsig();
+
+	pty_fd = data->pty_fd;
+	pipe_fd = data->pipe_fd;
+	count = write(pipe_fd, &c, sizeof(c));
+	if (count != sizeof(c))
+		os_info("winch_thread : failed to write synchronization byte, err = %d\n",
+			-count);
+
+	/*
+	 * We are not using SIG_IGN on purpose, so don't fix it as I thought to
+	 * do! If using SIG_IGN, the sigsuspend() call below would not stop on
+	 * SIGWINCH.
+	 */
+
+	signal(SIGWINCH, winch_handler);
+	sigfillset(&sigs);
+	/* Block all signals possible. */
+	if (sigprocmask(SIG_SETMASK, &sigs, NULL) < 0) {
+		os_info("winch_thread : sigprocmask failed, errno = %d\n",
+			errno);
+		goto wait_kill;
+	}
+	/* In sigsuspend(), block anything else than SIGWINCH. */
+	sigdelset(&sigs, SIGWINCH);
+
+	if (setsid() < 0) {
+		os_info("winch_thread : setsid failed, errno = %d\n",
+		       errno);
+		goto wait_kill;
+	}
+
+	if (ioctl(pty_fd, TIOCSCTTY, 0) < 0) {
+		os_info("winch_thread : TIOCSCTTY failed on "
+			"fd %d err = %d\n", pty_fd, errno);
+		goto wait_kill;
+	}
+
+	if (tcsetpgrp(pty_fd, os_getpid()) < 0) {
+		os_info("winch_thread : tcsetpgrp failed on fd %d err = %d\n",
+			pty_fd, errno);
+		goto wait_kill;
+	}
+
+	/*
+	 * These are synchronization calls between various UML threads on the
+	 * host - since they are not different kernel threads, we cannot use
+	 * kernel semaphores. We don't use SysV semaphores because they are
+	 * persistent.
+	 */
+	count = read(pipe_fd, &c, sizeof(c));
+	if (count != sizeof(c))
+		os_info("winch_thread : failed to read synchronization byte, err = %d\n",
+			errno);
+
+	while(1) {
+		/*
+		 * This will be interrupted by SIGWINCH only, since
+		 * other signals are blocked.
+		 */
+		sigsuspend(&sigs);
+
+		count = write(pipe_fd, &c, sizeof(c));
+		if (count != sizeof(c))
+			os_info("winch_thread : write failed, err = %d\n",
+				errno);
+	}
+
+wait_kill:
+	c = 2;
+	count = write(pipe_fd, &c, sizeof(c));
+	while (1)
+		pause();
+}
+
+static int winch_tramp(int fd, struct tty_port *port, int *fd_out,
+		       unsigned long *stack_out)
+{
+	struct winch_data data;
+	int fds[2], n, err, pid;
+	char c;
+
+	err = os_pipe(fds, 1, 1);
+	if (err < 0) {
+		printk(UM_KERN_ERR "winch_tramp : os_pipe failed, err = %d\n",
+		       -err);
+		goto out;
+	}
+
+	data = ((struct winch_data) { .pty_fd 		= fd,
+				      .pipe_fd 		= fds[1] } );
+	/*
+	 * CLONE_FILES so this thread doesn't hold open files which are open
+	 * now, but later closed in a different thread.  This is a
+	 * problem with /dev/net/tun, which if held open by this
+	 * thread, prevents the TUN/TAP device from being reused.
+	 */
+	pid = run_helper_thread(winch_thread, &data, CLONE_FILES, stack_out);
+	if (pid < 0) {
+		err = pid;
+		printk(UM_KERN_ERR "fork of winch_thread failed - errno = %d\n",
+		       -err);
+		goto out_close;
+	}
+
+	*fd_out = fds[0];
+	n = read(fds[0], &c, sizeof(c));
+	if (n != sizeof(c)) {
+		printk(UM_KERN_ERR "winch_tramp : failed to read "
+		       "synchronization byte\n");
+		printk(UM_KERN_ERR "read failed, err = %d\n", errno);
+		printk(UM_KERN_ERR "fd %d will not support SIGWINCH\n", fd);
+		err = -EINVAL;
+		goto out_close;
+	}
+
+	err = os_set_fd_block(*fd_out, 0);
+	if (err) {
+		printk(UM_KERN_ERR "winch_tramp: failed to set thread_fd "
+		       "non-blocking.\n");
+		goto out_close;
+	}
+
+	return pid;
+
+ out_close:
+	close(fds[1]);
+	close(fds[0]);
+ out:
+	return err;
+}
+
+void register_winch(int fd, struct tty_port *port)
+{
+	unsigned long stack;
+	int pid, thread, count, thread_fd = -1;
+	char c = 1;
+
+	if (!isatty(fd))
+		return;
+
+	pid = tcgetpgrp(fd);
+	if (is_skas_winch(pid, fd, port)) {
+		register_winch_irq(-1, fd, -1, port, 0);
+		return;
+	}
+
+	if (pid == -1) {
+		thread = winch_tramp(fd, port, &thread_fd, &stack);
+		if (thread < 0)
+			return;
+
+		register_winch_irq(thread_fd, fd, thread, port, stack);
+
+		count = write(thread_fd, &c, sizeof(c));
+		if (count != sizeof(c))
+			printk(UM_KERN_ERR "register_winch : failed to write "
+			       "synchronization byte, err = %d\n", errno);
+	}
+}

linux-7.2.3/arch/um/drivers/chan_user.h

@@ -0,0 +1,54 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __CHAN_USER_H__
+#define __CHAN_USER_H__
+
+#include <init.h>
+#include <linux/types.h>
+
+struct chan_opts {
+	void (*const announce)(char *dev_name, int dev);
+	char *xterm_title;
+	int raw;
+};
+
+struct chan_ops {
+	char *type;
+	void *(*init)(char *, int, const struct chan_opts *);
+	int (*open)(int, int, int, void *, char **);
+	void (*close)(int, void *);
+	int (*read)(int, __u8 *, void *);
+	int (*write)(int, const __u8 *, size_t, void *);
+	int (*console_write)(int, const char *, int);
+	int (*window_size)(int, void *, unsigned short *, unsigned short *);
+	void (*free)(void *);
+	int winch;
+};
+
+extern const struct chan_ops fd_ops, null_ops, port_ops, pts_ops, pty_ops,
+	tty_ops, xterm_ops;
+
+extern void generic_close(int fd, void *unused);
+extern int generic_read(int fd, __u8 *c_out, void *unused);
+extern int generic_write(int fd, const __u8 *buf, size_t n, void *unused);
+extern int generic_console_write(int fd, const char *buf, int n);
+extern int generic_window_size(int fd, void *unused, unsigned short *rows_out,
+			       unsigned short *cols_out);
+extern void generic_free(void *data);
+
+struct tty_port;
+extern void register_winch(int fd,  struct tty_port *port);
+extern void register_winch_irq(int fd, int tty_fd, int pid,
+			       struct tty_port *port, unsigned long stack);
+
+#define __channel_help(fn, prefix) \
+__uml_help(fn, prefix "[0-9]*=<channel description>\n" \
+"    Attach a console or serial line to a host channel.  See\n" \
+"    http://user-mode-linux.sourceforge.net/old/input.html for a complete\n" \
+"    description of this switch.\n\n" \
+);
+
+#endif

linux-7.2.3/arch/um/drivers/cow.h

@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __COW_H__
+#define __COW_H__
+
+#include <asm/types.h>
+
+extern int init_cow_file(int fd, char *cow_file, char *backing_file,
+			 int sectorsize, int alignment, int *bitmap_offset_out,
+			 unsigned long *bitmap_len_out, int *data_offset_out);
+
+extern int file_reader(__u64 offset, char *buf, int len, void *arg);
+extern int read_cow_header(int (*reader)(__u64, char *, int, void *),
+			   void *arg, __u32 *version_out,
+			   char **backing_file_out, long long *mtime_out,
+			   unsigned long long *size_out, int *sectorsize_out,
+			   __u32 *align_out, int *bitmap_offset_out);
+
+extern int write_cow_header(char *cow_file, int fd, char *backing_file,
+			    int sectorsize, int alignment,
+			    unsigned long long *size);
+
+extern void cow_sizes(int version, __u64 size, int sectorsize, int align,
+		      int bitmap_offset, unsigned long *bitmap_len_out,
+		      int *data_offset_out);
+
+#endif

linux-7.2.3/arch/um/drivers/cow_sys.h

@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __COW_SYS_H__
+#define __COW_SYS_H__
+
+#include <kern_util.h>
+#include <os.h>
+#include <um_malloc.h>
+
+static inline void *cow_malloc(int size)
+{
+	return uml_kmalloc(size, UM_GFP_KERNEL);
+}
+
+static inline void cow_free(void *ptr)
+{
+	kfree(ptr);
+}
+
+#define cow_printf printk
+
+static inline char *cow_strdup(char *str)
+{
+	return uml_strdup(str);
+}
+
+static inline int cow_seek_file(int fd, __u64 offset)
+{
+	return os_seek_file(fd, offset);
+}
+
+static inline int cow_file_size(char *file, unsigned long long *size_out)
+{
+	return os_file_size(file, size_out);
+}
+
+static inline int cow_write_file(int fd, void *buf, int size)
+{
+	return os_write_file(fd, buf, size);
+}
+
+#endif

linux-7.2.3/arch/um/drivers/cow_user.c

@@ -0,0 +1,450 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+/*
+ * _XOPEN_SOURCE is needed for pread, but we define _GNU_SOURCE, which defines
+ * that.
+ */
+#include <unistd.h>
+#include <errno.h>
+#include <string.h>
+#include <arpa/inet.h>
+#include <endian.h>
+#include "cow.h"
+#include "cow_sys.h"
+
+/*
+ * arch/um/Makefile remaps strrchr to kernel_strrchr; call the kernel
+ * name directly to avoid glibc >= 2.43's C23 strrchr macro.
+ */
+char *kernel_strrchr(const char *, int);
+
+#define PATH_LEN_V1 256
+
+/* unsigned time_t works until year 2106 */
+typedef __u32 time32_t;
+
+struct cow_header_v1 {
+	__s32 magic;
+	__s32 version;
+	char backing_file[PATH_LEN_V1];
+	time32_t mtime;
+	__u64 size;
+	__s32 sectorsize;
+} __attribute__((packed));
+
+/*
+ * Define PATH_LEN_V3 as the usual value of MAXPATHLEN, just hard-code it in
+ * case other systems have different values for MAXPATHLEN.
+ *
+ * The same must hold for V2 - we want file format compatibility, not anything
+ * else.
+ */
+#define PATH_LEN_V3 4096
+#define PATH_LEN_V2 PATH_LEN_V3
+
+struct cow_header_v2 {
+	__u32 magic;
+	__u32 version;
+	char backing_file[PATH_LEN_V2];
+	time32_t mtime;
+	__u64 size;
+	__s32 sectorsize;
+} __attribute__((packed));
+
+/*
+ * Changes from V2 -
+ *	PATH_LEN_V3 as described above
+ *	Explicitly specify field bit lengths for systems with different
+ *		lengths for the usual C types.  Not sure whether char or
+ *		time_t should be changed, this can be changed later without
+ *		breaking compatibility
+ *	Add alignment field so that different alignments can be used for the
+ *		bitmap and data
+ * 	Add cow_format field to allow for the possibility of different ways
+ *		of specifying the COW blocks.  For now, the only value is 0,
+ * 		for the traditional COW bitmap.
+ *	Move the backing_file field to the end of the header.  This allows
+ *		for the possibility of expanding it into the padding required
+ *		by the bitmap alignment.
+ * 	The bitmap and data portions of the file will be aligned as specified
+ * 		by the alignment field.  This is to allow COW files to be
+ *		put on devices with restrictions on access alignments, such as
+ *		/dev/raw, with a 512 byte alignment restriction.  This also
+ *		allows the data to be more aligned more strictly than on
+ *		sector boundaries.  This is needed for ubd-mmap, which needs
+ *		the data to be page aligned.
+ *	Fixed (finally!) the rounding bug
+ */
+
+/*
+ * Until Dec2005, __attribute__((packed)) was left out from the below
+ * definition, leading on 64-bit systems to 4 bytes of padding after mtime, to
+ * align size to 8-byte alignment.  This shifted all fields above (no padding
+ * was present on 32-bit, no other padding was added).
+ *
+ * However, this _can be detected_: it means that cow_format (always 0 until
+ * now) is shifted onto the first 4 bytes of backing_file, where it is otherwise
+ * impossible to find 4 zeros. -bb */
+
+struct cow_header_v3 {
+	__u32 magic;
+	__u32 version;
+	__u32 mtime;
+	__u64 size;
+	__u32 sectorsize;
+	__u32 alignment;
+	__u32 cow_format;
+	char backing_file[PATH_LEN_V3];
+} __attribute__((packed));
+
+/* This is the broken layout used by some 64-bit binaries. */
+struct cow_header_v3_broken {
+	__u32 magic;
+	__u32 version;
+	__s64 mtime;
+	__u64 size;
+	__u32 sectorsize;
+	__u32 alignment;
+	__u32 cow_format;
+	char backing_file[PATH_LEN_V3];
+};
+
+/* COW format definitions - for now, we have only the usual COW bitmap */
+#define COW_BITMAP 0
+
+union cow_header {
+	struct cow_header_v1 v1;
+	struct cow_header_v2 v2;
+	struct cow_header_v3 v3;
+	struct cow_header_v3_broken v3_b;
+};
+
+#define COW_MAGIC 0x4f4f4f4d  /* MOOO */
+#define COW_VERSION 3
+
+#define DIV_ROUND(x, len) (((x) + (len) - 1) / (len))
+#define ROUND_UP(x, align) DIV_ROUND(x, align) * (align)
+
+void cow_sizes(int version, __u64 size, int sectorsize, int align,
+	       int bitmap_offset, unsigned long *bitmap_len_out,
+	       int *data_offset_out)
+{
+	if (version < 3) {
+		*bitmap_len_out = (size + sectorsize - 1) / (8 * sectorsize);
+
+		*data_offset_out = bitmap_offset + *bitmap_len_out;
+		*data_offset_out = (*data_offset_out + sectorsize - 1) /
+			sectorsize;
+		*data_offset_out *= sectorsize;
+	}
+	else {
+		*bitmap_len_out = DIV_ROUND(size, sectorsize);
+		*bitmap_len_out = DIV_ROUND(*bitmap_len_out, 8);
+
+		*data_offset_out = bitmap_offset + *bitmap_len_out;
+		*data_offset_out = ROUND_UP(*data_offset_out, align);
+	}
+}
+
+static int absolutize(char *to, int size, char *from)
+{
+	char save_cwd[256], *slash;
+	int remaining;
+
+	if (getcwd(save_cwd, sizeof(save_cwd)) == NULL) {
+		cow_printf("absolutize : unable to get cwd - errno = %d\n",
+			   errno);
+		return -1;
+	}
+	slash = kernel_strrchr(from, '/');
+	if (slash != NULL) {
+		*slash = '\0';
+		if (chdir(from)) {
+			*slash = '/';
+			cow_printf("absolutize : Can't cd to '%s' - "
+				   "errno = %d\n", from, errno);
+			return -1;
+		}
+		*slash = '/';
+		if (getcwd(to, size) == NULL) {
+			cow_printf("absolutize : unable to get cwd of '%s' - "
+			       "errno = %d\n", from, errno);
+			return -1;
+		}
+		remaining = size - strlen(to);
+		if (strlen(slash) + 1 > remaining) {
+			cow_printf("absolutize : unable to fit '%s' into %d "
+			       "chars\n", from, size);
+			return -1;
+		}
+		strcat(to, slash);
+	}
+	else {
+		if (strlen(save_cwd) + 1 + strlen(from) + 1 > size) {
+			cow_printf("absolutize : unable to fit '%s' into %d "
+			       "chars\n", from, size);
+			return -1;
+		}
+		strcpy(to, save_cwd);
+		strcat(to, "/");
+		strcat(to, from);
+	}
+	if (chdir(save_cwd)) {
+		cow_printf("absolutize : Can't cd to '%s' - "
+			   "errno = %d\n", save_cwd, errno);
+		return -1;
+	}
+	return 0;
+}
+
+int write_cow_header(char *cow_file, int fd, char *backing_file,
+		     int sectorsize, int alignment, unsigned long long *size)
+{
+	struct cow_header_v3 *header;
+	long long modtime;
+	int err;
+
+	err = cow_seek_file(fd, 0);
+	if (err < 0) {
+		cow_printf("write_cow_header - lseek failed, err = %d\n", -err);
+		goto out;
+	}
+
+	err = -ENOMEM;
+	header = cow_malloc(sizeof(*header));
+	if (header == NULL) {
+		cow_printf("write_cow_header - failed to allocate COW V3 "
+			   "header\n");
+		goto out;
+	}
+	header->magic = htobe32(COW_MAGIC);
+	header->version = htobe32(COW_VERSION);
+
+	err = -EINVAL;
+	if (strlen(backing_file) > sizeof(header->backing_file) - 1) {
+		/* Below, %zd is for a size_t value */
+		cow_printf("Backing file name \"%s\" is too long - names are "
+			   "limited to %zd characters\n", backing_file,
+			   sizeof(header->backing_file) - 1);
+		goto out_free;
+	}
+
+	if (absolutize(header->backing_file, sizeof(header->backing_file),
+		      backing_file))
+		goto out_free;
+
+	err = os_file_modtime(header->backing_file, &modtime);
+	if (err < 0) {
+		cow_printf("write_cow_header - backing file '%s' mtime "
+			   "request failed, err = %d\n", header->backing_file,
+			   -err);
+		goto out_free;
+	}
+
+	err = cow_file_size(header->backing_file, size);
+	if (err < 0) {
+		cow_printf("write_cow_header - couldn't get size of "
+			   "backing file '%s', err = %d\n",
+			   header->backing_file, -err);
+		goto out_free;
+	}
+
+	header->mtime = htobe32(modtime);
+	header->size = htobe64(*size);
+	header->sectorsize = htobe32(sectorsize);
+	header->alignment = htobe32(alignment);
+	header->cow_format = COW_BITMAP;
+
+	err = cow_write_file(fd, header, sizeof(*header));
+	if (err != sizeof(*header)) {
+		cow_printf("write_cow_header - write of header to "
+			   "new COW file '%s' failed, err = %d\n", cow_file,
+			   -err);
+		goto out_free;
+	}
+	err = 0;
+ out_free:
+	cow_free(header);
+ out:
+	return err;
+}
+
+int file_reader(__u64 offset, char *buf, int len, void *arg)
+{
+	int fd = *((int *) arg);
+
+	return pread(fd, buf, len, offset);
+}
+
+/* XXX Need to sanity-check the values read from the header */
+
+int read_cow_header(int (*reader)(__u64, char *, int, void *), void *arg,
+		    __u32 *version_out, char **backing_file_out,
+		    long long *mtime_out, unsigned long long *size_out,
+		    int *sectorsize_out, __u32 *align_out,
+		    int *bitmap_offset_out)
+{
+	union cow_header *header;
+	char *file;
+	int err, n;
+	unsigned long version, magic;
+
+	header = cow_malloc(sizeof(*header));
+	if (header == NULL) {
+	        cow_printf("read_cow_header - Failed to allocate header\n");
+		return -ENOMEM;
+	}
+	err = -EINVAL;
+	n = (*reader)(0, (char *) header, sizeof(*header), arg);
+	if (n < offsetof(typeof(header->v1), backing_file)) {
+		cow_printf("read_cow_header - short header\n");
+		goto out;
+	}
+
+	magic = header->v1.magic;
+	if (magic == COW_MAGIC)
+		version = header->v1.version;
+	else if (magic == be32toh(COW_MAGIC))
+		version = be32toh(header->v1.version);
+	/* No error printed because the non-COW case comes through here */
+	else goto out;
+
+	*version_out = version;
+
+	if (version == 1) {
+		if (n < sizeof(header->v1)) {
+			cow_printf("read_cow_header - failed to read V1 "
+				   "header\n");
+			goto out;
+		}
+		*mtime_out = header->v1.mtime;
+		*size_out = header->v1.size;
+		*sectorsize_out = header->v1.sectorsize;
+		*bitmap_offset_out = sizeof(header->v1);
+		*align_out = *sectorsize_out;
+		file = header->v1.backing_file;
+	}
+	else if (version == 2) {
+		if (n < sizeof(header->v2)) {
+			cow_printf("read_cow_header - failed to read V2 "
+				   "header\n");
+			goto out;
+		}
+		*mtime_out = be32toh(header->v2.mtime);
+		*size_out = be64toh(header->v2.size);
+		*sectorsize_out = be32toh(header->v2.sectorsize);
+		*bitmap_offset_out = sizeof(header->v2);
+		*align_out = *sectorsize_out;
+		file = header->v2.backing_file;
+	}
+	/* This is very subtle - see above at union cow_header definition */
+	else if (version == 3 && (*((int*)header->v3.backing_file) != 0)) {
+		if (n < sizeof(header->v3)) {
+			cow_printf("read_cow_header - failed to read V3 "
+				   "header\n");
+			goto out;
+		}
+		*mtime_out = be32toh(header->v3.mtime);
+		*size_out = be64toh(header->v3.size);
+		*sectorsize_out = be32toh(header->v3.sectorsize);
+		*align_out = be32toh(header->v3.alignment);
+		if (*align_out == 0) {
+			cow_printf("read_cow_header - invalid COW header, "
+				   "align == 0\n");
+		}
+		*bitmap_offset_out = ROUND_UP(sizeof(header->v3), *align_out);
+		file = header->v3.backing_file;
+	}
+	else if (version == 3) {
+		cow_printf("read_cow_header - broken V3 file with"
+			   " 64-bit layout - recovering content.\n");
+
+		if (n < sizeof(header->v3_b)) {
+			cow_printf("read_cow_header - failed to read V3 "
+				   "header\n");
+			goto out;
+		}
+
+		/*
+		 * this was used until Dec2005 - 64bits are needed to represent
+		 * 2106+. I.e. we can safely do this truncating cast.
+		 *
+		 * Additionally, we must use be32toh() instead of be64toh(), since
+		 * the program used to use the former (tested - I got mtime
+		 * mismatch "0 vs whatever").
+		 *
+		 * Ever heard about bug-to-bug-compatibility ? ;-) */
+		*mtime_out = (time32_t) be32toh(header->v3_b.mtime);
+
+		*size_out = be64toh(header->v3_b.size);
+		*sectorsize_out = be32toh(header->v3_b.sectorsize);
+		*align_out = be32toh(header->v3_b.alignment);
+		if (*align_out == 0) {
+			cow_printf("read_cow_header - invalid COW header, "
+				   "align == 0\n");
+		}
+		*bitmap_offset_out = ROUND_UP(sizeof(header->v3_b), *align_out);
+		file = header->v3_b.backing_file;
+	}
+	else {
+		cow_printf("read_cow_header - invalid COW version\n");
+		goto out;
+	}
+	err = -ENOMEM;
+	*backing_file_out = cow_strdup(file);
+	if (*backing_file_out == NULL) {
+		cow_printf("read_cow_header - failed to allocate backing "
+			   "file\n");
+		goto out;
+	}
+	err = 0;
+ out:
+	cow_free(header);
+	return err;
+}
+
+int init_cow_file(int fd, char *cow_file, char *backing_file, int sectorsize,
+		  int alignment, int *bitmap_offset_out,
+		  unsigned long *bitmap_len_out, int *data_offset_out)
+{
+	unsigned long long size, offset;
+	char zero = 0;
+	int err;
+
+	err = write_cow_header(cow_file, fd, backing_file, sectorsize,
+			       alignment, &size);
+	if (err)
+		goto out;
+
+	*bitmap_offset_out = ROUND_UP(sizeof(struct cow_header_v3), alignment);
+	cow_sizes(COW_VERSION, size, sectorsize, alignment, *bitmap_offset_out,
+		  bitmap_len_out, data_offset_out);
+
+	offset = *data_offset_out + size - sizeof(zero);
+	err = cow_seek_file(fd, offset);
+	if (err < 0) {
+		cow_printf("cow bitmap lseek failed : err = %d\n", -err);
+		goto out;
+	}
+
+	/*
+	 * does not really matter how much we write it is just to set EOF
+	 * this also sets the entire COW bitmap
+	 * to zero without having to allocate it
+	 */
+	err = cow_write_file(fd, &zero, sizeof(zero));
+	if (err != sizeof(zero)) {
+		cow_printf("Write of bitmap to new COW file '%s' failed, "
+			   "err = %d\n", cow_file, -err);
+		if (err >= 0)
+			err = -EINVAL;
+		goto out;
+	}
+
+	return 0;
+ out:
+	return err;
+}

linux-7.2.3/arch/um/drivers/fd.c

@@ -0,0 +1,95 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <termios.h>
+#include "chan_user.h"
+#include <os.h>
+#include <um_malloc.h>
+
+struct fd_chan {
+	int fd;
+	int raw;
+	struct termios tt;
+	char str[sizeof("1234567890\0")];
+};
+
+static void *fd_init(char *str, int device, const struct chan_opts *opts)
+{
+	struct fd_chan *data;
+	char *end;
+	int n;
+
+	if (*str != ':') {
+		printk(UM_KERN_ERR "fd_init : channel type 'fd' must specify a "
+		       "file descriptor\n");
+		return NULL;
+	}
+	str++;
+	n = strtoul(str, &end, 0);
+	if ((*end != '\0') || (end == str)) {
+		printk(UM_KERN_ERR "fd_init : couldn't parse file descriptor "
+		       "'%s'\n", str);
+		return NULL;
+	}
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_KERNEL);
+	if (data == NULL)
+		return NULL;
+
+	*data = ((struct fd_chan) { .fd  	= n,
+				    .raw  	= opts->raw });
+	return data;
+}
+
+static int fd_open(int input, int output, int primary, void *d, char **dev_out)
+{
+	struct fd_chan *data = d;
+	int err;
+
+	if (data->raw && isatty(data->fd)) {
+		CATCH_EINTR(err = tcgetattr(data->fd, &data->tt));
+		if (err)
+			return err;
+
+		err = raw(data->fd);
+		if (err)
+			return err;
+	}
+	sprintf(data->str, "%d", data->fd);
+	*dev_out = data->str;
+	return data->fd;
+}
+
+static void fd_close(int fd, void *d)
+{
+	struct fd_chan *data = d;
+	int err;
+
+	if (!data->raw || !isatty(fd))
+		return;
+
+	CATCH_EINTR(err = tcsetattr(fd, TCSAFLUSH, &data->tt));
+	if (err)
+		printk(UM_KERN_ERR "Failed to restore terminal state - "
+		       "errno = %d\n", -err);
+	data->raw = 0;
+}
+
+const struct chan_ops fd_ops = {
+	.type		= "fd",
+	.init		= fd_init,
+	.open		= fd_open,
+	.close		= fd_close,
+	.read		= generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= generic_free,
+	.winch		= 1,
+};

linux-7.2.3/arch/um/drivers/harddog.h

@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef UM_WATCHDOG_H
+#define UM_WATCHDOG_H
+
+int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock);
+void stop_watchdog(int in_fd, int out_fd);
+int ping_watchdog(int fd);
+
+#endif /* UM_WATCHDOG_H */

linux-7.2.3/arch/um/drivers/harddog_kern.c

@@ -0,0 +1,174 @@
+/* UML hardware watchdog, shamelessly stolen from:
+ *
+ *	SoftDog	0.05:	A Software Watchdog Device
+ *
+ *	(c) Copyright 1996 Alan Cox <alan@redhat.com>, All Rights Reserved.
+ *				http://www.redhat.com
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	Neither Alan Cox nor CymruNet Ltd. admit liability nor provide
+ *	warranty for any of this software. This material is provided
+ *	"AS-IS" and at no charge.
+ *
+ *	(c) Copyright 1995    Alan Cox <alan@lxorguk.ukuu.org.uk>
+ *
+ *	Software only watchdog driver. Unlike its big brother the WDT501P
+ *	driver this won't always recover a failed machine.
+ *
+ *  03/96: Angelo Haritsis <ah@doc.ic.ac.uk> :
+ *	Modularised.
+ *	Added soft_margin; use upon insmod to change the timer delay.
+ *	NB: uses same minor as wdt (WATCHDOG_MINOR); we could use separate
+ *	    minors.
+ *
+ *  19980911 Alan Cox
+ *	Made SMP safe for 2.3.x
+ *
+ *  20011127 Joel Becker (jlbec@evilplan.org>
+ *	Added soft_noboot; Allows testing the softdog trigger without
+ *	requiring a recompile.
+ *	Added WDIOC_GETTIMEOUT and WDIOC_SETTIMOUT.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/reboot.h>
+#include <linux/mutex.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include "mconsole.h"
+#include "harddog.h"
+
+MODULE_DESCRIPTION("UML hardware watchdog");
+MODULE_LICENSE("GPL");
+
+static DEFINE_MUTEX(harddog_mutex);
+static DEFINE_SPINLOCK(lock);
+static int timer_alive;
+static int harddog_in_fd = -1;
+static int harddog_out_fd = -1;
+
+/*
+ *	Allow only one person to hold it open
+ */
+
+static int harddog_open(struct inode *inode, struct file *file)
+{
+	int err = -EBUSY;
+	char *sock = NULL;
+
+	mutex_lock(&harddog_mutex);
+	spin_lock(&lock);
+	if(timer_alive)
+		goto err;
+#ifdef CONFIG_WATCHDOG_NOWAYOUT
+	__module_get(THIS_MODULE);
+#endif
+
+#ifdef CONFIG_MCONSOLE
+	sock = mconsole_notify_socket();
+#endif
+	err = start_watchdog(&harddog_in_fd, &harddog_out_fd, sock);
+	if(err)
+		goto err;
+
+	timer_alive = 1;
+	spin_unlock(&lock);
+	mutex_unlock(&harddog_mutex);
+	return stream_open(inode, file);
+err:
+	spin_unlock(&lock);
+	mutex_unlock(&harddog_mutex);
+	return err;
+}
+
+static int harddog_release(struct inode *inode, struct file *file)
+{
+	/*
+	 *	Shut off the timer.
+	 */
+
+	spin_lock(&lock);
+
+	stop_watchdog(harddog_in_fd, harddog_out_fd);
+	harddog_in_fd = -1;
+	harddog_out_fd = -1;
+
+	timer_alive=0;
+	spin_unlock(&lock);
+
+	return 0;
+}
+
+static ssize_t harddog_write(struct file *file, const char __user *data, size_t len,
+			     loff_t *ppos)
+{
+	/*
+	 *	Refresh the timer.
+	 */
+	if(len)
+		return ping_watchdog(harddog_out_fd);
+	return 0;
+}
+
+static int harddog_ioctl_unlocked(struct file *file,
+				  unsigned int cmd, unsigned long arg)
+{
+	void __user *argp= (void __user *)arg;
+	static struct watchdog_info ident = {
+		WDIOC_SETTIMEOUT,
+		0,
+		"UML Hardware Watchdog"
+	};
+	switch (cmd) {
+		default:
+			return -ENOTTY;
+		case WDIOC_GETSUPPORT:
+			if(copy_to_user(argp, &ident, sizeof(ident)))
+				return -EFAULT;
+			return 0;
+		case WDIOC_GETSTATUS:
+		case WDIOC_GETBOOTSTATUS:
+			return put_user(0,(int __user *)argp);
+		case WDIOC_KEEPALIVE:
+			return ping_watchdog(harddog_out_fd);
+	}
+}
+
+static long harddog_ioctl(struct file *file,
+			  unsigned int cmd, unsigned long arg)
+{
+	long ret;
+
+	mutex_lock(&harddog_mutex);
+	ret = harddog_ioctl_unlocked(file, cmd, arg);
+	mutex_unlock(&harddog_mutex);
+
+	return ret;
+}
+
+static const struct file_operations harddog_fops = {
+	.owner		= THIS_MODULE,
+	.write		= harddog_write,
+	.unlocked_ioctl	= harddog_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
+	.open		= harddog_open,
+	.release	= harddog_release,
+};
+
+static struct miscdevice harddog_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &harddog_fops,
+};
+module_misc_device(harddog_miscdev);

linux-7.2.3/arch/um/drivers/harddog_user.c

@@ -0,0 +1,128 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <errno.h>
+#include <os.h>
+#include "harddog.h"
+
+struct dog_data {
+	int stdin_fd;
+	int stdout_fd;
+	int close_me[2];
+};
+
+static void pre_exec(void *d)
+{
+	struct dog_data *data = d;
+
+	dup2(data->stdin_fd, 0);
+	dup2(data->stdout_fd, 1);
+	dup2(data->stdout_fd, 2);
+	close(data->stdin_fd);
+	close(data->stdout_fd);
+	close(data->close_me[0]);
+	close(data->close_me[1]);
+}
+
+int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock)
+{
+	struct dog_data data;
+	int in_fds[2], out_fds[2], pid, n, err;
+	char pid_buf[sizeof("nnnnnnn\0")], c;
+	char *pid_args[] = { "/usr/bin/uml_watchdog", "-pid", pid_buf, NULL };
+	char *mconsole_args[] = { "/usr/bin/uml_watchdog", "-mconsole", NULL,
+				  NULL };
+	char **args = NULL;
+
+	err = os_pipe(in_fds, 1, 0);
+	if (err < 0) {
+		printk("harddog_open - os_pipe failed, err = %d\n", -err);
+		goto out;
+	}
+
+	err = os_pipe(out_fds, 1, 0);
+	if (err < 0) {
+		printk("harddog_open - os_pipe failed, err = %d\n", -err);
+		goto out_close_in;
+	}
+
+	data.stdin_fd = out_fds[0];
+	data.stdout_fd = in_fds[1];
+	data.close_me[0] = out_fds[1];
+	data.close_me[1] = in_fds[0];
+
+	if (sock != NULL) {
+		mconsole_args[2] = sock;
+		args = mconsole_args;
+	}
+	else {
+		/* XXX The os_getpid() is not SMP correct */
+		sprintf(pid_buf, "%d", os_getpid());
+		args = pid_args;
+	}
+
+	pid = run_helper(pre_exec, &data, args);
+
+	close(out_fds[0]);
+	close(in_fds[1]);
+
+	if (pid < 0) {
+		err = -pid;
+		printk("harddog_open - run_helper failed, errno = %d\n", -err);
+		goto out_close_out;
+	}
+
+	n = read(in_fds[0], &c, sizeof(c));
+	if (n == 0) {
+		printk("harddog_open - EOF on watchdog pipe\n");
+		helper_wait(pid);
+		err = -EIO;
+		goto out_close_out;
+	}
+	else if (n < 0) {
+		printk("harddog_open - read of watchdog pipe failed, "
+		       "err = %d\n", errno);
+		helper_wait(pid);
+		err = n;
+		goto out_close_out;
+	}
+	*in_fd_ret = in_fds[0];
+	*out_fd_ret = out_fds[1];
+	return 0;
+
+ out_close_in:
+	close(in_fds[0]);
+	close(in_fds[1]);
+ out_close_out:
+	close(out_fds[0]);
+	close(out_fds[1]);
+ out:
+	return err;
+}
+
+void stop_watchdog(int in_fd, int out_fd)
+{
+	close(in_fd);
+	close(out_fd);
+}
+
+int ping_watchdog(int fd)
+{
+	int n;
+	char c = '\n';
+
+	n = write(fd, &c, sizeof(c));
+	if (n != sizeof(c)) {
+		printk("ping_watchdog - write failed, ret = %d, err = %d\n",
+		       n, errno);
+		if (n < 0)
+			return n;
+		return -EIO;
+	}
+	return 1;
+
+}

linux-7.2.3/arch/um/drivers/harddog_user_exp.c

@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/export.h>
+#include "harddog.h"
+
+#if IS_MODULE(CONFIG_UML_WATCHDOG)
+EXPORT_SYMBOL(start_watchdog);
+EXPORT_SYMBOL(stop_watchdog);
+EXPORT_SYMBOL(ping_watchdog);
+#endif

linux-7.2.3/arch/um/drivers/hostaudio_kern.c

@@ -0,0 +1,353 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 Steve Schmidtke
+ */
+
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/sound.h>
+#include <linux/soundcard.h>
+#include <linux/mutex.h>
+#include <linux/uaccess.h>
+#include <init.h>
+#include <os.h>
+
+struct hostaudio_state {
+	int fd;
+};
+
+struct hostmixer_state {
+	int fd;
+};
+
+#define HOSTAUDIO_DEV_DSP "/dev/sound/dsp"
+#define HOSTAUDIO_DEV_MIXER "/dev/sound/mixer"
+
+/*
+ * Changed either at boot time or module load time.  At boot, this is
+ * single-threaded; at module load, multiple modules would each have
+ * their own copy of these variables.
+ */
+static char *dsp = HOSTAUDIO_DEV_DSP;
+static char *mixer = HOSTAUDIO_DEV_MIXER;
+
+#define DSP_HELP \
+"    This is used to specify the host dsp device to the hostaudio driver.\n" \
+"    The default is \"" HOSTAUDIO_DEV_DSP "\".\n\n"
+
+#define MIXER_HELP \
+"    This is used to specify the host mixer device to the hostaudio driver.\n"\
+"    The default is \"" HOSTAUDIO_DEV_MIXER "\".\n\n"
+
+module_param(dsp, charp, 0644);
+MODULE_PARM_DESC(dsp, DSP_HELP);
+module_param(mixer, charp, 0644);
+MODULE_PARM_DESC(mixer, MIXER_HELP);
+
+#ifndef MODULE
+static int set_dsp(char *name, int *add)
+{
+	*add = 0;
+	dsp = name;
+	return 0;
+}
+
+__uml_setup("dsp=", set_dsp, "dsp=<dsp device>\n" DSP_HELP);
+
+static int set_mixer(char *name, int *add)
+{
+	*add = 0;
+	mixer = name;
+	return 0;
+}
+
+__uml_setup("mixer=", set_mixer, "mixer=<mixer device>\n" MIXER_HELP);
+#endif
+
+static DEFINE_MUTEX(hostaudio_mutex);
+
+/* /dev/dsp file operations */
+
+static ssize_t hostaudio_read(struct file *file, char __user *buffer,
+			      size_t count, loff_t *ppos)
+{
+	struct hostaudio_state *state = file->private_data;
+	void *kbuf;
+	int err;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostaudio: read called, count = %d\n", count);
+#endif
+
+	kbuf = kmalloc(count, GFP_KERNEL);
+	if (kbuf == NULL)
+		return -ENOMEM;
+
+	err = os_read_file(state->fd, kbuf, count);
+	if (err < 0)
+		goto out;
+
+	if (copy_to_user(buffer, kbuf, err))
+		err = -EFAULT;
+
+out:
+	kfree(kbuf);
+	return err;
+}
+
+static ssize_t hostaudio_write(struct file *file, const char __user *buffer,
+			       size_t count, loff_t *ppos)
+{
+	struct hostaudio_state *state = file->private_data;
+	void *kbuf;
+	int err;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostaudio: write called, count = %d\n", count);
+#endif
+
+	kbuf = memdup_user(buffer, count);
+	if (IS_ERR(kbuf))
+		return PTR_ERR(kbuf);
+
+	err = os_write_file(state->fd, kbuf, count);
+	if (err < 0)
+		goto out;
+	*ppos += err;
+
+ out:
+	kfree(kbuf);
+	return err;
+}
+
+static __poll_t hostaudio_poll(struct file *file,
+				struct poll_table_struct *wait)
+{
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostaudio: poll called (unimplemented)\n");
+#endif
+
+	return 0;
+}
+
+static long hostaudio_ioctl(struct file *file,
+			   unsigned int cmd, unsigned long arg)
+{
+	struct hostaudio_state *state = file->private_data;
+	unsigned long data = 0;
+	int err;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostaudio: ioctl called, cmd = %u\n", cmd);
+#endif
+	switch(cmd){
+	case SNDCTL_DSP_SPEED:
+	case SNDCTL_DSP_STEREO:
+	case SNDCTL_DSP_GETBLKSIZE:
+	case SNDCTL_DSP_CHANNELS:
+	case SNDCTL_DSP_SUBDIVIDE:
+	case SNDCTL_DSP_SETFRAGMENT:
+		if (get_user(data, (int __user *) arg))
+			return -EFAULT;
+		break;
+	default:
+		break;
+	}
+
+	err = os_ioctl_generic(state->fd, cmd, (unsigned long) &data);
+
+	switch(cmd){
+	case SNDCTL_DSP_SPEED:
+	case SNDCTL_DSP_STEREO:
+	case SNDCTL_DSP_GETBLKSIZE:
+	case SNDCTL_DSP_CHANNELS:
+	case SNDCTL_DSP_SUBDIVIDE:
+	case SNDCTL_DSP_SETFRAGMENT:
+		if (put_user(data, (int __user *) arg))
+			return -EFAULT;
+		break;
+	default:
+		break;
+	}
+
+	return err;
+}
+
+static int hostaudio_open(struct inode *inode, struct file *file)
+{
+	struct hostaudio_state *state;
+	int r = 0, w = 0;
+	int ret;
+
+#ifdef DEBUG
+	kernel_param_lock(THIS_MODULE);
+	printk(KERN_DEBUG "hostaudio: open called (host: %s)\n", dsp);
+	kernel_param_unlock(THIS_MODULE);
+#endif
+
+	state = kmalloc_obj(struct hostaudio_state);
+	if (state == NULL)
+		return -ENOMEM;
+
+	if (file->f_mode & FMODE_READ)
+		r = 1;
+	if (file->f_mode & FMODE_WRITE)
+		w = 1;
+
+	kernel_param_lock(THIS_MODULE);
+	mutex_lock(&hostaudio_mutex);
+	ret = os_open_file(dsp, of_set_rw(OPENFLAGS(), r, w), 0);
+	mutex_unlock(&hostaudio_mutex);
+	kernel_param_unlock(THIS_MODULE);
+
+	if (ret < 0) {
+		kfree(state);
+		return ret;
+	}
+	state->fd = ret;
+	file->private_data = state;
+	return 0;
+}
+
+static int hostaudio_release(struct inode *inode, struct file *file)
+{
+	struct hostaudio_state *state = file->private_data;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostaudio: release called\n");
+#endif
+	os_close_file(state->fd);
+	kfree(state);
+
+	return 0;
+}
+
+/* /dev/mixer file operations */
+
+static long hostmixer_ioctl_mixdev(struct file *file,
+				  unsigned int cmd, unsigned long arg)
+{
+	struct hostmixer_state *state = file->private_data;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostmixer: ioctl called\n");
+#endif
+
+	return os_ioctl_generic(state->fd, cmd, arg);
+}
+
+static int hostmixer_open_mixdev(struct inode *inode, struct file *file)
+{
+	struct hostmixer_state *state;
+	int r = 0, w = 0;
+	int ret;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostmixer: open called (host: %s)\n", mixer);
+#endif
+
+	state = kmalloc_obj(struct hostmixer_state);
+	if (state == NULL)
+		return -ENOMEM;
+
+	if (file->f_mode & FMODE_READ)
+		r = 1;
+	if (file->f_mode & FMODE_WRITE)
+		w = 1;
+
+	kernel_param_lock(THIS_MODULE);
+	mutex_lock(&hostaudio_mutex);
+	ret = os_open_file(mixer, of_set_rw(OPENFLAGS(), r, w), 0);
+	mutex_unlock(&hostaudio_mutex);
+	kernel_param_unlock(THIS_MODULE);
+
+	if (ret < 0) {
+		kernel_param_lock(THIS_MODULE);
+		printk(KERN_ERR "hostaudio_open_mixdev failed to open '%s', "
+		       "err = %d\n", dsp, -ret);
+		kernel_param_unlock(THIS_MODULE);
+		kfree(state);
+		return ret;
+	}
+
+	file->private_data = state;
+	return 0;
+}
+
+static int hostmixer_release(struct inode *inode, struct file *file)
+{
+	struct hostmixer_state *state = file->private_data;
+
+#ifdef DEBUG
+	printk(KERN_DEBUG "hostmixer: release called\n");
+#endif
+
+	os_close_file(state->fd);
+	kfree(state);
+
+	return 0;
+}
+
+/* kernel module operations */
+
+static const struct file_operations hostaudio_fops = {
+	.owner          = THIS_MODULE,
+	.read           = hostaudio_read,
+	.write          = hostaudio_write,
+	.poll           = hostaudio_poll,
+	.unlocked_ioctl	= hostaudio_ioctl,
+	.compat_ioctl	= compat_ptr_ioctl,
+	.mmap           = NULL,
+	.open           = hostaudio_open,
+	.release        = hostaudio_release,
+};
+
+static const struct file_operations hostmixer_fops = {
+	.owner          = THIS_MODULE,
+	.unlocked_ioctl	= hostmixer_ioctl_mixdev,
+	.open           = hostmixer_open_mixdev,
+	.release        = hostmixer_release,
+};
+
+static struct {
+	int dev_audio;
+	int dev_mixer;
+} module_data;
+
+MODULE_AUTHOR("Steve Schmidtke");
+MODULE_DESCRIPTION("UML Audio Relay");
+MODULE_LICENSE("GPL");
+
+static int __init hostaudio_init_module(void)
+{
+	kernel_param_lock(THIS_MODULE);
+	printk(KERN_INFO "UML Audio Relay (host dsp = %s, host mixer = %s)\n",
+	       dsp, mixer);
+	kernel_param_unlock(THIS_MODULE);
+
+	module_data.dev_audio = register_sound_dsp(&hostaudio_fops, -1);
+	if (module_data.dev_audio < 0) {
+		printk(KERN_ERR "hostaudio: couldn't register DSP device!\n");
+		return -ENODEV;
+	}
+
+	module_data.dev_mixer = register_sound_mixer(&hostmixer_fops, -1);
+	if (module_data.dev_mixer < 0) {
+		printk(KERN_ERR "hostmixer: couldn't register mixer "
+		       "device!\n");
+		unregister_sound_dsp(module_data.dev_audio);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static void __exit hostaudio_cleanup_module (void)
+{
+	unregister_sound_mixer(module_data.dev_mixer);
+	unregister_sound_dsp(module_data.dev_audio);
+}
+
+module_init(hostaudio_init_module);
+module_exit(hostaudio_cleanup_module);

linux-7.2.3/arch/um/drivers/line.c

@@ -0,0 +1,771 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <linux/irqreturn.h>
+#include <linux/kd.h>
+#include <linux/sched/signal.h>
+#include <linux/slab.h>
+
+#include "chan.h"
+#include <irq_kern.h>
+#include <irq_user.h>
+#include <kern_util.h>
+#include <os.h>
+
+#define LINE_BUFSIZE 4096
+
+static irqreturn_t line_interrupt(int irq, void *data)
+{
+	struct chan *chan = data;
+	struct line *line = chan->line;
+
+	if (line)
+		chan_interrupt(line, irq);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * Returns the free space inside the ring buffer of this line.
+ *
+ * Should be called while holding line->lock (this does not modify data).
+ */
+static unsigned int write_room(struct line *line)
+{
+	int n;
+
+	if (line->buffer == NULL)
+		return LINE_BUFSIZE - 1;
+
+	/* This is for the case where the buffer is wrapped! */
+	n = line->head - line->tail;
+
+	if (n <= 0)
+		n += LINE_BUFSIZE; /* The other case */
+	return n - 1;
+}
+
+unsigned int line_write_room(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+	unsigned long flags;
+	unsigned int room;
+
+	spin_lock_irqsave(&line->lock, flags);
+	room = write_room(line);
+	spin_unlock_irqrestore(&line->lock, flags);
+
+	return room;
+}
+
+unsigned int line_chars_in_buffer(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+	unsigned long flags;
+	unsigned int ret;
+
+	spin_lock_irqsave(&line->lock, flags);
+	/* write_room subtracts 1 for the needed NULL, so we readd it.*/
+	ret = LINE_BUFSIZE - (write_room(line) + 1);
+	spin_unlock_irqrestore(&line->lock, flags);
+
+	return ret;
+}
+
+/*
+ * This copies the content of buf into the circular buffer associated with
+ * this line.
+ * The return value is the number of characters actually copied, i.e. the ones
+ * for which there was space: this function is not supposed to ever flush out
+ * the circular buffer.
+ *
+ * Must be called while holding line->lock!
+ */
+static int buffer_data(struct line *line, const u8 *buf, size_t len)
+{
+	int end, room;
+
+	if (line->buffer == NULL) {
+		line->buffer = kmalloc(LINE_BUFSIZE, GFP_ATOMIC);
+		if (line->buffer == NULL) {
+			printk(KERN_ERR "buffer_data - atomic allocation "
+			       "failed\n");
+			return 0;
+		}
+		line->head = line->buffer;
+		line->tail = line->buffer;
+	}
+
+	room = write_room(line);
+	len = (len > room) ? room : len;
+
+	end = line->buffer + LINE_BUFSIZE - line->tail;
+
+	if (len < end) {
+		memcpy(line->tail, buf, len);
+		line->tail += len;
+	}
+	else {
+		/* The circular buffer is wrapping */
+		memcpy(line->tail, buf, end);
+		buf += end;
+		memcpy(line->buffer, buf, len - end);
+		line->tail = line->buffer + len - end;
+	}
+
+	return len;
+}
+
+/*
+ * Flushes the ring buffer to the output channels. That is, write_chan is
+ * called, passing it line->head as buffer, and an appropriate count.
+ *
+ * On exit, returns 1 when the buffer is empty,
+ * 0 when the buffer is not empty on exit,
+ * and -errno when an error occurred.
+ *
+ * Must be called while holding line->lock!*/
+static int flush_buffer(struct line *line)
+{
+	int n, count;
+
+	if ((line->buffer == NULL) || (line->head == line->tail))
+		return 1;
+
+	if (line->tail < line->head) {
+		/* line->buffer + LINE_BUFSIZE is the end of the buffer! */
+		count = line->buffer + LINE_BUFSIZE - line->head;
+
+		n = write_chan(line->chan_out, line->head, count,
+			       line->write_irq);
+		if (n < 0)
+			return n;
+		if (n == count) {
+			/*
+			 * We have flushed from ->head to buffer end, now we
+			 * must flush only from the beginning to ->tail.
+			 */
+			line->head = line->buffer;
+		} else {
+			line->head += n;
+			return 0;
+		}
+	}
+
+	count = line->tail - line->head;
+	n = write_chan(line->chan_out, line->head, count,
+		       line->write_irq);
+
+	if (n < 0)
+		return n;
+
+	line->head += n;
+	return line->head == line->tail;
+}
+
+void line_flush_buffer(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&line->lock, flags);
+	flush_buffer(line);
+	spin_unlock_irqrestore(&line->lock, flags);
+}
+
+/*
+ * We map both ->flush_chars and ->put_char (which go in pair) onto
+ * ->flush_buffer and ->write. Hope it's not that bad.
+ */
+void line_flush_chars(struct tty_struct *tty)
+{
+	line_flush_buffer(tty);
+}
+
+ssize_t line_write(struct tty_struct *tty, const u8 *buf, size_t len)
+{
+	struct line *line = tty->driver_data;
+	unsigned long flags;
+	int n, ret = 0;
+
+	spin_lock_irqsave(&line->lock, flags);
+	if (line->head != line->tail)
+		ret = buffer_data(line, buf, len);
+	else {
+		n = write_chan(line->chan_out, buf, len,
+			       line->write_irq);
+		if (n < 0) {
+			ret = n;
+			goto out_up;
+		}
+
+		len -= n;
+		ret += n;
+		if (len > 0)
+			ret += buffer_data(line, buf + n, len);
+	}
+out_up:
+	spin_unlock_irqrestore(&line->lock, flags);
+	return ret;
+}
+
+void line_throttle(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+
+	deactivate_chan(line->chan_in, line->read_irq);
+	line->throttled = 1;
+}
+
+void line_unthrottle(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+
+	line->throttled = 0;
+	chan_interrupt(line, line->read_irq);
+}
+
+static irqreturn_t line_write_interrupt(int irq, void *data)
+{
+	struct chan *chan = data;
+	struct line *line = chan->line;
+	int err;
+
+	/*
+	 * Interrupts are disabled here because genirq keep irqs disabled when
+	 * calling the action handler.
+	 */
+
+	spin_lock(&line->lock);
+	err = flush_buffer(line);
+	if (err == 0) {
+		spin_unlock(&line->lock);
+		return IRQ_NONE;
+	} else if ((err < 0) && (err != -EAGAIN)) {
+		line->head = line->buffer;
+		line->tail = line->buffer;
+	}
+	spin_unlock(&line->lock);
+
+	tty_port_tty_wakeup(&line->port);
+
+	return IRQ_HANDLED;
+}
+
+int line_setup_irq(int fd, int input, int output, struct line *line, void *data)
+{
+	const struct line_driver *driver = line->driver;
+	int err;
+
+	if (input) {
+		err = um_request_irq(UM_IRQ_ALLOC, fd, IRQ_READ,
+				     line_interrupt, 0,
+				     driver->read_irq_name, data);
+		if (err < 0)
+			return err;
+
+		line->read_irq = err;
+	}
+
+	if (output) {
+		err = um_request_irq(UM_IRQ_ALLOC, fd, IRQ_WRITE,
+				     line_write_interrupt, 0,
+				     driver->write_irq_name, data);
+		if (err < 0)
+			return err;
+
+		line->write_irq = err;
+	}
+
+	return 0;
+}
+
+static int line_activate(struct tty_port *port, struct tty_struct *tty)
+{
+	int ret;
+	struct line *line = tty->driver_data;
+
+	ret = enable_chan(line);
+	if (ret)
+		return ret;
+
+	if (!line->sigio) {
+		chan_enable_winch(line->chan_out, port);
+		line->sigio = 1;
+	}
+
+	chan_window_size(line, &tty->winsize.ws_row,
+		&tty->winsize.ws_col);
+
+	return 0;
+}
+
+static void unregister_winch(struct tty_struct *tty);
+
+static void line_destruct(struct tty_port *port)
+{
+	struct tty_struct *tty = tty_port_tty_get(port);
+	struct line *line = tty->driver_data;
+
+	if (line->sigio) {
+		unregister_winch(tty);
+		line->sigio = 0;
+	}
+}
+
+static const struct tty_port_operations line_port_ops = {
+	.activate = line_activate,
+	.destruct = line_destruct,
+};
+
+int line_open(struct tty_struct *tty, struct file *filp)
+{
+	struct line *line = tty->driver_data;
+
+	return tty_port_open(&line->port, tty, filp);
+}
+
+int line_install(struct tty_driver *driver, struct tty_struct *tty,
+		 struct line *line)
+{
+	int ret;
+
+	ret = tty_standard_install(driver, tty);
+	if (ret)
+		return ret;
+
+	tty->driver_data = line;
+
+	return 0;
+}
+
+void line_close(struct tty_struct *tty, struct file * filp)
+{
+	struct line *line = tty->driver_data;
+
+	tty_port_close(&line->port, tty, filp);
+}
+
+void line_hangup(struct tty_struct *tty)
+{
+	struct line *line = tty->driver_data;
+
+	tty_port_hangup(&line->port);
+}
+
+void close_lines(struct line *lines, int nlines)
+{
+	int i;
+
+	for(i = 0; i < nlines; i++)
+		close_chan(&lines[i]);
+}
+
+int setup_one_line(struct line *lines, int n, char *init,
+		   const struct chan_opts *opts, char **error_out)
+{
+	struct line *line = &lines[n];
+	struct tty_driver *driver = line->driver->driver;
+	int err = -EINVAL;
+
+	if (line->port.count) {
+		*error_out = "Device is already open";
+		goto out;
+	}
+
+	if (!strcmp(init, "none")) {
+		if (line->valid) {
+			line->valid = 0;
+			kfree(line->init_str);
+			tty_unregister_device(driver, n);
+			parse_chan_pair(NULL, line, n, opts, error_out);
+			err = 0;
+		}
+		*error_out = "configured as 'none'";
+	} else {
+		char *new = kstrdup(init, GFP_KERNEL);
+		if (!new) {
+			*error_out = "Failed to allocate memory";
+			return -ENOMEM;
+		}
+		if (line->valid) {
+			tty_unregister_device(driver, n);
+			kfree(line->init_str);
+		}
+		line->init_str = new;
+		line->valid = 1;
+		err = parse_chan_pair(new, line, n, opts, error_out);
+		if (!err) {
+			struct device *d = tty_port_register_device(&line->port,
+					driver, n, NULL);
+			if (IS_ERR(d)) {
+				*error_out = "Failed to register device";
+				err = PTR_ERR(d);
+				parse_chan_pair(NULL, line, n, opts, error_out);
+			}
+		}
+		if (err) {
+			*error_out = "failed to parse channel pair";
+			line->init_str = NULL;
+			line->valid = 0;
+			kfree(new);
+		}
+	}
+out:
+	return err;
+}
+
+/*
+ * Common setup code for both startup command line and mconsole initialization.
+ * @lines contains the array (of size @num) to modify;
+ * @init is the setup string;
+ * @error_out is an error string in the case of failure;
+ */
+
+int line_setup(char **conf, unsigned int num, char **def,
+	       char *init, char *name)
+{
+	char *error;
+
+	if (*init == '=') {
+		/*
+		 * We said con=/ssl= instead of con#=, so we are configuring all
+		 * consoles at once.
+		 */
+		*def = init + 1;
+	} else {
+		char *end;
+		unsigned n = simple_strtoul(init, &end, 0);
+
+		if (*end != '=') {
+			error = "Couldn't parse device number";
+			goto out;
+		}
+		if (n >= num) {
+			error = "Device number out of range";
+			goto out;
+		}
+		conf[n] = end + 1;
+	}
+	return 0;
+
+out:
+	printk(KERN_ERR "Failed to set up %s with "
+	       "configuration string \"%s\" : %s\n", name, init, error);
+	return -EINVAL;
+}
+
+int line_config(struct line *lines, unsigned int num, char *str,
+		const struct chan_opts *opts, char **error_out)
+{
+	char *end;
+	int n;
+
+	if (*str == '=') {
+		*error_out = "Can't configure all devices from mconsole";
+		return -EINVAL;
+	}
+
+	n = simple_strtoul(str, &end, 0);
+	if (*end++ != '=') {
+		*error_out = "Couldn't parse device number";
+		return -EINVAL;
+	}
+	if (n >= num) {
+		*error_out = "Device number out of range";
+		return -EINVAL;
+	}
+
+	return setup_one_line(lines, n, end, opts, error_out);
+}
+
+int line_get_config(char *name, struct line *lines, unsigned int num, char *str,
+		    int size, char **error_out)
+{
+	struct line *line;
+	char *end;
+	int dev, n = 0;
+
+	dev = simple_strtoul(name, &end, 0);
+	if ((*end != '\0') || (end == name)) {
+		*error_out = "line_get_config failed to parse device number";
+		return 0;
+	}
+
+	if ((dev < 0) || (dev >= num)) {
+		*error_out = "device number out of range";
+		return 0;
+	}
+
+	line = &lines[dev];
+
+	if (!line->valid)
+		CONFIG_CHUNK(str, size, n, "none", 1);
+	else {
+		struct tty_struct *tty = tty_port_tty_get(&line->port);
+		if (tty == NULL) {
+			CONFIG_CHUNK(str, size, n, line->init_str, 1);
+		} else {
+			n = chan_config_string(line, str, size, error_out);
+			tty_kref_put(tty);
+		}
+	}
+
+	return n;
+}
+
+int line_id(char **str, int *start_out, int *end_out)
+{
+	char *end;
+	int n;
+
+	n = simple_strtoul(*str, &end, 0);
+	if ((*end != '\0') || (end == *str))
+		return -1;
+
+	*str = end;
+	*start_out = n;
+	*end_out = n;
+	return n;
+}
+
+int line_remove(struct line *lines, unsigned int num, int n, char **error_out)
+{
+	if (n >= num) {
+		*error_out = "Device number out of range";
+		return -EINVAL;
+	}
+	return setup_one_line(lines, n, "none", NULL, error_out);
+}
+
+int register_lines(struct line_driver *line_driver,
+		   const struct tty_operations *ops,
+		   struct line *lines, int nlines)
+{
+	struct tty_driver *driver;
+	int err;
+	int i;
+
+	driver = tty_alloc_driver(nlines, TTY_DRIVER_REAL_RAW |
+			TTY_DRIVER_DYNAMIC_DEV);
+	if (IS_ERR(driver))
+		return PTR_ERR(driver);
+
+	driver->driver_name = line_driver->name;
+	driver->name = line_driver->device_name;
+	driver->major = line_driver->major;
+	driver->minor_start = line_driver->minor_start;
+	driver->type = line_driver->type;
+	driver->subtype = line_driver->subtype;
+	driver->init_termios = tty_std_termios;
+
+	for (i = 0; i < nlines; i++) {
+		tty_port_init(&lines[i].port);
+		lines[i].port.ops = &line_port_ops;
+		spin_lock_init(&lines[i].lock);
+		lines[i].driver = line_driver;
+		INIT_LIST_HEAD(&lines[i].chan_list);
+	}
+	tty_set_operations(driver, ops);
+
+	err = tty_register_driver(driver);
+	if (err) {
+		printk(KERN_ERR "register_lines : can't register %s driver\n",
+		       line_driver->name);
+		tty_driver_kref_put(driver);
+		for (i = 0; i < nlines; i++)
+			tty_port_destroy(&lines[i].port);
+		return err;
+	}
+
+	line_driver->driver = driver;
+	mconsole_register_dev(&line_driver->mc);
+	return 0;
+}
+
+static DEFINE_SPINLOCK(winch_handler_lock);
+static LIST_HEAD(winch_handlers);
+
+struct winch {
+	struct list_head list;
+	int fd;
+	int tty_fd;
+	int pid;
+	struct tty_port *port;
+	unsigned long stack;
+	struct work_struct work;
+};
+
+static void __free_winch(struct work_struct *work)
+{
+	struct winch *winch = container_of(work, struct winch, work);
+	um_free_irq(WINCH_IRQ, winch);
+
+	if (winch->pid != -1)
+		os_kill_process(winch->pid, 1);
+	if (winch->stack != 0)
+		free_stack(winch->stack, 0);
+	kfree(winch);
+}
+
+static void free_winch(struct winch *winch)
+{
+	int fd = winch->fd;
+	winch->fd = -1;
+	if (fd != -1)
+		os_close_file(fd);
+	__free_winch(&winch->work);
+}
+
+static irqreturn_t winch_interrupt(int irq, void *data)
+{
+	struct winch *winch = data;
+	struct tty_struct *tty;
+	struct line *line;
+	int fd = winch->fd;
+	int err;
+	char c;
+	struct pid *pgrp;
+
+	if (fd != -1) {
+		err = generic_read(fd, &c, NULL);
+		/* A read of 2 means the winch thread failed and has warned */
+		if (err < 0 || (err == 1 && c == 2)) {
+			if (err != -EAGAIN) {
+				winch->fd = -1;
+				list_del(&winch->list);
+				os_close_file(fd);
+				if (err < 0) {
+					printk(KERN_ERR "winch_interrupt : read failed, errno = %d\n",
+					       -err);
+					printk(KERN_ERR "fd %d is losing SIGWINCH support\n",
+					       winch->tty_fd);
+				}
+				INIT_WORK(&winch->work, __free_winch);
+				schedule_work(&winch->work);
+				return IRQ_HANDLED;
+			}
+			goto out;
+		}
+	}
+	tty = tty_port_tty_get(winch->port);
+	if (tty != NULL) {
+		line = tty->driver_data;
+		if (line != NULL) {
+			chan_window_size(line, &tty->winsize.ws_row,
+					 &tty->winsize.ws_col);
+			pgrp = tty_get_pgrp(tty);
+			if (pgrp)
+				kill_pgrp(pgrp, SIGWINCH, 1);
+			put_pid(pgrp);
+		}
+		tty_kref_put(tty);
+	}
+ out:
+	return IRQ_HANDLED;
+}
+
+void register_winch_irq(int fd, int tty_fd, int pid, struct tty_port *port,
+			unsigned long stack)
+{
+	struct winch *winch;
+
+	winch = kmalloc_obj(*winch);
+	if (winch == NULL) {
+		printk(KERN_ERR "register_winch_irq - kmalloc failed\n");
+		goto cleanup;
+	}
+
+	*winch = ((struct winch) { .fd  	= fd,
+				   .tty_fd 	= tty_fd,
+				   .pid  	= pid,
+				   .port 	= port,
+				   .stack	= stack });
+
+	spin_lock(&winch_handler_lock);
+	list_add(&winch->list, &winch_handlers);
+	spin_unlock(&winch_handler_lock);
+
+	if (um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt,
+			   IRQF_SHARED, "winch", winch) < 0) {
+		printk(KERN_ERR "register_winch_irq - failed to register "
+		       "IRQ\n");
+		spin_lock(&winch_handler_lock);
+		list_del(&winch->list);
+		spin_unlock(&winch_handler_lock);
+		goto out_free;
+	}
+
+	return;
+
+ out_free:
+	kfree(winch);
+ cleanup:
+	os_kill_process(pid, 1);
+	os_close_file(fd);
+	if (stack != 0)
+		free_stack(stack, 0);
+}
+
+static void unregister_winch(struct tty_struct *tty)
+{
+	struct list_head *ele, *next;
+	struct winch *winch;
+	struct tty_struct *wtty;
+
+	spin_lock(&winch_handler_lock);
+
+	list_for_each_safe(ele, next, &winch_handlers) {
+		winch = list_entry(ele, struct winch, list);
+		wtty = tty_port_tty_get(winch->port);
+		if (wtty == tty) {
+			list_del(&winch->list);
+			spin_unlock(&winch_handler_lock);
+			free_winch(winch);
+			break;
+		}
+		tty_kref_put(wtty);
+	}
+	spin_unlock(&winch_handler_lock);
+}
+
+static void winch_cleanup(void)
+{
+	struct winch *winch;
+
+	spin_lock(&winch_handler_lock);
+	while ((winch = list_first_entry_or_null(&winch_handlers,
+						 struct winch, list))) {
+		list_del(&winch->list);
+		spin_unlock(&winch_handler_lock);
+
+		free_winch(winch);
+
+		spin_lock(&winch_handler_lock);
+	}
+
+	spin_unlock(&winch_handler_lock);
+}
+__uml_exitcall(winch_cleanup);
+
+char *add_xterm_umid(char *base)
+{
+	char *umid, *title;
+	int len;
+
+	umid = get_umid();
+	if (*umid == '\0')
+		return base;
+
+	len = strlen(base) + strlen(" ()") + strlen(umid) + 1;
+	title = kmalloc(len, GFP_KERNEL);
+	if (title == NULL) {
+		printk(KERN_ERR "Failed to allocate buffer for xterm title\n");
+		return base;
+	}
+
+	snprintf(title, len, "%s (%s)", base, umid);
+	return title;
+}

linux-7.2.3/arch/um/drivers/line.h

@@ -0,0 +1,96 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __LINE_H__
+#define __LINE_H__
+
+#include <linux/list.h>
+#include <linux/workqueue.h>
+#include <linux/tty.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/mutex.h>
+#include "chan_user.h"
+#include "mconsole_kern.h"
+
+/* There's only two modifiable fields in this - .mc.list and .driver */
+struct line_driver {
+	const char *name;
+	const char *device_name;
+	const short major;
+	const short minor_start;
+	const short type;
+	const short subtype;
+	const char *read_irq_name;
+	const char *write_irq_name;
+	struct mc_device mc;
+	struct tty_driver *driver;
+};
+
+struct line {
+	struct tty_port port;
+	int valid;
+
+	int read_irq, write_irq;
+
+	char *init_str;
+	struct list_head chan_list;
+	struct chan *chan_in, *chan_out;
+
+	/*This lock is actually, mostly, local to*/
+	spinlock_t lock;
+	int throttled;
+	/* Yes, this is a real circular buffer.
+	 * XXX: And this should become a struct kfifo!
+	 *
+	 * buffer points to a buffer allocated on demand, of length
+	 * LINE_BUFSIZE, head to the start of the ring, tail to the end.*/
+	u8 *buffer;
+	u8 *head;
+	u8 *tail;
+
+	int sigio;
+	struct delayed_work task;
+	const struct line_driver *driver;
+};
+
+extern void line_close(struct tty_struct *tty, struct file * filp);
+extern int line_open(struct tty_struct *tty, struct file *filp);
+extern int line_install(struct tty_driver *driver, struct tty_struct *tty,
+	struct line *line);
+extern void line_cleanup(struct tty_struct *tty);
+extern void line_hangup(struct tty_struct *tty);
+extern int line_setup(char **conf, unsigned nlines, char **def,
+		      char *init, char *name);
+extern ssize_t line_write(struct tty_struct *tty, const u8 *buf, size_t len);
+extern unsigned int line_chars_in_buffer(struct tty_struct *tty);
+extern void line_flush_buffer(struct tty_struct *tty);
+extern void line_flush_chars(struct tty_struct *tty);
+extern unsigned int line_write_room(struct tty_struct *tty);
+extern void line_throttle(struct tty_struct *tty);
+extern void line_unthrottle(struct tty_struct *tty);
+
+extern char *add_xterm_umid(char *base);
+extern int line_setup_irq(int fd, int input, int output, struct line *line,
+			  void *data);
+extern void line_close_chan(struct line *line);
+extern int register_lines(struct line_driver *line_driver,
+			  const struct tty_operations *driver,
+			  struct line *lines, int nlines);
+extern int setup_one_line(struct line *lines, int n, char *init,
+			  const struct chan_opts *opts, char **error_out);
+extern void close_lines(struct line *lines, int nlines);
+
+extern int line_config(struct line *lines, unsigned int sizeof_lines,
+		       char *str, const struct chan_opts *opts,
+		       char **error_out);
+extern int line_id(char **str, int *start_out, int *end_out);
+extern int line_remove(struct line *lines, unsigned int sizeof_lines, int n,
+		       char **error_out);
+extern int line_get_config(char *dev, struct line *lines,
+			   unsigned int sizeof_lines, char *str,
+			   int size, char **error_out);
+
+#endif

linux-7.2.3/arch/um/drivers/mconsole.h

@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org)
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#ifndef __MCONSOLE_H__
+#define __MCONSOLE_H__
+
+#ifdef __UM_HOST__
+#include <stdint.h>
+#define u32 uint32_t
+#endif
+
+#include <sysdep/ptrace.h>
+
+#define MCONSOLE_MAGIC (0xcafebabe)
+#define MCONSOLE_MAX_DATA (512)
+#define MCONSOLE_VERSION 2
+
+struct mconsole_request {
+	u32 magic;
+	u32 version;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mconsole_reply {
+	u32 err;
+	u32 more;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mconsole_notify {
+	u32 magic;
+	u32 version;
+	enum { MCONSOLE_SOCKET, MCONSOLE_PANIC, MCONSOLE_HANG,
+	       MCONSOLE_USER_NOTIFY } type;
+	u32 len;
+	char data[MCONSOLE_MAX_DATA];
+};
+
+struct mc_request;
+
+enum mc_context { MCONSOLE_INTR, MCONSOLE_PROC };
+
+struct mconsole_command
+{
+	char *command;
+	void (*handler)(struct mc_request *req);
+	enum mc_context context;
+};
+
+struct mc_request
+{
+	int len;
+	int as_interrupt;
+
+	int originating_fd;
+	unsigned int originlen;
+	unsigned char origin[128];			/* sockaddr_un */
+
+	struct mconsole_request request;
+	struct mconsole_command *cmd;
+	struct uml_pt_regs regs;
+};
+
+extern char mconsole_socket_name[];
+
+extern int mconsole_unlink_socket(void);
+extern int mconsole_reply_len(struct mc_request *req, const char *reply,
+			      int len, int err, int more);
+extern int mconsole_reply(struct mc_request *req, const char *str, int err,
+			  int more);
+
+extern void mconsole_version(struct mc_request *req);
+extern void mconsole_help(struct mc_request *req);
+extern void mconsole_halt(struct mc_request *req);
+extern void mconsole_reboot(struct mc_request *req);
+extern void mconsole_config(struct mc_request *req);
+extern void mconsole_remove(struct mc_request *req);
+extern void mconsole_sysrq(struct mc_request *req);
+extern void mconsole_cad(struct mc_request *req);
+extern void mconsole_stop(struct mc_request *req);
+extern void mconsole_go(struct mc_request *req);
+extern void mconsole_log(struct mc_request *req);
+extern void mconsole_proc(struct mc_request *req);
+extern void mconsole_stack(struct mc_request *req);
+
+extern int mconsole_get_request(int fd, struct mc_request *req);
+extern int mconsole_notify(char *sock_name, int type, const void *data,
+			   int len);
+extern char *mconsole_notify_socket(void);
+extern void lock_notify(void);
+extern void unlock_notify(void);
+
+#endif

linux-7.2.3/arch/um/drivers/mconsole_kern.c

@@ -0,0 +1,871 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org)
+ * Copyright (C) 2001 - 2008 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <linux/console.h>
+#include <linux/ctype.h>
+#include <linux/string.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
+#include <linux/panic_notifier.h>
+#include <linux/reboot.h>
+#include <linux/sched/debug.h>
+#include <linux/proc_fs.h>
+#include <linux/slab.h>
+#include <linux/syscalls.h>
+#include <linux/utsname.h>
+#include <linux/socket.h>
+#include <linux/un.h>
+#include <linux/workqueue.h>
+#include <linux/mutex.h>
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/file.h>
+#include <linux/uaccess.h>
+#include <asm/switch_to.h>
+
+#include <init.h>
+#include <irq_kern.h>
+#include <irq_user.h>
+#include <kern_util.h>
+#include "mconsole.h"
+#include "mconsole_kern.h"
+#include <os.h>
+
+static struct vfsmount *proc_mnt = NULL;
+
+static int do_unlink_socket(struct notifier_block *notifier,
+			    unsigned long what, void *data)
+{
+	return mconsole_unlink_socket();
+}
+
+
+static struct notifier_block reboot_notifier = {
+	.notifier_call		= do_unlink_socket,
+	.priority		= 0,
+};
+
+/* Safe without explicit locking for now.  Tasklets provide their own
+ * locking, and the interrupt handler is safe because it can't interrupt
+ * itself and it can only happen on CPU 0.
+ */
+
+static LIST_HEAD(mc_requests);
+
+static void mc_work_proc(struct work_struct *unused)
+{
+	struct mconsole_entry *req;
+	unsigned long flags;
+
+	while (!list_empty(&mc_requests)) {
+		local_irq_save(flags);
+		req = list_entry(mc_requests.next, struct mconsole_entry, list);
+		list_del(&req->list);
+		local_irq_restore(flags);
+		req->request.cmd->handler(&req->request);
+		kfree(req);
+	}
+}
+
+static DECLARE_WORK(mconsole_work, mc_work_proc);
+
+static irqreturn_t mconsole_interrupt(int irq, void *dev_id)
+{
+	/* long to avoid size mismatch warnings from gcc */
+	long fd;
+	struct mconsole_entry *new;
+	static struct mc_request req;	/* that's OK */
+
+	fd = (long) dev_id;
+	while (mconsole_get_request(fd, &req)) {
+		if (req.cmd->context == MCONSOLE_INTR)
+			(*req.cmd->handler)(&req);
+		else {
+			new = kmalloc_obj(*new, GFP_NOWAIT);
+			if (new == NULL)
+				mconsole_reply(&req, "Out of memory", 1, 0);
+			else {
+				new->request = req;
+				new->request.regs = get_irq_regs()->regs;
+				list_add(&new->list, &mc_requests);
+			}
+		}
+	}
+	if (!list_empty(&mc_requests))
+		schedule_work(&mconsole_work);
+	return IRQ_HANDLED;
+}
+
+void mconsole_version(struct mc_request *req)
+{
+	char version[256];
+
+	sprintf(version, "%s %s %s %s %s", utsname()->sysname,
+		utsname()->nodename, utsname()->release, utsname()->version,
+		utsname()->machine);
+	mconsole_reply(req, version, 0, 0);
+}
+
+void mconsole_log(struct mc_request *req)
+{
+	int len;
+	char *ptr = req->request.data;
+
+	ptr += strlen("log ");
+
+	len = req->len - (ptr - req->request.data);
+	printk(KERN_WARNING "%.*s", len, ptr);
+	mconsole_reply(req, "", 0, 0);
+}
+
+void mconsole_proc(struct mc_request *req)
+{
+	struct vfsmount *mnt = proc_mnt;
+	char *buf;
+	int len;
+	struct file *file;
+	int first_chunk = 1;
+	char *ptr = req->request.data;
+	loff_t pos = 0;
+
+	ptr += strlen("proc");
+	ptr = skip_spaces(ptr);
+
+	if (!mnt) {
+		mconsole_reply(req, "Proc not available", 1, 0);
+		goto out;
+	}
+	file = file_open_root_mnt(mnt, ptr, O_RDONLY, 0);
+	if (IS_ERR(file)) {
+		mconsole_reply(req, "Failed to open file", 1, 0);
+		printk(KERN_ERR "open /proc/%s: %ld\n", ptr, PTR_ERR(file));
+		goto out;
+	}
+
+	buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+	if (buf == NULL) {
+		mconsole_reply(req, "Failed to allocate buffer", 1, 0);
+		goto out_fput;
+	}
+
+	do {
+		len = kernel_read(file, buf, PAGE_SIZE - 1, &pos);
+		if (len < 0) {
+			mconsole_reply(req, "Read of file failed", 1, 0);
+			goto out_free;
+		}
+		/* Begin the file content on his own line. */
+		if (first_chunk) {
+			mconsole_reply(req, "\n", 0, 1);
+			first_chunk = 0;
+		}
+		buf[len] = '\0';
+		mconsole_reply(req, buf, 0, (len > 0));
+	} while (len > 0);
+ out_free:
+	kfree(buf);
+ out_fput:
+	fput(file);
+ out: ;
+}
+
+#define UML_MCONSOLE_HELPTEXT \
+"Commands: \n\
+    version - Get kernel version \n\
+    help - Print this message \n\
+    halt - Halt UML \n\
+    reboot - Reboot UML \n\
+    config <dev>=<config> - Add a new device to UML;  \n\
+	same syntax as command line \n\
+    config <dev> - Query the configuration of a device \n\
+    remove <dev> - Remove a device from UML \n\
+    sysrq <letter> - Performs the SysRq action controlled by the letter \n\
+    cad - invoke the Ctrl-Alt-Del handler \n\
+    stop - pause the UML; it will do nothing until it receives a 'go' \n\
+    go - continue the UML after a 'stop' \n\
+    log <string> - make UML enter <string> into the kernel log\n\
+    proc <file> - returns the contents of the UML's /proc/<file>\n\
+    stack <pid> - returns the stack of the specified pid\n\
+"
+
+void mconsole_help(struct mc_request *req)
+{
+	mconsole_reply(req, UML_MCONSOLE_HELPTEXT, 0, 0);
+}
+
+void mconsole_halt(struct mc_request *req)
+{
+	mconsole_reply(req, "", 0, 0);
+	machine_halt();
+}
+
+void mconsole_reboot(struct mc_request *req)
+{
+	mconsole_reply(req, "", 0, 0);
+	machine_restart(NULL);
+}
+
+void mconsole_cad(struct mc_request *req)
+{
+	mconsole_reply(req, "", 0, 0);
+	ctrl_alt_del();
+}
+
+void mconsole_go(struct mc_request *req)
+{
+	mconsole_reply(req, "Not stopped", 1, 0);
+}
+
+void mconsole_stop(struct mc_request *req)
+{
+	block_signals();
+	os_set_fd_block(req->originating_fd, 1);
+	mconsole_reply(req, "stopped", 0, 0);
+	for (;;) {
+		if (!mconsole_get_request(req->originating_fd, req))
+			continue;
+		if (req->cmd->handler == mconsole_go)
+			break;
+		if (req->cmd->handler == mconsole_stop) {
+			mconsole_reply(req, "Already stopped", 1, 0);
+			continue;
+		}
+		if (req->cmd->handler == mconsole_sysrq) {
+			struct pt_regs *old_regs;
+			old_regs = set_irq_regs((struct pt_regs *)&req->regs);
+			mconsole_sysrq(req);
+			set_irq_regs(old_regs);
+			continue;
+		}
+		(*req->cmd->handler)(req);
+	}
+	os_set_fd_block(req->originating_fd, 0);
+	mconsole_reply(req, "", 0, 0);
+	unblock_signals();
+}
+
+static DEFINE_SPINLOCK(mc_devices_lock);
+static LIST_HEAD(mconsole_devices);
+
+void mconsole_register_dev(struct mc_device *new)
+{
+	spin_lock(&mc_devices_lock);
+	BUG_ON(!list_empty(&new->list));
+	list_add(&new->list, &mconsole_devices);
+	spin_unlock(&mc_devices_lock);
+}
+
+static struct mc_device *mconsole_find_dev(char *name)
+{
+	struct list_head *ele;
+	struct mc_device *dev;
+
+	list_for_each(ele, &mconsole_devices) {
+		dev = list_entry(ele, struct mc_device, list);
+		if (!strncmp(name, dev->name, strlen(dev->name)))
+			return dev;
+	}
+	return NULL;
+}
+
+#define UNPLUGGED_PER_PAGE \
+	((PAGE_SIZE - sizeof(struct list_head)) / sizeof(unsigned long))
+
+struct unplugged_pages {
+	struct list_head list;
+	void *pages[UNPLUGGED_PER_PAGE];
+};
+
+static DEFINE_MUTEX(plug_mem_mutex);
+static unsigned long long unplugged_pages_count;
+static LIST_HEAD(unplugged_pages);
+static int unplug_index = UNPLUGGED_PER_PAGE;
+
+static int mem_config(char *str, char **error_out)
+{
+	unsigned long long diff;
+	int err = -EINVAL, i, add;
+	char *ret;
+
+	if (str[0] != '=') {
+		*error_out = "Expected '=' after 'mem'";
+		goto out;
+	}
+
+	str++;
+	if (str[0] == '-')
+		add = 0;
+	else if (str[0] == '+') {
+		add = 1;
+	}
+	else {
+		*error_out = "Expected increment to start with '-' or '+'";
+		goto out;
+	}
+
+	str++;
+	diff = memparse(str, &ret);
+	if (*ret != '\0') {
+		*error_out = "Failed to parse memory increment";
+		goto out;
+	}
+
+	diff /= PAGE_SIZE;
+
+	mutex_lock(&plug_mem_mutex);
+	for (i = 0; i < diff; i++) {
+		struct unplugged_pages *unplugged;
+		void *addr;
+
+		if (add) {
+			if (list_empty(&unplugged_pages))
+				break;
+
+			unplugged = list_entry(unplugged_pages.next,
+					       struct unplugged_pages, list);
+			if (unplug_index > 0)
+				addr = unplugged->pages[--unplug_index];
+			else {
+				list_del(&unplugged->list);
+				addr = unplugged;
+				unplug_index = UNPLUGGED_PER_PAGE;
+			}
+
+			free_page((unsigned long) addr);
+			unplugged_pages_count--;
+		}
+		else {
+			struct page *page;
+
+			page = alloc_page(GFP_ATOMIC);
+			if (page == NULL)
+				break;
+
+			unplugged = page_address(page);
+			if (unplug_index == UNPLUGGED_PER_PAGE) {
+				list_add(&unplugged->list, &unplugged_pages);
+				unplug_index = 0;
+			}
+			else {
+				struct list_head *entry = unplugged_pages.next;
+				addr = unplugged;
+
+				unplugged = list_entry(entry,
+						       struct unplugged_pages,
+						       list);
+				err = os_drop_memory(addr, PAGE_SIZE);
+				if (err) {
+					printk(KERN_ERR "Failed to release "
+					       "memory - errno = %d\n", err);
+					*error_out = "Failed to release memory";
+					goto out_unlock;
+				}
+				unplugged->pages[unplug_index++] = addr;
+			}
+
+			unplugged_pages_count++;
+		}
+	}
+
+	err = 0;
+out_unlock:
+	mutex_unlock(&plug_mem_mutex);
+out:
+	return err;
+}
+
+static int mem_get_config(char *name, char *str, int size, char **error_out)
+{
+	char buf[sizeof("18446744073709551615")];
+	int len = 0;
+
+	sprintf(buf, "%ld", uml_physmem);
+	CONFIG_CHUNK(str, size, len, buf, 1);
+
+	return len;
+}
+
+static int mem_id(char **str, int *start_out, int *end_out)
+{
+	*start_out = 0;
+	*end_out = 0;
+
+	return 0;
+}
+
+static int mem_remove(int n, char **error_out)
+{
+	*error_out = "Memory doesn't support the remove operation";
+	return -EBUSY;
+}
+
+static struct mc_device mem_mc = {
+	.list		= LIST_HEAD_INIT(mem_mc.list),
+	.name		= "mem",
+	.config		= mem_config,
+	.get_config	= mem_get_config,
+	.id		= mem_id,
+	.remove		= mem_remove,
+};
+
+static int __init mem_mc_init(void)
+{
+	if (can_drop_memory())
+		mconsole_register_dev(&mem_mc);
+	else printk(KERN_ERR "Can't release memory to the host - memory "
+		    "hotplug won't be supported\n");
+	return 0;
+}
+
+__initcall(mem_mc_init);
+
+#define CONFIG_BUF_SIZE 64
+
+static void mconsole_get_config(int (*get_config)(char *, char *, int,
+						  char **),
+				struct mc_request *req, char *name)
+{
+	char default_buf[CONFIG_BUF_SIZE], *error, *buf;
+	int n, size;
+
+	if (get_config == NULL) {
+		mconsole_reply(req, "No get_config routine defined", 1, 0);
+		return;
+	}
+
+	error = NULL;
+	size = ARRAY_SIZE(default_buf);
+	buf = default_buf;
+
+	while (1) {
+		n = (*get_config)(name, buf, size, &error);
+		if (error != NULL) {
+			mconsole_reply(req, error, 1, 0);
+			goto out;
+		}
+
+		if (n <= size) {
+			mconsole_reply(req, buf, 0, 0);
+			goto out;
+		}
+
+		if (buf != default_buf)
+			kfree(buf);
+
+		size = n;
+		buf = kmalloc(size, GFP_KERNEL);
+		if (buf == NULL) {
+			mconsole_reply(req, "Failed to allocate buffer", 1, 0);
+			return;
+		}
+	}
+ out:
+	if (buf != default_buf)
+		kfree(buf);
+}
+
+void mconsole_config(struct mc_request *req)
+{
+	struct mc_device *dev;
+	char *ptr = req->request.data, *name, *error_string = "";
+	int err;
+
+	ptr += strlen("config");
+	ptr = skip_spaces(ptr);
+	dev = mconsole_find_dev(ptr);
+	if (dev == NULL) {
+		mconsole_reply(req, "Bad configuration option", 1, 0);
+		return;
+	}
+
+	name = &ptr[strlen(dev->name)];
+	ptr = name;
+	while ((*ptr != '=') && (*ptr != '\0'))
+		ptr++;
+
+	if (*ptr == '=') {
+		err = (*dev->config)(name, &error_string);
+		mconsole_reply(req, error_string, err, 0);
+	}
+	else mconsole_get_config(dev->get_config, req, name);
+}
+
+void mconsole_remove(struct mc_request *req)
+{
+	struct mc_device *dev;
+	char *ptr = req->request.data, *err_msg = "";
+	char error[256];
+	int err, start, end, n;
+
+	ptr += strlen("remove");
+	ptr = skip_spaces(ptr);
+	dev = mconsole_find_dev(ptr);
+	if (dev == NULL) {
+		mconsole_reply(req, "Bad remove option", 1, 0);
+		return;
+	}
+
+	ptr = &ptr[strlen(dev->name)];
+
+	err = 1;
+	n = (*dev->id)(&ptr, &start, &end);
+	if (n < 0) {
+		err_msg = "Couldn't parse device number";
+		goto out;
+	}
+	else if ((n < start) || (n > end)) {
+		sprintf(error, "Invalid device number - must be between "
+			"%d and %d", start, end);
+		err_msg = error;
+		goto out;
+	}
+
+	err_msg = NULL;
+	err = (*dev->remove)(n, &err_msg);
+	switch(err) {
+	case 0:
+		err_msg = "";
+		break;
+	case -ENODEV:
+		if (err_msg == NULL)
+			err_msg = "Device doesn't exist";
+		break;
+	case -EBUSY:
+		if (err_msg == NULL)
+			err_msg = "Device is currently open";
+		break;
+	default:
+		break;
+	}
+out:
+	mconsole_reply(req, err_msg, err, 0);
+}
+
+struct mconsole_output {
+	struct list_head list;
+	struct mc_request *req;
+};
+
+static DEFINE_SPINLOCK(client_lock);
+static LIST_HEAD(clients);
+static char console_buf[MCONSOLE_MAX_DATA] __nonstring;
+
+static void console_write(struct console *console, const char *string,
+			  unsigned int len)
+{
+	struct list_head *ele;
+	int n;
+
+	if (list_empty(&clients))
+		return;
+
+	while (len > 0) {
+		n = min((size_t) len, ARRAY_SIZE(console_buf));
+		memcpy(console_buf, string, n);
+		string += n;
+		len -= n;
+
+		list_for_each(ele, &clients) {
+			struct mconsole_output *entry;
+
+			entry = list_entry(ele, struct mconsole_output, list);
+			mconsole_reply_len(entry->req, console_buf, n, 0, 1);
+		}
+	}
+}
+
+static struct console mc_console = { .name	= "mc",
+				     .write	= console_write,
+				     .flags	= CON_ENABLED,
+				     .index	= -1 };
+
+static int mc_add_console(void)
+{
+	register_console(&mc_console);
+	return 0;
+}
+
+late_initcall(mc_add_console);
+
+static void with_console(struct mc_request *req, void (*proc)(void *),
+			 void *arg)
+{
+	struct mconsole_output entry;
+	unsigned long flags;
+
+	entry.req = req;
+	spin_lock_irqsave(&client_lock, flags);
+	list_add(&entry.list, &clients);
+	spin_unlock_irqrestore(&client_lock, flags);
+
+	(*proc)(arg);
+
+	mconsole_reply_len(req, "", 0, 0, 0);
+
+	spin_lock_irqsave(&client_lock, flags);
+	list_del(&entry.list);
+	spin_unlock_irqrestore(&client_lock, flags);
+}
+
+#ifdef CONFIG_MAGIC_SYSRQ
+
+#include <linux/sysrq.h>
+
+static void sysrq_proc(void *arg)
+{
+	char *op = arg;
+	handle_sysrq(*op);
+}
+
+void mconsole_sysrq(struct mc_request *req)
+{
+	char *ptr = req->request.data;
+
+	ptr += strlen("sysrq");
+	ptr = skip_spaces(ptr);
+
+	/*
+	 * With 'b', the system will shut down without a chance to reply,
+	 * so in this case, we reply first.
+	 */
+	if (*ptr == 'b')
+		mconsole_reply(req, "", 0, 0);
+
+	with_console(req, sysrq_proc, ptr);
+}
+#else
+void mconsole_sysrq(struct mc_request *req)
+{
+	mconsole_reply(req, "Sysrq not compiled in", 1, 0);
+}
+#endif
+
+static void stack_proc(void *arg)
+{
+	struct task_struct *task = arg;
+
+	show_stack(task, NULL, KERN_INFO);
+}
+
+/*
+ * Mconsole stack trace
+ *  Added by Allan Graves, Jeff Dike
+ *  Dumps a stacks registers to the linux console.
+ *  Usage stack <pid>.
+ */
+void mconsole_stack(struct mc_request *req)
+{
+	char *ptr = req->request.data;
+	int pid_requested= -1;
+	struct task_struct *to = NULL;
+
+	/*
+	 * Would be nice:
+	 * 1) Send showregs output to mconsole.
+	 * 2) Add a way to stack dump all pids.
+	 */
+
+	ptr += strlen("stack");
+	ptr = skip_spaces(ptr);
+
+	/*
+	 * Should really check for multiple pids or reject bad args here
+	 */
+	/* What do the arguments in mconsole_reply mean? */
+	if (sscanf(ptr, "%d", &pid_requested) == 0) {
+		mconsole_reply(req, "Please specify a pid", 1, 0);
+		return;
+	}
+
+	to = find_task_by_pid_ns(pid_requested, &init_pid_ns);
+	if ((to == NULL) || (pid_requested == 0)) {
+		mconsole_reply(req, "Couldn't find that pid", 1, 0);
+		return;
+	}
+	with_console(req, stack_proc, to);
+}
+
+static int __init mount_proc(void)
+{
+	struct file_system_type *proc_fs_type;
+	struct vfsmount *mnt;
+
+	proc_fs_type = get_fs_type("proc");
+	if (!proc_fs_type)
+		return -ENODEV;
+
+	mnt = kern_mount(proc_fs_type);
+	put_filesystem(proc_fs_type);
+	if (IS_ERR(mnt))
+		return PTR_ERR(mnt);
+
+	proc_mnt = mnt;
+	return 0;
+}
+
+/*
+ * Changed by mconsole_setup, which is __setup, and called before SMP is
+ * active.
+ */
+static char *notify_socket = NULL;
+
+static int __init mconsole_init(void)
+{
+	/* long to avoid size mismatch warnings from gcc */
+	long sock;
+	int err;
+	char file[UNIX_PATH_MAX];
+
+	mount_proc();
+
+	if (umid_file_name("mconsole", file, sizeof(file)))
+		return -1;
+	snprintf(mconsole_socket_name, sizeof(file), "%s", file);
+
+	sock = os_create_unix_socket(file, sizeof(file), 1);
+	if (sock < 0) {
+		printk(KERN_ERR "Failed to initialize management console\n");
+		return 1;
+	}
+	if (os_set_fd_block(sock, 0))
+		goto out;
+
+	register_reboot_notifier(&reboot_notifier);
+
+	err = um_request_irq(MCONSOLE_IRQ, sock, IRQ_READ, mconsole_interrupt,
+			     IRQF_SHARED, "mconsole", (void *)sock);
+	if (err < 0) {
+		printk(KERN_ERR "Failed to get IRQ for management console\n");
+		goto out;
+	}
+
+	if (notify_socket != NULL) {
+		notify_socket = kstrdup(notify_socket, GFP_KERNEL);
+		if (notify_socket != NULL)
+			mconsole_notify(notify_socket, MCONSOLE_SOCKET,
+					mconsole_socket_name,
+					strlen(mconsole_socket_name) + 1);
+		else printk(KERN_ERR "mconsole_setup failed to strdup "
+			    "string\n");
+	}
+
+	printk(KERN_INFO "mconsole (version %d) initialized on %s\n",
+	       MCONSOLE_VERSION, mconsole_socket_name);
+	return 0;
+
+ out:
+	os_close_file(sock);
+	return 1;
+}
+
+__initcall(mconsole_init);
+
+static ssize_t mconsole_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *pos)
+{
+	char *buf;
+
+	buf = memdup_user_nul(buffer, count);
+	if (IS_ERR(buf))
+		return PTR_ERR(buf);
+
+	mconsole_notify(notify_socket, MCONSOLE_USER_NOTIFY, buf, count);
+	kfree(buf);
+	return count;
+}
+
+static const struct proc_ops mconsole_proc_ops = {
+	.proc_write	= mconsole_proc_write,
+	.proc_lseek	= noop_llseek,
+};
+
+static int create_proc_mconsole(void)
+{
+	struct proc_dir_entry *ent;
+
+	if (notify_socket == NULL)
+		return 0;
+
+	ent = proc_create("mconsole", 0200, NULL, &mconsole_proc_ops);
+	if (ent == NULL) {
+		printk(KERN_INFO "create_proc_mconsole : proc_create failed\n");
+		return 0;
+	}
+	return 0;
+}
+
+static DEFINE_SPINLOCK(notify_spinlock);
+
+void lock_notify(void)
+{
+	spin_lock(&notify_spinlock);
+}
+
+void unlock_notify(void)
+{
+	spin_unlock(&notify_spinlock);
+}
+
+__initcall(create_proc_mconsole);
+
+#define NOTIFY "notify:"
+
+static int mconsole_setup(char *str)
+{
+	if (!strncmp(str, NOTIFY, strlen(NOTIFY))) {
+		str += strlen(NOTIFY);
+		notify_socket = str;
+	}
+	else printk(KERN_ERR "mconsole_setup : Unknown option - '%s'\n", str);
+	return 1;
+}
+
+__setup("mconsole=", mconsole_setup);
+
+__uml_help(mconsole_setup,
+"mconsole=notify:<socket>\n"
+"    Requests that the mconsole driver send a message to the named Unix\n"
+"    socket containing the name of the mconsole socket.  This also serves\n"
+"    to notify outside processes when UML has booted far enough to respond\n"
+"    to mconsole requests.\n\n"
+);
+
+static int notify_panic(struct notifier_block *self, unsigned long unused1,
+			void *ptr)
+{
+	char *message = ptr;
+
+	if (notify_socket == NULL)
+		return 0;
+
+	mconsole_notify(notify_socket, MCONSOLE_PANIC, message,
+			strlen(message) + 1);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block panic_exit_notifier = {
+	.notifier_call	= notify_panic,
+	.priority	= INT_MAX, /* run as soon as possible */
+};
+
+static int add_notifier(void)
+{
+	atomic_notifier_chain_register(&panic_notifier_list,
+			&panic_exit_notifier);
+	return 0;
+}
+
+__initcall(add_notifier);
+
+char *mconsole_notify_socket(void)
+{
+	return notify_socket;
+}
+
+EXPORT_SYMBOL(mconsole_notify_socket);

linux-7.2.3/arch/um/drivers/mconsole_kern.h

@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __MCONSOLE_KERN_H__
+#define __MCONSOLE_KERN_H__
+
+#include <linux/list.h>
+#include "mconsole.h"
+
+struct mconsole_entry {
+	struct list_head list;
+	struct mc_request request;
+};
+
+/* All these methods are called in process context. */
+struct mc_device {
+	struct list_head list;
+	char *name;
+	int (*config)(char *, char **);
+	int (*get_config)(char *, char *, int, char **);
+	int (*id)(char **, int *, int *);
+	int (*remove)(int, char **);
+};
+
+#define CONFIG_CHUNK(str, size, current, chunk, end) \
+do { \
+	current += strlen(chunk); \
+	if(current >= size) \
+		str = NULL; \
+	if(str != NULL){ \
+		strcpy(str, chunk); \
+		str += strlen(chunk); \
+	} \
+	if(end) \
+		current++; \
+} while(0)
+
+#ifdef CONFIG_MCONSOLE
+
+extern void mconsole_register_dev(struct mc_device *new);
+
+#else
+
+static inline void mconsole_register_dev(struct mc_device *new)
+{
+}
+
+#endif
+
+#endif

linux-7.2.3/arch/um/drivers/mconsole_user.c

@@ -0,0 +1,221 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org)
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <errno.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/socket.h>
+#include <sys/uio.h>
+#include <sys/un.h>
+#include "mconsole.h"
+
+static struct mconsole_command commands[] = {
+	/*
+	 * With uts namespaces, uts information becomes process-specific, so
+	 * we need a process context.  If we try handling this in interrupt
+	 * context, we may hit an exiting process without a valid uts
+	 * namespace.
+	 */
+	{ "version", mconsole_version, MCONSOLE_PROC },
+	{ "halt", mconsole_halt, MCONSOLE_PROC },
+	{ "reboot", mconsole_reboot, MCONSOLE_PROC },
+	{ "config", mconsole_config, MCONSOLE_PROC },
+	{ "remove", mconsole_remove, MCONSOLE_PROC },
+	{ "sysrq", mconsole_sysrq, MCONSOLE_INTR },
+	{ "help", mconsole_help, MCONSOLE_INTR },
+	{ "cad", mconsole_cad, MCONSOLE_INTR },
+	{ "stop", mconsole_stop, MCONSOLE_PROC },
+	{ "go", mconsole_go, MCONSOLE_INTR },
+	{ "log", mconsole_log, MCONSOLE_INTR },
+	{ "proc", mconsole_proc, MCONSOLE_PROC },
+	{ "stack", mconsole_stack, MCONSOLE_INTR },
+};
+
+/* Initialized in mconsole_init, which is an initcall */
+char mconsole_socket_name[256];
+
+static int mconsole_reply_v0(struct mc_request *req, char *reply)
+{
+	struct iovec iov;
+	struct msghdr msg;
+
+	iov.iov_base = reply;
+	iov.iov_len = strlen(reply);
+
+	msg.msg_name = &(req->origin);
+	msg.msg_namelen = req->originlen;
+	msg.msg_iov = &iov;
+	msg.msg_iovlen = 1;
+	msg.msg_control = NULL;
+	msg.msg_controllen = 0;
+	msg.msg_flags = 0;
+
+	return sendmsg(req->originating_fd, &msg, 0);
+}
+
+static struct mconsole_command *mconsole_parse(struct mc_request *req)
+{
+	struct mconsole_command *cmd;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		cmd = &commands[i];
+		if (!strncmp(req->request.data, cmd->command,
+			    strlen(cmd->command))) {
+			return cmd;
+		}
+	}
+	return NULL;
+}
+
+#ifndef MIN
+#define MIN(a,b) ((a)<(b) ? (a):(b))
+#endif
+
+#define STRINGX(x) #x
+#define STRING(x) STRINGX(x)
+
+int mconsole_get_request(int fd, struct mc_request *req)
+{
+	int len;
+
+	req->originlen = sizeof(req->origin);
+	req->len = recvfrom(fd, &req->request, sizeof(req->request), 0,
+			    (struct sockaddr *) req->origin, &req->originlen);
+	if (req->len < 0)
+		return 0;
+
+	req->originating_fd = fd;
+
+	if (req->request.magic != MCONSOLE_MAGIC) {
+		/* Unversioned request */
+		len = MIN(sizeof(req->request.data) - 1,
+			  strlen((char *) &req->request));
+		memmove(req->request.data, &req->request, len);
+		req->request.data[len] = '\0';
+
+		req->request.magic = MCONSOLE_MAGIC;
+		req->request.version = 0;
+		req->request.len = len;
+
+		mconsole_reply_v0(req, "ERR Version 0 mconsole clients are "
+				  "not supported by this driver");
+		return 0;
+	}
+
+	if (req->request.len >= MCONSOLE_MAX_DATA) {
+		mconsole_reply(req, "Request too large", 1, 0);
+		return 0;
+	}
+	if (req->request.version != MCONSOLE_VERSION) {
+		mconsole_reply(req, "This driver only supports version "
+			       STRING(MCONSOLE_VERSION) " clients", 1, 0);
+	}
+
+	req->request.data[req->request.len] = '\0';
+	req->cmd = mconsole_parse(req);
+	if (req->cmd == NULL) {
+		mconsole_reply(req, "Unknown command", 1, 0);
+		return 0;
+	}
+
+	return 1;
+}
+
+int mconsole_reply_len(struct mc_request *req, const char *str, int total,
+		       int err, int more)
+{
+	/*
+	 * XXX This is a stack consumption problem.  It'd be nice to
+	 * make it global and serialize access to it, but there are a
+	 * ton of callers to this function.
+	 */
+	struct mconsole_reply reply;
+	int len, n;
+
+	do {
+		reply.err = err;
+
+		/* err can only be true on the first packet */
+		err = 0;
+
+		len = MIN(total, MCONSOLE_MAX_DATA - 1);
+
+		if (len == total) reply.more = more;
+		else reply.more = 1;
+
+		memcpy(reply.data, str, len);
+		reply.data[len] = '\0';
+		total -= len;
+		str += len;
+		reply.len = len + 1;
+
+		len = sizeof(reply) + reply.len - sizeof(reply.data);
+
+		n = sendto(req->originating_fd, &reply, len, 0,
+			   (struct sockaddr *) req->origin, req->originlen);
+
+		if (n < 0)
+			return -errno;
+	} while (total > 0);
+	return 0;
+}
+
+int mconsole_reply(struct mc_request *req, const char *str, int err, int more)
+{
+	return mconsole_reply_len(req, str, strlen(str), err, more);
+}
+
+
+int mconsole_unlink_socket(void)
+{
+	unlink(mconsole_socket_name);
+	return 0;
+}
+
+static int notify_sock = -1;
+
+int mconsole_notify(char *sock_name, int type, const void *data, int len)
+{
+	struct sockaddr_un target;
+	struct mconsole_notify packet;
+	int n, err = 0;
+
+	lock_notify();
+	if (notify_sock < 0) {
+		notify_sock = socket(PF_UNIX, SOCK_DGRAM, 0);
+		if (notify_sock < 0) {
+			err = -errno;
+			printk(UM_KERN_ERR "mconsole_notify - socket failed, "
+			       "errno = %d\n", errno);
+		}
+	}
+	unlock_notify();
+
+	if (err)
+		return err;
+
+	target.sun_family = AF_UNIX;
+	strcpy(target.sun_path, sock_name);
+
+	packet.magic = MCONSOLE_MAGIC;
+	packet.version = MCONSOLE_VERSION;
+	packet.type = type;
+	len = (len > sizeof(packet.data)) ? sizeof(packet.data) : len;
+	packet.len = len;
+	memcpy(packet.data, data, len);
+
+	err = 0;
+	len = sizeof(packet) + packet.len - sizeof(packet.data);
+	n = sendto(notify_sock, &packet, len, 0, (struct sockaddr *) &target,
+		   sizeof(target));
+	if (n < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "mconsole_notify - sendto failed, "
+		       "errno = %d\n", errno);
+	}
+	return err;
+}

linux-7.2.3/arch/um/drivers/null.c

@@ -0,0 +1,51 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <stddef.h>
+#include <errno.h>
+#include <fcntl.h>
+#include "chan_user.h"
+#include <os.h>
+
+/* This address is used only as a unique identifier */
+static int null_chan;
+
+static void *null_init(char *str, int device, const struct chan_opts *opts)
+{
+	return &null_chan;
+}
+
+static int null_open(int input, int output, int primary, void *d,
+		     char **dev_out)
+{
+	int fd;
+
+	*dev_out = NULL;
+
+	fd = open(DEV_NULL, O_RDWR);
+	return (fd < 0) ? -errno : fd;
+}
+
+static int null_read(int fd, __u8 *c_out, void *unused)
+{
+	return -ENODEV;
+}
+
+static void null_free(void *data)
+{
+}
+
+const struct chan_ops null_ops = {
+	.type		= "null",
+	.init		= null_init,
+	.open		= null_open,
+	.close		= generic_close,
+	.read		= null_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= null_free,
+	.winch		= 0,
+};

linux-7.2.3/arch/um/drivers/port.h

@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2001 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __PORT_H__
+#define __PORT_H__
+
+extern void *port_data(int port);
+extern int port_wait(void *data);
+extern void port_kern_close(void *d);
+extern int port_connection(int fd, int *socket_out, int *pid_out);
+extern int port_listen_fd(int port);
+extern void port_read(int fd, void *data);
+extern void port_kern_free(void *d);
+extern int port_rcv_fd(int fd);
+extern void port_remove_dev(void *d);
+
+#endif
+

linux-7.2.3/arch/um/drivers/port_kern.c

@@ -0,0 +1,305 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <linux/completion.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+#include <linux/workqueue.h>
+#include <asm/atomic.h>
+#include <init.h>
+#include <irq_kern.h>
+#include <os.h>
+#include "port.h"
+
+struct port_list {
+	struct list_head list;
+	atomic_t wait_count;
+	int has_connection;
+	struct completion done;
+	int port;
+	int fd;
+	spinlock_t lock;
+	struct list_head pending;
+	struct list_head connections;
+};
+
+struct port_dev {
+	struct port_list *port;
+	int helper_pid;
+	int telnetd_pid;
+};
+
+struct connection {
+	struct list_head list;
+	int fd;
+	int helper_pid;
+	int socket[2];
+	int telnetd_pid;
+	struct port_list *port;
+};
+
+static irqreturn_t pipe_interrupt(int irq, void *data)
+{
+	struct connection *conn = data;
+	int n_fds = 1, fd = -1;
+	ssize_t ret;
+
+	ret = os_rcv_fd_msg(conn->socket[0], &fd, n_fds, &conn->helper_pid,
+			    sizeof(conn->helper_pid));
+	if (ret != sizeof(conn->helper_pid)) {
+		if (ret == -EAGAIN)
+			return IRQ_NONE;
+
+		printk(KERN_ERR "pipe_interrupt : os_rcv_fd_msg returned %zd\n",
+		       ret);
+		os_close_file(conn->fd);
+	}
+
+	list_del(&conn->list);
+
+	conn->fd = fd;
+	list_add(&conn->list, &conn->port->connections);
+
+	complete(&conn->port->done);
+	return IRQ_HANDLED;
+}
+
+#define NO_WAITER_MSG \
+    "****\n" \
+    "There are currently no UML consoles waiting for port connections.\n" \
+    "Either disconnect from one to make it available or activate some more\n" \
+    "by enabling more consoles in the UML /etc/inittab.\n" \
+    "****\n"
+
+static int port_accept(struct port_list *port)
+{
+	struct connection *conn;
+	int fd, socket[2], pid;
+
+	fd = port_connection(port->fd, socket, &pid);
+	if (fd < 0) {
+		if (fd != -EAGAIN)
+			printk(KERN_ERR "port_accept : port_connection "
+			       "returned %d\n", -fd);
+		goto out;
+	}
+
+	conn = kmalloc_obj(*conn, GFP_ATOMIC);
+	if (conn == NULL) {
+		printk(KERN_ERR "port_accept : failed to allocate "
+		       "connection\n");
+		goto out_close;
+	}
+	*conn = ((struct connection)
+		{ .list 	= LIST_HEAD_INIT(conn->list),
+		  .fd 		= fd,
+		  .socket  	= { socket[0], socket[1] },
+		  .telnetd_pid 	= pid,
+		  .port 	= port });
+
+	if (um_request_irq(TELNETD_IRQ, socket[0], IRQ_READ, pipe_interrupt,
+			  IRQF_SHARED, "telnetd", conn) < 0) {
+		printk(KERN_ERR "port_accept : failed to get IRQ for "
+		       "telnetd\n");
+		goto out_free;
+	}
+
+	if (atomic_read(&port->wait_count) == 0) {
+		os_write_file(fd, NO_WAITER_MSG, sizeof(NO_WAITER_MSG));
+		printk(KERN_ERR "No one waiting for port\n");
+	}
+	list_add(&conn->list, &port->pending);
+	return 1;
+
+ out_free:
+	kfree(conn);
+ out_close:
+	os_close_file(fd);
+	os_kill_process(pid, 1);
+ out:
+	return 0;
+}
+
+static DEFINE_MUTEX(ports_mutex);
+static LIST_HEAD(ports);
+
+static void port_work_proc(struct work_struct *unused)
+{
+	struct port_list *port;
+	struct list_head *ele;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	list_for_each(ele, &ports) {
+		port = list_entry(ele, struct port_list, list);
+		if (!port->has_connection)
+			continue;
+
+		while (port_accept(port))
+			;
+		port->has_connection = 0;
+	}
+	local_irq_restore(flags);
+}
+
+static DECLARE_WORK(port_work, port_work_proc);
+
+static irqreturn_t port_interrupt(int irq, void *data)
+{
+	struct port_list *port = data;
+
+	port->has_connection = 1;
+	schedule_work(&port_work);
+	return IRQ_HANDLED;
+}
+
+void *port_data(int port_num)
+{
+	struct list_head *ele;
+	struct port_list *port;
+	struct port_dev *dev = NULL;
+	int fd;
+
+	mutex_lock(&ports_mutex);
+	list_for_each(ele, &ports) {
+		port = list_entry(ele, struct port_list, list);
+		if (port->port == port_num)
+			goto found;
+	}
+	port = kmalloc_obj(struct port_list);
+	if (port == NULL) {
+		printk(KERN_ERR "Allocation of port list failed\n");
+		goto out;
+	}
+
+	fd = port_listen_fd(port_num);
+	if (fd < 0) {
+		printk(KERN_ERR "binding to port %d failed, errno = %d\n",
+		       port_num, -fd);
+		goto out_free;
+	}
+
+	if (um_request_irq(ACCEPT_IRQ, fd, IRQ_READ, port_interrupt,
+			  IRQF_SHARED, "port", port) < 0) {
+		printk(KERN_ERR "Failed to get IRQ for port %d\n", port_num);
+		goto out_close;
+	}
+
+	*port = ((struct port_list)
+		{ .list 	 	= LIST_HEAD_INIT(port->list),
+		  .wait_count		= ATOMIC_INIT(0),
+		  .has_connection 	= 0,
+		  .port 	 	= port_num,
+		  .fd  			= fd,
+		  .pending 		= LIST_HEAD_INIT(port->pending),
+		  .connections 		= LIST_HEAD_INIT(port->connections) });
+	spin_lock_init(&port->lock);
+	init_completion(&port->done);
+	list_add(&port->list, &ports);
+
+ found:
+	dev = kmalloc_obj(struct port_dev);
+	if (dev == NULL) {
+		printk(KERN_ERR "Allocation of port device entry failed\n");
+		goto out;
+	}
+
+	*dev = ((struct port_dev) { .port  		= port,
+				    .helper_pid  	= -1,
+				    .telnetd_pid  	= -1 });
+	goto out;
+
+ out_close:
+	os_close_file(fd);
+ out_free:
+	kfree(port);
+ out:
+	mutex_unlock(&ports_mutex);
+	return dev;
+}
+
+int port_wait(void *data)
+{
+	struct port_dev *dev = data;
+	struct connection *conn;
+	struct port_list *port = dev->port;
+	int fd;
+
+	atomic_inc(&port->wait_count);
+	while (1) {
+		fd = -ERESTARTSYS;
+		if (wait_for_completion_interruptible(&port->done))
+			goto out;
+
+		spin_lock(&port->lock);
+
+		conn = list_entry(port->connections.next, struct connection,
+				  list);
+		list_del(&conn->list);
+		spin_unlock(&port->lock);
+
+		os_shutdown_socket(conn->socket[0], 1, 1);
+		os_close_file(conn->socket[0]);
+		os_shutdown_socket(conn->socket[1], 1, 1);
+		os_close_file(conn->socket[1]);
+
+		/* This is done here because freeing an IRQ can't be done
+		 * within the IRQ handler.  So, pipe_interrupt always ups
+		 * the semaphore regardless of whether it got a successful
+		 * connection.  Then we loop here throwing out failed
+		 * connections until a good one is found.
+		 */
+		um_free_irq(TELNETD_IRQ, conn);
+
+		if (conn->fd >= 0)
+			break;
+		os_close_file(conn->fd);
+		kfree(conn);
+	}
+
+	fd = conn->fd;
+	dev->helper_pid = conn->helper_pid;
+	dev->telnetd_pid = conn->telnetd_pid;
+	kfree(conn);
+ out:
+	atomic_dec(&port->wait_count);
+	return fd;
+}
+
+void port_remove_dev(void *d)
+{
+	struct port_dev *dev = d;
+
+	if (dev->helper_pid != -1)
+		os_kill_process(dev->helper_pid, 0);
+	if (dev->telnetd_pid != -1)
+		os_kill_process(dev->telnetd_pid, 1);
+	dev->helper_pid = -1;
+	dev->telnetd_pid = -1;
+}
+
+void port_kern_free(void *d)
+{
+	struct port_dev *dev = d;
+
+	port_remove_dev(dev);
+	kfree(dev);
+}
+
+static void free_port(void)
+{
+	struct list_head *ele;
+	struct port_list *port;
+
+	list_for_each(ele, &ports) {
+		port = list_entry(ele, struct port_list, list);
+		free_irq_by_fd(port->fd);
+		os_close_file(port->fd);
+	}
+}
+
+__uml_exitcall(free_port);

linux-7.2.3/arch/um/drivers/port_user.c

@@ -0,0 +1,217 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <termios.h>
+#include <unistd.h>
+#include <netinet/in.h>
+#include "chan_user.h"
+#include <os.h>
+#include "port.h"
+#include <um_malloc.h>
+
+struct port_chan {
+	int raw;
+	struct termios tt;
+	void *kernel_data;
+	char dev[sizeof("32768\0")];
+};
+
+static void *port_init(char *str, int device, const struct chan_opts *opts)
+{
+	struct port_chan *data;
+	void *kern_data;
+	char *end;
+	int port;
+
+	if (*str != ':') {
+		printk(UM_KERN_ERR "port_init : channel type 'port' must "
+		       "specify a port number\n");
+		return NULL;
+	}
+	str++;
+	port = strtoul(str, &end, 0);
+	if ((*end != '\0') || (end == str)) {
+		printk(UM_KERN_ERR "port_init : couldn't parse port '%s'\n",
+		       str);
+		return NULL;
+	}
+
+	kern_data = port_data(port);
+	if (kern_data == NULL)
+		return NULL;
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_KERNEL);
+	if (data == NULL)
+		goto err;
+
+	*data = ((struct port_chan) { .raw  		= opts->raw,
+				      .kernel_data 	= kern_data });
+	sprintf(data->dev, "%d", port);
+
+	return data;
+ err:
+	port_kern_free(kern_data);
+	return NULL;
+}
+
+static void port_free(void *d)
+{
+	struct port_chan *data = d;
+
+	port_kern_free(data->kernel_data);
+	kfree(data);
+}
+
+static int port_open(int input, int output, int primary, void *d,
+		     char **dev_out)
+{
+	struct port_chan *data = d;
+	int fd, err;
+
+	fd = port_wait(data->kernel_data);
+	if ((fd >= 0) && data->raw) {
+		CATCH_EINTR(err = tcgetattr(fd, &data->tt));
+		if (err)
+			return err;
+
+		err = raw(fd);
+		if (err)
+			return err;
+	}
+	*dev_out = data->dev;
+	return fd;
+}
+
+static void port_close(int fd, void *d)
+{
+	struct port_chan *data = d;
+
+	port_remove_dev(data->kernel_data);
+	os_close_file(fd);
+}
+
+const struct chan_ops port_ops = {
+	.type		= "port",
+	.init		= port_init,
+	.open		= port_open,
+	.close		= port_close,
+	.read	        = generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= port_free,
+	.winch		= 1,
+};
+
+int port_listen_fd(int port)
+{
+	struct sockaddr_in addr;
+	int fd, err, arg;
+
+	fd = socket(PF_INET, SOCK_STREAM, 0);
+	if (fd == -1)
+		return -errno;
+
+	arg = 1;
+	if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &arg, sizeof(arg)) < 0) {
+		err = -errno;
+		goto out;
+	}
+
+	addr.sin_family = AF_INET;
+	addr.sin_port = htons(port);
+	addr.sin_addr.s_addr = htonl(INADDR_ANY);
+	if (bind(fd, (struct sockaddr *) &addr, sizeof(addr)) < 0) {
+		err = -errno;
+		goto out;
+	}
+
+	if (listen(fd, 1) < 0) {
+		err = -errno;
+		goto out;
+	}
+
+	err = os_set_fd_block(fd, 0);
+	if (err < 0)
+		goto out;
+
+	return fd;
+ out:
+	close(fd);
+	return err;
+}
+
+struct port_pre_exec_data {
+	int sock_fd;
+	int pipe_fd;
+};
+
+static void port_pre_exec(void *arg)
+{
+	struct port_pre_exec_data *data = arg;
+
+	dup2(data->sock_fd, 0);
+	dup2(data->sock_fd, 1);
+	dup2(data->sock_fd, 2);
+	close(data->sock_fd);
+	dup2(data->pipe_fd, 3);
+	shutdown(3, SHUT_RD);
+	close(data->pipe_fd);
+}
+
+int port_connection(int fd, int *socket, int *pid_out)
+{
+	int new, err;
+	char *env;
+	char *argv[] = { "in.telnetd", "-L",
+			 OS_LIB_PATH "/uml/port-helper", NULL };
+	struct port_pre_exec_data data;
+
+	if ((env = getenv("UML_PORT_HELPER")))
+		argv[2] = env;
+
+	new = accept(fd, NULL, 0);
+	if (new < 0)
+		return -errno;
+
+	err = os_access(argv[2], X_OK);
+	if (err < 0) {
+		printk(UM_KERN_ERR "port_connection : error accessing port-helper "
+		       "executable at %s: %s\n", argv[2], strerror(-err));
+		if (env == NULL)
+			printk(UM_KERN_ERR "Set UML_PORT_HELPER environment "
+				"variable to path to uml-utilities port-helper "
+				"binary\n");
+		goto out_close;
+	}
+
+	err = os_pipe(socket, 0, 0);
+	if (err < 0)
+		goto out_close;
+
+	data = ((struct port_pre_exec_data)
+		{ .sock_fd  		= new,
+		  .pipe_fd 		= socket[1] });
+
+	err = run_helper(port_pre_exec, &data, argv);
+	if (err < 0)
+		goto out_shutdown;
+
+	*pid_out = err;
+	return new;
+
+ out_shutdown:
+	shutdown(socket[0], SHUT_RDWR);
+	close(socket[0]);
+	shutdown(socket[1], SHUT_RDWR);
+	close(socket[1]);
+ out_close:
+	close(new);
+	return err;
+}

linux-7.2.3/arch/um/drivers/pty.c

@@ -0,0 +1,165 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <string.h>
+#include <termios.h>
+#include <sys/stat.h>
+#include "chan_user.h"
+#include <os.h>
+#include <um_malloc.h>
+
+struct pty_chan {
+	void (*announce)(char *dev_name, int dev);
+	int dev;
+	int raw;
+	struct termios tt;
+	char dev_name[sizeof("/dev/pts/0123456\0")];
+};
+
+static void *pty_chan_init(char *str, int device, const struct chan_opts *opts)
+{
+	struct pty_chan *data;
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_KERNEL);
+	if (data == NULL)
+		return NULL;
+
+	*data = ((struct pty_chan) { .announce  	= opts->announce,
+				     .dev  		= device,
+				     .raw  		= opts->raw });
+	return data;
+}
+
+static int pts_open(int input, int output, int primary, void *d,
+		    char **dev_out)
+{
+	struct pty_chan *data = d;
+	char *dev;
+	int fd, err;
+
+	fd = get_pty();
+	if (fd < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "open_pts : Failed to open pts\n");
+		return err;
+	}
+
+	if (data->raw) {
+		CATCH_EINTR(err = tcgetattr(fd, &data->tt));
+		if (err)
+			goto out_close;
+
+		err = raw(fd);
+		if (err)
+			goto out_close;
+	}
+
+	dev = ptsname(fd);
+	sprintf(data->dev_name, "%s", dev);
+	*dev_out = data->dev_name;
+
+	if (data->announce)
+		(*data->announce)(dev, data->dev);
+
+	return fd;
+
+out_close:
+	close(fd);
+	return err;
+}
+
+static int getmaster(char *line)
+{
+	struct stat buf;
+	char *pty, *bank, *cp;
+	int master, err;
+
+	pty = &line[strlen("/dev/ptyp")];
+	for (bank = "pqrs"; *bank; bank++) {
+		line[strlen("/dev/pty")] = *bank;
+		*pty = '0';
+		/* Did we hit the end ? */
+		if ((stat(line, &buf) < 0) && (errno == ENOENT))
+			break;
+
+		for (cp = "0123456789abcdef"; *cp; cp++) {
+			*pty = *cp;
+			master = open(line, O_RDWR);
+			if (master >= 0) {
+				char *tp = &line[strlen("/dev/")];
+
+				/* verify slave side is usable */
+				*tp = 't';
+				err = access(line, R_OK | W_OK);
+				*tp = 'p';
+				if (!err)
+					return master;
+				close(master);
+			}
+		}
+	}
+
+	printk(UM_KERN_ERR "getmaster - no usable host pty devices\n");
+	return -ENOENT;
+}
+
+static int pty_open(int input, int output, int primary, void *d,
+		    char **dev_out)
+{
+	struct pty_chan *data = d;
+	int fd, err;
+	char dev[sizeof("/dev/ptyxx\0")] = "/dev/ptyxx";
+
+	fd = getmaster(dev);
+	if (fd < 0)
+		return fd;
+
+	if (data->raw) {
+		err = raw(fd);
+		if (err) {
+			close(fd);
+			return err;
+		}
+	}
+
+	if (data->announce)
+		(*data->announce)(dev, data->dev);
+
+	sprintf(data->dev_name, "%s", dev);
+	*dev_out = data->dev_name;
+
+	return fd;
+}
+
+const struct chan_ops pty_ops = {
+	.type		= "pty",
+	.init		= pty_chan_init,
+	.open		= pty_open,
+	.close		= generic_close,
+	.read		= generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= generic_free,
+	.winch		= 0,
+};
+
+const struct chan_ops pts_ops = {
+	.type		= "pts",
+	.init		= pty_chan_init,
+	.open		= pts_open,
+	.close		= generic_close,
+	.read		= generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= generic_free,
+	.winch		= 0,
+};

linux-7.2.3/arch/um/drivers/random.c

@@ -0,0 +1,121 @@
+/* Copyright (C) 2005 - 2008 Jeff Dike <jdike@{linux.intel,addtoit}.com> */
+
+/* Much of this ripped from drivers/char/hw_random.c, see there for other
+ * copyright.
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ */
+#include <linux/sched/signal.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/interrupt.h>
+#include <linux/miscdevice.h>
+#include <linux/hw_random.h>
+#include <linux/delay.h>
+#include <linux/uaccess.h>
+#include <init.h>
+#include <irq_kern.h>
+#include <os.h>
+
+/*
+ * core module information
+ */
+#define RNG_MODULE_NAME "hw_random"
+
+/* Changed at init time, in the non-modular case, and at module load
+ * time, in the module case.  Presumably, the module subsystem
+ * protects against a module being loaded twice at the same time.
+ */
+static int random_fd = -1;
+static struct hwrng hwrng;
+static DECLARE_COMPLETION(have_data);
+
+static int rng_dev_read(struct hwrng *rng, void *buf, size_t max, bool block)
+{
+	int ret;
+
+	for (;;) {
+		ret = os_read_file(random_fd, buf, max);
+		if (block && ret == -EAGAIN) {
+			add_sigio_fd(random_fd);
+
+			ret = wait_for_completion_killable(&have_data);
+
+			ignore_sigio_fd(random_fd);
+			deactivate_fd(random_fd, RANDOM_IRQ);
+
+			if (ret < 0)
+				break;
+		} else {
+			break;
+		}
+	}
+
+	return ret != -EAGAIN ? ret : 0;
+}
+
+static irqreturn_t random_interrupt(int irq, void *data)
+{
+	complete(&have_data);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * rng_init - initialize RNG module
+ */
+static int __init rng_init (void)
+{
+	int err;
+
+	err = os_open_file("/dev/random", of_read(OPENFLAGS()), 0);
+	if (err < 0)
+		goto out;
+
+	random_fd = err;
+	err = um_request_irq(RANDOM_IRQ, random_fd, IRQ_READ, random_interrupt,
+			     0, "random", NULL);
+	if (err < 0)
+		goto err_out_cleanup_hw;
+
+	sigio_broken();
+	hwrng.name = RNG_MODULE_NAME;
+	hwrng.read = rng_dev_read;
+
+	err = hwrng_register(&hwrng);
+	if (err) {
+		pr_err(RNG_MODULE_NAME " registering failed (%d)\n", err);
+		goto err_out_cleanup_hw;
+	}
+out:
+	return err;
+
+err_out_cleanup_hw:
+	os_close_file(random_fd);
+	random_fd = -1;
+	goto out;
+}
+
+/*
+ * rng_cleanup - shutdown RNG module
+ */
+
+static void cleanup(void)
+{
+	free_irq_by_fd(random_fd);
+	os_close_file(random_fd);
+}
+
+static void __exit rng_cleanup(void)
+{
+	hwrng_unregister(&hwrng);
+	os_close_file(random_fd);
+}
+
+module_init (rng_init);
+module_exit (rng_cleanup);
+__uml_exitcall(cleanup);
+
+MODULE_DESCRIPTION("UML Host Random Number Generator (RNG) driver");
+MODULE_LICENSE("GPL");

linux-7.2.3/arch/um/drivers/rtc.h

@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2020 Intel Corporation
+ * Author: Johannes Berg <johannes@sipsolutions.net>
+ */
+#ifndef __UM_RTC_H__
+#define __UM_RTC_H__
+
+int uml_rtc_start(bool timetravel);
+int uml_rtc_enable_alarm(unsigned long long delta_seconds);
+void uml_rtc_disable_alarm(void);
+void uml_rtc_stop(bool timetravel);
+void uml_rtc_send_timetravel_alarm(void);
+
+#endif /* __UM_RTC_H__ */

linux-7.2.3/arch/um/drivers/rtc_kern.c

@@ -0,0 +1,213 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020 Intel Corporation
+ * Author: Johannes Berg <johannes@sipsolutions.net>
+ */
+#include <linux/platform_device.h>
+#include <linux/time-internal.h>
+#include <linux/suspend.h>
+#include <linux/err.h>
+#include <linux/rtc.h>
+#include <kern_util.h>
+#include <irq_kern.h>
+#include <os.h>
+#include "rtc.h"
+
+static time64_t uml_rtc_alarm_time;
+static bool uml_rtc_alarm_enabled;
+static struct rtc_device *uml_rtc;
+static int uml_rtc_irq_fd, uml_rtc_irq;
+
+#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
+
+static void uml_rtc_time_travel_alarm(struct time_travel_event *ev)
+{
+	uml_rtc_send_timetravel_alarm();
+}
+
+static struct time_travel_event uml_rtc_alarm_event = {
+	.fn = uml_rtc_time_travel_alarm,
+};
+#endif
+
+static int uml_rtc_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct timespec64 ts;
+
+	/* Use this to get correct time in time-travel mode */
+	read_persistent_clock64(&ts);
+	rtc_time64_to_tm(timespec64_to_ktime(ts) / NSEC_PER_SEC, tm);
+
+	return 0;
+}
+
+static int uml_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	rtc_time64_to_tm(uml_rtc_alarm_time, &alrm->time);
+	alrm->enabled = uml_rtc_alarm_enabled;
+
+	return 0;
+}
+
+static int uml_rtc_alarm_irq_enable(struct device *dev, unsigned int enable)
+{
+	struct timespec64 ts;
+	unsigned long long secs;
+
+	if (!enable && !uml_rtc_alarm_enabled)
+		return 0;
+
+	uml_rtc_alarm_enabled = enable;
+
+	read_persistent_clock64(&ts);
+	secs = uml_rtc_alarm_time - ts.tv_sec;
+
+	if (time_travel_mode == TT_MODE_OFF) {
+		if (!enable) {
+			uml_rtc_disable_alarm();
+			return 0;
+		}
+
+		/* enable or update */
+		return uml_rtc_enable_alarm(secs);
+	} else {
+		time_travel_del_event(&uml_rtc_alarm_event);
+
+		if (enable)
+			time_travel_add_event_rel(&uml_rtc_alarm_event,
+						  secs * NSEC_PER_SEC -
+						  ts.tv_nsec);
+	}
+
+	return 0;
+}
+
+static int uml_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	uml_rtc_alarm_irq_enable(dev, 0);
+	uml_rtc_alarm_time = rtc_tm_to_time64(&alrm->time);
+	uml_rtc_alarm_irq_enable(dev, alrm->enabled);
+
+	return 0;
+}
+
+static const struct rtc_class_ops uml_rtc_ops = {
+	.read_time = uml_rtc_read_time,
+	.read_alarm = uml_rtc_read_alarm,
+	.alarm_irq_enable = uml_rtc_alarm_irq_enable,
+	.set_alarm = uml_rtc_set_alarm,
+};
+
+static irqreturn_t uml_rtc_interrupt(int irq, void *data)
+{
+	unsigned long long c = 0;
+
+	/* alarm triggered, it's now off */
+	uml_rtc_alarm_enabled = false;
+
+	os_read_file(uml_rtc_irq_fd, &c, sizeof(c));
+	WARN_ON(c == 0);
+
+	pm_system_wakeup();
+	rtc_update_irq(uml_rtc, 1, RTC_IRQF | RTC_AF);
+
+	return IRQ_HANDLED;
+}
+
+static int uml_rtc_setup(void)
+{
+	int err;
+
+	err = uml_rtc_start(time_travel_mode != TT_MODE_OFF);
+	if (WARN(err < 0, "err = %d\n", err))
+		return err;
+
+	uml_rtc_irq_fd = err;
+
+	err = um_request_irq(UM_IRQ_ALLOC, uml_rtc_irq_fd, IRQ_READ,
+			     uml_rtc_interrupt, 0, "rtc", NULL);
+	if (err < 0) {
+		uml_rtc_stop(time_travel_mode != TT_MODE_OFF);
+		return err;
+	}
+
+	irq_set_irq_wake(err, 1);
+
+	uml_rtc_irq = err;
+	return 0;
+}
+
+static void uml_rtc_cleanup(void)
+{
+	um_free_irq(uml_rtc_irq, NULL);
+	uml_rtc_stop(time_travel_mode != TT_MODE_OFF);
+}
+
+static int uml_rtc_probe(struct platform_device *pdev)
+{
+	int err;
+
+	err = uml_rtc_setup();
+	if (err)
+		return err;
+
+	uml_rtc = devm_rtc_allocate_device(&pdev->dev);
+	if (IS_ERR(uml_rtc)) {
+		err = PTR_ERR(uml_rtc);
+		goto cleanup;
+	}
+
+	uml_rtc->ops = &uml_rtc_ops;
+
+	device_init_wakeup(&pdev->dev, 1);
+
+	err = devm_rtc_register_device(uml_rtc);
+	if (err)
+		goto cleanup;
+
+	return 0;
+cleanup:
+	uml_rtc_cleanup();
+	return err;
+}
+
+static void uml_rtc_remove(struct platform_device *pdev)
+{
+	device_init_wakeup(&pdev->dev, 0);
+	uml_rtc_cleanup();
+}
+
+static struct platform_driver uml_rtc_driver = {
+	.probe = uml_rtc_probe,
+	.remove = uml_rtc_remove,
+	.driver = {
+		.name = "uml-rtc",
+	},
+};
+
+static int __init uml_rtc_init(void)
+{
+	struct platform_device *pdev;
+	int err;
+
+	err = platform_driver_register(&uml_rtc_driver);
+	if (err)
+		return err;
+
+	pdev = platform_device_alloc("uml-rtc", 0);
+	if (!pdev) {
+		err = -ENOMEM;
+		goto unregister;
+	}
+
+	err = platform_device_add(pdev);
+	if (err)
+		goto unregister;
+	return 0;
+
+unregister:
+	platform_device_put(pdev);
+	platform_driver_unregister(&uml_rtc_driver);
+	return err;
+}
+device_initcall(uml_rtc_init);

linux-7.2.3/arch/um/drivers/rtc_user.c

@@ -0,0 +1,81 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020 Intel Corporation
+ * Author: Johannes Berg <johannes@sipsolutions.net>
+ */
+#include <stdbool.h>
+#include <os.h>
+#include <errno.h>
+#include <sched.h>
+#include <unistd.h>
+#include <kern_util.h>
+#include <sys/select.h>
+#include <stdio.h>
+#include <sys/timerfd.h>
+#include "rtc.h"
+
+static int uml_rtc_irq_fds[2];
+
+void uml_rtc_send_timetravel_alarm(void)
+{
+	unsigned long long c = 1;
+
+	CATCH_EINTR(write(uml_rtc_irq_fds[1], &c, sizeof(c)));
+}
+
+int uml_rtc_start(bool timetravel)
+{
+	int err;
+
+	if (timetravel) {
+		err = os_pipe(uml_rtc_irq_fds, 1, 1);
+		if (err)
+			goto fail;
+	} else {
+		uml_rtc_irq_fds[0] = timerfd_create(CLOCK_REALTIME, TFD_CLOEXEC);
+		if (uml_rtc_irq_fds[0] < 0) {
+			err = -errno;
+			goto fail;
+		}
+
+		/* apparently timerfd won't send SIGIO, use workaround */
+		sigio_broken();
+		err = add_sigio_fd(uml_rtc_irq_fds[0]);
+		if (err < 0) {
+			close(uml_rtc_irq_fds[0]);
+			goto fail;
+		}
+	}
+
+	return uml_rtc_irq_fds[0];
+fail:
+	uml_rtc_stop(timetravel);
+	return err;
+}
+
+int uml_rtc_enable_alarm(unsigned long long delta_seconds)
+{
+	struct itimerspec it = {
+		.it_value = {
+			.tv_sec = delta_seconds,
+		},
+	};
+
+	if (timerfd_settime(uml_rtc_irq_fds[0], 0, &it, NULL))
+		return -errno;
+	return 0;
+}
+
+void uml_rtc_disable_alarm(void)
+{
+	uml_rtc_enable_alarm(0);
+}
+
+void uml_rtc_stop(bool timetravel)
+{
+	if (timetravel)
+		os_close_file(uml_rtc_irq_fds[1]);
+	else
+		ignore_sigio_fd(uml_rtc_irq_fds[0]);
+	os_close_file(uml_rtc_irq_fds[0]);
+}

linux-7.2.3/arch/um/drivers/ssl.c

@@ -0,0 +1,205 @@
+// SPDX-License-Identifier: GPL-2.0
+/* 
+ * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com)
+ */
+
+#include <linux/fs.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/major.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/console.h>
+#include <asm/termbits.h>
+#include <asm/irq.h>
+#include "chan.h"
+#include <init.h>
+#include <irq_user.h>
+#include "mconsole_kern.h"
+
+static const int ssl_version = 1;
+
+#define NR_PORTS 64
+
+static void ssl_announce(char *dev_name, int dev)
+{
+	printk(KERN_INFO "Serial line %d assigned device '%s'\n", dev,
+	       dev_name);
+}
+
+/* Almost const, except that xterm_title may be changed in an initcall */
+static struct chan_opts opts = {
+	.announce 	= ssl_announce,
+	.xterm_title	= "Serial Line #%d",
+	.raw		= 1,
+};
+
+static int ssl_config(char *str, char **error_out);
+static int ssl_get_config(char *dev, char *str, int size, char **error_out);
+static int ssl_remove(int n, char **error_out);
+
+
+/* Const, except for .mc.list */
+static struct line_driver driver = {
+	.name 			= "UML serial line",
+	.device_name 		= "ttyS",
+	.major 			= TTY_MAJOR,
+	.minor_start 		= 64,
+	.type 		 	= TTY_DRIVER_TYPE_SERIAL,
+	.subtype 	 	= 0,
+	.read_irq_name 		= "ssl",
+	.write_irq_name 	= "ssl-write",
+	.mc  = {
+		.list		= LIST_HEAD_INIT(driver.mc.list),
+		.name  		= "ssl",
+		.config 	= ssl_config,
+		.get_config 	= ssl_get_config,
+		.id		= line_id,
+		.remove 	= ssl_remove,
+	},
+};
+
+/* The array is initialized by line_init, at initcall time.  The
+ * elements are locked individually as needed.
+ */
+static char *conf[NR_PORTS];
+static char *def_conf = CONFIG_SSL_CHAN;
+static struct line serial_lines[NR_PORTS];
+
+static int ssl_config(char *str, char **error_out)
+{
+	return line_config(serial_lines, ARRAY_SIZE(serial_lines), str, &opts,
+			   error_out);
+}
+
+static int ssl_get_config(char *dev, char *str, int size, char **error_out)
+{
+	return line_get_config(dev, serial_lines, ARRAY_SIZE(serial_lines), str,
+			       size, error_out);
+}
+
+static int ssl_remove(int n, char **error_out)
+{
+	return line_remove(serial_lines, ARRAY_SIZE(serial_lines), n,
+			   error_out);
+}
+
+static int ssl_install(struct tty_driver *driver, struct tty_struct *tty)
+{
+	return line_install(driver, tty, &serial_lines[tty->index]);
+}
+
+static const struct tty_operations ssl_ops = {
+	.open 	 		= line_open,
+	.close 	 		= line_close,
+	.write 	 		= line_write,
+	.write_room		= line_write_room,
+	.chars_in_buffer 	= line_chars_in_buffer,
+	.flush_buffer 		= line_flush_buffer,
+	.flush_chars 		= line_flush_chars,
+	.throttle 		= line_throttle,
+	.unthrottle 		= line_unthrottle,
+	.install		= ssl_install,
+	.hangup			= line_hangup,
+};
+
+/* Changed by ssl_init and referenced by ssl_exit, which are both serialized
+ * by being an initcall and exitcall, respectively.
+ */
+static int ssl_init_done;
+
+static void ssl_console_write(struct console *c, const char *string,
+			      unsigned len)
+{
+	struct line *line = &serial_lines[c->index];
+	unsigned long flags;
+
+	spin_lock_irqsave(&line->lock, flags);
+	console_write_chan(line->chan_out, string, len);
+	spin_unlock_irqrestore(&line->lock, flags);
+}
+
+static struct tty_driver *ssl_console_device(struct console *c, int *index)
+{
+	*index = c->index;
+	return driver.driver;
+}
+
+static int ssl_console_setup(struct console *co, char *options)
+{
+	struct line *line = &serial_lines[co->index];
+
+	return console_open_chan(line, co);
+}
+
+/* No locking for register_console call - relies on single-threaded initcalls */
+static struct console ssl_cons = {
+	.name		= "ttyS",
+	.write		= ssl_console_write,
+	.device		= ssl_console_device,
+	.setup		= ssl_console_setup,
+	.flags		= CON_PRINTBUFFER|CON_ANYTIME,
+	.index		= -1,
+};
+
+static int ssl_init(void)
+{
+	char *new_title;
+	int err;
+	int i;
+
+	printk(KERN_INFO "Initializing software serial port version %d\n",
+	       ssl_version);
+
+	err = register_lines(&driver, &ssl_ops, serial_lines,
+				    ARRAY_SIZE(serial_lines));
+	if (err)
+		return err;
+
+	new_title = add_xterm_umid(opts.xterm_title);
+	if (new_title != NULL)
+		opts.xterm_title = new_title;
+
+	for (i = 0; i < NR_PORTS; i++) {
+		char *error;
+		char *s = conf[i];
+		if (!s)
+			s = def_conf;
+		if (setup_one_line(serial_lines, i, s, &opts, &error))
+			printk(KERN_ERR "setup_one_line failed for "
+			       "device %d : %s\n", i, error);
+	}
+
+	ssl_init_done = 1;
+	register_console(&ssl_cons);
+	return 0;
+}
+late_initcall(ssl_init);
+
+static void ssl_exit(void)
+{
+	if (!ssl_init_done)
+		return;
+	close_lines(serial_lines, ARRAY_SIZE(serial_lines));
+}
+__uml_exitcall(ssl_exit);
+
+static int ssl_chan_setup(char *str)
+{
+	line_setup(conf, NR_PORTS, &def_conf, str, "serial line");
+	return 1;
+}
+
+__setup("ssl", ssl_chan_setup);
+__channel_help(ssl_chan_setup, "ssl");
+
+static int ssl_non_raw_setup(char *str)
+{
+	opts.raw = 0;
+	return 1;
+}
+__setup("ssl-non-raw", ssl_non_raw_setup);
+__uml_help(ssl_non_raw_setup,
+"ssl-non-raw\n"
+"    Set serial lines to non-raw mode.\n\n"
+);

linux-7.2.3/arch/um/drivers/stderr_console.c

@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/console.h>
+
+#include "chan_user.h"
+
+/* ----------------------------------------------------------------------------- */
+/* trivial console driver -- simply dump everything to stderr                    */
+
+/*
+ * Don't register by default -- as this registers very early in the
+ * boot process it becomes the default console.
+ *
+ * Initialized at init time.
+ */
+static int use_stderr_console = 0;
+
+static void stderr_console_write(struct console *console, const char *string,
+				 unsigned len)
+{
+	generic_write(2 /* stderr */, string, len, NULL);
+}
+
+static struct console stderr_console = {
+	.name		= "stderr",
+	.write		= stderr_console_write,
+	.flags		= CON_PRINTBUFFER,
+};
+
+static int __init stderr_console_init(void)
+{
+	if (use_stderr_console)
+		register_console(&stderr_console);
+	return 0;
+}
+console_initcall(stderr_console_init);
+
+static int stderr_setup(char *str)
+{
+	if (!str)
+		return 0;
+	use_stderr_console = simple_strtoul(str,&str,0);
+	return 1;
+}
+__setup("stderr=", stderr_setup);
+
+/* The previous behavior of not unregistering led to /dev/console being
+ * impossible to open.  My FC5 filesystem started having init die, and the
+ * system panicing because of this.  Unregistering causes the real
+ * console to become the default console, and /dev/console can then be
+ * opened.  Making this an initcall makes this happen late enough that
+ * there is no added value in dumping everything to stderr, and the
+ * normal console is good enough to show you all available output.
+ */
+static int __init unregister_stderr(void)
+{
+	unregister_console(&stderr_console);
+
+	return 0;
+}
+
+__initcall(unregister_stderr);

linux-7.2.3/arch/um/drivers/stdio_console.c

@@ -0,0 +1,198 @@
+// SPDX-License-Identifier: GPL-2.0
+/* 
+ * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
+ */
+
+#include <linux/posix_types.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/types.h>
+#include <linux/major.h>
+#include <linux/kdev_t.h>
+#include <linux/console.h>
+#include <linux/string.h>
+#include <linux/sched.h>
+#include <linux/list.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/hardirq.h>
+#include <asm/current.h>
+#include <asm/irq.h>
+#include "stdio_console.h"
+#include "chan.h"
+#include <irq_user.h>
+#include "mconsole_kern.h"
+#include <init.h>
+
+#define MAX_TTYS (16)
+
+static void stdio_announce(char *dev_name, int dev)
+{
+	printk(KERN_INFO "Virtual console %d assigned device '%s'\n", dev,
+	       dev_name);
+}
+
+/* Almost const, except that xterm_title may be changed in an initcall */
+static struct chan_opts opts = {
+	.announce 	= stdio_announce,
+	.xterm_title	= "Virtual Console #%d",
+	.raw		= 1,
+};
+
+static int con_config(char *str, char **error_out);
+static int con_get_config(char *dev, char *str, int size, char **error_out);
+static int con_remove(int n, char **con_remove);
+
+
+/* Const, except for .mc.list */
+static struct line_driver driver = {
+	.name 			= "UML console",
+	.device_name 		= "tty",
+	.major 			= TTY_MAJOR,
+	.minor_start 		= 0,
+	.type 		 	= TTY_DRIVER_TYPE_CONSOLE,
+	.subtype 	 	= SYSTEM_TYPE_CONSOLE,
+	.read_irq_name 		= "console",
+	.write_irq_name 	= "console-write",
+	.mc  = {
+		.list		= LIST_HEAD_INIT(driver.mc.list),
+		.name  		= "con",
+		.config 	= con_config,
+		.get_config 	= con_get_config,
+		.id		= line_id,
+		.remove 	= con_remove,
+	},
+};
+
+/* The array is initialized by line_init, at initcall time.  The
+ * elements are locked individually as needed.
+ */
+static char *vt_conf[MAX_TTYS];
+static char *def_conf;
+static struct line vts[MAX_TTYS];
+
+static int con_config(char *str, char **error_out)
+{
+	return line_config(vts, ARRAY_SIZE(vts), str, &opts, error_out);
+}
+
+static int con_get_config(char *dev, char *str, int size, char **error_out)
+{
+	return line_get_config(dev, vts, ARRAY_SIZE(vts), str, size, error_out);
+}
+
+static int con_remove(int n, char **error_out)
+{
+	return line_remove(vts, ARRAY_SIZE(vts), n, error_out);
+}
+
+/* Set in an initcall, checked in an exitcall */
+static int con_init_done;
+
+static int con_install(struct tty_driver *driver, struct tty_struct *tty)
+{
+	return line_install(driver, tty, &vts[tty->index]);
+}
+
+static const struct tty_operations console_ops = {
+	.open 	 		= line_open,
+	.install		= con_install,
+	.close 	 		= line_close,
+	.write 	 		= line_write,
+	.write_room		= line_write_room,
+	.chars_in_buffer 	= line_chars_in_buffer,
+	.flush_buffer 		= line_flush_buffer,
+	.flush_chars 		= line_flush_chars,
+	.throttle 		= line_throttle,
+	.unthrottle 		= line_unthrottle,
+	.hangup			= line_hangup,
+};
+
+static void uml_console_write(struct console *console, const char *string,
+			      unsigned len)
+{
+	struct line *line = &vts[console->index];
+	unsigned long flags;
+
+	spin_lock_irqsave(&line->lock, flags);
+	console_write_chan(line->chan_out, string, len);
+	spin_unlock_irqrestore(&line->lock, flags);
+}
+
+static struct tty_driver *uml_console_device(struct console *c, int *index)
+{
+	*index = c->index;
+	return driver.driver;
+}
+
+static int uml_console_setup(struct console *co, char *options)
+{
+	struct line *line = &vts[co->index];
+
+	return console_open_chan(line, co);
+}
+
+/* No locking for register_console call - relies on single-threaded initcalls */
+static struct console stdiocons = {
+	.name		= "tty",
+	.write		= uml_console_write,
+	.device		= uml_console_device,
+	.setup		= uml_console_setup,
+	.flags		= CON_PRINTBUFFER|CON_ANYTIME,
+	.index		= -1,
+};
+
+static int stdio_init(void)
+{
+	char *new_title;
+	int err;
+	int i;
+
+	err = register_lines(&driver, &console_ops, vts,
+					ARRAY_SIZE(vts));
+	if (err)
+		return err;
+
+	printk(KERN_INFO "Initialized stdio console driver\n");
+
+	new_title = add_xterm_umid(opts.xterm_title);
+	if(new_title != NULL)
+		opts.xterm_title = new_title;
+
+	for (i = 0; i < MAX_TTYS; i++) {
+		char *error;
+		char *s = vt_conf[i];
+		if (!s)
+			s = def_conf;
+		if (!s)
+			s = i ? CONFIG_CON_CHAN : CONFIG_CON_ZERO_CHAN;
+		if (setup_one_line(vts, i, s, &opts, &error))
+			printk(KERN_ERR "setup_one_line failed for "
+			       "device %d : %s\n", i, error);
+	}
+
+	con_init_done = 1;
+	register_console(&stdiocons);
+	return 0;
+}
+late_initcall(stdio_init);
+
+static void console_exit(void)
+{
+	if (!con_init_done)
+		return;
+	close_lines(vts, ARRAY_SIZE(vts));
+}
+__uml_exitcall(console_exit);
+
+static int console_chan_setup(char *str)
+{
+	if (!strncmp(str, "sole=", 5))	/* console= option specifies tty */
+		return 0;
+
+	line_setup(vt_conf, MAX_TTYS, &def_conf, str, "console");
+	return 1;
+}
+__setup("con", console_chan_setup);
+__channel_help(console_chan_setup, "con");

linux-7.2.3/arch/um/drivers/stdio_console.h

@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __STDIO_CONSOLE_H
+#define __STDIO_CONSOLE_H
+
+extern void save_console_flags(void);
+#endif
+

linux-7.2.3/arch/um/drivers/tty.c

@@ -0,0 +1,81 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <termios.h>
+#include "chan_user.h"
+#include <os.h>
+#include <um_malloc.h>
+
+struct tty_chan {
+	char *dev;
+	int raw;
+	struct termios tt;
+};
+
+static void *tty_chan_init(char *str, int device, const struct chan_opts *opts)
+{
+	struct tty_chan *data;
+
+	if (*str != ':') {
+		printk(UM_KERN_ERR "tty_init : channel type 'tty' must specify "
+		       "a device\n");
+		return NULL;
+	}
+	str++;
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_KERNEL);
+	if (data == NULL)
+		return NULL;
+	*data = ((struct tty_chan) { .dev 	= str,
+				     .raw 	= opts->raw });
+
+	return data;
+}
+
+static int tty_open(int input, int output, int primary, void *d,
+		    char **dev_out)
+{
+	struct tty_chan *data = d;
+	int fd, err, mode = 0;
+
+	if (input && output)
+		mode = O_RDWR;
+	else if (input)
+		mode = O_RDONLY;
+	else if (output)
+		mode = O_WRONLY;
+
+	fd = open(data->dev, mode);
+	if (fd < 0)
+		return -errno;
+
+	if (data->raw) {
+		CATCH_EINTR(err = tcgetattr(fd, &data->tt));
+		if (err)
+			return err;
+
+		err = raw(fd);
+		if (err)
+			return err;
+	}
+
+	*dev_out = data->dev;
+	return fd;
+}
+
+const struct chan_ops tty_ops = {
+	.type		= "tty",
+	.init		= tty_chan_init,
+	.open		= tty_open,
+	.close		= generic_close,
+	.read		= generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= generic_free,
+	.winch		= 0,
+};

linux-7.2.3/arch/um/drivers/ubd.h

@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2000 Jeff Dike (jdike@karaya.com)
+ * Copyright (C) 2001 RidgeRun, Inc (glonnon@ridgerun.com)
+ */
+
+#ifndef __UM_UBD_USER_H
+#define __UM_UBD_USER_H
+
+#include <os.h>
+
+int start_io_thread(struct os_helper_thread **td_out, int *fd_out);
+void *io_thread(void *arg);
+extern int kernel_fd;
+
+extern int ubd_read_poll(int timeout);
+extern int ubd_write_poll(int timeout);
+
+#define UBD_REQ_BUFFER_SIZE 64
+
+#endif
+

linux-7.2.3/arch/um/drivers/ubd_user.c

@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2016 Anton Ivanov (aivanov@brocade.com)
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright (C) 2001 Ridgerun,Inc (glonnon@ridgerun.com)
+ */
+
+#include <stddef.h>
+#include <unistd.h>
+#include <errno.h>
+#include <sched.h>
+#include <signal.h>
+#include <string.h>
+#include <netinet/in.h>
+#include <sys/time.h>
+#include <sys/socket.h>
+#include <sys/mman.h>
+#include <sys/param.h>
+#include <endian.h>
+#include <byteswap.h>
+
+#include "ubd.h"
+#include <os.h>
+#include <poll.h>
+
+static struct pollfd kernel_pollfd;
+
+int start_io_thread(struct os_helper_thread **td_out, int *fd_out)
+{
+	int fds[2], err;
+
+	err = os_pipe(fds, 1, 1);
+	if(err < 0){
+		printk("start_io_thread - os_pipe failed, err = %d\n", -err);
+		goto out;
+	}
+
+	kernel_fd = fds[0];
+	kernel_pollfd.fd = kernel_fd;
+	kernel_pollfd.events = POLLIN;
+	*fd_out = fds[1];
+
+	err = os_set_fd_block(*fd_out, 0);
+	err |= os_set_fd_block(kernel_fd, 0);
+	if (err) {
+		printk("start_io_thread - failed to set nonblocking I/O.\n");
+		goto out_close;
+	}
+
+	err = os_run_helper_thread(td_out, io_thread, NULL);
+	if (err < 0) {
+		printk("%s - failed to run helper thread, err = %d\n",
+		       __func__, -err);
+		goto out_close;
+	}
+
+	return 0;
+
+ out_close:
+	os_close_file(fds[0]);
+	os_close_file(fds[1]);
+	kernel_fd = -1;
+	*fd_out = -1;
+ out:
+	return err;
+}
+
+int ubd_read_poll(int timeout)
+{
+	kernel_pollfd.events = POLLIN;
+	return poll(&kernel_pollfd, 1, timeout);
+}
+int ubd_write_poll(int timeout)
+{
+	kernel_pollfd.events = POLLOUT;
+	return poll(&kernel_pollfd, 1, timeout);
+}
+

linux-7.2.3/arch/um/drivers/vector_kern.h

@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2002 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#ifndef __UM_VECTOR_KERN_H
+#define __UM_VECTOR_KERN_H
+
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+#include <linux/skbuff.h>
+#include <linux/socket.h>
+#include <linux/list.h>
+#include <linux/ctype.h>
+#include <linux/workqueue.h>
+#include <linux/interrupt.h>
+#include <asm/atomic.h>
+
+#include "vector_user.h"
+
+/* Queue structure specially adapted for multiple enqueue/dequeue
+ * in a mmsgrecv/mmsgsend context
+ */
+
+/* Dequeue method */
+
+#define QUEUE_SENDMSG 0
+#define QUEUE_SENDMMSG 1
+
+#define VECTOR_RX 1
+#define VECTOR_TX (1 << 1)
+#define VECTOR_BPF (1 << 2)
+#define VECTOR_QDISC_BYPASS (1 << 3)
+#define VECTOR_BPF_FLASH (1 << 4)
+
+#define ETH_MAX_PACKET 1500
+#define ETH_HEADER_OTHER 32 /* just in case someone decides to go mad on QnQ */
+
+#define MAX_FILTER_PROG (2 << 16)
+
+struct vector_queue {
+	struct mmsghdr *mmsg_vector;
+	void **skbuff_vector;
+	 /* backlink to device which owns us */
+	struct net_device *dev;
+	spinlock_t head_lock;
+	spinlock_t tail_lock;
+	atomic_t queue_depth;
+	int head, tail, max_depth, max_iov_frags;
+	short options;
+};
+
+struct vector_estats {
+	uint64_t rx_queue_max;
+	uint64_t rx_queue_running_average;
+	uint64_t tx_queue_max;
+	uint64_t tx_queue_running_average;
+	uint64_t rx_encaps_errors;
+	uint64_t tx_timeout_count;
+	uint64_t tx_restart_queue;
+	uint64_t tx_kicks;
+	uint64_t tx_flow_control_xon;
+	uint64_t tx_flow_control_xoff;
+	uint64_t rx_csum_offload_good;
+	uint64_t rx_csum_offload_errors;
+	uint64_t sg_ok;
+	uint64_t sg_linearized;
+};
+
+#define VERIFY_HEADER_NOK -1
+#define VERIFY_HEADER_OK 0
+#define VERIFY_CSUM_OK 1
+
+struct vector_private {
+	struct list_head list;
+	struct net_device *dev;
+	struct napi_struct		napi	____cacheline_aligned;
+
+	int unit;
+
+	/* Timeout timer in TX */
+
+	struct timer_list tl;
+
+	/* Scheduled "remove device" work */
+	struct work_struct reset_tx;
+	struct vector_fds *fds;
+
+	struct vector_queue *rx_queue;
+	struct vector_queue *tx_queue;
+
+	int rx_irq;
+	int tx_irq;
+
+	struct arglist *parsed;
+
+	void *transport_data; /* transport specific params if needed */
+
+	int max_packet;
+	int req_size; /* different from max packet - used for TSO */
+	int headroom;
+
+	int options;
+
+	/* remote address if any - some transports will leave this as null */
+
+	int header_size;
+	int rx_header_size;
+	int coalesce;
+
+	void *header_rxbuffer;
+	void *header_txbuffer;
+
+	int (*form_header)(uint8_t *header,
+		struct sk_buff *skb, struct vector_private *vp);
+	int (*verify_header)(uint8_t *header,
+		struct sk_buff *skb, struct vector_private *vp);
+
+	spinlock_t stats_lock;
+
+	bool rexmit_scheduled;
+	bool opened;
+	bool in_write_poll;
+	bool in_error;
+
+	/* guest allowed to use ethtool flash to load bpf */
+	bool bpf_via_flash;
+
+	/* ethtool stats */
+
+	struct vector_estats estats;
+	struct sock_fprog *bpf;
+
+	char user[];
+};
+
+extern int build_transport_data(struct vector_private *vp);
+
+#endif

linux-7.2.3/arch/um/drivers/vector_transports.c

@@ -0,0 +1,494 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2017 - Cambridge Greys Limited
+ * Copyright (C) 2011 - 2014 Cisco Systems Inc
+ */
+
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <asm/byteorder.h>
+#include <uapi/linux/ip.h>
+#include <uapi/linux/virtio_net.h>
+#include <linux/virtio_net.h>
+#include <linux/virtio_byteorder.h>
+#include <linux/netdev_features.h>
+#include "vector_user.h"
+#include "vector_kern.h"
+
+#define GOOD_LINEAR 512
+#define GSO_ERROR "Incoming GSO frames and GRO disabled on the interface"
+
+struct gre_minimal_header {
+	uint16_t header;
+	uint16_t arptype;
+};
+
+
+struct uml_gre_data {
+	uint32_t rx_key;
+	uint32_t tx_key;
+	uint32_t sequence;
+
+	bool ipv6;
+	bool has_sequence;
+	bool pin_sequence;
+	bool checksum;
+	bool key;
+	struct gre_minimal_header expected_header;
+
+	uint32_t checksum_offset;
+	uint32_t key_offset;
+	uint32_t sequence_offset;
+
+};
+
+struct uml_l2tpv3_data {
+	uint64_t rx_cookie;
+	uint64_t tx_cookie;
+	uint64_t rx_session;
+	uint64_t tx_session;
+	uint32_t counter;
+
+	bool udp;
+	bool ipv6;
+	bool has_counter;
+	bool pin_counter;
+	bool cookie;
+	bool cookie_is_64;
+
+	uint32_t cookie_offset;
+	uint32_t session_offset;
+	uint32_t counter_offset;
+};
+
+static int l2tpv3_form_header(uint8_t *header,
+	struct sk_buff *skb, struct vector_private *vp)
+{
+	struct uml_l2tpv3_data *td = vp->transport_data;
+	uint32_t *counter;
+
+	if (td->udp)
+		*(uint32_t *) header = cpu_to_be32(L2TPV3_DATA_PACKET);
+	(*(uint32_t *) (header + td->session_offset)) = td->tx_session;
+
+	if (td->cookie) {
+		if (td->cookie_is_64)
+			(*(uint64_t *)(header + td->cookie_offset)) =
+				td->tx_cookie;
+		else
+			(*(uint32_t *)(header + td->cookie_offset)) =
+				td->tx_cookie;
+	}
+	if (td->has_counter) {
+		counter = (uint32_t *)(header + td->counter_offset);
+		if (td->pin_counter) {
+			*counter = 0;
+		} else {
+			td->counter++;
+			*counter = cpu_to_be32(td->counter);
+		}
+	}
+	return 0;
+}
+
+static int gre_form_header(uint8_t *header,
+		struct sk_buff *skb, struct vector_private *vp)
+{
+	struct uml_gre_data *td = vp->transport_data;
+	uint32_t *sequence;
+	*((uint32_t *) header) = *((uint32_t *) &td->expected_header);
+	if (td->key)
+		(*(uint32_t *) (header + td->key_offset)) = td->tx_key;
+	if (td->has_sequence) {
+		sequence = (uint32_t *)(header + td->sequence_offset);
+		if (td->pin_sequence)
+			*sequence = 0;
+		else
+			*sequence = cpu_to_be32(++td->sequence);
+	}
+	return 0;
+}
+
+static int raw_form_header(uint8_t *header,
+		struct sk_buff *skb, struct vector_private *vp)
+{
+	struct virtio_net_hdr *vheader = (struct virtio_net_hdr *) header;
+
+	virtio_net_hdr_from_skb(
+		skb,
+		vheader,
+		virtio_legacy_is_little_endian(),
+		false,
+		0
+	);
+
+	return 0;
+}
+
+static int l2tpv3_verify_header(
+	uint8_t *header, struct sk_buff *skb, struct vector_private *vp)
+{
+	struct uml_l2tpv3_data *td = vp->transport_data;
+	uint32_t *session;
+	uint64_t cookie;
+
+	if ((!td->udp) && (!td->ipv6))
+		header += sizeof(struct iphdr) /* fix for ipv4 raw */;
+
+	/* we do not do a strict check for "data" packets as per
+	 * the RFC spec because the pure IP spec does not have
+	 * that anyway.
+	 */
+
+	if (td->cookie) {
+		if (td->cookie_is_64)
+			cookie = *(uint64_t *)(header + td->cookie_offset);
+		else
+			cookie = *(uint32_t *)(header + td->cookie_offset);
+		if (cookie != td->rx_cookie) {
+			if (net_ratelimit())
+				netdev_err(vp->dev, "uml_l2tpv3: unknown cookie id");
+			return -1;
+		}
+	}
+	session = (uint32_t *) (header + td->session_offset);
+	if (*session != td->rx_session) {
+		if (net_ratelimit())
+			netdev_err(vp->dev, "uml_l2tpv3: session mismatch");
+		return -1;
+	}
+	return 0;
+}
+
+static int gre_verify_header(
+	uint8_t *header, struct sk_buff *skb, struct vector_private *vp)
+{
+
+	uint32_t key;
+	struct uml_gre_data *td = vp->transport_data;
+
+	if (!td->ipv6)
+		header += sizeof(struct iphdr) /* fix for ipv4 raw */;
+
+	if (*((uint32_t *) header) != *((uint32_t *) &td->expected_header)) {
+		if (net_ratelimit())
+			netdev_err(vp->dev, "header type disagreement, expecting %0x, got %0x",
+				*((uint32_t *) &td->expected_header),
+				*((uint32_t *) header)
+			);
+		return -1;
+	}
+
+	if (td->key) {
+		key = (*(uint32_t *)(header + td->key_offset));
+		if (key != td->rx_key) {
+			if (net_ratelimit())
+				netdev_err(vp->dev, "unknown key id %0x, expecting %0x",
+						key, td->rx_key);
+			return -1;
+		}
+	}
+	return 0;
+}
+
+static int raw_verify_header(
+	uint8_t *header, struct sk_buff *skb, struct vector_private *vp)
+{
+	struct virtio_net_hdr *vheader = (struct virtio_net_hdr *) header;
+
+	if ((vheader->gso_type != VIRTIO_NET_HDR_GSO_NONE) &&
+		(vp->req_size != 65536)) {
+		if (net_ratelimit())
+			netdev_err(
+				vp->dev,
+				GSO_ERROR
+		);
+	}
+	if ((vheader->flags & VIRTIO_NET_HDR_F_DATA_VALID) > 0)
+		return 1;
+
+	virtio_net_hdr_to_skb(skb, vheader, virtio_legacy_is_little_endian());
+	return 0;
+}
+
+static bool get_uint_param(
+	struct arglist *def, char *param, unsigned int *result)
+{
+	char *arg = uml_vector_fetch_arg(def, param);
+
+	if (arg != NULL) {
+		if (kstrtoint(arg, 0, result) == 0)
+			return true;
+	}
+	return false;
+}
+
+static bool get_ulong_param(
+	struct arglist *def, char *param, unsigned long *result)
+{
+	char *arg = uml_vector_fetch_arg(def, param);
+
+	if (arg != NULL) {
+		if (kstrtoul(arg, 0, result) == 0)
+			return true;
+		return true;
+	}
+	return false;
+}
+
+static int build_gre_transport_data(struct vector_private *vp)
+{
+	struct uml_gre_data *td;
+	int temp_int;
+	int temp_rx;
+	int temp_tx;
+
+	vp->transport_data = kmalloc_obj(struct uml_gre_data);
+	if (vp->transport_data == NULL)
+		return -ENOMEM;
+	td = vp->transport_data;
+	td->sequence = 0;
+
+	td->expected_header.arptype = GRE_IRB;
+	td->expected_header.header = 0;
+
+	vp->form_header = &gre_form_header;
+	vp->verify_header = &gre_verify_header;
+	vp->header_size = 4;
+	td->key_offset = 4;
+	td->sequence_offset = 4;
+	td->checksum_offset = 4;
+
+	td->ipv6 = false;
+	if (get_uint_param(vp->parsed, "v6", &temp_int)) {
+		if (temp_int > 0)
+			td->ipv6 = true;
+	}
+	td->key = false;
+	if (get_uint_param(vp->parsed, "rx_key", &temp_rx)) {
+		if (get_uint_param(vp->parsed, "tx_key", &temp_tx)) {
+			td->key = true;
+			td->expected_header.header |= GRE_MODE_KEY;
+			td->rx_key = cpu_to_be32(temp_rx);
+			td->tx_key = cpu_to_be32(temp_tx);
+			vp->header_size += 4;
+			td->sequence_offset += 4;
+		} else {
+			return -EINVAL;
+		}
+	}
+
+	td->sequence = false;
+	if (get_uint_param(vp->parsed, "sequence", &temp_int)) {
+		if (temp_int > 0) {
+			vp->header_size += 4;
+			td->has_sequence = true;
+			td->expected_header.header |= GRE_MODE_SEQUENCE;
+			if (get_uint_param(
+				vp->parsed, "pin_sequence", &temp_int)) {
+				if (temp_int > 0)
+					td->pin_sequence = true;
+			}
+		}
+	}
+	vp->rx_header_size = vp->header_size;
+	if (!td->ipv6)
+		vp->rx_header_size += sizeof(struct iphdr);
+	return 0;
+}
+
+static int build_l2tpv3_transport_data(struct vector_private *vp)
+{
+
+	struct uml_l2tpv3_data *td;
+	int temp_int, temp_rxs, temp_txs;
+	unsigned long temp_rx;
+	unsigned long temp_tx;
+
+	vp->transport_data = kmalloc_obj(struct uml_l2tpv3_data);
+
+	if (vp->transport_data == NULL)
+		return -ENOMEM;
+
+	td = vp->transport_data;
+
+	vp->form_header = &l2tpv3_form_header;
+	vp->verify_header = &l2tpv3_verify_header;
+	td->counter = 0;
+
+	vp->header_size = 4;
+	td->session_offset = 0;
+	td->cookie_offset = 4;
+	td->counter_offset = 4;
+
+
+	td->ipv6 = false;
+	if (get_uint_param(vp->parsed, "v6", &temp_int)) {
+		if (temp_int > 0)
+			td->ipv6 = true;
+	}
+
+	if (get_uint_param(vp->parsed, "rx_session", &temp_rxs)) {
+		if (get_uint_param(vp->parsed, "tx_session", &temp_txs)) {
+			td->tx_session = cpu_to_be32(temp_txs);
+			td->rx_session = cpu_to_be32(temp_rxs);
+		} else {
+			return -EINVAL;
+		}
+	} else {
+		return -EINVAL;
+	}
+
+	td->cookie_is_64  = false;
+	if (get_uint_param(vp->parsed, "cookie64", &temp_int)) {
+		if (temp_int > 0)
+			td->cookie_is_64  = true;
+	}
+	td->cookie = false;
+	if (get_ulong_param(vp->parsed, "rx_cookie", &temp_rx)) {
+		if (get_ulong_param(vp->parsed, "tx_cookie", &temp_tx)) {
+			td->cookie = true;
+			if (td->cookie_is_64) {
+				td->rx_cookie = cpu_to_be64(temp_rx);
+				td->tx_cookie = cpu_to_be64(temp_tx);
+				vp->header_size += 8;
+				td->counter_offset += 8;
+			} else {
+				td->rx_cookie = cpu_to_be32(temp_rx);
+				td->tx_cookie = cpu_to_be32(temp_tx);
+				vp->header_size += 4;
+				td->counter_offset += 4;
+			}
+		} else {
+			return -EINVAL;
+		}
+	}
+
+	td->has_counter = false;
+	if (get_uint_param(vp->parsed, "counter", &temp_int)) {
+		if (temp_int > 0) {
+			td->has_counter = true;
+			vp->header_size += 4;
+			if (get_uint_param(
+				vp->parsed, "pin_counter", &temp_int)) {
+				if (temp_int > 0)
+					td->pin_counter = true;
+			}
+		}
+	}
+
+	if (get_uint_param(vp->parsed, "udp", &temp_int)) {
+		if (temp_int > 0) {
+			td->udp = true;
+			vp->header_size += 4;
+			td->counter_offset += 4;
+			td->session_offset += 4;
+			td->cookie_offset += 4;
+		}
+	}
+
+	vp->rx_header_size = vp->header_size;
+	if ((!td->ipv6) && (!td->udp))
+		vp->rx_header_size += sizeof(struct iphdr);
+
+	return 0;
+}
+
+static int build_raw_transport_data(struct vector_private *vp)
+{
+	if (uml_raw_enable_vnet_headers(vp->fds->rx_fd)) {
+		if (!uml_raw_enable_vnet_headers(vp->fds->tx_fd))
+			return -1;
+		vp->form_header = &raw_form_header;
+		vp->verify_header = &raw_verify_header;
+		vp->header_size = sizeof(struct virtio_net_hdr);
+		vp->rx_header_size = sizeof(struct virtio_net_hdr);
+		vp->dev->hw_features |= (NETIF_F_TSO | NETIF_F_GRO);
+		vp->dev->features |=
+			(NETIF_F_RXCSUM | NETIF_F_HW_CSUM |
+				NETIF_F_TSO | NETIF_F_GRO);
+		netdev_info(
+			vp->dev,
+			"raw: using vnet headers for tso and tx/rx checksum"
+		);
+	}
+	return 0;
+}
+
+static int build_hybrid_transport_data(struct vector_private *vp)
+{
+	if (uml_raw_enable_vnet_headers(vp->fds->rx_fd)) {
+		vp->form_header = &raw_form_header;
+		vp->verify_header = &raw_verify_header;
+		vp->header_size = sizeof(struct virtio_net_hdr);
+		vp->rx_header_size = sizeof(struct virtio_net_hdr);
+		vp->dev->hw_features |=
+			(NETIF_F_TSO | NETIF_F_GSO | NETIF_F_GRO);
+		vp->dev->features |=
+			(NETIF_F_RXCSUM | NETIF_F_HW_CSUM |
+				NETIF_F_TSO | NETIF_F_GSO | NETIF_F_GRO);
+		netdev_info(
+			vp->dev,
+			"tap/raw hybrid: using vnet headers for tso and tx/rx checksum"
+		);
+	} else {
+		return 0; /* do not try to enable tap too if raw failed */
+	}
+	if (uml_tap_enable_vnet_headers(vp->fds->tx_fd))
+		return 0;
+	return -1;
+}
+
+static int build_tap_transport_data(struct vector_private *vp)
+{
+	/* "Pure" tap uses the same fd for rx and tx */
+	if (uml_tap_enable_vnet_headers(vp->fds->tx_fd)) {
+		vp->form_header = &raw_form_header;
+		vp->verify_header = &raw_verify_header;
+		vp->header_size = sizeof(struct virtio_net_hdr);
+		vp->rx_header_size = sizeof(struct virtio_net_hdr);
+		vp->dev->hw_features |=
+			(NETIF_F_TSO | NETIF_F_GSO | NETIF_F_GRO);
+		vp->dev->features |=
+			(NETIF_F_RXCSUM | NETIF_F_HW_CSUM |
+				NETIF_F_TSO | NETIF_F_GSO | NETIF_F_GRO);
+		netdev_info(
+			vp->dev,
+			"tap: using vnet headers for tso and tx/rx checksum"
+		);
+		return 0;
+	}
+	return -1;
+}
+
+
+static int build_bess_transport_data(struct vector_private *vp)
+{
+	vp->form_header = NULL;
+	vp->verify_header = NULL;
+	vp->header_size = 0;
+	vp->rx_header_size = 0;
+	return 0;
+}
+
+int build_transport_data(struct vector_private *vp)
+{
+	char *transport = uml_vector_fetch_arg(vp->parsed, "transport");
+
+	if (strncmp(transport, TRANS_GRE, TRANS_GRE_LEN) == 0)
+		return build_gre_transport_data(vp);
+	if (strncmp(transport, TRANS_L2TPV3, TRANS_L2TPV3_LEN) == 0)
+		return build_l2tpv3_transport_data(vp);
+	if (strncmp(transport, TRANS_RAW, TRANS_RAW_LEN) == 0)
+		return build_raw_transport_data(vp);
+	if (strncmp(transport, TRANS_TAP, TRANS_TAP_LEN) == 0)
+		return build_tap_transport_data(vp);
+	if (strncmp(transport, TRANS_HYBRID, TRANS_HYBRID_LEN) == 0)
+		return build_hybrid_transport_data(vp);
+	if (strncmp(transport, TRANS_BESS, TRANS_BESS_LEN) == 0)
+		return build_bess_transport_data(vp);
+	return 0;
+}
+

linux-7.2.3/arch/um/drivers/vector_user.c

@@ -0,0 +1,941 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdbool.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <stdarg.h>
+#include <errno.h>
+#include <stddef.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <net/if.h>
+#include <linux/if_tun.h>
+#include <arpa/inet.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <sys/socket.h>
+#include <sys/un.h>
+#include <netinet/ip.h>
+#include <linux/if_ether.h>
+#include <linux/if_packet.h>
+#include <sys/wait.h>
+#include <sys/uio.h>
+#include <linux/virtio_net.h>
+#include <netdb.h>
+#include <stdlib.h>
+#include <os.h>
+#include <limits.h>
+#include <um_malloc.h>
+#include "vector_user.h"
+
+#define ID_GRE 0
+#define ID_L2TPV3 1
+#define ID_BESS 2
+#define ID_MAX 2
+
+#define TOKEN_IFNAME "ifname"
+#define TOKEN_SCRIPT "ifup"
+
+#define TRANS_RAW "raw"
+#define TRANS_RAW_LEN strlen(TRANS_RAW)
+
+#define TRANS_FD "fd"
+#define TRANS_FD_LEN strlen(TRANS_FD)
+
+#define TRANS_VDE "vde"
+#define TRANS_VDE_LEN strlen(TRANS_VDE)
+
+#define VNET_HDR_FAIL "could not enable vnet headers on fd %d"
+#define TUN_GET_F_FAIL "tapraw: TUNGETFEATURES failed: %s"
+#define L2TPV3_BIND_FAIL "l2tpv3_open : could not bind socket err=%i"
+#define UNIX_BIND_FAIL "unix_open : could not bind socket err=%i"
+#define BPF_ATTACH_FAIL "Failed to attach filter size %d prog %px to %d, err %d\n"
+#define BPF_DETACH_FAIL "Failed to detach filter size %d prog %px to %d, err %d\n"
+
+#define MAX_UN_LEN 107
+
+static const char padchar[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
+static const char *template = "tapXXXXXX";
+
+/* This is very ugly and brute force lookup, but it is done
+ * only once at initialization so not worth doing hashes or
+ * anything more intelligent
+ */
+
+char *uml_vector_fetch_arg(struct arglist *ifspec, char *token)
+{
+	int i;
+
+	for (i = 0; i < ifspec->numargs; i++) {
+		if (strcmp(ifspec->tokens[i], token) == 0)
+			return ifspec->values[i];
+	}
+	return NULL;
+
+}
+
+struct arglist *uml_parse_vector_ifspec(char *arg)
+{
+	struct arglist *result;
+	int pos, len;
+	bool parsing_token = true, next_starts = true;
+
+	if (arg == NULL)
+		return NULL;
+	result = uml_kmalloc(sizeof(struct arglist), UM_GFP_KERNEL);
+	if (result == NULL)
+		return NULL;
+	result->numargs = 0;
+	len = strlen(arg);
+	for (pos = 0; pos < len; pos++) {
+		if (next_starts) {
+			if (parsing_token) {
+				result->tokens[result->numargs] = arg + pos;
+			} else {
+				result->values[result->numargs] = arg + pos;
+				result->numargs++;
+			}
+			next_starts = false;
+		}
+		if (*(arg + pos) == '=') {
+			if (parsing_token)
+				parsing_token = false;
+			else
+				goto cleanup;
+			next_starts = true;
+			(*(arg + pos)) = '\0';
+		}
+		if (*(arg + pos) == ',') {
+			parsing_token = true;
+			next_starts = true;
+			(*(arg + pos)) = '\0';
+		}
+	}
+	return result;
+cleanup:
+	printk(UM_KERN_ERR "vector_setup - Couldn't parse '%s'\n", arg);
+	kfree(result);
+	return NULL;
+}
+
+/*
+ * Socket/FD configuration functions. These return an structure
+ * of rx and tx descriptors to cover cases where these are not
+ * the same (f.e. read via raw socket and write via tap).
+ */
+
+#define PATH_NET_TUN "/dev/net/tun"
+
+
+static int create_tap_fd(char *iface)
+{
+	struct ifreq ifr;
+	int fd = -1;
+	int err = -ENOMEM, offload;
+
+	fd = open(PATH_NET_TUN, O_RDWR);
+	if (fd < 0) {
+		printk(UM_KERN_ERR "uml_tap: failed to open tun device\n");
+		goto tap_fd_cleanup;
+	}
+	memset(&ifr, 0, sizeof(ifr));
+	ifr.ifr_flags = IFF_TAP | IFF_NO_PI | IFF_VNET_HDR;
+	strscpy(ifr.ifr_name, iface);
+
+	err = ioctl(fd, TUNSETIFF, (void *) &ifr);
+	if (err != 0) {
+		printk(UM_KERN_ERR "uml_tap: failed to select tap interface\n");
+		goto tap_fd_cleanup;
+	}
+
+	offload = TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6;
+	ioctl(fd, TUNSETOFFLOAD, offload);
+	return fd;
+tap_fd_cleanup:
+	if (fd >= 0)
+		os_close_file(fd);
+	return err;
+}
+
+static int create_raw_fd(char *iface, int flags, int proto)
+{
+	struct ifreq ifr;
+	int fd = -1;
+	struct sockaddr_ll sock;
+	int err = -ENOMEM;
+
+	fd = socket(AF_PACKET, SOCK_RAW, flags);
+	if (fd == -1) {
+		err = -errno;
+		goto raw_fd_cleanup;
+	}
+	memset(&ifr, 0, sizeof(ifr));
+	strscpy(ifr.ifr_name, iface);
+	if (ioctl(fd, SIOCGIFINDEX, (void *) &ifr) < 0) {
+		err = -errno;
+		goto raw_fd_cleanup;
+	}
+
+	sock.sll_family = AF_PACKET;
+	sock.sll_protocol = htons(proto);
+	sock.sll_ifindex = ifr.ifr_ifindex;
+
+	if (bind(fd,
+		(struct sockaddr *) &sock, sizeof(struct sockaddr_ll)) < 0) {
+		err = -errno;
+		goto raw_fd_cleanup;
+	}
+	return fd;
+raw_fd_cleanup:
+	printk(UM_KERN_ERR "user_init_raw: init failed, error %d", err);
+	if (fd >= 0)
+		os_close_file(fd);
+	return err;
+}
+
+
+static struct vector_fds *user_init_tap_fds(struct arglist *ifspec)
+{
+	int fd = -1, i;
+	char *iface;
+	struct vector_fds *result = NULL;
+	bool dynamic = false;
+	char dynamic_ifname[IFNAMSIZ];
+	char *argv[] = {NULL, NULL, NULL, NULL};
+
+	iface = uml_vector_fetch_arg(ifspec, TOKEN_IFNAME);
+	if (iface == NULL) {
+		dynamic = true;
+		iface = dynamic_ifname;
+		srand(getpid());
+	}
+
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result == NULL) {
+		printk(UM_KERN_ERR "uml_tap: failed to allocate file descriptors\n");
+		goto tap_cleanup;
+	}
+	result->rx_fd = -1;
+	result->tx_fd = -1;
+	result->remote_addr = NULL;
+	result->remote_addr_size = 0;
+
+	/* TAP */
+	do {
+		if (dynamic) {
+			strcpy(iface, template);
+			for (i = 0; i < strlen(iface); i++) {
+				if (iface[i] == 'X') {
+					iface[i] = padchar[rand() % strlen(padchar)];
+				}
+			}
+		}
+		fd = create_tap_fd(iface);
+		if ((fd < 0) && (!dynamic)) {
+			printk(UM_KERN_ERR "uml_tap: failed to create tun interface\n");
+			goto tap_cleanup;
+		}
+		result->tx_fd = fd;
+		result->rx_fd = fd;
+	} while (fd < 0);
+
+	argv[0] = uml_vector_fetch_arg(ifspec, TOKEN_SCRIPT);
+	if (argv[0]) {
+		argv[1] = iface;
+		run_helper(NULL, NULL, argv);
+	}
+
+	return result;
+tap_cleanup:
+	printk(UM_KERN_ERR "user_init_tap: init failed, error %d", fd);
+	kfree(result);
+	return NULL;
+}
+
+static struct vector_fds *user_init_hybrid_fds(struct arglist *ifspec)
+{
+	char *iface;
+	struct vector_fds *result = NULL;
+	char *argv[] = {NULL, NULL, NULL, NULL};
+
+	iface = uml_vector_fetch_arg(ifspec, TOKEN_IFNAME);
+	if (iface == NULL) {
+		printk(UM_KERN_ERR "uml_tap: failed to parse interface spec\n");
+		goto hybrid_cleanup;
+	}
+
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result == NULL) {
+		printk(UM_KERN_ERR "uml_tap: failed to allocate file descriptors\n");
+		goto hybrid_cleanup;
+	}
+	result->rx_fd = -1;
+	result->tx_fd = -1;
+	result->remote_addr = NULL;
+	result->remote_addr_size = 0;
+
+	/* TAP */
+
+	result->tx_fd = create_tap_fd(iface);
+	if (result->tx_fd < 0) {
+		printk(UM_KERN_ERR "uml_tap: failed to create tun interface: %i\n", result->tx_fd);
+		goto hybrid_cleanup;
+	}
+
+	/* RAW */
+
+	result->rx_fd = create_raw_fd(iface, ETH_P_ALL, ETH_P_ALL);
+	if (result->rx_fd == -1) {
+		printk(UM_KERN_ERR
+			"uml_tap: failed to create paired raw socket: %i\n", result->rx_fd);
+		goto hybrid_cleanup;
+	}
+
+	argv[0] = uml_vector_fetch_arg(ifspec, TOKEN_SCRIPT);
+	if (argv[0]) {
+		argv[1] = iface;
+		run_helper(NULL, NULL, argv);
+	}
+	return result;
+hybrid_cleanup:
+	printk(UM_KERN_ERR "user_init_hybrid: init failed");
+	kfree(result);
+	return NULL;
+}
+
+static struct vector_fds *user_init_unix_fds(struct arglist *ifspec, int id)
+{
+	int fd = -1;
+	int socktype;
+	char *src, *dst;
+	struct vector_fds *result = NULL;
+	struct sockaddr_un *local_addr = NULL, *remote_addr = NULL;
+
+	src = uml_vector_fetch_arg(ifspec, "src");
+	dst = uml_vector_fetch_arg(ifspec, "dst");
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result == NULL) {
+		printk(UM_KERN_ERR "unix open:cannot allocate remote addr");
+		goto unix_cleanup;
+	}
+	remote_addr = uml_kmalloc(sizeof(struct sockaddr_un), UM_GFP_KERNEL);
+	if (remote_addr == NULL) {
+		printk(UM_KERN_ERR "unix open:cannot allocate remote addr");
+		goto unix_cleanup;
+	}
+
+	switch (id) {
+	case ID_BESS:
+		socktype = SOCK_SEQPACKET;
+		if ((src != NULL) && (strlen(src) <= MAX_UN_LEN)) {
+			local_addr = uml_kmalloc(sizeof(struct sockaddr_un), UM_GFP_KERNEL);
+			if (local_addr == NULL) {
+				printk(UM_KERN_ERR "bess open:cannot allocate local addr");
+				goto unix_cleanup;
+			}
+			local_addr->sun_family = AF_UNIX;
+			memcpy(local_addr->sun_path, src, strlen(src) + 1);
+		}
+		if ((dst == NULL) || (strlen(dst) > MAX_UN_LEN))
+			goto unix_cleanup;
+		remote_addr->sun_family = AF_UNIX;
+		memcpy(remote_addr->sun_path, dst, strlen(dst) + 1);
+		break;
+	default:
+		printk(KERN_ERR "Unsupported unix socket type\n");
+		return NULL;
+	}
+
+	fd = socket(AF_UNIX, socktype, 0);
+	if (fd == -1) {
+		printk(UM_KERN_ERR
+			"unix open: could not open socket, error = %d",
+			-errno
+		);
+		goto unix_cleanup;
+	}
+	if (local_addr != NULL) {
+		if (bind(fd, (struct sockaddr *) local_addr, sizeof(struct sockaddr_un))) {
+			printk(UM_KERN_ERR UNIX_BIND_FAIL, errno);
+			goto unix_cleanup;
+		}
+	}
+	switch (id) {
+	case ID_BESS:
+		if (connect(fd, (const struct sockaddr *) remote_addr, sizeof(struct sockaddr_un)) < 0) {
+			printk(UM_KERN_ERR "bess open:cannot connect to %s %i", remote_addr->sun_path, -errno);
+			goto unix_cleanup;
+		}
+		break;
+	}
+	result->rx_fd = fd;
+	result->tx_fd = fd;
+	result->remote_addr_size = sizeof(struct sockaddr_un);
+	result->remote_addr = remote_addr;
+	return result;
+unix_cleanup:
+	if (fd >= 0)
+		os_close_file(fd);
+	kfree(remote_addr);
+	kfree(result);
+	return NULL;
+}
+
+static int strtofd(const char *nptr)
+{
+	long fd;
+	char *endptr;
+
+	if (nptr == NULL)
+		return -1;
+
+	errno = 0;
+	fd = strtol(nptr, &endptr, 10);
+	if (nptr == endptr ||
+		errno != 0 ||
+		*endptr != '\0' ||
+		fd < 0 ||
+		fd > INT_MAX) {
+		return -1;
+	}
+	return fd;
+}
+
+static struct vector_fds *user_init_fd_fds(struct arglist *ifspec)
+{
+	int fd = -1;
+	char *fdarg = NULL;
+	struct vector_fds *result = NULL;
+
+	fdarg = uml_vector_fetch_arg(ifspec, "fd");
+	fd = strtofd(fdarg);
+	if (fd == -1) {
+		printk(UM_KERN_ERR "fd open: bad or missing fd argument");
+		goto fd_cleanup;
+	}
+
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result == NULL) {
+		printk(UM_KERN_ERR "fd open: allocation failed");
+		goto fd_cleanup;
+	}
+
+	result->rx_fd = fd;
+	result->tx_fd = fd;
+	result->remote_addr_size = 0;
+	result->remote_addr = NULL;
+	return result;
+
+fd_cleanup:
+	if (fd >= 0)
+		os_close_file(fd);
+	kfree(result);
+	return NULL;
+}
+
+/* enough char to store an int type */
+#define ENOUGH(type) ((CHAR_BIT * sizeof(type) - 1) / 3 + 2)
+#define ENOUGH_OCTAL(type) ((CHAR_BIT * sizeof(type) + 2) / 3)
+/* vde_plug --descr xx --port2 xx --mod2 xx --group2 xx seqpacket://NN vnl (NULL) */
+#define VDE_MAX_ARGC 12
+#define VDE_SEQPACKET_HEAD "seqpacket://"
+#define VDE_SEQPACKET_HEAD_LEN (sizeof(VDE_SEQPACKET_HEAD) - 1)
+#define VDE_DEFAULT_DESCRIPTION "UML"
+
+static struct vector_fds *user_init_vde_fds(struct arglist *ifspec)
+{
+	char seqpacketvnl[VDE_SEQPACKET_HEAD_LEN + ENOUGH(int) + 1];
+	char *argv[VDE_MAX_ARGC] = {"vde_plug"};
+	int argc = 1;
+	int rv;
+	int sv[2];
+	struct vector_fds *result = NULL;
+
+	char *vnl = uml_vector_fetch_arg(ifspec,"vnl");
+	char *descr = uml_vector_fetch_arg(ifspec,"descr");
+	char *port = uml_vector_fetch_arg(ifspec,"port");
+	char *mode = uml_vector_fetch_arg(ifspec,"mode");
+	char *group = uml_vector_fetch_arg(ifspec,"group");
+	if (descr == NULL) descr = VDE_DEFAULT_DESCRIPTION;
+
+	argv[argc++] = "--descr";
+	argv[argc++] = descr;
+	if (port != NULL) {
+		argv[argc++] = "--port2";
+		argv[argc++] = port;
+	}
+	if (mode != NULL) {
+		argv[argc++] = "--mod2";
+		argv[argc++] = mode;
+	}
+	if (group != NULL) {
+		argv[argc++] = "--group2";
+		argv[argc++] = group;
+	}
+	argv[argc++] = seqpacketvnl;
+	argv[argc++] = vnl;
+	argv[argc++] = NULL;
+
+	rv = socketpair(AF_UNIX, SOCK_SEQPACKET, 0, sv);
+	if (rv  < 0) {
+		printk(UM_KERN_ERR "vde: seqpacket socketpair err %d", -errno);
+		return NULL;
+	}
+	rv = os_set_exec_close(sv[0]);
+	if (rv  < 0) {
+		printk(UM_KERN_ERR "vde: seqpacket socketpair cloexec err %d", -errno);
+		goto vde_cleanup_sv;
+	}
+	snprintf(seqpacketvnl, sizeof(seqpacketvnl), VDE_SEQPACKET_HEAD "%d", sv[1]);
+
+	run_helper(NULL, NULL, argv);
+
+	close(sv[1]);
+
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result == NULL) {
+		printk(UM_KERN_ERR "fd open: allocation failed");
+		goto vde_cleanup;
+	}
+
+	result->rx_fd = sv[0];
+	result->tx_fd = sv[0];
+	result->remote_addr_size = 0;
+	result->remote_addr = NULL;
+	return result;
+
+vde_cleanup_sv:
+	close(sv[1]);
+vde_cleanup:
+	close(sv[0]);
+	return NULL;
+}
+
+static struct vector_fds *user_init_raw_fds(struct arglist *ifspec)
+{
+	int rxfd = -1, txfd = -1;
+	int err = -ENOMEM;
+	char *iface;
+	struct vector_fds *result = NULL;
+	char *argv[] = {NULL, NULL, NULL, NULL};
+
+	iface = uml_vector_fetch_arg(ifspec, TOKEN_IFNAME);
+	if (iface == NULL)
+		goto raw_cleanup;
+
+	rxfd = create_raw_fd(iface, ETH_P_ALL, ETH_P_ALL);
+	if (rxfd == -1) {
+		err = -errno;
+		goto raw_cleanup;
+	}
+	txfd = create_raw_fd(iface, 0, ETH_P_IP); /* Turn off RX on this fd */
+	if (txfd == -1) {
+		err = -errno;
+		goto raw_cleanup;
+	}
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result != NULL) {
+		result->rx_fd = rxfd;
+		result->tx_fd = txfd;
+		result->remote_addr = NULL;
+		result->remote_addr_size = 0;
+	}
+	argv[0] = uml_vector_fetch_arg(ifspec, TOKEN_SCRIPT);
+	if (argv[0]) {
+		argv[1] = iface;
+		run_helper(NULL, NULL, argv);
+	}
+	return result;
+raw_cleanup:
+	printk(UM_KERN_ERR "user_init_raw: init failed, error %d", err);
+	kfree(result);
+	return NULL;
+}
+
+
+bool uml_raw_enable_qdisc_bypass(int fd)
+{
+	int optval = 1;
+
+	if (setsockopt(fd,
+		SOL_PACKET, PACKET_QDISC_BYPASS,
+		&optval, sizeof(optval)) != 0) {
+		return false;
+	}
+	return true;
+}
+
+bool uml_raw_enable_vnet_headers(int fd)
+{
+	int optval = 1;
+
+	if (setsockopt(fd,
+		SOL_PACKET, PACKET_VNET_HDR,
+		&optval, sizeof(optval)) != 0) {
+		printk(UM_KERN_INFO VNET_HDR_FAIL, fd);
+		return false;
+	}
+	return true;
+}
+bool uml_tap_enable_vnet_headers(int fd)
+{
+	unsigned int features;
+	int len = sizeof(struct virtio_net_hdr);
+
+	if (ioctl(fd, TUNGETFEATURES, &features) == -1) {
+		printk(UM_KERN_INFO TUN_GET_F_FAIL, strerror(errno));
+		return false;
+	}
+	if ((features & IFF_VNET_HDR) == 0) {
+		printk(UM_KERN_INFO "tapraw: No VNET HEADER support");
+		return false;
+	}
+	ioctl(fd, TUNSETVNETHDRSZ, &len);
+	return true;
+}
+
+static struct vector_fds *user_init_socket_fds(struct arglist *ifspec, int id)
+{
+	int err = -ENOMEM;
+	int fd = -1, gairet;
+	struct addrinfo srchints;
+	struct addrinfo dsthints;
+	bool v6, udp;
+	char *value;
+	char *src, *dst, *srcport, *dstport;
+	struct addrinfo *gairesult = NULL;
+	struct vector_fds *result = NULL;
+
+
+	value = uml_vector_fetch_arg(ifspec, "v6");
+	v6 = false;
+	udp = false;
+	if (value != NULL) {
+		if (strtol((const char *) value, NULL, 10) > 0)
+			v6 = true;
+	}
+
+	value = uml_vector_fetch_arg(ifspec, "udp");
+	if (value != NULL) {
+		if (strtol((const char *) value, NULL, 10) > 0)
+			udp = true;
+	}
+	src = uml_vector_fetch_arg(ifspec, "src");
+	dst = uml_vector_fetch_arg(ifspec, "dst");
+	srcport = uml_vector_fetch_arg(ifspec, "srcport");
+	dstport = uml_vector_fetch_arg(ifspec, "dstport");
+
+	memset(&dsthints, 0, sizeof(dsthints));
+
+	if (v6)
+		dsthints.ai_family = AF_INET6;
+	else
+		dsthints.ai_family = AF_INET;
+
+	switch (id) {
+	case ID_GRE:
+		dsthints.ai_socktype = SOCK_RAW;
+		dsthints.ai_protocol = IPPROTO_GRE;
+		break;
+	case ID_L2TPV3:
+		if (udp) {
+			dsthints.ai_socktype = SOCK_DGRAM;
+			dsthints.ai_protocol = 0;
+		} else {
+			dsthints.ai_socktype = SOCK_RAW;
+			dsthints.ai_protocol = IPPROTO_L2TP;
+		}
+		break;
+	default:
+		printk(KERN_ERR "Unsupported socket type\n");
+		return NULL;
+	}
+	memcpy(&srchints, &dsthints, sizeof(struct addrinfo));
+
+	gairet = getaddrinfo(src, srcport, &dsthints, &gairesult);
+	if ((gairet != 0) || (gairesult == NULL)) {
+		printk(UM_KERN_ERR
+			"socket_open : could not resolve src, error = %s",
+			gai_strerror(gairet)
+		);
+		return NULL;
+	}
+	fd = socket(gairesult->ai_family,
+		gairesult->ai_socktype, gairesult->ai_protocol);
+	if (fd == -1) {
+		printk(UM_KERN_ERR
+			"socket_open : could not open socket, error = %d",
+			-errno
+		);
+		goto cleanup;
+	}
+	if (bind(fd,
+		(struct sockaddr *) gairesult->ai_addr,
+		gairesult->ai_addrlen)) {
+		printk(UM_KERN_ERR L2TPV3_BIND_FAIL, errno);
+		goto cleanup;
+	}
+
+	if (gairesult != NULL)
+		freeaddrinfo(gairesult);
+
+	gairesult = NULL;
+
+	gairet = getaddrinfo(dst, dstport, &dsthints, &gairesult);
+	if ((gairet != 0) || (gairesult == NULL)) {
+		printk(UM_KERN_ERR
+			"socket_open : could not resolve dst, error = %s",
+			gai_strerror(gairet)
+		);
+		return NULL;
+	}
+
+	result = uml_kmalloc(sizeof(struct vector_fds), UM_GFP_KERNEL);
+	if (result != NULL) {
+		result->rx_fd = fd;
+		result->tx_fd = fd;
+		result->remote_addr = uml_kmalloc(
+			gairesult->ai_addrlen, UM_GFP_KERNEL);
+		if (result->remote_addr == NULL)
+			goto cleanup;
+		result->remote_addr_size = gairesult->ai_addrlen;
+		memcpy(
+			result->remote_addr,
+			gairesult->ai_addr,
+			gairesult->ai_addrlen
+		);
+	}
+	freeaddrinfo(gairesult);
+	return result;
+cleanup:
+	if (gairesult != NULL)
+		freeaddrinfo(gairesult);
+	printk(UM_KERN_ERR "user_init_socket: init failed, error %d", err);
+	if (fd >= 0)
+		os_close_file(fd);
+	if (result != NULL) {
+		kfree(result->remote_addr);
+		kfree(result);
+	}
+	return NULL;
+}
+
+struct vector_fds *uml_vector_user_open(
+	int unit,
+	struct arglist *parsed
+)
+{
+	char *transport;
+
+	if (parsed == NULL) {
+		printk(UM_KERN_ERR "no parsed config for unit %d\n", unit);
+		return NULL;
+	}
+	transport = uml_vector_fetch_arg(parsed, "transport");
+	if (transport == NULL) {
+		printk(UM_KERN_ERR "missing transport for unit %d\n", unit);
+		return NULL;
+	}
+	if (strncmp(transport, TRANS_RAW, TRANS_RAW_LEN) == 0)
+		return user_init_raw_fds(parsed);
+	if (strncmp(transport, TRANS_HYBRID, TRANS_HYBRID_LEN) == 0)
+		return user_init_hybrid_fds(parsed);
+	if (strncmp(transport, TRANS_TAP, TRANS_TAP_LEN) == 0)
+		return user_init_tap_fds(parsed);
+	if (strncmp(transport, TRANS_GRE, TRANS_GRE_LEN) == 0)
+		return user_init_socket_fds(parsed, ID_GRE);
+	if (strncmp(transport, TRANS_L2TPV3, TRANS_L2TPV3_LEN) == 0)
+		return user_init_socket_fds(parsed, ID_L2TPV3);
+	if (strncmp(transport, TRANS_BESS, TRANS_BESS_LEN) == 0)
+		return user_init_unix_fds(parsed, ID_BESS);
+	if (strncmp(transport, TRANS_FD, TRANS_FD_LEN) == 0)
+		return user_init_fd_fds(parsed);
+	if (strncmp(transport, TRANS_VDE, TRANS_VDE_LEN) == 0)
+		return user_init_vde_fds(parsed);
+	return NULL;
+}
+
+
+int uml_vector_sendmsg(int fd, void *hdr, int flags)
+{
+	int n;
+
+	CATCH_EINTR(n = sendmsg(fd, (struct msghdr *) hdr,  flags));
+	if ((n < 0) && (errno == EAGAIN))
+		return 0;
+	if (n >= 0)
+		return n;
+	else
+		return -errno;
+}
+
+int uml_vector_recvmsg(int fd, void *hdr, int flags)
+{
+	int n;
+	struct msghdr *msg = (struct msghdr *) hdr;
+
+	CATCH_EINTR(n = readv(fd, msg->msg_iov, msg->msg_iovlen));
+	if ((n < 0) && (errno == EAGAIN))
+		return 0;
+	if (n >= 0)
+		return n;
+	else
+		return -errno;
+}
+
+int uml_vector_writev(int fd, void *hdr, int iovcount)
+{
+	int n;
+
+	CATCH_EINTR(n = writev(fd, (struct iovec *) hdr,  iovcount));
+	if ((n < 0) && ((errno == EAGAIN) || (errno == ENOBUFS)))
+		return 0;
+	if (n >= 0)
+		return n;
+	else
+		return -errno;
+}
+
+int uml_vector_sendmmsg(
+	int fd,
+	void *msgvec,
+	unsigned int vlen,
+	unsigned int flags)
+{
+	int n;
+
+	CATCH_EINTR(n = sendmmsg(fd, (struct mmsghdr *) msgvec, vlen, flags));
+	if ((n < 0) && ((errno == EAGAIN) || (errno == ENOBUFS)))
+		return 0;
+	if (n >= 0)
+		return n;
+	else
+		return -errno;
+}
+
+int uml_vector_recvmmsg(
+	int fd,
+	void *msgvec,
+	unsigned int vlen,
+	unsigned int flags)
+{
+	int n;
+
+	CATCH_EINTR(
+		n = recvmmsg(fd, (struct mmsghdr *) msgvec, vlen, flags, 0));
+	if ((n < 0) && (errno == EAGAIN))
+		return 0;
+	if (n >= 0)
+		return n;
+	else
+		return -errno;
+}
+int uml_vector_attach_bpf(int fd, void *bpf)
+{
+	struct sock_fprog *prog = bpf;
+
+	int err = setsockopt(fd, SOL_SOCKET, SO_ATTACH_FILTER, bpf, sizeof(struct sock_fprog));
+
+	if (err < 0)
+		printk(KERN_ERR BPF_ATTACH_FAIL, prog->len, prog->filter, fd, -errno);
+	return err;
+}
+
+int uml_vector_detach_bpf(int fd, void *bpf)
+{
+	struct sock_fprog *prog = bpf;
+
+	int err = setsockopt(fd, SOL_SOCKET, SO_DETACH_FILTER, bpf, sizeof(struct sock_fprog));
+	if (err < 0)
+		printk(KERN_ERR BPF_DETACH_FAIL, prog->len, prog->filter, fd, -errno);
+	return err;
+}
+void *uml_vector_default_bpf(const void *mac)
+{
+	struct sock_filter *bpf;
+	uint32_t *mac1 = (uint32_t *)(mac + 2);
+	uint16_t *mac2 = (uint16_t *) mac;
+	struct sock_fprog *bpf_prog;
+
+	bpf_prog = uml_kmalloc(sizeof(struct sock_fprog), UM_GFP_KERNEL);
+	if (bpf_prog) {
+		bpf_prog->len = DEFAULT_BPF_LEN;
+		bpf_prog->filter = NULL;
+	} else {
+		return NULL;
+	}
+	bpf = uml_kmalloc(
+		sizeof(struct sock_filter) * DEFAULT_BPF_LEN, UM_GFP_KERNEL);
+	if (bpf) {
+		bpf_prog->filter = bpf;
+		/* ld	[8] */
+		bpf[0] = (struct sock_filter){ 0x20, 0, 0, 0x00000008 };
+		/* jeq	#0xMAC[2-6] jt 2 jf 5*/
+		bpf[1] = (struct sock_filter){ 0x15, 0, 3, ntohl(*mac1)};
+		/* ldh	[6] */
+		bpf[2] = (struct sock_filter){ 0x28, 0, 0, 0x00000006 };
+		/* jeq	#0xMAC[0-1] jt 4 jf 5 */
+		bpf[3] = (struct sock_filter){ 0x15, 0, 1, ntohs(*mac2)};
+		/* ret	#0 */
+		bpf[4] = (struct sock_filter){ 0x6, 0, 0, 0x00000000 };
+		/* ret	#0x40000 */
+		bpf[5] = (struct sock_filter){ 0x6, 0, 0, 0x00040000 };
+	} else {
+		kfree(bpf_prog);
+		bpf_prog = NULL;
+	}
+	return bpf_prog;
+}
+
+/* Note - this function requires a valid mac being passed as an arg */
+
+void *uml_vector_user_bpf(char *filename)
+{
+	struct sock_filter *bpf;
+	struct sock_fprog *bpf_prog;
+	struct stat statbuf;
+	int res, ffd = -1;
+
+	if (filename == NULL)
+		return NULL;
+
+	if (stat(filename, &statbuf) < 0) {
+		printk(KERN_ERR "Error %d reading bpf file", -errno);
+		return false;
+	}
+	bpf_prog = uml_kmalloc(sizeof(struct sock_fprog), UM_GFP_KERNEL);
+	if (bpf_prog == NULL) {
+		printk(KERN_ERR "Failed to allocate bpf prog buffer");
+		return NULL;
+	}
+	bpf_prog->len = statbuf.st_size / sizeof(struct sock_filter);
+	bpf_prog->filter = NULL;
+	ffd = os_open_file(filename, of_read(OPENFLAGS()), 0);
+	if (ffd < 0) {
+		printk(KERN_ERR "Error %d opening bpf file", -errno);
+		goto bpf_failed;
+	}
+	bpf = uml_kmalloc(statbuf.st_size, UM_GFP_KERNEL);
+	if (bpf == NULL) {
+		printk(KERN_ERR "Failed to allocate bpf buffer");
+		goto bpf_failed;
+	}
+	bpf_prog->filter = bpf;
+	res = os_read_file(ffd, bpf, statbuf.st_size);
+	if (res < statbuf.st_size) {
+		printk(KERN_ERR "Failed to read bpf program %s, error %d", filename, res);
+		kfree(bpf);
+		goto bpf_failed;
+	}
+	os_close_file(ffd);
+	return bpf_prog;
+bpf_failed:
+	if (ffd > 0)
+		os_close_file(ffd);
+	kfree(bpf_prog);
+	return NULL;
+}

linux-7.2.3/arch/um/drivers/vector_user.h

@@ -0,0 +1,107 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2002 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#ifndef __UM_VECTOR_USER_H
+#define __UM_VECTOR_USER_H
+
+#define MAXVARGS	20
+
+#define TOKEN_IFNAME "ifname"
+
+#define TRANS_RAW "raw"
+#define TRANS_RAW_LEN strlen(TRANS_RAW)
+
+#define TRANS_TAP "tap"
+#define TRANS_TAP_LEN strlen(TRANS_TAP)
+
+#define TRANS_GRE "gre"
+#define TRANS_GRE_LEN strlen(TRANS_GRE)
+
+#define TRANS_L2TPV3 "l2tpv3"
+#define TRANS_L2TPV3_LEN strlen(TRANS_L2TPV3)
+
+#define TRANS_HYBRID "hybrid"
+#define TRANS_HYBRID_LEN strlen(TRANS_HYBRID)
+
+#define TRANS_BESS "bess"
+#define TRANS_BESS_LEN strlen(TRANS_BESS)
+
+#define DEFAULT_BPF_LEN 6
+
+#ifndef IPPROTO_GRE
+#define IPPROTO_GRE 0x2F
+#endif
+
+#define GRE_MODE_CHECKSUM	cpu_to_be16(8 << 12)	/* checksum */
+#define GRE_MODE_RESERVED	cpu_to_be16(4 << 12)	/* unused */
+#define GRE_MODE_KEY		cpu_to_be16(2 << 12)	/* KEY present */
+#define GRE_MODE_SEQUENCE	cpu_to_be16(1 << 12)	/* sequence */
+
+#define GRE_IRB cpu_to_be16(0x6558)
+
+#define L2TPV3_DATA_PACKET 0x30000
+
+/* IANA-assigned IP protocol ID for L2TPv3 */
+
+#ifndef IPPROTO_L2TP
+#define IPPROTO_L2TP 0x73
+#endif
+
+struct arglist {
+	int	numargs;
+	char	*tokens[MAXVARGS];
+	char	*values[MAXVARGS];
+};
+
+/* Separating read and write FDs allows us to have different
+ * rx and tx method. Example - read tap via raw socket using
+ * recvmmsg, write using legacy tap write calls
+ */
+
+struct vector_fds {
+	int rx_fd;
+	int tx_fd;
+	void *remote_addr;
+	int remote_addr_size;
+};
+
+#define VECTOR_READ	1
+
+extern struct arglist *uml_parse_vector_ifspec(char *arg);
+
+extern struct vector_fds *uml_vector_user_open(
+	int unit,
+	struct arglist *parsed
+);
+
+extern char *uml_vector_fetch_arg(
+	struct arglist *ifspec,
+	char *token
+);
+
+extern int uml_vector_recvmsg(int fd, void *hdr, int flags);
+extern int uml_vector_sendmsg(int fd, void *hdr, int flags);
+extern int uml_vector_writev(int fd, void *hdr, int iovcount);
+extern int uml_vector_sendmmsg(
+	int fd, void *msgvec,
+	unsigned int vlen,
+	unsigned int flags
+);
+extern int uml_vector_recvmmsg(
+	int fd,
+	void *msgvec,
+	unsigned int vlen,
+	unsigned int flags
+);
+extern void *uml_vector_default_bpf(const void *mac);
+extern void *uml_vector_user_bpf(char *filename);
+extern int uml_vector_attach_bpf(int fd, void *bpf);
+extern int uml_vector_detach_bpf(int fd, void *bpf);
+extern bool uml_raw_enable_qdisc_bypass(int fd);
+extern bool uml_raw_enable_vnet_headers(int fd);
+extern bool uml_tap_enable_vnet_headers(int fd);
+
+
+#endif

linux-7.2.3/arch/um/drivers/vfio_kern.c

@@ -0,0 +1,707 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025 Ant Group
+ * Author: Tiwei Bie <tiwei.btw@antgroup.com>
+ */
+
+#define pr_fmt(fmt) "vfio-uml: " fmt
+
+#include <linux/module.h>
+#include <linux/logic_iomem.h>
+#include <linux/mutex.h>
+#include <linux/list.h>
+#include <linux/string.h>
+#include <linux/unaligned.h>
+#include <irq_kern.h>
+#include <init.h>
+#include <os.h>
+
+#include "mconsole_kern.h"
+#include "virt-pci.h"
+#include "vfio_user.h"
+
+#define to_vdev(_pdev) container_of(_pdev, struct uml_vfio_device, pdev)
+
+struct uml_vfio_intr_ctx {
+	struct uml_vfio_device *dev;
+	int irq;
+};
+
+struct uml_vfio_device {
+	const char *name;
+	int group;
+
+	struct um_pci_device pdev;
+	struct uml_vfio_user_device udev;
+	struct uml_vfio_intr_ctx *intr_ctx;
+
+	int msix_cap;
+	int msix_bar;
+	int msix_offset;
+	int msix_size;
+	u32 *msix_data;
+
+	struct list_head list;
+};
+
+struct uml_vfio_group {
+	int id;
+	int fd;
+	int users;
+	struct list_head list;
+};
+
+static struct {
+	int fd;
+	int users;
+} uml_vfio_container = { .fd = -1 };
+static DEFINE_MUTEX(uml_vfio_container_mtx);
+
+static LIST_HEAD(uml_vfio_groups);
+static DEFINE_MUTEX(uml_vfio_groups_mtx);
+
+static LIST_HEAD(uml_vfio_devices);
+static DEFINE_MUTEX(uml_vfio_devices_mtx);
+
+static int uml_vfio_set_container(int group_fd)
+{
+	int err;
+
+	guard(mutex)(&uml_vfio_container_mtx);
+
+	err = uml_vfio_user_set_container(uml_vfio_container.fd, group_fd);
+	if (err)
+		return err;
+
+	uml_vfio_container.users++;
+	if (uml_vfio_container.users > 1)
+		return 0;
+
+	err = uml_vfio_user_setup_iommu(uml_vfio_container.fd);
+	if (err) {
+		uml_vfio_user_unset_container(uml_vfio_container.fd, group_fd);
+		uml_vfio_container.users--;
+	}
+	return err;
+}
+
+static void uml_vfio_unset_container(int group_fd)
+{
+	guard(mutex)(&uml_vfio_container_mtx);
+
+	uml_vfio_user_unset_container(uml_vfio_container.fd, group_fd);
+	uml_vfio_container.users--;
+}
+
+static int uml_vfio_open_group(int group_id)
+{
+	struct uml_vfio_group *group;
+	int err;
+
+	guard(mutex)(&uml_vfio_groups_mtx);
+
+	list_for_each_entry(group, &uml_vfio_groups, list) {
+		if (group->id == group_id) {
+			group->users++;
+			return group->fd;
+		}
+	}
+
+	group = kzalloc_obj(*group);
+	if (!group)
+		return -ENOMEM;
+
+	group->fd = uml_vfio_user_open_group(group_id);
+	if (group->fd < 0) {
+		err = group->fd;
+		goto free_group;
+	}
+
+	err = uml_vfio_set_container(group->fd);
+	if (err)
+		goto close_group;
+
+	group->id = group_id;
+	group->users = 1;
+
+	list_add(&group->list, &uml_vfio_groups);
+
+	return group->fd;
+
+close_group:
+	os_close_file(group->fd);
+free_group:
+	kfree(group);
+	return err;
+}
+
+static int uml_vfio_release_group(int group_fd)
+{
+	struct uml_vfio_group *group;
+
+	guard(mutex)(&uml_vfio_groups_mtx);
+
+	list_for_each_entry(group, &uml_vfio_groups, list) {
+		if (group->fd == group_fd) {
+			group->users--;
+			if (group->users == 0) {
+				uml_vfio_unset_container(group_fd);
+				os_close_file(group_fd);
+				list_del(&group->list);
+				kfree(group);
+			}
+			return 0;
+		}
+	}
+
+	return -ENOENT;
+}
+
+static irqreturn_t uml_vfio_interrupt(int unused, void *opaque)
+{
+	struct uml_vfio_intr_ctx *ctx = opaque;
+	struct uml_vfio_device *dev = ctx->dev;
+	int index = ctx - dev->intr_ctx;
+	int irqfd = dev->udev.irqfd[index];
+	int irq = dev->msix_data[index];
+	uint64_t v;
+	int r;
+
+	do {
+		r = os_read_file(irqfd, &v, sizeof(v));
+		if (r == sizeof(v))
+			generic_handle_irq(irq);
+	} while (r == sizeof(v) || r == -EINTR);
+	WARN(r != -EAGAIN, "read returned %d\n", r);
+
+	return IRQ_HANDLED;
+}
+
+static int uml_vfio_activate_irq(struct uml_vfio_device *dev, int index)
+{
+	struct uml_vfio_intr_ctx *ctx = &dev->intr_ctx[index];
+	int err, irqfd;
+
+	if (ctx->irq >= 0)
+		return 0;
+
+	irqfd = uml_vfio_user_activate_irq(&dev->udev, index);
+	if (irqfd < 0)
+		return irqfd;
+
+	ctx->irq = um_request_irq(UM_IRQ_ALLOC, irqfd, IRQ_READ,
+				  uml_vfio_interrupt, 0,
+				  "vfio-uml", ctx);
+	if (ctx->irq < 0) {
+		err = ctx->irq;
+		goto deactivate;
+	}
+
+	err = add_sigio_fd(irqfd);
+	if (err)
+		goto free_irq;
+
+	return 0;
+
+free_irq:
+	um_free_irq(ctx->irq, ctx);
+	ctx->irq = -1;
+deactivate:
+	uml_vfio_user_deactivate_irq(&dev->udev, index);
+	return err;
+}
+
+static int uml_vfio_deactivate_irq(struct uml_vfio_device *dev, int index)
+{
+	struct uml_vfio_intr_ctx *ctx = &dev->intr_ctx[index];
+
+	if (ctx->irq >= 0) {
+		ignore_sigio_fd(dev->udev.irqfd[index]);
+		um_free_irq(ctx->irq, ctx);
+		uml_vfio_user_deactivate_irq(&dev->udev, index);
+		ctx->irq = -1;
+	}
+	return 0;
+}
+
+static int uml_vfio_update_msix_cap(struct uml_vfio_device *dev,
+				    unsigned int offset, int size,
+				    unsigned long val)
+{
+	/*
+	 * Here, we handle only the operations we care about,
+	 * ignoring the rest.
+	 */
+	if (size == 2 && offset == dev->msix_cap + PCI_MSIX_FLAGS) {
+		switch (val & ~PCI_MSIX_FLAGS_QSIZE) {
+		case PCI_MSIX_FLAGS_ENABLE:
+		case 0:
+			return uml_vfio_user_update_irqs(&dev->udev);
+		}
+	}
+	return 0;
+}
+
+static int uml_vfio_update_msix_table(struct uml_vfio_device *dev,
+				      unsigned int offset, int size,
+				      unsigned long val)
+{
+	int index;
+
+	/*
+	 * Here, we handle only the operations we care about,
+	 * ignoring the rest.
+	 */
+	offset -= dev->msix_offset + PCI_MSIX_ENTRY_DATA;
+
+	if (size != 4 || offset % PCI_MSIX_ENTRY_SIZE != 0)
+		return 0;
+
+	index = offset / PCI_MSIX_ENTRY_SIZE;
+	if (index >= dev->udev.irq_count)
+		return -EINVAL;
+
+	dev->msix_data[index] = val;
+
+	return val ? uml_vfio_activate_irq(dev, index) :
+		uml_vfio_deactivate_irq(dev, index);
+}
+
+static unsigned long __uml_vfio_cfgspace_read(struct uml_vfio_device *dev,
+					      unsigned int offset, int size)
+{
+	u8 data[8];
+
+	memset(data, 0xff, sizeof(data));
+
+	if (uml_vfio_user_cfgspace_read(&dev->udev, offset, data, size))
+		return ULONG_MAX;
+
+	switch (size) {
+	case 1:
+		return data[0];
+	case 2:
+		return le16_to_cpup((void *)data);
+	case 4:
+		return le32_to_cpup((void *)data);
+#ifdef CONFIG_64BIT
+	case 8:
+		return le64_to_cpup((void *)data);
+#endif
+	default:
+		return ULONG_MAX;
+	}
+}
+
+static unsigned long uml_vfio_cfgspace_read(struct um_pci_device *pdev,
+					    unsigned int offset, int size)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+
+	return __uml_vfio_cfgspace_read(dev, offset, size);
+}
+
+static void __uml_vfio_cfgspace_write(struct uml_vfio_device *dev,
+				      unsigned int offset, int size,
+				      unsigned long val)
+{
+	u8 data[8];
+
+	switch (size) {
+	case 1:
+		data[0] = (u8)val;
+		break;
+	case 2:
+		put_unaligned_le16(val, (void *)data);
+		break;
+	case 4:
+		put_unaligned_le32(val, (void *)data);
+		break;
+#ifdef CONFIG_64BIT
+	case 8:
+		put_unaligned_le64(val, (void *)data);
+		break;
+#endif
+	}
+
+	WARN_ON(uml_vfio_user_cfgspace_write(&dev->udev, offset, data, size));
+}
+
+static void uml_vfio_cfgspace_write(struct um_pci_device *pdev,
+				    unsigned int offset, int size,
+				    unsigned long val)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+
+	if (offset < dev->msix_cap + PCI_CAP_MSIX_SIZEOF &&
+	    offset + size > dev->msix_cap)
+		WARN_ON(uml_vfio_update_msix_cap(dev, offset, size, val));
+
+	__uml_vfio_cfgspace_write(dev, offset, size, val);
+}
+
+static void uml_vfio_bar_copy_from(struct um_pci_device *pdev, int bar,
+				   void *buffer, unsigned int offset, int size)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+
+	memset(buffer, 0xff, size);
+	uml_vfio_user_bar_read(&dev->udev, bar, offset, buffer, size);
+}
+
+static unsigned long uml_vfio_bar_read(struct um_pci_device *pdev, int bar,
+				       unsigned int offset, int size)
+{
+	u8 data[8];
+
+	uml_vfio_bar_copy_from(pdev, bar, data, offset, size);
+
+	switch (size) {
+	case 1:
+		return data[0];
+	case 2:
+		return le16_to_cpup((void *)data);
+	case 4:
+		return le32_to_cpup((void *)data);
+#ifdef CONFIG_64BIT
+	case 8:
+		return le64_to_cpup((void *)data);
+#endif
+	default:
+		return ULONG_MAX;
+	}
+}
+
+static void uml_vfio_bar_copy_to(struct um_pci_device *pdev, int bar,
+				 unsigned int offset, const void *buffer,
+				 int size)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+
+	uml_vfio_user_bar_write(&dev->udev, bar, offset, buffer, size);
+}
+
+static void uml_vfio_bar_write(struct um_pci_device *pdev, int bar,
+			       unsigned int offset, int size,
+			       unsigned long val)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+	u8 data[8];
+
+	if (bar == dev->msix_bar && offset + size > dev->msix_offset &&
+	    offset < dev->msix_offset + dev->msix_size)
+		WARN_ON(uml_vfio_update_msix_table(dev, offset, size, val));
+
+	switch (size) {
+	case 1:
+		data[0] = (u8)val;
+		break;
+	case 2:
+		put_unaligned_le16(val, (void *)data);
+		break;
+	case 4:
+		put_unaligned_le32(val, (void *)data);
+		break;
+#ifdef CONFIG_64BIT
+	case 8:
+		put_unaligned_le64(val, (void *)data);
+		break;
+#endif
+	}
+
+	uml_vfio_bar_copy_to(pdev, bar, offset, data, size);
+}
+
+static void uml_vfio_bar_set(struct um_pci_device *pdev, int bar,
+			     unsigned int offset, u8 value, int size)
+{
+	struct uml_vfio_device *dev = to_vdev(pdev);
+	int i;
+
+	for (i = 0; i < size; i++)
+		uml_vfio_user_bar_write(&dev->udev, bar, offset + i, &value, 1);
+}
+
+static const struct um_pci_ops uml_vfio_um_pci_ops = {
+	.cfgspace_read	= uml_vfio_cfgspace_read,
+	.cfgspace_write	= uml_vfio_cfgspace_write,
+	.bar_read	= uml_vfio_bar_read,
+	.bar_write	= uml_vfio_bar_write,
+	.bar_copy_from	= uml_vfio_bar_copy_from,
+	.bar_copy_to	= uml_vfio_bar_copy_to,
+	.bar_set	= uml_vfio_bar_set,
+};
+
+static u8 uml_vfio_find_capability(struct uml_vfio_device *dev, u8 cap)
+{
+	u8 id, pos;
+	u16 ent;
+	int ttl = 48; /* PCI_FIND_CAP_TTL */
+
+	pos = __uml_vfio_cfgspace_read(dev, PCI_CAPABILITY_LIST, sizeof(pos));
+
+	while (pos && ttl--) {
+		ent = __uml_vfio_cfgspace_read(dev, pos, sizeof(ent));
+
+		id = ent & 0xff;
+		if (id == 0xff)
+			break;
+		if (id == cap)
+			return pos;
+
+		pos = ent >> 8;
+	}
+
+	return 0;
+}
+
+static int uml_vfio_read_msix_table(struct uml_vfio_device *dev)
+{
+	unsigned int off;
+	u16 flags;
+	u32 tbl;
+
+	off = uml_vfio_find_capability(dev, PCI_CAP_ID_MSIX);
+	if (!off)
+		return -ENOTSUPP;
+
+	dev->msix_cap = off;
+
+	tbl = __uml_vfio_cfgspace_read(dev, off + PCI_MSIX_TABLE, sizeof(tbl));
+	flags = __uml_vfio_cfgspace_read(dev, off + PCI_MSIX_FLAGS, sizeof(flags));
+
+	dev->msix_bar = tbl & PCI_MSIX_TABLE_BIR;
+	dev->msix_offset = tbl & PCI_MSIX_TABLE_OFFSET;
+	dev->msix_size = ((flags & PCI_MSIX_FLAGS_QSIZE) + 1) * PCI_MSIX_ENTRY_SIZE;
+
+	dev->msix_data = kzalloc(dev->msix_size, GFP_KERNEL);
+	if (!dev->msix_data)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void uml_vfio_open_device(struct uml_vfio_device *dev)
+{
+	struct uml_vfio_intr_ctx *ctx;
+	int err, group_id, i;
+
+	group_id = uml_vfio_user_get_group_id(dev->name);
+	if (group_id < 0) {
+		pr_err("Failed to get group id (%s), error %d\n",
+		       dev->name, group_id);
+		goto free_dev;
+	}
+
+	dev->group = uml_vfio_open_group(group_id);
+	if (dev->group < 0) {
+		pr_err("Failed to open group %d (%s), error %d\n",
+		       group_id, dev->name, dev->group);
+		goto free_dev;
+	}
+
+	err = uml_vfio_user_setup_device(&dev->udev, dev->group, dev->name);
+	if (err) {
+		pr_err("Failed to setup device (%s), error %d\n",
+		       dev->name, err);
+		goto release_group;
+	}
+
+	err = uml_vfio_read_msix_table(dev);
+	if (err) {
+		pr_err("Failed to read MSI-X table (%s), error %d\n",
+		       dev->name, err);
+		goto teardown_udev;
+	}
+
+	dev->intr_ctx = kmalloc_objs(struct uml_vfio_intr_ctx,
+				     dev->udev.irq_count);
+	if (!dev->intr_ctx) {
+		pr_err("Failed to allocate interrupt context (%s)\n",
+		       dev->name);
+		goto free_msix;
+	}
+
+	for (i = 0; i < dev->udev.irq_count; i++) {
+		ctx = &dev->intr_ctx[i];
+		ctx->dev = dev;
+		ctx->irq = -1;
+	}
+
+	dev->pdev.ops = &uml_vfio_um_pci_ops;
+
+	err = um_pci_device_register(&dev->pdev);
+	if (err) {
+		pr_err("Failed to register UM PCI device (%s), error %d\n",
+		       dev->name, err);
+		goto free_intr_ctx;
+	}
+
+	return;
+
+free_intr_ctx:
+	kfree(dev->intr_ctx);
+free_msix:
+	kfree(dev->msix_data);
+teardown_udev:
+	uml_vfio_user_teardown_device(&dev->udev);
+release_group:
+	uml_vfio_release_group(dev->group);
+free_dev:
+	list_del(&dev->list);
+	kfree(dev->name);
+	kfree(dev);
+}
+
+static void uml_vfio_release_device(struct uml_vfio_device *dev)
+{
+	int i;
+
+	for (i = 0; i < dev->udev.irq_count; i++)
+		uml_vfio_deactivate_irq(dev, i);
+	uml_vfio_user_update_irqs(&dev->udev);
+
+	um_pci_device_unregister(&dev->pdev);
+	kfree(dev->intr_ctx);
+	kfree(dev->msix_data);
+	uml_vfio_user_teardown_device(&dev->udev);
+	uml_vfio_release_group(dev->group);
+	list_del(&dev->list);
+	kfree(dev->name);
+	kfree(dev);
+}
+
+static struct uml_vfio_device *uml_vfio_find_device(const char *device)
+{
+	struct uml_vfio_device *dev;
+
+	list_for_each_entry(dev, &uml_vfio_devices, list) {
+		if (!strcmp(dev->name, device))
+			return dev;
+	}
+	return NULL;
+}
+
+static struct uml_vfio_device *uml_vfio_add_device(const char *device)
+{
+	struct uml_vfio_device *dev;
+	int fd;
+
+	guard(mutex)(&uml_vfio_devices_mtx);
+
+	if (uml_vfio_container.fd < 0) {
+		fd = uml_vfio_user_open_container();
+		if (fd < 0)
+			return ERR_PTR(fd);
+		uml_vfio_container.fd = fd;
+	}
+
+	if (uml_vfio_find_device(device))
+		return ERR_PTR(-EEXIST);
+
+	dev = kzalloc_obj(*dev);
+	if (!dev)
+		return ERR_PTR(-ENOMEM);
+
+	dev->name = kstrdup(device, GFP_KERNEL);
+	if (!dev->name) {
+		kfree(dev);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	list_add_tail(&dev->list, &uml_vfio_devices);
+	return dev;
+}
+
+static int uml_vfio_cmdline_set(const char *device, const struct kernel_param *kp)
+{
+	struct uml_vfio_device *dev;
+
+	dev = uml_vfio_add_device(device);
+	if (IS_ERR(dev))
+		return PTR_ERR(dev);
+	return 0;
+}
+
+static int uml_vfio_cmdline_get(char *buffer, const struct kernel_param *kp)
+{
+	return 0;
+}
+
+static const struct kernel_param_ops uml_vfio_cmdline_param_ops = {
+	.set = uml_vfio_cmdline_set,
+	.get = uml_vfio_cmdline_get,
+};
+
+device_param_cb(device, &uml_vfio_cmdline_param_ops, NULL, 0400);
+__uml_help(uml_vfio_cmdline_param_ops,
+"vfio_uml.device=<domain:bus:slot.function>\n"
+"    Pass through a PCI device to UML via VFIO. Currently, only MSI-X\n"
+"    capable devices are supported, and it is assumed that drivers will\n"
+"    use MSI-X. This parameter can be specified multiple times to pass\n"
+"    through multiple PCI devices to UML.\n\n"
+);
+
+static int uml_vfio_mc_config(char *str, char **error_out)
+{
+	struct uml_vfio_device *dev;
+
+	if (*str != '=') {
+		*error_out = "Invalid config";
+		return -EINVAL;
+	}
+	str += 1;
+
+	dev = uml_vfio_add_device(str);
+	if (IS_ERR(dev))
+		return PTR_ERR(dev);
+	uml_vfio_open_device(dev);
+	return 0;
+}
+
+static int uml_vfio_mc_id(char **str, int *start_out, int *end_out)
+{
+	return -EOPNOTSUPP;
+}
+
+static int uml_vfio_mc_remove(int n, char **error_out)
+{
+	return -EOPNOTSUPP;
+}
+
+static struct mc_device uml_vfio_mc = {
+	.list           = LIST_HEAD_INIT(uml_vfio_mc.list),
+	.name           = "vfio_uml.device",
+	.config         = uml_vfio_mc_config,
+	.get_config     = NULL,
+	.id             = uml_vfio_mc_id,
+	.remove         = uml_vfio_mc_remove,
+};
+
+static int __init uml_vfio_init(void)
+{
+	struct uml_vfio_device *dev, *n;
+
+	sigio_broken();
+
+	/* If the opening fails, the device will be released. */
+	list_for_each_entry_safe(dev, n, &uml_vfio_devices, list)
+		uml_vfio_open_device(dev);
+
+	mconsole_register_dev(&uml_vfio_mc);
+
+	return 0;
+}
+late_initcall(uml_vfio_init);
+
+static void __exit uml_vfio_exit(void)
+{
+	struct uml_vfio_device *dev, *n;
+
+	list_for_each_entry_safe(dev, n, &uml_vfio_devices, list)
+		uml_vfio_release_device(dev);
+
+	if (uml_vfio_container.fd >= 0)
+		os_close_file(uml_vfio_container.fd);
+}
+module_exit(uml_vfio_exit);

linux-7.2.3/arch/um/drivers/vfio_user.c

@@ -0,0 +1,327 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025 Ant Group
+ * Author: Tiwei Bie <tiwei.btw@antgroup.com>
+ */
+#include <errno.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/eventfd.h>
+#include <linux/limits.h>
+#include <linux/vfio.h>
+#include <linux/pci_regs.h>
+#include <as-layout.h>
+#include <um_malloc.h>
+
+#include "vfio_user.h"
+
+int uml_vfio_user_open_container(void)
+{
+	int r, fd;
+
+	fd = open("/dev/vfio/vfio", O_RDWR);
+	if (fd < 0)
+		return -errno;
+
+	r = ioctl(fd, VFIO_GET_API_VERSION);
+	if (r != VFIO_API_VERSION) {
+		r = r < 0 ? -errno : -EINVAL;
+		goto error;
+	}
+
+	r = ioctl(fd, VFIO_CHECK_EXTENSION, VFIO_TYPE1_IOMMU);
+	if (r <= 0) {
+		r = r < 0 ? -errno : -EINVAL;
+		goto error;
+	}
+
+	return fd;
+
+error:
+	close(fd);
+	return r;
+}
+
+int uml_vfio_user_setup_iommu(int container)
+{
+	/*
+	 * This is a bit tricky. See the big comment in
+	 * vhost_user_set_mem_table() in virtio_uml.c.
+	 */
+	unsigned long reserved = uml_reserved - uml_physmem;
+	struct vfio_iommu_type1_dma_map dma_map = {
+		.argsz = sizeof(dma_map),
+		.flags = VFIO_DMA_MAP_FLAG_READ | VFIO_DMA_MAP_FLAG_WRITE,
+		.vaddr = uml_reserved,
+		.iova = reserved,
+		.size = physmem_size - reserved,
+	};
+
+	if (ioctl(container, VFIO_SET_IOMMU, VFIO_TYPE1_IOMMU) < 0)
+		return -errno;
+
+	if (ioctl(container, VFIO_IOMMU_MAP_DMA, &dma_map) < 0)
+		return -errno;
+
+	return 0;
+}
+
+int uml_vfio_user_get_group_id(const char *device)
+{
+	char *path, *buf, *end;
+	const char *name;
+	int r;
+
+	path = uml_kmalloc(PATH_MAX, UM_GFP_KERNEL);
+	if (!path)
+		return -ENOMEM;
+
+	sprintf(path, "/sys/bus/pci/devices/%s/iommu_group", device);
+
+	buf = uml_kmalloc(PATH_MAX + 1, UM_GFP_KERNEL);
+	if (!buf) {
+		r = -ENOMEM;
+		goto free_path;
+	}
+
+	r = readlink(path, buf, PATH_MAX);
+	if (r < 0) {
+		r = -errno;
+		goto free_buf;
+	}
+	buf[r] = '\0';
+
+	name = basename(buf);
+
+	r = strtoul(name, &end, 10);
+	if (*end != '\0' || end == name) {
+		r = -EINVAL;
+		goto free_buf;
+	}
+
+free_buf:
+	kfree(buf);
+free_path:
+	kfree(path);
+	return r;
+}
+
+int uml_vfio_user_open_group(int group_id)
+{
+	char *path;
+	int fd;
+
+	path = uml_kmalloc(PATH_MAX, UM_GFP_KERNEL);
+	if (!path)
+		return -ENOMEM;
+
+	sprintf(path, "/dev/vfio/%d", group_id);
+
+	fd = open(path, O_RDWR);
+	if (fd < 0) {
+		fd = -errno;
+		goto out;
+	}
+
+out:
+	kfree(path);
+	return fd;
+}
+
+int uml_vfio_user_set_container(int container, int group)
+{
+	if (ioctl(group, VFIO_GROUP_SET_CONTAINER, &container) < 0)
+		return -errno;
+	return 0;
+}
+
+int uml_vfio_user_unset_container(int container, int group)
+{
+	if (ioctl(group, VFIO_GROUP_UNSET_CONTAINER, &container) < 0)
+		return -errno;
+	return 0;
+}
+
+static int vfio_set_irqs(int device, int start, int count, int *irqfd)
+{
+	struct vfio_irq_set *irq_set;
+	int argsz = sizeof(*irq_set) + sizeof(*irqfd) * count;
+	int err = 0;
+
+	irq_set = uml_kmalloc(argsz, UM_GFP_KERNEL);
+	if (!irq_set)
+		return -ENOMEM;
+
+	irq_set->argsz = argsz;
+	irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX;
+	irq_set->start = start;
+	irq_set->count = count;
+	memcpy(irq_set->data, irqfd, sizeof(*irqfd) * count);
+
+	if (ioctl(device, VFIO_DEVICE_SET_IRQS, irq_set) < 0) {
+		err = -errno;
+		goto out;
+	}
+
+out:
+	kfree(irq_set);
+	return err;
+}
+
+int uml_vfio_user_setup_device(struct uml_vfio_user_device *dev,
+			       int group, const char *device)
+{
+	struct vfio_device_info device_info = { .argsz = sizeof(device_info) };
+	struct vfio_irq_info irq_info = { .argsz = sizeof(irq_info) };
+	int err, i;
+
+	dev->device = ioctl(group, VFIO_GROUP_GET_DEVICE_FD, device);
+	if (dev->device < 0)
+		return -errno;
+
+	if (ioctl(dev->device, VFIO_DEVICE_GET_INFO, &device_info) < 0) {
+		err = -errno;
+		goto close_device;
+	}
+
+	dev->num_regions = device_info.num_regions;
+	if (dev->num_regions > VFIO_PCI_CONFIG_REGION_INDEX + 1)
+		dev->num_regions = VFIO_PCI_CONFIG_REGION_INDEX + 1;
+
+	dev->region = uml_kmalloc(sizeof(*dev->region) * dev->num_regions,
+				  UM_GFP_KERNEL);
+	if (!dev->region) {
+		err = -ENOMEM;
+		goto close_device;
+	}
+
+	for (i = 0; i < dev->num_regions; i++) {
+		struct vfio_region_info region = {
+			.argsz = sizeof(region),
+			.index = i,
+		};
+		if (ioctl(dev->device, VFIO_DEVICE_GET_REGION_INFO, &region) < 0) {
+			err = -errno;
+			goto free_region;
+		}
+		dev->region[i].size = region.size;
+		dev->region[i].offset = region.offset;
+	}
+
+	/* Only MSI-X is supported currently. */
+	irq_info.index = VFIO_PCI_MSIX_IRQ_INDEX;
+	if (ioctl(dev->device, VFIO_DEVICE_GET_IRQ_INFO, &irq_info) < 0) {
+		err = -errno;
+		goto free_region;
+	}
+
+	dev->irq_count = irq_info.count;
+
+	dev->irqfd = uml_kmalloc(sizeof(int) * dev->irq_count, UM_GFP_KERNEL);
+	if (!dev->irqfd) {
+		err = -ENOMEM;
+		goto free_region;
+	}
+
+	memset(dev->irqfd, -1, sizeof(int) * dev->irq_count);
+
+	err = vfio_set_irqs(dev->device, 0, dev->irq_count, dev->irqfd);
+	if (err)
+		goto free_irqfd;
+
+	return 0;
+
+free_irqfd:
+	kfree(dev->irqfd);
+free_region:
+	kfree(dev->region);
+close_device:
+	close(dev->device);
+	return err;
+}
+
+void uml_vfio_user_teardown_device(struct uml_vfio_user_device *dev)
+{
+	kfree(dev->irqfd);
+	kfree(dev->region);
+	close(dev->device);
+}
+
+int uml_vfio_user_activate_irq(struct uml_vfio_user_device *dev, int index)
+{
+	int irqfd;
+
+	irqfd = eventfd(0, EFD_NONBLOCK | EFD_CLOEXEC);
+	if (irqfd < 0)
+		return -errno;
+
+	dev->irqfd[index] = irqfd;
+	return irqfd;
+}
+
+void uml_vfio_user_deactivate_irq(struct uml_vfio_user_device *dev, int index)
+{
+	close(dev->irqfd[index]);
+	dev->irqfd[index] = -1;
+}
+
+int uml_vfio_user_update_irqs(struct uml_vfio_user_device *dev)
+{
+	return vfio_set_irqs(dev->device, 0, dev->irq_count, dev->irqfd);
+}
+
+static int vfio_region_read(struct uml_vfio_user_device *dev, unsigned int index,
+			    uint64_t offset, void *buf, uint64_t size)
+{
+	if (index >= dev->num_regions || offset + size > dev->region[index].size)
+		return -EINVAL;
+
+	if (pread(dev->device, buf, size, dev->region[index].offset + offset) < 0)
+		return -errno;
+
+	return 0;
+}
+
+static int vfio_region_write(struct uml_vfio_user_device *dev, unsigned int index,
+			     uint64_t offset, const void *buf, uint64_t size)
+{
+	if (index >= dev->num_regions || offset + size > dev->region[index].size)
+		return -EINVAL;
+
+	if (pwrite(dev->device, buf, size, dev->region[index].offset + offset) < 0)
+		return -errno;
+
+	return 0;
+}
+
+int uml_vfio_user_cfgspace_read(struct uml_vfio_user_device *dev,
+				unsigned int offset, void *buf, int size)
+{
+	return vfio_region_read(dev, VFIO_PCI_CONFIG_REGION_INDEX,
+				offset, buf, size);
+}
+
+int uml_vfio_user_cfgspace_write(struct uml_vfio_user_device *dev,
+				 unsigned int offset, const void *buf, int size)
+{
+	return vfio_region_write(dev, VFIO_PCI_CONFIG_REGION_INDEX,
+				 offset, buf, size);
+}
+
+int uml_vfio_user_bar_read(struct uml_vfio_user_device *dev, int bar,
+			   unsigned int offset, void *buf, int size)
+{
+	return vfio_region_read(dev, bar, offset, buf, size);
+}
+
+int uml_vfio_user_bar_write(struct uml_vfio_user_device *dev, int bar,
+			    unsigned int offset, const void *buf, int size)
+{
+	return vfio_region_write(dev, bar, offset, buf, size);
+}

linux-7.2.3/arch/um/drivers/vfio_user.h

@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __UM_VFIO_USER_H
+#define __UM_VFIO_USER_H
+
+struct uml_vfio_user_device {
+	int device;
+
+	struct {
+		uint64_t size;
+		uint64_t offset;
+	} *region;
+	int num_regions;
+
+	int32_t *irqfd;
+	int irq_count;
+};
+
+int uml_vfio_user_open_container(void);
+int uml_vfio_user_setup_iommu(int container);
+
+int uml_vfio_user_get_group_id(const char *device);
+int uml_vfio_user_open_group(int group_id);
+int uml_vfio_user_set_container(int container, int group);
+int uml_vfio_user_unset_container(int container, int group);
+
+int uml_vfio_user_setup_device(struct uml_vfio_user_device *dev,
+			       int group, const char *device);
+void uml_vfio_user_teardown_device(struct uml_vfio_user_device *dev);
+
+int uml_vfio_user_activate_irq(struct uml_vfio_user_device *dev, int index);
+void uml_vfio_user_deactivate_irq(struct uml_vfio_user_device *dev, int index);
+int uml_vfio_user_update_irqs(struct uml_vfio_user_device *dev);
+
+int uml_vfio_user_cfgspace_read(struct uml_vfio_user_device *dev,
+				unsigned int offset, void *buf, int size);
+int uml_vfio_user_cfgspace_write(struct uml_vfio_user_device *dev,
+				 unsigned int offset, const void *buf, int size);
+
+int uml_vfio_user_bar_read(struct uml_vfio_user_device *dev, int bar,
+			   unsigned int offset, void *buf, int size);
+int uml_vfio_user_bar_write(struct uml_vfio_user_device *dev, int bar,
+			    unsigned int offset, const void *buf, int size);
+
+#endif /* __UM_VFIO_USER_H */

linux-7.2.3/arch/um/drivers/vhost_user.h

@@ -0,0 +1,123 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* Vhost-user protocol */
+
+#ifndef __VHOST_USER_H__
+#define __VHOST_USER_H__
+
+/* Message flags */
+#define VHOST_USER_FLAG_REPLY		BIT(2)
+#define VHOST_USER_FLAG_NEED_REPLY	BIT(3)
+/* Feature bits */
+#define VHOST_USER_F_PROTOCOL_FEATURES	30
+/* Protocol feature bits */
+#define VHOST_USER_PROTOCOL_F_MQ			0
+#define VHOST_USER_PROTOCOL_F_REPLY_ACK			3
+#define VHOST_USER_PROTOCOL_F_SLAVE_REQ			5
+#define VHOST_USER_PROTOCOL_F_CONFIG			9
+#define VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS	14
+/* Vring state index masks */
+#define VHOST_USER_VRING_INDEX_MASK	0xff
+#define VHOST_USER_VRING_POLL_MASK	BIT(8)
+
+/* Supported version */
+#define VHOST_USER_VERSION		1
+/* Supported transport features */
+#define VHOST_USER_SUPPORTED_F		BIT_ULL(VHOST_USER_F_PROTOCOL_FEATURES)
+/* Supported protocol features */
+#define VHOST_USER_SUPPORTED_PROTOCOL_F	(BIT_ULL(VHOST_USER_PROTOCOL_F_MQ) | \
+					 BIT_ULL(VHOST_USER_PROTOCOL_F_REPLY_ACK) | \
+					 BIT_ULL(VHOST_USER_PROTOCOL_F_SLAVE_REQ) | \
+					 BIT_ULL(VHOST_USER_PROTOCOL_F_CONFIG) | \
+					 BIT_ULL(VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS))
+
+enum vhost_user_request {
+	VHOST_USER_GET_FEATURES = 1,
+	VHOST_USER_SET_FEATURES = 2,
+	VHOST_USER_SET_OWNER = 3,
+	VHOST_USER_RESET_OWNER = 4,
+	VHOST_USER_SET_MEM_TABLE = 5,
+	VHOST_USER_SET_LOG_BASE = 6,
+	VHOST_USER_SET_LOG_FD = 7,
+	VHOST_USER_SET_VRING_NUM = 8,
+	VHOST_USER_SET_VRING_ADDR = 9,
+	VHOST_USER_SET_VRING_BASE = 10,
+	VHOST_USER_GET_VRING_BASE = 11,
+	VHOST_USER_SET_VRING_KICK = 12,
+	VHOST_USER_SET_VRING_CALL = 13,
+	VHOST_USER_SET_VRING_ERR = 14,
+	VHOST_USER_GET_PROTOCOL_FEATURES = 15,
+	VHOST_USER_SET_PROTOCOL_FEATURES = 16,
+	VHOST_USER_GET_QUEUE_NUM = 17,
+	VHOST_USER_SET_VRING_ENABLE = 18,
+	VHOST_USER_SEND_RARP = 19,
+	VHOST_USER_NET_SEND_MTU = 20,
+	VHOST_USER_SET_SLAVE_REQ_FD = 21,
+	VHOST_USER_IOTLB_MSG = 22,
+	VHOST_USER_SET_VRING_ENDIAN = 23,
+	VHOST_USER_GET_CONFIG = 24,
+	VHOST_USER_SET_CONFIG = 25,
+	VHOST_USER_VRING_KICK = 35,
+};
+
+enum vhost_user_slave_request {
+	VHOST_USER_SLAVE_IOTLB_MSG = 1,
+	VHOST_USER_SLAVE_CONFIG_CHANGE_MSG = 2,
+	VHOST_USER_SLAVE_VRING_HOST_NOTIFIER_MSG = 3,
+	VHOST_USER_SLAVE_VRING_CALL = 4,
+};
+
+struct vhost_user_header {
+	/*
+	 * Use enum vhost_user_request for outgoing messages,
+	 * uses enum vhost_user_slave_request for incoming ones.
+	 */
+	u32 request;
+	u32 flags;
+	u32 size;
+} __packed;
+
+struct vhost_user_config {
+	u32 offset;
+	u32 size;
+	u32 flags;
+	u8 payload[]; /* Variable length */
+} __packed;
+
+struct vhost_user_vring_state {
+	u32 index;
+	u32 num;
+} __packed;
+
+struct vhost_user_vring_addr {
+	u32 index;
+	u32 flags;
+	u64 desc, used, avail, log;
+} __packed;
+
+struct vhost_user_mem_region {
+	u64 guest_addr;
+	u64 size;
+	u64 user_addr;
+	u64 mmap_offset;
+} __packed;
+
+struct vhost_user_mem_regions {
+	u32 num;
+	u32 padding;
+	struct vhost_user_mem_region regions[2]; /* Currently supporting 2 */
+} __packed;
+
+union vhost_user_payload {
+	u64 integer;
+	struct vhost_user_config config;
+	struct vhost_user_vring_state vring_state;
+	struct vhost_user_vring_addr vring_addr;
+	struct vhost_user_mem_regions mem_regions;
+};
+
+struct vhost_user_msg {
+	struct vhost_user_header header;
+	union vhost_user_payload payload;
+} __packed;
+
+#endif

linux-7.2.3/arch/um/drivers/virt-pci.c

@@ -0,0 +1,618 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020 Intel Corporation
+ * Author: Johannes Berg <johannes@sipsolutions.net>
+ */
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/logic_iomem.h>
+#include <linux/of_platform.h>
+#include <linux/irqchip/irq-msi-lib.h>
+#include <linux/irqdomain.h>
+#include <linux/msi.h>
+#include <linux/unaligned.h>
+#include <irq_kern.h>
+
+#include "virt-pci.h"
+
+#define MAX_DEVICES 8
+#define MAX_MSI_VECTORS 32
+#define CFG_SPACE_SIZE 4096
+
+struct um_pci_device_reg {
+	struct um_pci_device *dev;
+	void __iomem *iomem;
+};
+
+static struct pci_host_bridge *bridge;
+static DEFINE_MUTEX(um_pci_mtx);
+static struct um_pci_device *um_pci_platform_device;
+static struct um_pci_device_reg um_pci_devices[MAX_DEVICES];
+static struct fwnode_handle *um_pci_fwnode;
+static struct irq_domain *um_pci_inner_domain;
+static unsigned long um_pci_msi_used[BITS_TO_LONGS(MAX_MSI_VECTORS)];
+
+static unsigned long um_pci_cfgspace_read(void *priv, unsigned int offset,
+					  int size)
+{
+	struct um_pci_device_reg *reg = priv;
+	struct um_pci_device *dev = reg->dev;
+
+	if (!dev)
+		return ULONG_MAX;
+
+	switch (size) {
+	case 1:
+	case 2:
+	case 4:
+#ifdef CONFIG_64BIT
+	case 8:
+#endif
+		break;
+	default:
+		WARN(1, "invalid config space read size %d\n", size);
+		return ULONG_MAX;
+	}
+
+	return dev->ops->cfgspace_read(dev, offset, size);
+}
+
+static void um_pci_cfgspace_write(void *priv, unsigned int offset, int size,
+				  unsigned long val)
+{
+	struct um_pci_device_reg *reg = priv;
+	struct um_pci_device *dev = reg->dev;
+
+	if (!dev)
+		return;
+
+	switch (size) {
+	case 1:
+	case 2:
+	case 4:
+#ifdef CONFIG_64BIT
+	case 8:
+#endif
+		break;
+	default:
+		WARN(1, "invalid config space write size %d\n", size);
+		return;
+	}
+
+	dev->ops->cfgspace_write(dev, offset, size, val);
+}
+
+static const struct logic_iomem_ops um_pci_device_cfgspace_ops = {
+	.read = um_pci_cfgspace_read,
+	.write = um_pci_cfgspace_write,
+};
+
+static unsigned long um_pci_bar_read(void *priv, unsigned int offset,
+				     int size)
+{
+	u8 *resptr = priv;
+	struct um_pci_device *dev = container_of(resptr - *resptr,
+						 struct um_pci_device,
+						 resptr[0]);
+	u8 bar = *resptr;
+
+	switch (size) {
+	case 1:
+	case 2:
+	case 4:
+#ifdef CONFIG_64BIT
+	case 8:
+#endif
+		break;
+	default:
+		WARN(1, "invalid bar read size %d\n", size);
+		return ULONG_MAX;
+	}
+
+	return dev->ops->bar_read(dev, bar, offset, size);
+}
+
+static void um_pci_bar_write(void *priv, unsigned int offset, int size,
+			     unsigned long val)
+{
+	u8 *resptr = priv;
+	struct um_pci_device *dev = container_of(resptr - *resptr,
+						 struct um_pci_device,
+						 resptr[0]);
+	u8 bar = *resptr;
+
+	switch (size) {
+	case 1:
+	case 2:
+	case 4:
+#ifdef CONFIG_64BIT
+	case 8:
+#endif
+		break;
+	default:
+		WARN(1, "invalid bar write size %d\n", size);
+		return;
+	}
+
+	dev->ops->bar_write(dev, bar, offset, size, val);
+}
+
+static void um_pci_bar_copy_from(void *priv, void *buffer,
+				 unsigned int offset, int size)
+{
+	u8 *resptr = priv;
+	struct um_pci_device *dev = container_of(resptr - *resptr,
+						 struct um_pci_device,
+						 resptr[0]);
+	u8 bar = *resptr;
+
+	dev->ops->bar_copy_from(dev, bar, buffer, offset, size);
+}
+
+static void um_pci_bar_copy_to(void *priv, unsigned int offset,
+			       const void *buffer, int size)
+{
+	u8 *resptr = priv;
+	struct um_pci_device *dev = container_of(resptr - *resptr,
+						 struct um_pci_device,
+						 resptr[0]);
+	u8 bar = *resptr;
+
+	dev->ops->bar_copy_to(dev, bar, offset, buffer, size);
+}
+
+static void um_pci_bar_set(void *priv, unsigned int offset, u8 value, int size)
+{
+	u8 *resptr = priv;
+	struct um_pci_device *dev = container_of(resptr - *resptr,
+						 struct um_pci_device,
+						 resptr[0]);
+	u8 bar = *resptr;
+
+	dev->ops->bar_set(dev, bar, offset, value, size);
+}
+
+static const struct logic_iomem_ops um_pci_device_bar_ops = {
+	.read = um_pci_bar_read,
+	.write = um_pci_bar_write,
+	.set = um_pci_bar_set,
+	.copy_from = um_pci_bar_copy_from,
+	.copy_to = um_pci_bar_copy_to,
+};
+
+static void __iomem *um_pci_map_bus(struct pci_bus *bus, unsigned int devfn,
+				    int where)
+{
+	struct um_pci_device_reg *dev;
+	unsigned int busn = bus->number;
+
+	if (busn > 0)
+		return NULL;
+
+	/* not allowing functions for now ... */
+	if (devfn % 8)
+		return NULL;
+
+	if (devfn / 8 >= ARRAY_SIZE(um_pci_devices))
+		return NULL;
+
+	dev = &um_pci_devices[devfn / 8];
+	if (!dev)
+		return NULL;
+
+	return (void __iomem *)((unsigned long)dev->iomem + where);
+}
+
+static struct pci_ops um_pci_ops = {
+	.map_bus = um_pci_map_bus,
+	.read = pci_generic_config_read,
+	.write = pci_generic_config_write,
+};
+
+static void um_pci_rescan(void)
+{
+	pci_lock_rescan_remove();
+	pci_rescan_bus(bridge->bus);
+	pci_unlock_rescan_remove();
+}
+
+#ifdef CONFIG_OF
+/* Copied from arch/x86/kernel/devicetree.c */
+struct device_node *pcibios_get_phb_of_node(struct pci_bus *bus)
+{
+	struct device_node *np;
+
+	for_each_node_by_type(np, "pci") {
+		const void *prop;
+		unsigned int bus_min;
+
+		prop = of_get_property(np, "bus-range", NULL);
+		if (!prop)
+			continue;
+		bus_min = be32_to_cpup(prop);
+		if (bus->number == bus_min)
+			return np;
+	}
+	return NULL;
+}
+#endif
+
+static struct resource virt_cfgspace_resource = {
+	.name = "PCI config space",
+	.start = 0xf0000000 - MAX_DEVICES * CFG_SPACE_SIZE,
+	.end = 0xf0000000 - 1,
+	.flags = IORESOURCE_MEM,
+};
+
+static long um_pci_map_cfgspace(unsigned long offset, size_t size,
+				const struct logic_iomem_ops **ops,
+				void **priv)
+{
+	if (WARN_ON(size > CFG_SPACE_SIZE || offset % CFG_SPACE_SIZE))
+		return -EINVAL;
+
+	if (offset / CFG_SPACE_SIZE < MAX_DEVICES) {
+		*ops = &um_pci_device_cfgspace_ops;
+		*priv = &um_pci_devices[offset / CFG_SPACE_SIZE];
+		return 0;
+	}
+
+	WARN(1, "cannot map offset 0x%lx/0x%zx\n", offset, size);
+	return -ENOENT;
+}
+
+static const struct logic_iomem_region_ops um_pci_cfgspace_ops = {
+	.map = um_pci_map_cfgspace,
+};
+
+static struct resource virt_iomem_resource = {
+	.name = "PCI iomem",
+	.start = 0xf0000000,
+	.end = 0xffffffff,
+	.flags = IORESOURCE_MEM,
+};
+
+struct um_pci_map_iomem_data {
+	unsigned long offset;
+	size_t size;
+	const struct logic_iomem_ops **ops;
+	void **priv;
+	long ret;
+};
+
+static int um_pci_map_iomem_walk(struct pci_dev *pdev, void *_data)
+{
+	struct um_pci_map_iomem_data *data = _data;
+	struct um_pci_device_reg *reg = &um_pci_devices[pdev->devfn / 8];
+	struct um_pci_device *dev;
+	int i;
+
+	if (!reg->dev)
+		return 0;
+
+	for (i = 0; i < ARRAY_SIZE(dev->resptr); i++) {
+		struct resource *r = &pdev->resource[i];
+
+		if ((r->flags & IORESOURCE_TYPE_BITS) != IORESOURCE_MEM)
+			continue;
+
+		/*
+		 * must be the whole or part of the resource,
+		 * not allowed to only overlap
+		 */
+		if (data->offset < r->start || data->offset > r->end)
+			continue;
+		if (data->offset + data->size - 1 > r->end)
+			continue;
+
+		dev = reg->dev;
+		*data->ops = &um_pci_device_bar_ops;
+		dev->resptr[i] = i;
+		*data->priv = &dev->resptr[i];
+		data->ret = data->offset - r->start;
+
+		/* no need to continue */
+		return 1;
+	}
+
+	return 0;
+}
+
+static long um_pci_map_iomem(unsigned long offset, size_t size,
+			     const struct logic_iomem_ops **ops,
+			     void **priv)
+{
+	struct um_pci_map_iomem_data data = {
+		/* we want the full address here */
+		.offset = offset + virt_iomem_resource.start,
+		.size = size,
+		.ops = ops,
+		.priv = priv,
+		.ret = -ENOENT,
+	};
+
+	pci_walk_bus(bridge->bus, um_pci_map_iomem_walk, &data);
+	return data.ret;
+}
+
+static const struct logic_iomem_region_ops um_pci_iomem_ops = {
+	.map = um_pci_map_iomem,
+};
+
+static void um_pci_compose_msi_msg(struct irq_data *data, struct msi_msg *msg)
+{
+	/*
+	 * This is a very low address and not actually valid 'physical' memory
+	 * in UML, so we can simply map MSI(-X) vectors to there, it cannot be
+	 * legitimately written to by the device in any other way.
+	 * We use the (virtual) IRQ number here as the message to simplify the
+	 * code that receives the message, where for now we simply trust the
+	 * device to send the correct message.
+	 */
+	msg->address_hi = 0;
+	msg->address_lo = 0xa0000;
+	msg->data = data->irq;
+}
+
+static struct irq_chip um_pci_msi_bottom_irq_chip = {
+	.name = "UM virtual MSI",
+	.irq_compose_msi_msg = um_pci_compose_msi_msg,
+};
+
+static int um_pci_inner_domain_alloc(struct irq_domain *domain,
+				     unsigned int virq, unsigned int nr_irqs,
+				     void *args)
+{
+	unsigned long bit;
+
+	WARN_ON(nr_irqs != 1);
+
+	mutex_lock(&um_pci_mtx);
+	bit = find_first_zero_bit(um_pci_msi_used, MAX_MSI_VECTORS);
+	if (bit >= MAX_MSI_VECTORS) {
+		mutex_unlock(&um_pci_mtx);
+		return -ENOSPC;
+	}
+
+	set_bit(bit, um_pci_msi_used);
+	mutex_unlock(&um_pci_mtx);
+
+	irq_domain_set_info(domain, virq, bit, &um_pci_msi_bottom_irq_chip,
+			    domain->host_data, handle_simple_irq,
+			    NULL, NULL);
+
+	return 0;
+}
+
+static void um_pci_inner_domain_free(struct irq_domain *domain,
+				     unsigned int virq, unsigned int nr_irqs)
+{
+	struct irq_data *d = irq_domain_get_irq_data(domain, virq);
+
+	mutex_lock(&um_pci_mtx);
+
+	if (!test_bit(d->hwirq, um_pci_msi_used))
+		pr_err("trying to free unused MSI#%lu\n", d->hwirq);
+	else
+		__clear_bit(d->hwirq, um_pci_msi_used);
+
+	mutex_unlock(&um_pci_mtx);
+}
+
+static const struct irq_domain_ops um_pci_inner_domain_ops = {
+	.select = msi_lib_irq_domain_select,
+	.alloc = um_pci_inner_domain_alloc,
+	.free = um_pci_inner_domain_free,
+};
+
+#define UM_PCI_MSI_FLAGS_REQUIRED (MSI_FLAG_USE_DEF_DOM_OPS		| \
+				   MSI_FLAG_USE_DEF_CHIP_OPS		| \
+				   MSI_FLAG_NO_AFFINITY)
+#define UM_PCI_MSI_FLAGS_SUPPORTED (MSI_GENERIC_FLAGS_MASK		| \
+				    MSI_FLAG_PCI_MSIX)
+
+static const struct msi_parent_ops um_pci_msi_parent_ops = {
+	.required_flags		= UM_PCI_MSI_FLAGS_REQUIRED,
+	.supported_flags	= UM_PCI_MSI_FLAGS_SUPPORTED,
+	.bus_select_token	= DOMAIN_BUS_NEXUS,
+	.bus_select_mask	= MATCH_PCI_MSI,
+	.prefix			= "UM-virtual-",
+	.init_dev_msi_info	= msi_lib_init_dev_msi_info,
+};
+
+static struct resource busn_resource = {
+	.name	= "PCI busn",
+	.start	= 0,
+	.end	= 0,
+	.flags	= IORESOURCE_BUS,
+};
+
+static int um_pci_map_irq(const struct pci_dev *pdev, u8 slot, u8 pin)
+{
+	struct um_pci_device_reg *reg = &um_pci_devices[pdev->devfn / 8];
+
+	if (WARN_ON(!reg->dev))
+		return -EINVAL;
+
+	/* Yes, we map all pins to the same IRQ ... doesn't matter for now. */
+	return reg->dev->irq;
+}
+
+void *pci_root_bus_fwnode(struct pci_bus *bus)
+{
+	return um_pci_fwnode;
+}
+
+static long um_pci_map_platform(unsigned long offset, size_t size,
+				const struct logic_iomem_ops **ops,
+				void **priv)
+{
+	if (!um_pci_platform_device)
+		return -ENOENT;
+
+	*ops = &um_pci_device_bar_ops;
+	*priv = &um_pci_platform_device->resptr[0];
+
+	return offset;
+}
+
+static const struct logic_iomem_region_ops um_pci_platform_ops = {
+	.map = um_pci_map_platform,
+};
+
+static struct resource virt_platform_resource = {
+	.name = "platform",
+	.start = 0x10000000,
+	.end = 0x1fffffff,
+	.flags = IORESOURCE_MEM,
+};
+
+int um_pci_device_register(struct um_pci_device *dev)
+{
+	int i, free = -1;
+	int err = 0;
+
+	mutex_lock(&um_pci_mtx);
+	for (i = 0; i < MAX_DEVICES; i++) {
+		if (um_pci_devices[i].dev)
+			continue;
+		free = i;
+		break;
+	}
+
+	if (free < 0) {
+		err = -ENOSPC;
+		goto out;
+	}
+
+	dev->irq = irq_alloc_desc(numa_node_id());
+	if (dev->irq < 0) {
+		err = dev->irq;
+		goto out;
+	}
+
+	um_pci_devices[free].dev = dev;
+
+out:
+	mutex_unlock(&um_pci_mtx);
+	if (!err)
+		um_pci_rescan();
+	return err;
+}
+
+void um_pci_device_unregister(struct um_pci_device *dev)
+{
+	int i;
+
+	mutex_lock(&um_pci_mtx);
+	for (i = 0; i < MAX_DEVICES; i++) {
+		if (um_pci_devices[i].dev != dev)
+			continue;
+		um_pci_devices[i].dev = NULL;
+		irq_free_desc(dev->irq);
+		break;
+	}
+	mutex_unlock(&um_pci_mtx);
+
+	if (i < MAX_DEVICES) {
+		struct pci_dev *pci_dev;
+
+		pci_dev = pci_get_slot(bridge->bus, i);
+		if (pci_dev)
+			pci_stop_and_remove_bus_device_locked(pci_dev);
+	}
+}
+
+int um_pci_platform_device_register(struct um_pci_device *dev)
+{
+	guard(mutex)(&um_pci_mtx);
+	if (um_pci_platform_device)
+		return -EBUSY;
+	um_pci_platform_device = dev;
+	return 0;
+}
+
+void um_pci_platform_device_unregister(struct um_pci_device *dev)
+{
+	guard(mutex)(&um_pci_mtx);
+	if (um_pci_platform_device == dev)
+		um_pci_platform_device = NULL;
+}
+
+static int __init um_pci_init(void)
+{
+	int err, i;
+
+	WARN_ON(logic_iomem_add_region(&virt_cfgspace_resource,
+				       &um_pci_cfgspace_ops));
+	WARN_ON(logic_iomem_add_region(&virt_iomem_resource,
+				       &um_pci_iomem_ops));
+	WARN_ON(logic_iomem_add_region(&virt_platform_resource,
+				       &um_pci_platform_ops));
+
+	bridge = pci_alloc_host_bridge(0);
+	if (!bridge) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	um_pci_fwnode = irq_domain_alloc_named_fwnode("um-pci");
+	if (!um_pci_fwnode) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	struct irq_domain_info info = {
+		.fwnode		= um_pci_fwnode,
+		.ops		= &um_pci_inner_domain_ops,
+		.size		= MAX_MSI_VECTORS,
+	};
+
+	um_pci_inner_domain = msi_create_parent_irq_domain(&info, &um_pci_msi_parent_ops);
+	if (!um_pci_inner_domain) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	pci_add_resource(&bridge->windows, &virt_iomem_resource);
+	pci_add_resource(&bridge->windows, &busn_resource);
+	bridge->ops = &um_pci_ops;
+	bridge->map_irq = um_pci_map_irq;
+
+	for (i = 0; i < MAX_DEVICES; i++) {
+		resource_size_t start;
+
+		start = virt_cfgspace_resource.start + i * CFG_SPACE_SIZE;
+		um_pci_devices[i].iomem = ioremap(start, CFG_SPACE_SIZE);
+		if (WARN(!um_pci_devices[i].iomem, "failed to map %d\n", i)) {
+			err = -ENOMEM;
+			goto free;
+		}
+	}
+
+	err = pci_host_probe(bridge);
+	if (err)
+		goto free;
+
+	return 0;
+
+free:
+	if (um_pci_inner_domain)
+		irq_domain_remove(um_pci_inner_domain);
+	if (um_pci_fwnode)
+		irq_domain_free_fwnode(um_pci_fwnode);
+	if (bridge) {
+		pci_free_resource_list(&bridge->windows);
+		pci_free_host_bridge(bridge);
+	}
+	return err;
+}
+device_initcall(um_pci_init);
+
+static void __exit um_pci_exit(void)
+{
+	irq_domain_remove(um_pci_inner_domain);
+	pci_free_resource_list(&bridge->windows);
+	pci_free_host_bridge(bridge);
+}
+module_exit(um_pci_exit);

linux-7.2.3/arch/um/drivers/virt-pci.h

@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __UM_VIRT_PCI_H
+#define __UM_VIRT_PCI_H
+
+#include <linux/pci.h>
+
+struct um_pci_device {
+	const struct um_pci_ops *ops;
+
+	/* for now just standard BARs */
+	u8 resptr[PCI_STD_NUM_BARS];
+
+	int irq;
+};
+
+struct um_pci_ops {
+	unsigned long (*cfgspace_read)(struct um_pci_device *dev,
+				       unsigned int offset, int size);
+	void (*cfgspace_write)(struct um_pci_device *dev, unsigned int offset,
+			       int size, unsigned long val);
+
+	unsigned long (*bar_read)(struct um_pci_device *dev, int bar,
+				  unsigned int offset, int size);
+	void (*bar_write)(struct um_pci_device *dev, int bar,
+			  unsigned int offset, int size, unsigned long val);
+
+	void (*bar_copy_from)(struct um_pci_device *dev, int bar, void *buffer,
+			      unsigned int offset, int size);
+	void (*bar_copy_to)(struct um_pci_device *dev, int bar,
+			    unsigned int offset, const void *buffer, int size);
+	void (*bar_set)(struct um_pci_device *dev, int bar,
+			unsigned int offset, u8 value, int size);
+};
+
+int um_pci_device_register(struct um_pci_device *dev);
+void um_pci_device_unregister(struct um_pci_device *dev);
+
+int um_pci_platform_device_register(struct um_pci_device *dev);
+void um_pci_platform_device_unregister(struct um_pci_device *dev);
+
+#endif /* __UM_VIRT_PCI_H */

linux-7.2.3/arch/um/drivers/virtio_pcidev.c

@@ -0,0 +1,634 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020 Intel Corporation
+ * Author: Johannes Berg <johannes@sipsolutions.net>
+ */
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/virtio.h>
+#include <linux/virtio_config.h>
+#include <linux/logic_iomem.h>
+#include <linux/of_platform.h>
+#include <linux/irqdomain.h>
+#include <linux/virtio_pcidev.h>
+#include <linux/virtio-uml.h>
+#include <linux/delay.h>
+#include <linux/msi.h>
+#include <linux/unaligned.h>
+#include <irq_kern.h>
+
+#include "virt-pci.h"
+
+#define to_virtio_pcidev(_pdev) \
+	container_of(_pdev, struct virtio_pcidev_device, pdev)
+
+/* for MSI-X we have a 32-bit payload */
+#define MAX_IRQ_MSG_SIZE (sizeof(struct virtio_pcidev_msg) + sizeof(u32))
+#define NUM_IRQ_MSGS	10
+
+struct virtio_pcidev_message_buffer {
+	struct virtio_pcidev_msg hdr;
+	u8 data[8];
+};
+
+struct virtio_pcidev_device {
+	struct um_pci_device pdev;
+	struct virtio_device *vdev;
+
+	struct virtqueue *cmd_vq, *irq_vq;
+
+#define VIRTIO_PCIDEV_WRITE_BUFS	20
+	struct virtio_pcidev_message_buffer bufs[VIRTIO_PCIDEV_WRITE_BUFS + 1];
+	void *extra_ptrs[VIRTIO_PCIDEV_WRITE_BUFS + 1];
+	DECLARE_BITMAP(used_bufs, VIRTIO_PCIDEV_WRITE_BUFS);
+
+#define VIRTIO_PCIDEV_STAT_WAITING	0
+	unsigned long status;
+
+	bool platform;
+};
+
+static unsigned int virtio_pcidev_max_delay_us = 40000;
+module_param_named(max_delay_us, virtio_pcidev_max_delay_us, uint, 0644);
+
+static int virtio_pcidev_get_buf(struct virtio_pcidev_device *dev, bool *posted)
+{
+	int i;
+
+	for (i = 0; i < VIRTIO_PCIDEV_WRITE_BUFS; i++) {
+		if (!test_and_set_bit(i, dev->used_bufs))
+			return i;
+	}
+
+	*posted = false;
+	return VIRTIO_PCIDEV_WRITE_BUFS;
+}
+
+static void virtio_pcidev_free_buf(struct virtio_pcidev_device *dev, void *buf)
+{
+	int i;
+
+	if (buf == &dev->bufs[VIRTIO_PCIDEV_WRITE_BUFS]) {
+		kfree(dev->extra_ptrs[VIRTIO_PCIDEV_WRITE_BUFS]);
+		dev->extra_ptrs[VIRTIO_PCIDEV_WRITE_BUFS] = NULL;
+		return;
+	}
+
+	for (i = 0; i < VIRTIO_PCIDEV_WRITE_BUFS; i++) {
+		if (buf == &dev->bufs[i]) {
+			kfree(dev->extra_ptrs[i]);
+			dev->extra_ptrs[i] = NULL;
+			WARN_ON(!test_and_clear_bit(i, dev->used_bufs));
+			return;
+		}
+	}
+
+	WARN_ON(1);
+}
+
+static int virtio_pcidev_send_cmd(struct virtio_pcidev_device *dev,
+				  struct virtio_pcidev_msg *cmd,
+				  unsigned int cmd_size,
+				  const void *extra, unsigned int extra_size,
+				  void *out, unsigned int out_size)
+{
+	struct scatterlist out_sg, extra_sg, in_sg;
+	struct scatterlist *sgs_list[] = {
+		[0] = &out_sg,
+		[1] = extra ? &extra_sg : &in_sg,
+		[2] = extra ? &in_sg : NULL,
+	};
+	struct virtio_pcidev_message_buffer *buf;
+	int delay_count = 0;
+	bool bounce_out;
+	int ret, len;
+	int buf_idx;
+	bool posted;
+
+	if (WARN_ON(cmd_size < sizeof(*cmd) || cmd_size > sizeof(*buf)))
+		return -EINVAL;
+
+	switch (cmd->op) {
+	case VIRTIO_PCIDEV_OP_CFG_WRITE:
+	case VIRTIO_PCIDEV_OP_MMIO_WRITE:
+	case VIRTIO_PCIDEV_OP_MMIO_MEMSET:
+		/* in PCI, writes are posted, so don't wait */
+		posted = !out;
+		WARN_ON(!posted);
+		break;
+	default:
+		posted = false;
+		break;
+	}
+
+	bounce_out = !posted && cmd_size <= sizeof(*cmd) &&
+		     out && out_size <= sizeof(buf->data);
+
+	buf_idx = virtio_pcidev_get_buf(dev, &posted);
+	buf = &dev->bufs[buf_idx];
+	memcpy(buf, cmd, cmd_size);
+
+	if (posted && extra && extra_size > sizeof(buf) - cmd_size) {
+		dev->extra_ptrs[buf_idx] = kmemdup(extra, extra_size,
+						   GFP_ATOMIC);
+
+		if (!dev->extra_ptrs[buf_idx]) {
+			virtio_pcidev_free_buf(dev, buf);
+			return -ENOMEM;
+		}
+		extra = dev->extra_ptrs[buf_idx];
+	} else if (extra && extra_size <= sizeof(buf) - cmd_size) {
+		memcpy((u8 *)buf + cmd_size, extra, extra_size);
+		cmd_size += extra_size;
+		extra_size = 0;
+		extra = NULL;
+		cmd = (void *)buf;
+	} else {
+		cmd = (void *)buf;
+	}
+
+	sg_init_one(&out_sg, cmd, cmd_size);
+	if (extra)
+		sg_init_one(&extra_sg, extra, extra_size);
+	/* allow stack for small buffers */
+	if (bounce_out)
+		sg_init_one(&in_sg, buf->data, out_size);
+	else if (out)
+		sg_init_one(&in_sg, out, out_size);
+
+	/* add to internal virtio queue */
+	ret = virtqueue_add_sgs(dev->cmd_vq, sgs_list,
+				extra ? 2 : 1,
+				out ? 1 : 0,
+				cmd, GFP_ATOMIC);
+	if (ret) {
+		virtio_pcidev_free_buf(dev, buf);
+		return ret;
+	}
+
+	if (posted) {
+		virtqueue_kick(dev->cmd_vq);
+		return 0;
+	}
+
+	/* kick and poll for getting a response on the queue */
+	set_bit(VIRTIO_PCIDEV_STAT_WAITING, &dev->status);
+	virtqueue_kick(dev->cmd_vq);
+	ret = 0;
+
+	while (1) {
+		void *completed = virtqueue_get_buf(dev->cmd_vq, &len);
+
+		if (completed == buf)
+			break;
+
+		if (completed)
+			virtio_pcidev_free_buf(dev, completed);
+
+		if (WARN_ONCE(virtqueue_is_broken(dev->cmd_vq) ||
+			      ++delay_count > virtio_pcidev_max_delay_us,
+			      "um virt-pci delay: %d", delay_count)) {
+			ret = -EIO;
+			break;
+		}
+		udelay(1);
+	}
+	clear_bit(VIRTIO_PCIDEV_STAT_WAITING, &dev->status);
+
+	if (bounce_out)
+		memcpy(out, buf->data, out_size);
+
+	virtio_pcidev_free_buf(dev, buf);
+
+	return ret;
+}
+
+static unsigned long virtio_pcidev_cfgspace_read(struct um_pci_device *pdev,
+						 unsigned int offset, int size)
+{
+	struct virtio_pcidev_device *dev = to_virtio_pcidev(pdev);
+	struct virtio_pcidev_msg hdr = {
+		.op = VIRTIO_PCIDEV_OP_CFG_READ,
+		.size = size,
+		.addr = offset,
+	};
+	/* max 8, we might not use it all */
+	u8 data[8];
+
+	memset(data, 0xff, sizeof(data));
+
+	/* size has been checked in um_pci_cfgspace_read() */
+	if (virtio_pcidev_send_cmd(dev, &hdr, sizeof(hdr), NULL, 0, data, size))
+		return ULONG_MAX;
+
+	switch (size) {
+	case 1:
+		return data[0];
+	case 2:
+		return le16_to_cpup((void *)data);
+	case 4:
+		return le32_to_cpup((void *)data);
+#ifdef CONFIG_64BIT
+	case 8:
+		return le64_to_cpup((void *)data);
+#endif
+	default:
+		return ULONG_MAX;
+	}
+}
+
+static void virtio_pcidev_cfgspace_write(struct um_pci_device *pdev,
+					 unsigned int offset, int size,
+					 unsigned long val)
+{
+	struct virtio_pcidev_device *dev = to_virtio_pcidev(pdev);
+	struct {
+		struct virtio_pcidev_msg hdr;
+		/* maximum size - we may only use parts of it */
+		u8 data[8];
+	} msg = {
+		.hdr = {
+			.op = VIRTIO_PCIDEV_OP_CFG_WRITE,
+			.size = size,
+			.addr = offset,
+		},
+	};
+
+	/* size has been checked in um_pci_cfgspace_write() */
+	switch (size) {
+	case 1:
+		msg.data[0] = (u8)val;
+		break;
+	case 2:
+		put_unaligned_le16(val, (void *)msg.data);
+		break;
+	case 4:
+		put_unaligned_le32(val, (void *)msg.data);
+		break;
+#ifdef CONFIG_64BIT
+	case 8:
+		put_unaligned_le64(val, (void *)msg.data);
+		break;
+#endif
+	}
+
+	WARN_ON(virtio_pcidev_send_cmd(dev, &msg.hdr, sizeof(msg), NULL, 0, NULL, 0));
+}
+
+static void virtio_pcidev_bar_copy_from(struct um_pci_device *pdev,
+					int bar, void *buffer,
+					unsigned int offset, int size)
+{
+	struct virtio_pcidev_device *dev = to_virtio_pcidev(pdev);
+	struct virtio_pcidev_msg hdr = {
+		.op = VIRTIO_PCIDEV_OP_MMIO_READ,
+		.bar = bar,
+		.size = size,
+		.addr = offset,
+	};
+
+	memset(buffer, 0xff, size);
+
+	virtio_pcidev_send_cmd(dev, &hdr, sizeof(hdr), NULL, 0, buffer, size);
+}
+
+static unsigned long virtio_pcidev_bar_read(struct um_pci_device *pdev, int bar,
+					    unsigned int offset, int size)
+{
+	/* 8 is maximum size - we may only use parts of it */
+	u8 data[8];
+
+	/* size has been checked in um_pci_bar_read() */
+	virtio_pcidev_bar_copy_from(pdev, bar, data, offset, size);
+
+	switch (size) {
+	case 1:
+		return data[0];
+	case 2:
+		return le16_to_cpup((void *)data);
+	case 4:
+		return le32_to_cpup((void *)data);
+#ifdef CONFIG_64BIT
+	case 8:
+		return le64_to_cpup((void *)data);
+#endif
+	default:
+		return ULONG_MAX;
+	}
+}
+
+static void virtio_pcidev_bar_copy_to(struct um_pci_device *pdev,
+				      int bar, unsigned int offset,
+				      const void *buffer, int size)
+{
+	struct virtio_pcidev_device *dev = to_virtio_pcidev(pdev);
+	struct virtio_pcidev_msg hdr = {
+		.op = VIRTIO_PCIDEV_OP_MMIO_WRITE,
+		.bar = bar,
+		.size = size,
+		.addr = offset,
+	};
+
+	virtio_pcidev_send_cmd(dev, &hdr, sizeof(hdr), buffer, size, NULL, 0);
+}
+
+static void virtio_pcidev_bar_write(struct um_pci_device *pdev, int bar,
+				    unsigned int offset, int size,
+				    unsigned long val)
+{
+	/* maximum size - we may only use parts of it */
+	u8 data[8];
+
+	/* size has been checked in um_pci_bar_write() */
+	switch (size) {
+	case 1:
+		data[0] = (u8)val;
+		break;
+	case 2:
+		put_unaligned_le16(val, (void *)data);
+		break;
+	case 4:
+		put_unaligned_le32(val, (void *)data);
+		break;
+#ifdef CONFIG_64BIT
+	case 8:
+		put_unaligned_le64(val, (void *)data);
+		break;
+#endif
+	}
+
+	virtio_pcidev_bar_copy_to(pdev, bar, offset, data, size);
+}
+
+static void virtio_pcidev_bar_set(struct um_pci_device *pdev, int bar,
+				  unsigned int offset, u8 value, int size)
+{
+	struct virtio_pcidev_device *dev = to_virtio_pcidev(pdev);
+	struct {
+		struct virtio_pcidev_msg hdr;
+		u8 data;
+	} msg = {
+		.hdr = {
+			.op = VIRTIO_PCIDEV_OP_CFG_WRITE,
+			.bar = bar,
+			.size = size,
+			.addr = offset,
+		},
+		.data = value,
+	};
+
+	virtio_pcidev_send_cmd(dev, &msg.hdr, sizeof(msg), NULL, 0, NULL, 0);
+}
+
+static const struct um_pci_ops virtio_pcidev_um_pci_ops = {
+	.cfgspace_read	= virtio_pcidev_cfgspace_read,
+	.cfgspace_write	= virtio_pcidev_cfgspace_write,
+	.bar_read	= virtio_pcidev_bar_read,
+	.bar_write	= virtio_pcidev_bar_write,
+	.bar_copy_from	= virtio_pcidev_bar_copy_from,
+	.bar_copy_to	= virtio_pcidev_bar_copy_to,
+	.bar_set	= virtio_pcidev_bar_set,
+};
+
+static void virtio_pcidev_irq_vq_addbuf(struct virtqueue *vq, void *buf, bool kick)
+{
+	struct scatterlist sg[1];
+
+	sg_init_one(sg, buf, MAX_IRQ_MSG_SIZE);
+	if (virtqueue_add_inbuf(vq, sg, 1, buf, GFP_ATOMIC))
+		kfree(buf);
+	else if (kick)
+		virtqueue_kick(vq);
+}
+
+static void virtio_pcidev_handle_irq_message(struct virtqueue *vq,
+					     struct virtio_pcidev_msg *msg)
+{
+	struct virtio_device *vdev = vq->vdev;
+	struct virtio_pcidev_device *dev = vdev->priv;
+
+	if (!dev->pdev.irq)
+		return;
+
+	/* we should properly chain interrupts, but on ARCH=um we don't care */
+
+	switch (msg->op) {
+	case VIRTIO_PCIDEV_OP_INT:
+		generic_handle_irq(dev->pdev.irq);
+		break;
+	case VIRTIO_PCIDEV_OP_MSI:
+		/* our MSI message is just the interrupt number */
+		if (msg->size == sizeof(u32))
+			generic_handle_irq(le32_to_cpup((void *)msg->data));
+		else
+			generic_handle_irq(le16_to_cpup((void *)msg->data));
+		break;
+	case VIRTIO_PCIDEV_OP_PME:
+		/* nothing to do - we already woke up due to the message */
+		break;
+	default:
+		dev_err(&vdev->dev, "unexpected virt-pci message %d\n", msg->op);
+		break;
+	}
+}
+
+static void virtio_pcidev_cmd_vq_cb(struct virtqueue *vq)
+{
+	struct virtio_device *vdev = vq->vdev;
+	struct virtio_pcidev_device *dev = vdev->priv;
+	void *cmd;
+	int len;
+
+	if (test_bit(VIRTIO_PCIDEV_STAT_WAITING, &dev->status))
+		return;
+
+	while ((cmd = virtqueue_get_buf(vq, &len)))
+		virtio_pcidev_free_buf(dev, cmd);
+}
+
+static void virtio_pcidev_irq_vq_cb(struct virtqueue *vq)
+{
+	struct virtio_pcidev_msg *msg;
+	int len;
+
+	while ((msg = virtqueue_get_buf(vq, &len))) {
+		if (len >= sizeof(*msg))
+			virtio_pcidev_handle_irq_message(vq, msg);
+
+		/* recycle the message buffer */
+		virtio_pcidev_irq_vq_addbuf(vq, msg, true);
+	}
+}
+
+static int virtio_pcidev_init_vqs(struct virtio_pcidev_device *dev)
+{
+	struct virtqueue_info vqs_info[] = {
+		{ "cmd", virtio_pcidev_cmd_vq_cb },
+		{ "irq", virtio_pcidev_irq_vq_cb },
+	};
+	struct virtqueue *vqs[2];
+	int err, i;
+
+	err = virtio_find_vqs(dev->vdev, 2, vqs, vqs_info, NULL);
+	if (err)
+		return err;
+
+	dev->cmd_vq = vqs[0];
+	dev->irq_vq = vqs[1];
+
+	virtio_device_ready(dev->vdev);
+
+	for (i = 0; i < NUM_IRQ_MSGS; i++) {
+		void *msg = kzalloc(MAX_IRQ_MSG_SIZE, GFP_KERNEL);
+
+		if (msg)
+			virtio_pcidev_irq_vq_addbuf(dev->irq_vq, msg, false);
+	}
+
+	virtqueue_kick(dev->irq_vq);
+
+	return 0;
+}
+
+static void __virtio_pcidev_virtio_platform_remove(struct virtio_device *vdev,
+						   struct virtio_pcidev_device *dev)
+{
+	um_pci_platform_device_unregister(&dev->pdev);
+
+	virtio_reset_device(vdev);
+	vdev->config->del_vqs(vdev);
+
+	kfree(dev);
+}
+
+static int virtio_pcidev_virtio_platform_probe(struct virtio_device *vdev,
+					       struct virtio_pcidev_device *dev)
+{
+	int err;
+
+	dev->platform = true;
+
+	err = virtio_pcidev_init_vqs(dev);
+	if (err)
+		goto err_free;
+
+	err = um_pci_platform_device_register(&dev->pdev);
+	if (err)
+		goto err_reset;
+
+	err = of_platform_default_populate(vdev->dev.of_node, NULL, &vdev->dev);
+	if (err)
+		goto err_unregister;
+
+	return 0;
+
+err_unregister:
+	um_pci_platform_device_unregister(&dev->pdev);
+err_reset:
+	virtio_reset_device(vdev);
+	vdev->config->del_vqs(vdev);
+err_free:
+	kfree(dev);
+	return err;
+}
+
+static int virtio_pcidev_virtio_probe(struct virtio_device *vdev)
+{
+	struct virtio_pcidev_device *dev;
+	int err;
+
+	dev = kzalloc_obj(*dev);
+	if (!dev)
+		return -ENOMEM;
+
+	dev->vdev = vdev;
+	vdev->priv = dev;
+
+	dev->pdev.ops = &virtio_pcidev_um_pci_ops;
+
+	if (of_device_is_compatible(vdev->dev.of_node, "simple-bus"))
+		return virtio_pcidev_virtio_platform_probe(vdev, dev);
+
+	err = virtio_pcidev_init_vqs(dev);
+	if (err)
+		goto err_free;
+
+	err = um_pci_device_register(&dev->pdev);
+	if (err)
+		goto err_reset;
+
+	device_set_wakeup_enable(&vdev->dev, true);
+
+	/*
+	 * In order to do suspend-resume properly, don't allow VQs
+	 * to be suspended.
+	 */
+	virtio_uml_set_no_vq_suspend(vdev, true);
+
+	return 0;
+
+err_reset:
+	virtio_reset_device(vdev);
+	vdev->config->del_vqs(vdev);
+err_free:
+	kfree(dev);
+	return err;
+}
+
+static void virtio_pcidev_virtio_remove(struct virtio_device *vdev)
+{
+	struct virtio_pcidev_device *dev = vdev->priv;
+
+	if (dev->platform) {
+		of_platform_depopulate(&vdev->dev);
+		__virtio_pcidev_virtio_platform_remove(vdev, dev);
+		return;
+	}
+
+	device_set_wakeup_enable(&vdev->dev, false);
+
+	um_pci_device_unregister(&dev->pdev);
+
+	/* Stop all virtqueues */
+	virtio_reset_device(vdev);
+	dev->cmd_vq = NULL;
+	dev->irq_vq = NULL;
+	vdev->config->del_vqs(vdev);
+
+	kfree(dev);
+}
+
+static void virtio_pcidev_virtio_shutdown(struct virtio_device *vdev)
+{
+	/* nothing to do, we just don't want queue shutdown */
+}
+
+static struct virtio_device_id id_table[] = {
+	{ CONFIG_UML_PCI_OVER_VIRTIO_DEVICE_ID, VIRTIO_DEV_ANY_ID },
+	{ 0 },
+};
+MODULE_DEVICE_TABLE(virtio, id_table);
+
+static struct virtio_driver virtio_pcidev_virtio_driver = {
+	.driver.name = "virtio-pci",
+	.id_table = id_table,
+	.probe = virtio_pcidev_virtio_probe,
+	.remove = virtio_pcidev_virtio_remove,
+	.shutdown = virtio_pcidev_virtio_shutdown,
+};
+
+static int __init virtio_pcidev_init(void)
+{
+	if (WARN(CONFIG_UML_PCI_OVER_VIRTIO_DEVICE_ID < 0,
+		 "No virtio device ID configured for PCI - no PCI support\n"))
+		return 0;
+
+	return register_virtio_driver(&virtio_pcidev_virtio_driver);
+}
+late_initcall(virtio_pcidev_init);
+
+static void __exit virtio_pcidev_exit(void)
+{
+	unregister_virtio_driver(&virtio_pcidev_virtio_driver);
+}
+module_exit(virtio_pcidev_exit);

linux-7.2.3/arch/um/drivers/xterm.c

@@ -0,0 +1,226 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stddef.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <string.h>
+#include <termios.h>
+#include "chan_user.h"
+#include <os.h>
+#include <um_malloc.h>
+#include "xterm.h"
+
+struct xterm_chan {
+	int pid;
+	int helper_pid;
+	int chan_fd;
+	char *title;
+	int device;
+	int raw;
+	struct termios tt;
+};
+
+static void *xterm_init(char *str, int device, const struct chan_opts *opts)
+{
+	struct xterm_chan *data;
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_KERNEL);
+	if (data == NULL)
+		return NULL;
+	*data = ((struct xterm_chan) { .pid 		= -1,
+				       .helper_pid 	= -1,
+				       .chan_fd		= -1,
+				       .device 		= device,
+				       .title 		= opts->xterm_title,
+				       .raw  		= opts->raw } );
+	return data;
+}
+
+/* Only changed by xterm_setup, which is a setup */
+static char *terminal_emulator = CONFIG_XTERM_CHAN_DEFAULT_EMULATOR;
+static char *title_switch = "-T";
+static char *exec_switch = "-e";
+
+static int __init xterm_setup(char *line, int *add)
+{
+	*add = 0;
+	terminal_emulator = line;
+
+	line = strchr(line, ',');
+	if (line == NULL)
+		return 0;
+
+	*line++ = '\0';
+	if (*line)
+		title_switch = line;
+
+	line = strchr(line, ',');
+	if (line == NULL)
+		return 0;
+
+	*line++ = '\0';
+	if (*line)
+		exec_switch = line;
+
+	return 0;
+}
+
+__uml_setup("xterm=", xterm_setup,
+"xterm=<terminal emulator>,<title switch>,<exec switch>\n"
+"    Specifies an alternate terminal emulator to use for the debugger,\n"
+"    consoles, and serial lines when they are attached to the xterm channel.\n"
+"    The values are the terminal emulator binary, the switch it uses to set\n"
+"    its title, and the switch it uses to execute a subprocess,\n"
+"    respectively.  The title switch must have the form '<switch> title',\n"
+"    not '<switch>=title'.  Similarly, the exec switch must have the form\n"
+"    '<switch> command arg1 arg2 ...'.\n"
+"    The default values are 'xterm=" CONFIG_XTERM_CHAN_DEFAULT_EMULATOR
+     ",-T,-e'.\n"
+"    Values for gnome-terminal are 'xterm=gnome-terminal,-t,--'.\n\n"
+);
+
+static int xterm_open(int input, int output, int primary, void *d,
+		      char **dev_out)
+{
+	struct xterm_chan *data = d;
+	int pid, fd, new, err;
+	char title[256], file[] = "/tmp/xterm-pipeXXXXXX";
+	char *argv[] = { terminal_emulator, title_switch, title, exec_switch,
+			 OS_LIB_PATH "/uml/port-helper", "-uml-socket",
+			 file, NULL };
+
+	if (access(argv[4], X_OK) < 0)
+		argv[4] = "port-helper";
+
+	/* Ensure we are running on Xorg or Wayland. */
+	if (!getenv("DISPLAY") && !getenv("WAYLAND_DISPLAY")) {
+		printk(UM_KERN_ERR "xterm_open : neither $DISPLAY nor $WAYLAND_DISPLAY is set.\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * This business of getting a descriptor to a temp file,
+	 * deleting the file and closing the descriptor is just to get
+	 * a known-unused name for the Unix socket that we really
+	 * want.
+	 */
+	fd = mkstemp(file);
+	if (fd < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "xterm_open : mkstemp failed, errno = %d\n",
+		       errno);
+		return err;
+	}
+
+	if (unlink(file)) {
+		err = -errno;
+		printk(UM_KERN_ERR "xterm_open : unlink failed, errno = %d\n",
+		       errno);
+		close(fd);
+		return err;
+	}
+	close(fd);
+
+	fd = os_create_unix_socket(file, sizeof(file), 1);
+	if (fd < 0) {
+		printk(UM_KERN_ERR "xterm_open : create_unix_socket failed, "
+		       "errno = %d\n", -fd);
+		return fd;
+	}
+
+	sprintf(title, data->title, data->device);
+	pid = run_helper(NULL, NULL, argv);
+	if (pid < 0) {
+		err = pid;
+		printk(UM_KERN_ERR "xterm_open : run_helper failed, "
+		       "errno = %d\n", -err);
+		goto out_close1;
+	}
+
+	err = os_set_fd_block(fd, 0);
+	if (err < 0) {
+		printk(UM_KERN_ERR "xterm_open : failed to set descriptor "
+		       "non-blocking, err = %d\n", -err);
+		goto out_kill;
+	}
+
+	data->chan_fd = fd;
+	new = xterm_fd(fd, &data->helper_pid);
+	if (new < 0) {
+		err = new;
+		printk(UM_KERN_ERR "xterm_open : xterm_fd failed, err = %d\n",
+		       -err);
+		goto out_kill;
+	}
+
+	err = os_set_fd_block(new, 0);
+	if (err) {
+		printk(UM_KERN_ERR "xterm_open : failed to set xterm "
+		       "descriptor non-blocking, err = %d\n", -err);
+		goto out_close2;
+	}
+
+	CATCH_EINTR(err = tcgetattr(new, &data->tt));
+	if (err) {
+		new = err;
+		goto out_close2;
+	}
+
+	if (data->raw) {
+		err = raw(new);
+		if (err) {
+			new = err;
+			goto out_close2;
+		}
+	}
+
+	unlink(file);
+	data->pid = pid;
+	*dev_out = NULL;
+
+	return new;
+
+ out_close2:
+	close(new);
+ out_kill:
+	os_kill_process(pid, 1);
+ out_close1:
+	close(fd);
+
+	return err;
+}
+
+static void xterm_close(int fd, void *d)
+{
+	struct xterm_chan *data = d;
+
+	if (data->pid != -1)
+		os_kill_process(data->pid, 1);
+	data->pid = -1;
+
+	if (data->helper_pid != -1)
+		os_kill_process(data->helper_pid, 0);
+	data->helper_pid = -1;
+
+	if (data->chan_fd != -1)
+		os_close_file(data->chan_fd);
+	os_close_file(fd);
+}
+
+const struct chan_ops xterm_ops = {
+	.type		= "xterm",
+	.init		= xterm_init,
+	.open		= xterm_open,
+	.close		= xterm_close,
+	.read		= generic_read,
+	.write		= generic_write,
+	.console_write	= generic_console_write,
+	.window_size	= generic_window_size,
+	.free		= generic_free,
+	.winch		= 1,
+};

linux-7.2.3/arch/um/drivers/xterm.h

@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* 
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ */
+
+#ifndef __XTERM_H__
+#define __XTERM_H__
+
+extern int xterm_fd(int socket, int *pid_out);
+
+#endif
+

linux-7.2.3/arch/um/drivers/xterm_kern.c

@@ -0,0 +1,83 @@
+// SPDX-License-Identifier: GPL-2.0
+/* 
+ * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <linux/slab.h>
+#include <linux/completion.h>
+#include <linux/irqreturn.h>
+#include <asm/irq.h>
+#include <irq_kern.h>
+#include <os.h>
+#include "xterm.h"
+
+struct xterm_wait {
+	struct completion ready;
+	int fd;
+	int pid;
+	int new_fd;
+};
+
+static irqreturn_t xterm_interrupt(int irq, void *data)
+{
+	struct xterm_wait *xterm = data;
+	int fd = -1, n_fds = 1;
+	ssize_t ret;
+
+	ret = os_rcv_fd_msg(xterm->fd, &fd, n_fds,
+			    &xterm->pid, sizeof(xterm->pid));
+	if (ret == -EAGAIN)
+		return IRQ_NONE;
+
+	if (ret < 0)
+		fd = ret;
+	else if (ret != sizeof(xterm->pid))
+		fd = -EMSGSIZE;
+
+	xterm->new_fd = fd;
+	complete(&xterm->ready);
+
+	return IRQ_HANDLED;
+}
+
+int xterm_fd(int socket, int *pid_out)
+{
+	struct xterm_wait *data;
+	int err, ret;
+
+	data = kmalloc_obj(*data);
+	if (data == NULL) {
+		printk(KERN_ERR "xterm_fd : failed to allocate xterm_wait\n");
+		return -ENOMEM;
+	}
+
+	/* This is a locked semaphore... */
+	*data = ((struct xterm_wait) { .fd 		= socket,
+				       .pid 		= -1,
+				       .new_fd	 	= -1 });
+	init_completion(&data->ready);
+
+	err = um_request_irq(XTERM_IRQ, socket, IRQ_READ, xterm_interrupt,
+			     IRQF_SHARED, "xterm", data);
+	if (err < 0) {
+		printk(KERN_ERR "xterm_fd : failed to get IRQ for xterm, "
+		       "err = %d\n",  err);
+		ret = err;
+		goto out;
+	}
+
+	/* ... so here we wait for an xterm interrupt.
+	 *
+	 * XXX Note, if the xterm doesn't work for some reason (eg. DISPLAY
+	 * isn't set) this will hang... */
+	wait_for_completion(&data->ready);
+
+	um_free_irq(XTERM_IRQ, data);
+
+	ret = data->new_fd;
+	*pid_out = data->pid;
+ out:
+	kfree(data);
+
+	return ret;
+}