Resguardo inicial

linux-7.2.3/arch/um/os-Linux/Makefile

@@ -0,0 +1,23 @@
+# SPDX-License-Identifier: GPL-2.0
+# 
+# Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+#
+
+# Don't instrument UML-specific code
+KCOV_INSTRUMENT                := n
+
+obj-y = elf_aux.o execvp.o file.o helper.o irq.o main.o mem.o process.o \
+	registers.o sigio.o signal.o start_up.o time.o tty.o \
+	umid.o user_syms.o util.o skas/
+
+CFLAGS_signal.o += -Wframe-larger-than=4096
+
+CFLAGS_main.o += -Wno-frame-larger-than
+
+obj-$(CONFIG_SMP) += smp.o
+
+USER_OBJS := $(user-objs-y) elf_aux.o execvp.o file.o helper.o irq.o \
+	main.o mem.o process.o registers.o sigio.o signal.o start_up.o time.o \
+	tty.o umid.o util.o smp.o
+
+include $(srctree)/arch/um/scripts/Makefile.rules

linux-7.2.3/arch/um/os-Linux/elf_aux.c

@@ -0,0 +1,50 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *  arch/um/kernel/elf_aux.c
+ *
+ *  Scan the ELF auxiliary vector provided by the host to extract
+ *  information about vsyscall-page, etc.
+ *
+ *  Copyright (C) 2004 Fujitsu Siemens Computers GmbH
+ *  Author: Bodo Stroesser (bodo.stroesser@fujitsu-siemens.com)
+ */
+#include <elf.h>
+#include <stddef.h>
+#include <init.h>
+#include <elf_user.h>
+#include <mem_user.h>
+#include "internal.h"
+#include <linux/swab.h>
+
+#if __BITS_PER_LONG == 64
+typedef Elf64_auxv_t elf_auxv_t;
+#else
+typedef Elf32_auxv_t elf_auxv_t;
+#endif
+
+/* These are initialized very early in boot and never changed */
+char * elf_aux_platform;
+long elf_aux_hwcap;
+
+__init void scan_elf_aux( char **envp)
+{
+	elf_auxv_t * auxv;
+
+	while ( *envp++ != NULL) ;
+
+	for ( auxv = (elf_auxv_t *)envp; auxv->a_type != AT_NULL; auxv++) {
+		switch ( auxv->a_type ) {
+			case AT_HWCAP:
+				elf_aux_hwcap = auxv->a_un.a_val;
+				break;
+			case AT_PLATFORM:
+                                /* elf.h removed the pointer elements from
+                                 * a_un, so we have to use a_val, which is
+                                 * all that's left.
+                                 */
+				elf_aux_platform =
+					(char *) (long) auxv->a_un.a_val;
+				break;
+		}
+	}
+}

linux-7.2.3/arch/um/os-Linux/execvp.c

@@ -0,0 +1,150 @@
+/* Copyright (C) 2006 by Paolo Giarrusso - modified from glibc' execvp.c.
+   Original copyright notice follows:
+
+   Copyright (C) 1991,92,1995-99,2002,2004 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Lesser General Public
+   License as published by the Free Software Foundation; either
+   version 2.1 of the License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public
+   License along with the GNU C Library; if not, write to the Free
+   Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+   02111-1307 USA.  */
+#include <unistd.h>
+
+#include <stdbool.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <limits.h>
+
+#ifndef TEST
+#include <um_malloc.h>
+#else
+#include <stdio.h>
+#define um_kmalloc malloc
+#endif
+#include <os.h>
+
+/* Execute FILE, searching in the `PATH' environment variable if it contains
+   no slashes, with arguments ARGV and environment from `environ'.  */
+int execvp_noalloc(char *buf, const char *file, char *const argv[])
+{
+	if (*file == '\0') {
+		return -ENOENT;
+	}
+
+	if (strchr (file, '/') != NULL) {
+		/* Don't search when it contains a slash.  */
+		execv(file, argv);
+	} else {
+		int got_eacces;
+		size_t len, pathlen;
+		char *name, *p;
+		char *path = getenv("PATH");
+		if (path == NULL)
+			path = ":/bin:/usr/bin";
+
+		len = strlen(file) + 1;
+		pathlen = strlen(path);
+		/* Copy the file name at the top.  */
+		name = memcpy(buf + pathlen + 1, file, len);
+		/* And add the slash.  */
+		*--name = '/';
+
+		got_eacces = 0;
+		p = path;
+		do {
+			char *startp;
+
+			path = p;
+			//Let's avoid this GNU extension.
+			//p = strchrnul (path, ':');
+			p = strchr(path, ':');
+			if (!p)
+				p = strchr(path, '\0');
+
+			if (p == path)
+				/* Two adjacent colons, or a colon at the beginning or the end
+				   of `PATH' means to search the current directory.  */
+				startp = name + 1;
+			else
+				startp = memcpy(name - (p - path), path, p - path);
+
+			/* Try to execute this name.  If it works, execv will not return.  */
+			execv(startp, argv);
+
+			/*
+			if (errno == ENOEXEC) {
+			}
+			*/
+
+			switch (errno) {
+				case EACCES:
+					/* Record the we got a `Permission denied' error.  If we end
+					   up finding no executable we can use, we want to diagnose
+					   that we did find one but were denied access.  */
+					got_eacces = 1;
+					break;
+				case ENOENT:
+				case ESTALE:
+				case ENOTDIR:
+					/* Those errors indicate the file is missing or not executable
+					   by us, in which case we want to just try the next path
+					   directory.  */
+				case ENODEV:
+				case ETIMEDOUT:
+					/* Some strange filesystems like AFS return even
+					   stranger error numbers.  They cannot reasonably mean
+					   anything else so ignore those, too.  */
+				case ENOEXEC:
+					/* We won't go searching for the shell
+					 * if it is not executable - the Linux
+					 * kernel already handles this enough,
+					 * for us. */
+					break;
+
+				default:
+					/* Some other error means we found an executable file, but
+					   something went wrong executing it; return the error to our
+					   caller.  */
+					return -errno;
+			}
+		} while (*p++ != '\0');
+
+		/* We tried every element and none of them worked.  */
+		if (got_eacces)
+			/* At least one failure was due to permissions, so report that
+			   error.  */
+			return -EACCES;
+	}
+
+	/* Return the error from the last attempt (probably ENOENT).  */
+	return -errno;
+}
+#ifdef TEST
+int main(int argc, char**argv)
+{
+	char buf[PATH_MAX];
+	int ret;
+	argc--;
+	if (!argc) {
+		os_warn("Not enough arguments\n");
+		return 1;
+	}
+	argv++;
+	if (ret = execvp_noalloc(buf, argv[0], argv)) {
+		errno = -ret;
+		perror("execvp_noalloc");
+	}
+	return 0;
+}
+#endif

linux-7.2.3/arch/um/os-Linux/file.c

@@ -0,0 +1,726 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <linux/falloc.h>
+#include <sys/ioctl.h>
+#include <sys/mount.h>
+#include <sys/socket.h>
+#include <sys/stat.h>
+#include <sys/sysmacros.h>
+#include <sys/un.h>
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/eventfd.h>
+#include <poll.h>
+#include <os.h>
+
+static void copy_stat(struct uml_stat *dst, const struct stat64 *src)
+{
+	*dst = ((struct uml_stat) {
+		.ust_dev     = src->st_dev,     /* device */
+		.ust_ino     = src->st_ino,     /* inode */
+		.ust_mode    = src->st_mode,    /* protection */
+		.ust_nlink   = src->st_nlink,   /* number of hard links */
+		.ust_uid     = src->st_uid,     /* user ID of owner */
+		.ust_gid     = src->st_gid,     /* group ID of owner */
+		.ust_size    = src->st_size,    /* total size, in bytes */
+		.ust_blksize = src->st_blksize, /* blocksize for filesys I/O */
+		.ust_blocks  = src->st_blocks,  /* number of blocks allocated */
+		.ust_atime   = src->st_atime,   /* time of last access */
+		.ust_mtime   = src->st_mtime,   /* time of last modification */
+		.ust_ctime   = src->st_ctime,   /* time of last change */
+	});
+}
+
+int os_stat_fd(const int fd, struct uml_stat *ubuf)
+{
+	struct stat64 sbuf;
+	int err;
+
+	CATCH_EINTR(err = fstat64(fd, &sbuf));
+	if (err < 0)
+		return -errno;
+
+	if (ubuf != NULL)
+		copy_stat(ubuf, &sbuf);
+	return err;
+}
+
+int os_stat_file(const char *file_name, struct uml_stat *ubuf)
+{
+	struct stat64 sbuf;
+	int err;
+
+	CATCH_EINTR(err = stat64(file_name, &sbuf));
+	if (err < 0)
+		return -errno;
+
+	if (ubuf != NULL)
+		copy_stat(ubuf, &sbuf);
+	return err;
+}
+
+int os_access(const char *file, int mode)
+{
+	int amode, err;
+
+	amode = (mode & OS_ACC_R_OK ? R_OK : 0) |
+		(mode & OS_ACC_W_OK ? W_OK : 0) |
+		(mode & OS_ACC_X_OK ? X_OK : 0) |
+		(mode & OS_ACC_F_OK ? F_OK : 0);
+
+	err = access(file, amode);
+	if (err < 0)
+		return -errno;
+
+	return 0;
+}
+
+/* FIXME? required only by hostaudio (because it passes ioctls verbatim) */
+int os_ioctl_generic(int fd, unsigned int cmd, unsigned long arg)
+{
+	int err;
+
+	err = ioctl(fd, cmd, arg);
+	if (err < 0)
+		return -errno;
+
+	return err;
+}
+
+/* FIXME: ensure namebuf in os_get_if_name is big enough */
+int os_get_ifname(int fd, char* namebuf)
+{
+	if (ioctl(fd, SIOCGIFNAME, namebuf) < 0)
+		return -errno;
+
+	return 0;
+}
+
+int os_mode_fd(int fd, int mode)
+{
+	int err;
+
+	CATCH_EINTR(err = fchmod(fd, mode));
+	if (err < 0)
+		return -errno;
+
+	return 0;
+}
+
+int os_file_type(char *file)
+{
+	struct uml_stat buf;
+	int err;
+
+	err = os_stat_file(file, &buf);
+	if (err < 0)
+		return err;
+
+	if (S_ISDIR(buf.ust_mode))
+		return OS_TYPE_DIR;
+	else if (S_ISLNK(buf.ust_mode))
+		return OS_TYPE_SYMLINK;
+	else if (S_ISCHR(buf.ust_mode))
+		return OS_TYPE_CHARDEV;
+	else if (S_ISBLK(buf.ust_mode))
+		return OS_TYPE_BLOCKDEV;
+	else if (S_ISFIFO(buf.ust_mode))
+		return OS_TYPE_FIFO;
+	else if (S_ISSOCK(buf.ust_mode))
+		return OS_TYPE_SOCK;
+	else return OS_TYPE_FILE;
+}
+
+int os_file_mode(const char *file, struct openflags *mode_out)
+{
+	int err;
+
+	*mode_out = OPENFLAGS();
+
+	err = access(file, W_OK);
+	if (err && (errno != EACCES))
+		return -errno;
+	else if (!err)
+		*mode_out = of_write(*mode_out);
+
+	err = access(file, R_OK);
+	if (err && (errno != EACCES))
+		return -errno;
+	else if (!err)
+		*mode_out = of_read(*mode_out);
+
+	return err;
+}
+
+int os_open_file(const char *file, struct openflags flags, int mode)
+{
+	int fd, err, f = 0;
+
+	if (flags.r && flags.w)
+		f = O_RDWR;
+	else if (flags.r)
+		f = O_RDONLY;
+	else if (flags.w)
+		f = O_WRONLY;
+	else f = 0;
+
+	if (flags.s)
+		f |= O_SYNC;
+	if (flags.c)
+		f |= O_CREAT;
+	if (flags.t)
+		f |= O_TRUNC;
+	if (flags.e)
+		f |= O_EXCL;
+	if (flags.a)
+		f |= O_APPEND;
+
+	fd = open64(file, f, mode);
+	if (fd < 0)
+		return -errno;
+
+	if (flags.cl && fcntl(fd, F_SETFD, 1)) {
+		err = -errno;
+		close(fd);
+		return err;
+	}
+
+	return fd;
+}
+
+int os_connect_socket(const char *name)
+{
+	struct sockaddr_un sock;
+	int fd, err;
+
+	sock.sun_family = AF_UNIX;
+	snprintf(sock.sun_path, sizeof(sock.sun_path), "%s", name);
+
+	fd = socket(AF_UNIX, SOCK_STREAM, 0);
+	if (fd < 0) {
+		err = -errno;
+		goto out;
+	}
+
+	err = connect(fd, (struct sockaddr *) &sock, sizeof(sock));
+	if (err) {
+		err = -errno;
+		goto out_close;
+	}
+
+	return fd;
+
+out_close:
+	close(fd);
+out:
+	return err;
+}
+
+int os_dup_file(int fd)
+{
+	int new_fd = dup(fd);
+
+	if (new_fd < 0)
+		return -errno;
+
+	return new_fd;
+}
+
+void os_close_file(int fd)
+{
+	close(fd);
+}
+
+int os_seek_file(int fd, unsigned long long offset)
+{
+	unsigned long long actual;
+
+	actual = lseek64(fd, offset, SEEK_SET);
+	if (actual != offset)
+		return -errno;
+	return 0;
+}
+
+int os_read_file(int fd, void *buf, int len)
+{
+	int n = read(fd, buf, len);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_pread_file(int fd, void *buf, int len, unsigned long long offset)
+{
+	int n = pread(fd, buf, len, offset);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_write_file(int fd, const void *buf, int len)
+{
+	int n = write(fd, (void *) buf, len);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_sync_file(int fd)
+{
+	int n = fdatasync(fd);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_pwrite_file(int fd, const void *buf, int len, unsigned long long offset)
+{
+	int n = pwrite(fd, (void *) buf, len, offset);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+
+int os_file_size(const char *file, unsigned long long *size_out)
+{
+	struct uml_stat buf;
+	int err;
+
+	err = os_stat_file(file, &buf);
+	if (err < 0) {
+		printk(UM_KERN_ERR "Couldn't stat \"%s\" : err = %d\n", file,
+		       -err);
+		return err;
+	}
+
+	if (S_ISBLK(buf.ust_mode)) {
+		int fd;
+		long blocks;
+
+		fd = open(file, O_RDONLY, 0);
+		if (fd < 0) {
+			err = -errno;
+			printk(UM_KERN_ERR "Couldn't open \"%s\", "
+			       "errno = %d\n", file, errno);
+			return err;
+		}
+		if (ioctl(fd, BLKGETSIZE, &blocks) < 0) {
+			err = -errno;
+			printk(UM_KERN_ERR "Couldn't get the block size of "
+			       "\"%s\", errno = %d\n", file, errno);
+			close(fd);
+			return err;
+		}
+		*size_out = ((long long) blocks) * 512;
+		close(fd);
+	}
+	else *size_out = buf.ust_size;
+
+	return 0;
+}
+
+int os_file_modtime(const char *file, long long *modtime)
+{
+	struct uml_stat buf;
+	int err;
+
+	err = os_stat_file(file, &buf);
+	if (err < 0) {
+		printk(UM_KERN_ERR "Couldn't stat \"%s\" : err = %d\n", file,
+		       -err);
+		return err;
+	}
+
+	*modtime = buf.ust_mtime;
+	return 0;
+}
+
+int os_set_exec_close(int fd)
+{
+	int err;
+
+	CATCH_EINTR(err = fcntl(fd, F_SETFD, FD_CLOEXEC));
+
+	if (err < 0)
+		return -errno;
+	return err;
+}
+
+int os_pipe(int *fds, int stream, int close_on_exec)
+{
+	int err, type = stream ? SOCK_STREAM : SOCK_DGRAM;
+
+	err = socketpair(AF_UNIX, type, 0, fds);
+	if (err < 0)
+		return -errno;
+
+	if (!close_on_exec)
+		return 0;
+
+	err = os_set_exec_close(fds[0]);
+	if (err < 0)
+		goto error;
+
+	err = os_set_exec_close(fds[1]);
+	if (err < 0)
+		goto error;
+
+	return 0;
+
+ error:
+	printk(UM_KERN_ERR "os_pipe : Setting FD_CLOEXEC failed, err = %d\n",
+	       -err);
+	close(fds[1]);
+	close(fds[0]);
+	return err;
+}
+
+int os_set_fd_async(int fd)
+{
+	int err, flags;
+
+	flags = fcntl(fd, F_GETFL);
+	if (flags < 0)
+		return -errno;
+
+	flags |= O_ASYNC | O_NONBLOCK;
+	if (fcntl(fd, F_SETFL, flags) < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "os_set_fd_async : failed to set O_ASYNC "
+		       "and O_NONBLOCK on fd # %d, errno = %d\n", fd, errno);
+		return err;
+	}
+
+	if ((fcntl(fd, F_SETSIG, SIGIO) < 0) ||
+	    (fcntl(fd, F_SETOWN, os_getpid()) < 0)) {
+		err = -errno;
+		printk(UM_KERN_ERR "os_set_fd_async : Failed to fcntl F_SETOWN "
+		       "(or F_SETSIG) fd %d, errno = %d\n", fd, errno);
+		return err;
+	}
+
+	return 0;
+}
+
+int os_clear_fd_async(int fd)
+{
+	int flags;
+
+	flags = fcntl(fd, F_GETFL);
+	if (flags < 0)
+		return -errno;
+
+	flags &= ~(O_ASYNC | O_NONBLOCK);
+	if (fcntl(fd, F_SETFL, flags) < 0)
+		return -errno;
+	return 0;
+}
+
+int os_set_fd_block(int fd, int blocking)
+{
+	int flags;
+
+	flags = fcntl(fd, F_GETFL);
+	if (flags < 0)
+		return -errno;
+
+	if (blocking)
+		flags &= ~O_NONBLOCK;
+	else
+		flags |= O_NONBLOCK;
+
+	if (fcntl(fd, F_SETFL, flags) < 0)
+		return -errno;
+
+	return 0;
+}
+
+int os_accept_connection(int fd)
+{
+	int new;
+
+	new = accept(fd, NULL, 0);
+	if (new < 0)
+		return -errno;
+	return new;
+}
+
+#ifndef SHUT_RD
+#define SHUT_RD 0
+#endif
+
+#ifndef SHUT_WR
+#define SHUT_WR 1
+#endif
+
+#ifndef SHUT_RDWR
+#define SHUT_RDWR 2
+#endif
+
+int os_shutdown_socket(int fd, int r, int w)
+{
+	int what, err;
+
+	if (r && w)
+		what = SHUT_RDWR;
+	else if (r)
+		what = SHUT_RD;
+	else if (w)
+		what = SHUT_WR;
+	else
+		return -EINVAL;
+
+	err = shutdown(fd, what);
+	if (err < 0)
+		return -errno;
+	return 0;
+}
+
+/**
+ * os_rcv_fd_msg - receive message with (optional) FDs
+ * @fd: the FD to receive from
+ * @fds: the array for FDs to write to
+ * @n_fds: number of FDs to receive (@fds array size)
+ * @data: the message buffer
+ * @data_len: the size of the message to receive
+ *
+ * Receive a message with FDs.
+ *
+ * Returns: the size of the received message, or an error code
+ */
+ssize_t os_rcv_fd_msg(int fd, int *fds, unsigned int n_fds,
+		      void *data, size_t data_len)
+{
+#define MAX_RCV_FDS	2
+	char buf[CMSG_SPACE(sizeof(*fds) * MAX_RCV_FDS)];
+	struct cmsghdr *cmsg;
+	struct iovec iov = {
+		.iov_base = data,
+		.iov_len = data_len,
+	};
+	struct msghdr msg = {
+		.msg_iov = &iov,
+		.msg_iovlen = 1,
+		.msg_control = buf,
+		.msg_controllen = CMSG_SPACE(sizeof(*fds) * n_fds),
+	};
+	int n;
+
+	if (n_fds > MAX_RCV_FDS)
+		return -EINVAL;
+
+	n = recvmsg(fd, &msg, 0);
+	if (n < 0)
+		return -errno;
+
+	cmsg = CMSG_FIRSTHDR(&msg);
+	if (!cmsg ||
+	    cmsg->cmsg_level != SOL_SOCKET ||
+	    cmsg->cmsg_type != SCM_RIGHTS)
+		return n;
+
+	memcpy(fds, CMSG_DATA(cmsg), cmsg->cmsg_len - CMSG_LEN(0));
+	return n;
+}
+
+int os_create_unix_socket(const char *file, int len, int close_on_exec)
+{
+	struct sockaddr_un addr;
+	int sock, err;
+
+	sock = socket(PF_UNIX, SOCK_DGRAM, 0);
+	if (sock < 0)
+		return -errno;
+
+	if (close_on_exec) {
+		err = os_set_exec_close(sock);
+		if (err < 0)
+			printk(UM_KERN_ERR "create_unix_socket : "
+			       "close_on_exec failed, err = %d", -err);
+	}
+
+	addr.sun_family = AF_UNIX;
+
+	snprintf(addr.sun_path, len, "%s", file);
+
+	err = bind(sock, (struct sockaddr *) &addr, sizeof(addr));
+	if (err < 0)
+		return -errno;
+
+	return sock;
+}
+
+void os_flush_stdout(void)
+{
+	fflush(stdout);
+}
+
+int os_lock_file(int fd, int excl)
+{
+	int type = excl ? F_WRLCK : F_RDLCK;
+	struct flock lock = ((struct flock) { .l_type	= type,
+					      .l_whence	= SEEK_SET,
+					      .l_start	= 0,
+					      .l_len	= 0 } );
+	int err, save;
+
+	err = fcntl(fd, F_SETLK, &lock);
+	if (!err)
+		goto out;
+
+	save = -errno;
+	err = fcntl(fd, F_GETLK, &lock);
+	if (err) {
+		err = -errno;
+		goto out;
+	}
+
+	printk(UM_KERN_ERR "F_SETLK failed, file already locked by pid %d\n",
+	       lock.l_pid);
+	err = save;
+ out:
+	return err;
+}
+
+unsigned os_major(unsigned long long dev)
+{
+	return major(dev);
+}
+
+unsigned os_minor(unsigned long long dev)
+{
+	return minor(dev);
+}
+
+unsigned long long os_makedev(unsigned major, unsigned minor)
+{
+	return makedev(major, minor);
+}
+
+int os_falloc_punch(int fd, unsigned long long offset, int len)
+{
+	int n = fallocate(fd, FALLOC_FL_PUNCH_HOLE|FALLOC_FL_KEEP_SIZE, offset, len);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_falloc_zeroes(int fd, unsigned long long offset, int len)
+{
+	int n = fallocate(fd, FALLOC_FL_ZERO_RANGE|FALLOC_FL_KEEP_SIZE, offset, len);
+
+	if (n < 0)
+		return -errno;
+	return n;
+}
+
+int os_eventfd(unsigned int initval, int flags)
+{
+	int fd = eventfd(initval, flags);
+
+	if (fd < 0)
+		return -errno;
+	return fd;
+}
+
+int os_sendmsg_fds(int fd, const void *buf, unsigned int len, const int *fds,
+		   unsigned int fds_num)
+{
+	struct iovec iov = {
+		.iov_base = (void *) buf,
+		.iov_len = len,
+	};
+	union {
+		char control[CMSG_SPACE(sizeof(*fds) * OS_SENDMSG_MAX_FDS)];
+		struct cmsghdr align;
+	} u;
+	unsigned int fds_size = sizeof(*fds) * fds_num;
+	struct msghdr msg = {
+		.msg_iov = &iov,
+		.msg_iovlen = 1,
+		.msg_control = u.control,
+		.msg_controllen = CMSG_SPACE(fds_size),
+	};
+	struct cmsghdr *cmsg = CMSG_FIRSTHDR(&msg);
+	int err;
+
+	if (fds_num > OS_SENDMSG_MAX_FDS)
+		return -EINVAL;
+	memset(u.control, 0, sizeof(u.control));
+	cmsg->cmsg_level = SOL_SOCKET;
+	cmsg->cmsg_type = SCM_RIGHTS;
+	cmsg->cmsg_len = CMSG_LEN(fds_size);
+	memcpy(CMSG_DATA(cmsg), fds, fds_size);
+	err = sendmsg(fd, &msg, 0);
+
+	if (err < 0)
+		return -errno;
+	return err;
+}
+
+int os_poll(unsigned int n, const int *fds)
+{
+	/* currently need 2 FDs at most so avoid dynamic allocation */
+	struct pollfd pollfds[2] = {};
+	unsigned int i;
+	int ret;
+
+	if (n > ARRAY_SIZE(pollfds))
+		return -EINVAL;
+
+	for (i = 0; i < n; i++) {
+		pollfds[i].fd = fds[i];
+		pollfds[i].events = POLLIN;
+	}
+
+	ret = poll(pollfds, n, -1);
+	if (ret < 0)
+		return -errno;
+
+	/* Return the index of the available FD */
+	for (i = 0; i < n; i++) {
+		if (pollfds[i].revents)
+			return i;
+	}
+
+	return -EIO;
+}
+
+void *os_mmap_rw_shared(int fd, size_t size)
+{
+	void *res = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
+
+	if (res == MAP_FAILED)
+		return NULL;
+
+	return res;
+}
+
+void *os_mremap_rw_shared(void *old_addr, size_t old_size, size_t new_size)
+{
+	void *res;
+
+	res = mremap(old_addr, old_size, new_size, MREMAP_MAYMOVE, NULL);
+
+	if (res == MAP_FAILED)
+		return NULL;
+
+	return res;
+}

linux-7.2.3/arch/um/os-Linux/helper.c

@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <errno.h>
+#include <sched.h>
+#include <pthread.h>
+#include <linux/limits.h>
+#include <sys/socket.h>
+#include <sys/wait.h>
+#include <kern_util.h>
+#include <os.h>
+#include <um_malloc.h>
+
+struct helper_data {
+	void (*pre_exec)(void*);
+	void *pre_data;
+	char **argv;
+	int fd;
+	char *buf;
+};
+
+static int helper_child(void *arg)
+{
+	struct helper_data *data = arg;
+	char **argv = data->argv;
+	int err, ret;
+
+	if (data->pre_exec != NULL)
+		(*data->pre_exec)(data->pre_data);
+	err = execvp_noalloc(data->buf, argv[0], argv);
+
+	/* If the exec succeeds, we don't get here */
+	CATCH_EINTR(ret = write(data->fd, &err, sizeof(err)));
+
+	return 0;
+}
+
+/* Returns either the pid of the child process we run or -E* on failure. */
+int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv)
+{
+	struct helper_data data;
+	unsigned long stack, sp;
+	int pid, fds[2], ret, n;
+
+	stack = alloc_stack(0, __uml_cant_sleep());
+	if (stack == 0)
+		return -ENOMEM;
+
+	ret = socketpair(AF_UNIX, SOCK_STREAM, 0, fds);
+	if (ret < 0) {
+		ret = -errno;
+		printk(UM_KERN_ERR "run_helper : pipe failed, errno = %d\n",
+		       errno);
+		goto out_free;
+	}
+
+	ret = os_set_exec_close(fds[1]);
+	if (ret < 0) {
+		printk(UM_KERN_ERR "run_helper : setting FD_CLOEXEC failed, "
+		       "ret = %d\n", -ret);
+		goto out_close;
+	}
+
+	sp = stack + UM_KERN_PAGE_SIZE;
+	data.pre_exec = pre_exec;
+	data.pre_data = pre_data;
+	data.argv = argv;
+	data.fd = fds[1];
+	data.buf = __uml_cant_sleep() ? uml_kmalloc(PATH_MAX, UM_GFP_ATOMIC) :
+					uml_kmalloc(PATH_MAX, UM_GFP_KERNEL);
+	pid = clone(helper_child, (void *) sp, CLONE_VM, &data);
+	if (pid < 0) {
+		ret = -errno;
+		printk(UM_KERN_ERR "run_helper : clone failed, errno = %d\n",
+		       errno);
+		goto out_free2;
+	}
+
+	close(fds[1]);
+	fds[1] = -1;
+
+	/*
+	 * Read the errno value from the child, if the exec failed, or get 0 if
+	 * the exec succeeded because the pipe fd was set as close-on-exec.
+	 */
+	n = read(fds[0], &ret, sizeof(ret));
+	if (n == 0) {
+		ret = pid;
+	} else {
+		if (n < 0) {
+			n = -errno;
+			printk(UM_KERN_ERR "run_helper : read on pipe failed, "
+			       "ret = %d\n", -n);
+			ret = n;
+		}
+		CATCH_EINTR(waitpid(pid, NULL, __WALL));
+	}
+
+	if (ret < 0)
+		printk(UM_KERN_ERR "run_helper : failed to exec %s on host: %s\n",
+		       argv[0], strerror(-ret));
+
+out_free2:
+	kfree(data.buf);
+out_close:
+	if (fds[1] != -1)
+		close(fds[1]);
+	close(fds[0]);
+out_free:
+	free_stack(stack, 0);
+	return ret;
+}
+
+int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags,
+		      unsigned long *stack_out)
+{
+	unsigned long stack, sp;
+	int pid, status, err;
+
+	/* To share memory space, use os_run_helper_thread() instead. */
+	if (flags & CLONE_VM)
+		return -EINVAL;
+
+	stack = alloc_stack(0, __uml_cant_sleep());
+	if (stack == 0)
+		return -ENOMEM;
+
+	sp = stack + UM_KERN_PAGE_SIZE;
+	pid = clone(proc, (void *) sp, flags, arg);
+	if (pid < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "run_helper_thread : clone failed, "
+		       "errno = %d\n", errno);
+		return err;
+	}
+	if (stack_out == NULL) {
+		CATCH_EINTR(pid = waitpid(pid, &status, __WALL));
+		if (pid < 0) {
+			err = -errno;
+			printk(UM_KERN_ERR "run_helper_thread - wait failed, "
+			       "errno = %d\n", errno);
+			pid = err;
+		}
+		if (!WIFEXITED(status) || (WEXITSTATUS(status) != 0))
+			printk(UM_KERN_ERR "run_helper_thread - thread "
+			       "returned status 0x%x\n", status);
+		free_stack(stack, 0);
+	} else
+		*stack_out = stack;
+	return pid;
+}
+
+int helper_wait(int pid)
+{
+	int ret, status;
+	int wflags = __WALL;
+
+	CATCH_EINTR(ret = waitpid(pid, &status, wflags));
+	if (ret < 0) {
+		printk(UM_KERN_ERR "helper_wait : waitpid process %d failed, "
+		       "errno = %d\n", pid, errno);
+		return -errno;
+	} else if (!WIFEXITED(status) || WEXITSTATUS(status) != 0) {
+		printk(UM_KERN_ERR "helper_wait : process %d exited with "
+		       "status 0x%x\n", pid, status);
+		return -ECHILD;
+	} else
+		return 0;
+}
+
+struct os_helper_thread {
+	pthread_t handle;
+};
+
+int os_run_helper_thread(struct os_helper_thread **td_out,
+			 void *(*routine)(void *), void *arg)
+{
+	struct os_helper_thread *td;
+	sigset_t sigset, oset;
+	int err, flags;
+
+	flags = __uml_cant_sleep() ? UM_GFP_ATOMIC : UM_GFP_KERNEL;
+	td = uml_kmalloc(sizeof(*td), flags);
+	if (!td)
+		return -ENOMEM;
+
+	sigfillset(&sigset);
+	if (sigprocmask(SIG_SETMASK, &sigset, &oset) < 0) {
+		err = -errno;
+		kfree(td);
+		return err;
+	}
+
+	err = pthread_create(&td->handle, NULL, routine, arg);
+
+	if (sigprocmask(SIG_SETMASK, &oset, NULL) < 0)
+		panic("Failed to restore the signal mask: %d", errno);
+
+	if (err != 0)
+		kfree(td);
+	else
+		*td_out = td;
+
+	return -err;
+}
+
+void os_kill_helper_thread(struct os_helper_thread *td)
+{
+	pthread_cancel(td->handle);
+	pthread_join(td->handle, NULL);
+	kfree(td);
+}
+
+void os_fix_helper_thread_signals(void)
+{
+	sigset_t sigset;
+
+	sigemptyset(&sigset);
+
+	sigaddset(&sigset, SIGWINCH);
+	sigaddset(&sigset, SIGPIPE);
+	sigaddset(&sigset, SIGPROF);
+	sigaddset(&sigset, SIGINT);
+	sigaddset(&sigset, SIGTERM);
+	sigaddset(&sigset, SIGCHLD);
+	sigaddset(&sigset, SIGALRM);
+	sigaddset(&sigset, SIGIO);
+	sigaddset(&sigset, SIGUSR1);
+
+	pthread_sigmask(SIG_SETMASK, &sigset, NULL);
+}

linux-7.2.3/arch/um/os-Linux/internal.h

@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __UM_OS_LINUX_INTERNAL_H
+#define __UM_OS_LINUX_INTERNAL_H
+
+#include <mm_id.h>
+#include <stub-data.h>
+#include <signal.h>
+
+/*
+ * elf_aux.c
+ */
+void scan_elf_aux(char **envp);
+
+/*
+ * mem.c
+ */
+void check_tmpexec(void);
+
+/*
+ * signal.c
+ */
+extern __thread int signals_enabled;
+int timer_alarm_pending(void);
+
+/*
+ * skas/process.c
+ */
+void wait_stub_done(int pid);
+void wait_stub_done_seccomp(struct mm_id *mm_idp, int running, int wait_sigsys);
+
+/*
+ * smp.c
+ */
+#define IPI_SIGNAL	SIGRTMIN
+
+#endif /* __UM_OS_LINUX_INTERNAL_H */

linux-7.2.3/arch/um/os-Linux/irq.c

@@ -0,0 +1,145 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2017 - Cambridge Greys Ltd
+ * Copyright (C) 2011 - 2014 Cisco Systems Inc
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdlib.h>
+#include <errno.h>
+#include <sys/epoll.h>
+#include <signal.h>
+#include <string.h>
+#include <irq_user.h>
+#include <os.h>
+#include <um_malloc.h>
+
+/* Epoll support */
+
+static int epollfd = -1;
+
+#define MAX_EPOLL_EVENTS 64
+
+static struct epoll_event epoll_events[MAX_EPOLL_EVENTS];
+
+/* Helper to return an Epoll data pointer from an epoll event structure.
+ * We need to keep this one on the userspace side to keep includes separate
+ */
+
+void *os_epoll_get_data_pointer(int index)
+{
+	return epoll_events[index].data.ptr;
+}
+
+/* Helper to compare events versus the events in the epoll structure.
+ * Same as above - needs to be on the userspace side
+ */
+
+
+int os_epoll_triggered(int index, int events)
+{
+	return epoll_events[index].events & events;
+}
+/* Helper to set the event mask.
+ * The event mask is opaque to the kernel side, because it does not have
+ * access to the right includes/defines for EPOLL constants.
+ */
+
+int os_event_mask(enum um_irq_type irq_type)
+{
+	if (irq_type == IRQ_READ)
+		return EPOLLIN | EPOLLPRI | EPOLLERR | EPOLLHUP | EPOLLRDHUP;
+	if (irq_type == IRQ_WRITE)
+		return EPOLLOUT;
+	return 0;
+}
+
+/*
+ * Initial Epoll Setup
+ */
+int os_setup_epoll(void)
+{
+	epollfd = epoll_create(MAX_EPOLL_EVENTS);
+	return epollfd;
+}
+
+/*
+ * Helper to run the actual epoll_wait
+ */
+int os_waiting_for_events_epoll(void)
+{
+	int n, err;
+
+	n = epoll_wait(epollfd,
+		(struct epoll_event *) &epoll_events, MAX_EPOLL_EVENTS, 0);
+	if (n < 0) {
+		err = -errno;
+		if (errno != EINTR)
+			printk(
+				UM_KERN_ERR "os_waiting_for_events:"
+				" epoll returned %d, error = %s\n", n,
+				strerror(errno)
+			);
+		return err;
+	}
+	return n;
+}
+
+
+/*
+ * Helper to add a fd to epoll
+ */
+int os_add_epoll_fd(int events, int fd, void *data)
+{
+	struct epoll_event event;
+	int result;
+
+	event.data.ptr = data;
+	event.events = events | EPOLLET;
+	result = epoll_ctl(epollfd, EPOLL_CTL_ADD, fd, &event);
+	if ((result) && (errno == EEXIST))
+		result = os_mod_epoll_fd(events, fd, data);
+	if (result)
+		printk("epollctl add err fd %d, %s\n", fd, strerror(errno));
+	return result;
+}
+
+/*
+ * Helper to mod the fd event mask and/or data backreference
+ */
+int os_mod_epoll_fd(int events, int fd, void *data)
+{
+	struct epoll_event event;
+	int result;
+
+	event.data.ptr = data;
+	event.events = events;
+	result = epoll_ctl(epollfd, EPOLL_CTL_MOD, fd, &event);
+	if (result)
+		printk(UM_KERN_ERR
+			"epollctl mod err fd %d, %s\n", fd, strerror(errno));
+	return result;
+}
+
+/*
+ * Helper to delete the epoll fd
+ */
+int os_del_epoll_fd(int fd)
+{
+	struct epoll_event event;
+	/* This is quiet as we use this as IO ON/OFF - so it is often
+	 * invoked on a non-existent fd
+	 */
+	return epoll_ctl(epollfd, EPOLL_CTL_DEL, fd, &event);
+}
+
+void os_set_ioignore(void)
+{
+	signal(SIGIO, SIG_IGN);
+}
+
+void os_close_epoll_fd(void)
+{
+	/* Needed so we do not leak an fd when rebooting */
+	os_close_file(epollfd);
+}

linux-7.2.3/arch/um/os-Linux/main.c

@@ -0,0 +1,265 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015 Thomas Meyer (thomas@m3y3r.de)
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <signal.h>
+#include <string.h>
+#include <sys/resource.h>
+#include <sys/personality.h>
+#include <as-layout.h>
+#include <init.h>
+#include <kern_util.h>
+#include <os.h>
+#include <um_malloc.h>
+#include "internal.h"
+
+#define STACKSIZE (8 * 1024 * 1024)
+
+static void __init set_stklim(void)
+{
+	struct rlimit lim;
+
+	if (getrlimit(RLIMIT_STACK, &lim) < 0) {
+		perror("getrlimit");
+		exit(1);
+	}
+	if ((lim.rlim_cur == RLIM_INFINITY) || (lim.rlim_cur > STACKSIZE)) {
+		lim.rlim_cur = STACKSIZE;
+		if (setrlimit(RLIMIT_STACK, &lim) < 0) {
+			perror("setrlimit");
+			exit(1);
+		}
+	}
+}
+
+static void last_ditch_exit(int sig)
+{
+	uml_cleanup();
+	exit(1);
+}
+
+static void __init install_fatal_handler(int sig)
+{
+	struct sigaction action;
+
+	/* All signals are enabled in this handler ... */
+	sigemptyset(&action.sa_mask);
+
+	/*
+	 * ... including the signal being handled, plus we want the
+	 * handler reset to the default behavior, so that if an exit
+	 * handler is hanging for some reason, the UML will just die
+	 * after this signal is sent a second time.
+	 */
+	action.sa_flags = SA_RESETHAND | SA_NODEFER;
+	action.sa_restorer = NULL;
+	action.sa_handler = last_ditch_exit;
+	if (sigaction(sig, &action, NULL) < 0) {
+		os_warn("failed to install handler for signal %d "
+			"- errno = %d\n", sig, errno);
+		exit(1);
+	}
+}
+
+#define UML_LIB_PATH	":" OS_LIB_PATH "/uml"
+
+static void __init setup_env_path(void)
+{
+	char *new_path = NULL;
+	char *old_path = NULL;
+	int path_len = 0;
+
+	old_path = getenv("PATH");
+	/*
+	 * if no PATH variable is set or it has an empty value
+	 * just use the default + /usr/lib/uml
+	 */
+	if (!old_path || (path_len = strlen(old_path)) == 0) {
+		if (putenv("PATH=:/bin:/usr/bin/" UML_LIB_PATH))
+			perror("couldn't putenv");
+		return;
+	}
+
+	/* append /usr/lib/uml to the existing path */
+	path_len += strlen("PATH=" UML_LIB_PATH) + 1;
+	new_path = malloc(path_len);
+	if (!new_path) {
+		perror("couldn't malloc to set a new PATH");
+		return;
+	}
+	snprintf(new_path, path_len, "PATH=%s" UML_LIB_PATH, old_path);
+	if (putenv(new_path)) {
+		perror("couldn't putenv to set a new PATH");
+		free(new_path);
+	}
+}
+
+int __init main(int argc, char **argv, char **envp)
+{
+	char **new_argv;
+	int ret, i, err;
+
+	/* Disable randomization and re-exec if it was changed successfully */
+	ret = personality(PER_LINUX | ADDR_NO_RANDOMIZE);
+	if (ret >= 0 && (ret & (PER_LINUX | ADDR_NO_RANDOMIZE)) !=
+			 (PER_LINUX | ADDR_NO_RANDOMIZE)) {
+		char buf[4096] = {};
+		ssize_t ret;
+
+		ret = readlink("/proc/self/exe", buf, sizeof(buf));
+		if (ret < 0 || ret >= sizeof(buf)) {
+			perror("readlink failure");
+			exit(1);
+		}
+		execve(buf, argv, envp);
+	}
+
+	set_stklim();
+
+	setup_env_path();
+
+	setsid();
+
+	new_argv = malloc((argc + 1) * sizeof(char *));
+	if (new_argv == NULL) {
+		perror("Mallocing argv");
+		exit(1);
+	}
+	for (i = 0; i < argc; i++) {
+		new_argv[i] = strdup(argv[i]);
+		if (new_argv[i] == NULL) {
+			perror("Mallocing an arg");
+			exit(1);
+		}
+	}
+	new_argv[argc] = NULL;
+
+	/*
+	 * Allow these signals to bring down a UML if all other
+	 * methods of control fail.
+	 */
+	install_fatal_handler(SIGINT);
+	install_fatal_handler(SIGTERM);
+
+	scan_elf_aux(envp);
+
+	change_sig(SIGPIPE, 0);
+	ret = linux_main(argc, argv, envp);
+
+	/*
+	 * Disable SIGPROF - I have no idea why libc doesn't do this or turn
+	 * off the profiling time, but UML dies with a SIGPROF just before
+	 * exiting when profiling is active.
+	 */
+	change_sig(SIGPROF, 0);
+
+	/*
+	 * This signal stuff used to be in the reboot case.  However,
+	 * sometimes a timer signal can come in when we're halting (reproducably
+	 * when writing out gcov information, presumably because that takes
+	 * some time) and cause a segfault.
+	 */
+
+	/* stop timers and set timer signal to be ignored */
+	os_timer_disable(0);
+
+	/* disable SIGIO for the fds and set SIGIO to be ignored */
+	err = deactivate_all_fds();
+	if (err)
+		os_warn("deactivate_all_fds failed, errno = %d\n", -err);
+
+	/*
+	 * Let any pending signals fire now.  This ensures
+	 * that they won't be delivered after the exec, when
+	 * they are definitely not expected.
+	 */
+	unblock_signals();
+
+	os_info("\n");
+	/* Reboot */
+	if (ret) {
+		execvp(new_argv[0], new_argv);
+		perror("Failed to exec kernel");
+		ret = 1;
+	}
+	return uml_exitcode;
+}
+
+extern void *__real_malloc(int);
+extern void __real_free(void *);
+
+/* workaround for -Wmissing-prototypes warnings */
+void *__wrap_malloc(int size);
+void *__wrap_calloc(int n, int size);
+void __wrap_free(void *ptr);
+
+void *__wrap_malloc(int size)
+{
+	void *ret;
+
+	if (!kmalloc_ok)
+		return __real_malloc(size);
+	else if (size <= UM_KERN_PAGE_SIZE)
+		/* finding contiguous pages can be hard*/
+		ret = uml_kmalloc(size, UM_GFP_KERNEL);
+	else ret = vmalloc(size);
+
+	/*
+	 * glibc people insist that if malloc fails, errno should be
+	 * set by malloc as well. So we do.
+	 */
+	if (ret == NULL)
+		errno = ENOMEM;
+
+	return ret;
+}
+
+void *__wrap_calloc(int n, int size)
+{
+	void *ptr = __wrap_malloc(n * size);
+
+	if (ptr == NULL)
+		return NULL;
+	memset(ptr, 0, n * size);
+	return ptr;
+}
+
+void __wrap_free(void *ptr)
+{
+	unsigned long addr = (unsigned long) ptr;
+
+	/*
+	 * We need to know how the allocation happened, so it can be correctly
+	 * freed.  This is done by seeing what region of memory the pointer is
+	 * in -
+	 * 	physical memory - kmalloc/kfree
+	 *	kernel virtual memory - vmalloc/vfree
+	 * 	anywhere else - malloc/free
+	 * If kmalloc is not yet possible, then either high_physmem and/or
+	 * end_vm are still 0 (as at startup), in which case we call free, or
+	 * we have set them, but anyway addr has not been allocated from those
+	 * areas. So, in both cases __real_free is called.
+	 *
+	 * CAN_KMALLOC is checked because it would be bad to free a buffer
+	 * with kmalloc/vmalloc after they have been turned off during
+	 * shutdown.
+	 * XXX: However, we sometimes shutdown CAN_KMALLOC temporarily, so
+	 * there is a possibility for memory leaks.
+	 */
+
+	if ((addr >= uml_physmem) && (addr < high_physmem)) {
+		if (kmalloc_ok)
+			kfree(ptr);
+	}
+	else if ((addr >= start_vm) && (addr < end_vm)) {
+		if (kmalloc_ok)
+			vfree(ptr);
+	}
+	else __real_free(ptr);
+}

linux-7.2.3/arch/um/os-Linux/mem.c

@@ -0,0 +1,246 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stddef.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <sys/mman.h>
+#include <sys/vfs.h>
+#include <linux/magic.h>
+#include <init.h>
+#include <kern_util.h>
+#include <os.h>
+#include "internal.h"
+
+/*
+ * kasan_map_memory - maps memory from @start with a size of @len.
+ * The allocated memory is filled with zeroes upon success.
+ * @start: the start address of the memory to be mapped
+ * @len: the length of the memory to be mapped
+ *
+ * This function is used to map shadow memory for KASAN in uml
+ */
+void kasan_map_memory(void *start, size_t len)
+{
+	if (mmap(start,
+		 len,
+		 PROT_READ|PROT_WRITE,
+		 MAP_FIXED|MAP_ANONYMOUS|MAP_PRIVATE|MAP_NORESERVE,
+		 -1,
+		 0) == MAP_FAILED) {
+		os_info("Couldn't allocate shadow memory: %s\n.",
+			strerror(errno));
+		exit(1);
+	}
+
+	if (madvise(start, len, MADV_DONTDUMP)) {
+		os_info("Couldn't set MAD_DONTDUMP on shadow memory: %s\n.",
+			strerror(errno));
+		exit(1);
+	}
+
+	if (madvise(start, len, MADV_DONTFORK)) {
+		os_info("Couldn't set MADV_DONTFORK on shadow memory: %s\n.",
+			strerror(errno));
+		exit(1);
+	}
+}
+
+/* Set by make_tempfile() during early boot. */
+char *tempdir = NULL;
+
+/* Check if dir is on tmpfs. Return 0 if yes, -1 if no or error. */
+static int __init check_tmpfs(const char *dir)
+{
+	struct statfs st;
+
+	os_info("Checking if %s is on tmpfs...", dir);
+	if (statfs(dir, &st) < 0) {
+		os_info("%s\n", strerror(errno));
+	} else if (st.f_type != TMPFS_MAGIC) {
+		os_info("no\n");
+	} else {
+		os_info("OK\n");
+		return 0;
+	}
+	return -1;
+}
+
+/*
+ * Choose the tempdir to use. We want something on tmpfs so that our memory is
+ * not subject to the host's vm.dirty_ratio. If a tempdir is specified in the
+ * environment, we use that even if it's not on tmpfs, but we warn the user.
+ * Otherwise, we try common tmpfs locations, and if no tmpfs directory is found
+ * then we fall back to /tmp.
+ */
+static char * __init choose_tempdir(void)
+{
+	static const char * const vars[] = {
+		"TMPDIR",
+		"TMP",
+		"TEMP",
+		NULL
+	};
+	static const char fallback_dir[] = "/tmp";
+	static const char * const tmpfs_dirs[] = {
+		"/dev/shm",
+		fallback_dir,
+		NULL
+	};
+	int i;
+	const char *dir;
+
+	os_info("Checking environment variables for a tempdir...");
+	for (i = 0; vars[i]; i++) {
+		dir = getenv(vars[i]);
+		if ((dir != NULL) && (*dir != '\0')) {
+			os_info("%s\n", dir);
+			if (check_tmpfs(dir) >= 0)
+				goto done;
+			else
+				goto warn;
+		}
+	}
+	os_info("none found\n");
+
+	for (i = 0; tmpfs_dirs[i]; i++) {
+		dir = tmpfs_dirs[i];
+		if (check_tmpfs(dir) >= 0)
+			goto done;
+	}
+
+	dir = fallback_dir;
+warn:
+	os_warn("Warning: tempdir %s is not on tmpfs\n", dir);
+done:
+	/* Make a copy since getenv results may not remain valid forever. */
+	return strdup(dir);
+}
+
+/*
+ * Create an unlinked tempfile in a suitable tempdir. template must be the
+ * basename part of the template with a leading '/'.
+ */
+static int __init make_tempfile(const char *template)
+{
+	char *tempname;
+	int fd;
+
+	if (tempdir == NULL) {
+		tempdir = choose_tempdir();
+		if (tempdir == NULL) {
+			os_warn("Failed to choose tempdir: %s\n",
+				strerror(errno));
+			return -1;
+		}
+	}
+
+#ifdef O_TMPFILE
+	fd = open(tempdir, O_CLOEXEC | O_RDWR | O_EXCL | O_TMPFILE, 0700);
+	/*
+	 * If the running system does not support O_TMPFILE flag then retry
+	 * without it.
+	 */
+	if (fd != -1 || (errno != EINVAL && errno != EISDIR &&
+			errno != EOPNOTSUPP))
+		return fd;
+#endif
+
+	tempname = malloc(strlen(tempdir) + strlen(template) + 1);
+	if (tempname == NULL)
+		return -1;
+
+	strcpy(tempname, tempdir);
+	strcat(tempname, template);
+	fd = mkstemp(tempname);
+	if (fd < 0) {
+		os_warn("open - cannot create %s: %s\n", tempname,
+			strerror(errno));
+		goto out;
+	}
+	if (unlink(tempname) < 0) {
+		perror("unlink");
+		goto close;
+	}
+	free(tempname);
+	return fd;
+close:
+	close(fd);
+out:
+	free(tempname);
+	return -1;
+}
+
+#define TEMPNAME_TEMPLATE "/vm_file-XXXXXX"
+
+static int __init create_tmp_file(unsigned long long len)
+{
+	int fd, err;
+	char zero;
+
+	fd = make_tempfile(TEMPNAME_TEMPLATE);
+	if (fd < 0)
+		exit(1);
+
+	/*
+	 * Seek to len - 1 because writing a character there will
+	 * increase the file size by one byte, to the desired length.
+	 */
+	if (lseek64(fd, len - 1, SEEK_SET) < 0) {
+		perror("lseek64");
+		exit(1);
+	}
+
+	zero = 0;
+
+	err = write(fd, &zero, 1);
+	if (err != 1) {
+		perror("write");
+		exit(1);
+	}
+
+	return fd;
+}
+
+int __init create_mem_file(unsigned long long len)
+{
+	int err, fd;
+
+	fd = create_tmp_file(len);
+
+	err = os_set_exec_close(fd);
+	if (err < 0) {
+		errno = -err;
+		perror("exec_close");
+	}
+	return fd;
+}
+
+void __init check_tmpexec(void)
+{
+	void *addr;
+	int err, fd = create_tmp_file(UM_KERN_PAGE_SIZE);
+
+	addr = mmap(NULL, UM_KERN_PAGE_SIZE,
+		    PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE, fd, 0);
+	os_info("Checking PROT_EXEC mmap in %s...", tempdir);
+	if (addr == MAP_FAILED) {
+		err = errno;
+		os_warn("%s\n", strerror(err));
+		close(fd);
+		if (err == EPERM)
+			os_warn("%s must be not mounted noexec\n", tempdir);
+		exit(1);
+	}
+	os_info("OK\n");
+	munmap(addr, UM_KERN_PAGE_SIZE);
+
+	close(fd);
+}

linux-7.2.3/arch/um/os-Linux/process.c

@@ -0,0 +1,211 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015 Thomas Meyer (thomas@m3y3r.de)
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <signal.h>
+#include <fcntl.h>
+#include <limits.h>
+#include <linux/futex.h>
+#include <sys/mman.h>
+#include <sys/ptrace.h>
+#include <sys/prctl.h>
+#include <sys/wait.h>
+#include <asm/unistd.h>
+#include <init.h>
+#include <longjmp.h>
+#include <os.h>
+#include <skas/skas.h>
+
+void os_alarm_process(int pid)
+{
+	if (pid <= 0)
+		return;
+
+	kill(pid, SIGALRM);
+}
+
+void os_kill_process(int pid, int reap_child)
+{
+	if (pid <= 0)
+		return;
+
+	/* Block signals until child is reaped */
+	block_signals();
+
+	kill(pid, SIGKILL);
+	if (reap_child)
+		CATCH_EINTR(waitpid(pid, NULL, __WALL));
+
+	unblock_signals();
+}
+
+/* Kill off a ptraced child by all means available.  kill it normally first,
+ * then PTRACE_KILL it, then PTRACE_CONT it in case it's in a run state from
+ * which it can't exit directly.
+ */
+
+void os_kill_ptraced_process(int pid, int reap_child)
+{
+	if (pid <= 0)
+		return;
+
+	/* Block signals until child is reaped */
+	block_signals();
+
+	kill(pid, SIGKILL);
+	ptrace(PTRACE_KILL, pid);
+	ptrace(PTRACE_CONT, pid);
+	if (reap_child)
+		CATCH_EINTR(waitpid(pid, NULL, __WALL));
+
+	unblock_signals();
+}
+
+pid_t os_reap_child(void)
+{
+	int status;
+
+	/* Try to reap a child */
+	return waitpid(-1, &status, WNOHANG);
+}
+
+/* Don't use the glibc version, which caches the result in TLS. It misses some
+ * syscalls, and also breaks with clone(), which does not unshare the TLS.
+ */
+
+int os_getpid(void)
+{
+	return syscall(__NR_getpid);
+}
+
+int os_map_memory(void *virt, int fd, unsigned long long off, unsigned long len,
+		  int r, int w, int x)
+{
+	void *loc;
+	int prot;
+
+	prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) |
+		(x ? PROT_EXEC : 0);
+
+	loc = mmap64((void *) virt, len, prot, MAP_SHARED | MAP_FIXED,
+		     fd, off);
+	if (loc == MAP_FAILED)
+		return -errno;
+	return 0;
+}
+
+int os_protect_memory(void *addr, unsigned long len, int r, int w, int x)
+{
+	int prot = ((r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) |
+		    (x ? PROT_EXEC : 0));
+
+	if (mprotect(addr, len, prot) < 0)
+		return -errno;
+
+	return 0;
+}
+
+int os_unmap_memory(void *addr, int len)
+{
+	int err;
+
+	err = munmap(addr, len);
+	if (err < 0)
+		return -errno;
+	return 0;
+}
+
+#ifndef MADV_REMOVE
+#define MADV_REMOVE KERNEL_MADV_REMOVE
+#endif
+
+int os_drop_memory(void *addr, int length)
+{
+	int err;
+
+	err = madvise(addr, length, MADV_REMOVE);
+	if (err < 0)
+		err = -errno;
+	return err;
+}
+
+int __init can_drop_memory(void)
+{
+	void *addr;
+	int fd, ok = 0;
+
+	printk(UM_KERN_INFO "Checking host MADV_REMOVE support...");
+	fd = create_mem_file(UM_KERN_PAGE_SIZE);
+	if (fd < 0) {
+		printk(UM_KERN_ERR "Creating test memory file failed, "
+		       "err = %d\n", -fd);
+		goto out;
+	}
+
+	addr = mmap64(NULL, UM_KERN_PAGE_SIZE, PROT_READ | PROT_WRITE,
+		      MAP_SHARED, fd, 0);
+	if (addr == MAP_FAILED) {
+		printk(UM_KERN_ERR "Mapping test memory file failed, "
+		       "err = %d\n", -errno);
+		goto out_close;
+	}
+
+	if (madvise(addr, UM_KERN_PAGE_SIZE, MADV_REMOVE) != 0) {
+		printk(UM_KERN_ERR "MADV_REMOVE failed, err = %d\n", -errno);
+		goto out_unmap;
+	}
+
+	printk(UM_KERN_CONT "OK\n");
+	ok = 1;
+
+out_unmap:
+	munmap(addr, UM_KERN_PAGE_SIZE);
+out_close:
+	close(fd);
+out:
+	return ok;
+}
+
+void init_new_thread_signals(void)
+{
+	set_handler(SIGSEGV);
+	set_handler(SIGTRAP);
+	set_handler(SIGFPE);
+	set_handler(SIGILL);
+	set_handler(SIGBUS);
+	signal(SIGHUP, SIG_IGN);
+	set_handler(SIGIO);
+	/* We (currently) only use the child reaper IRQ in seccomp mode */
+	if (using_seccomp)
+		set_handler(SIGCHLD);
+	signal(SIGWINCH, SIG_IGN);
+}
+
+void os_set_pdeathsig(void)
+{
+	prctl(PR_SET_PDEATHSIG, SIGKILL);
+}
+
+int os_futex_wait(void *uaddr, unsigned int val)
+{
+	int r;
+
+	CATCH_EINTR(r = syscall(__NR_futex, uaddr, FUTEX_WAIT, val,
+				NULL, NULL, 0));
+	return r < 0 ? -errno : r;
+}
+
+int os_futex_wake(void *uaddr)
+{
+	int r;
+
+	CATCH_EINTR(r = syscall(__NR_futex, uaddr, FUTEX_WAKE, INT_MAX,
+				NULL, NULL, 0));
+	return r < 0 ? -errno : r;
+}

linux-7.2.3/arch/um/os-Linux/registers.c

@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2004 PathScale, Inc
+ * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <errno.h>
+#include <string.h>
+#include <sys/ptrace.h>
+#include <sysdep/ptrace.h>
+#include <sysdep/ptrace_user.h>
+#include <registers.h>
+#include <stdlib.h>
+
+/* This is set once at boot time and not changed thereafter */
+
+unsigned long exec_regs[MAX_REG_NR];
+unsigned long *exec_fp_regs;
+
+int init_pid_registers(int pid)
+{
+	int err;
+
+	err = ptrace(PTRACE_GETREGS, pid, 0, exec_regs);
+	if (err < 0)
+		return -errno;
+
+	err = arch_init_registers(pid);
+	if (err < 0)
+		return err;
+
+	exec_fp_regs = malloc(host_fp_size);
+	get_fp_registers(pid, exec_fp_regs);
+	return 0;
+}
+
+void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
+{
+	memcpy(regs, exec_regs, sizeof(exec_regs));
+
+	if (fp_regs)
+		memcpy(fp_regs, exec_fp_regs, host_fp_size);
+}

linux-7.2.3/arch/um/os-Linux/sigio.c

@@ -0,0 +1,304 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2008 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <unistd.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <poll.h>
+#include <pty.h>
+#include <sched.h>
+#include <signal.h>
+#include <string.h>
+#include <sys/epoll.h>
+#include <asm/unistd.h>
+#include <kern_util.h>
+#include <init.h>
+#include <os.h>
+#include <sigio.h>
+#include <um_malloc.h>
+
+/*
+ * Protected by sigio_lock(), also used by sigio_cleanup, which is an
+ * exitcall.
+ */
+static struct os_helper_thread *write_sigio_td;
+
+static int epollfd = -1;
+
+#define MAX_EPOLL_EVENTS 64
+
+static struct epoll_event epoll_events[MAX_EPOLL_EVENTS];
+
+static void *write_sigio_thread(void *unused)
+{
+	int pid = getpid();
+	int r;
+
+	os_fix_helper_thread_signals();
+
+	while (1) {
+		r = epoll_wait(epollfd, epoll_events, MAX_EPOLL_EVENTS, -1);
+		if (r < 0) {
+			if (errno == EINTR)
+				continue;
+			printk(UM_KERN_ERR "%s: epoll_wait failed, errno = %d\n",
+			       __func__, errno);
+		}
+
+		CATCH_EINTR(r = syscall(__NR_tgkill, pid, pid, SIGIO));
+		if (r < 0)
+			printk(UM_KERN_ERR "%s: tgkill failed, errno = %d\n",
+			       __func__, errno);
+	}
+
+	return NULL;
+}
+
+int __add_sigio_fd(int fd)
+{
+	struct epoll_event event = {
+		.data.fd = fd,
+		.events = EPOLLIN | EPOLLET,
+	};
+	int r;
+
+	CATCH_EINTR(r = epoll_ctl(epollfd, EPOLL_CTL_ADD, fd, &event));
+	return r < 0 ? -errno : 0;
+}
+
+int add_sigio_fd(int fd)
+{
+	int err;
+
+	sigio_lock();
+	err = __add_sigio_fd(fd);
+	sigio_unlock();
+
+	return err;
+}
+
+int __ignore_sigio_fd(int fd)
+{
+	struct epoll_event event;
+	int r;
+
+	CATCH_EINTR(r = epoll_ctl(epollfd, EPOLL_CTL_DEL, fd, &event));
+	return r < 0 ? -errno : 0;
+}
+
+int ignore_sigio_fd(int fd)
+{
+	int err;
+
+	sigio_lock();
+	err = __ignore_sigio_fd(fd);
+	sigio_unlock();
+
+	return err;
+}
+
+static void write_sigio_workaround(void)
+{
+	int err;
+
+	sigio_lock();
+	if (write_sigio_td)
+		goto out;
+
+	epollfd = epoll_create(MAX_EPOLL_EVENTS);
+	if (epollfd < 0) {
+		printk(UM_KERN_ERR "%s: epoll_create failed, errno = %d\n",
+		       __func__, errno);
+		goto out;
+	}
+
+	err = os_run_helper_thread(&write_sigio_td, write_sigio_thread, NULL);
+	if (err < 0) {
+		printk(UM_KERN_ERR "%s: os_run_helper_thread failed, errno = %d\n",
+		       __func__, -err);
+		close(epollfd);
+		epollfd = -1;
+		goto out;
+	}
+
+out:
+	sigio_unlock();
+}
+
+void sigio_broken(void)
+{
+	write_sigio_workaround();
+}
+
+/* Changed during early boot */
+static int pty_output_sigio;
+
+void maybe_sigio_broken(int fd)
+{
+	if (!isatty(fd))
+		return;
+
+	if (pty_output_sigio)
+		return;
+
+	sigio_broken();
+}
+
+static void sigio_cleanup(void)
+{
+	if (!write_sigio_td)
+		return;
+
+	os_kill_helper_thread(write_sigio_td);
+	write_sigio_td = NULL;
+}
+
+__uml_exitcall(sigio_cleanup);
+
+/* Used as a flag during SIGIO testing early in boot */
+static int got_sigio;
+
+static void __init handler(int sig)
+{
+	got_sigio = 1;
+}
+
+struct openpty_arg {
+	int master;
+	int slave;
+	int err;
+};
+
+static void openpty_cb(void *arg)
+{
+	struct openpty_arg *info = arg;
+
+	info->err = 0;
+	if (openpty(&info->master, &info->slave, NULL, NULL, NULL))
+		info->err = -errno;
+}
+
+static int async_pty(int master, int slave)
+{
+	int flags;
+
+	flags = fcntl(master, F_GETFL);
+	if (flags < 0)
+		return -errno;
+
+	if ((fcntl(master, F_SETFL, flags | O_NONBLOCK | O_ASYNC) < 0) ||
+	    (fcntl(master, F_SETOWN, os_getpid()) < 0))
+		return -errno;
+
+	if ((fcntl(slave, F_SETFL, flags | O_NONBLOCK) < 0))
+		return -errno;
+
+	return 0;
+}
+
+static void __init check_one_sigio(void (*proc)(int, int))
+{
+	struct sigaction old, new;
+	struct openpty_arg pty = { .master = -1, .slave = -1 };
+	int master, slave, err;
+
+	initial_thread_cb(openpty_cb, &pty);
+	if (pty.err) {
+		printk(UM_KERN_ERR "check_one_sigio failed, errno = %d\n",
+		       -pty.err);
+		return;
+	}
+
+	master = pty.master;
+	slave = pty.slave;
+
+	if ((master == -1) || (slave == -1)) {
+		printk(UM_KERN_ERR "check_one_sigio failed to allocate a "
+		       "pty\n");
+		return;
+	}
+
+	/* Not now, but complain so we now where we failed. */
+	err = raw(master);
+	if (err < 0) {
+		printk(UM_KERN_ERR "check_one_sigio : raw failed, errno = %d\n",
+		      -err);
+		return;
+	}
+
+	err = async_pty(master, slave);
+	if (err < 0) {
+		printk(UM_KERN_ERR "check_one_sigio : sigio_async failed, "
+		       "err = %d\n", -err);
+		return;
+	}
+
+	if (sigaction(SIGIO, NULL, &old) < 0) {
+		printk(UM_KERN_ERR "check_one_sigio : sigaction 1 failed, "
+		       "errno = %d\n", errno);
+		return;
+	}
+
+	new = old;
+	new.sa_handler = handler;
+	if (sigaction(SIGIO, &new, NULL) < 0) {
+		printk(UM_KERN_ERR "check_one_sigio : sigaction 2 failed, "
+		       "errno = %d\n", errno);
+		return;
+	}
+
+	got_sigio = 0;
+	(*proc)(master, slave);
+
+	close(master);
+	close(slave);
+
+	if (sigaction(SIGIO, &old, NULL) < 0)
+		printk(UM_KERN_ERR "check_one_sigio : sigaction 3 failed, "
+		       "errno = %d\n", errno);
+}
+
+static void tty_output(int master, int slave)
+{
+	int n;
+	char buf[512];
+
+	printk(UM_KERN_INFO "Checking that host ptys support output SIGIO...");
+
+	memset(buf, 0, sizeof(buf));
+
+	while (write(master, buf, sizeof(buf)) > 0) ;
+	if (errno != EAGAIN)
+		printk(UM_KERN_ERR "tty_output : write failed, errno = %d\n",
+		       errno);
+	while (((n = read(slave, buf, sizeof(buf))) > 0) &&
+	       !({ barrier(); got_sigio; }))
+		;
+
+	if (got_sigio) {
+		printk(UM_KERN_CONT "Yes\n");
+		pty_output_sigio = 1;
+	} else if (n == -EAGAIN)
+		printk(UM_KERN_CONT "No, enabling workaround\n");
+	else
+		printk(UM_KERN_CONT "tty_output : read failed, err = %d\n", n);
+}
+
+static void __init check_sigio(void)
+{
+	if ((access("/dev/ptmx", R_OK) < 0) &&
+	    (access("/dev/ptyp0", R_OK) < 0)) {
+		printk(UM_KERN_WARNING "No pseudo-terminals available - "
+		       "skipping pty SIGIO check\n");
+		return;
+	}
+	check_one_sigio(tty_output);
+}
+
+/* Here because it only does the SIGIO testing for now */
+void __init os_check_bugs(void)
+{
+	check_sigio();
+}

linux-7.2.3/arch/um/os-Linux/signal.c

@@ -0,0 +1,506 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015 Anton Ivanov (aivanov@{brocade.com,kot-begemot.co.uk})
+ * Copyright (C) 2015 Thomas Meyer (thomas@m3y3r.de)
+ * Copyright (C) 2004 PathScale, Inc
+ * Copyright (C) 2004 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdlib.h>
+#include <stdarg.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <signal.h>
+#include <string.h>
+#include <strings.h>
+#include <as-layout.h>
+#include <kern_util.h>
+#include <os.h>
+#include <skas.h>
+#include <sysdep/mcontext.h>
+#include <um_malloc.h>
+#include <sys/ucontext.h>
+#include <timetravel.h>
+#include "internal.h"
+
+void (*sig_info[NSIG])(int, struct siginfo *, struct uml_pt_regs *, void *mc) = {
+	[SIGTRAP]	= relay_signal,
+	[SIGFPE]	= relay_signal,
+	[SIGILL]	= relay_signal,
+	[SIGWINCH]	= winch,
+	[SIGBUS]	= relay_signal,
+	[SIGSEGV]	= segv_handler,
+	[SIGIO]		= sigio_handler,
+	[SIGCHLD]	= sigchld_handler,
+};
+
+static void sig_handler_common(int sig, struct siginfo *si, mcontext_t *mc)
+{
+	struct uml_pt_regs r;
+
+	r.is_user = 0;
+	if (sig == SIGSEGV) {
+		/* For segfaults, we want the data from the sigcontext. */
+		get_regs_from_mc(&r, mc);
+		GET_FAULTINFO_FROM_MC(r.faultinfo, mc);
+	}
+
+	/* enable signals if sig isn't IRQ signal */
+	if ((sig != SIGIO) && (sig != SIGWINCH) && (sig != SIGCHLD))
+		unblock_signals_trace();
+
+	(*sig_info[sig])(sig, si, &r, mc);
+}
+
+/*
+ * These are the asynchronous signals.  SIGPROF is excluded because we want to
+ * be able to profile all of UML, not just the non-critical sections.  If
+ * profiling is not thread-safe, then that is not my problem.  We can disable
+ * profiling when SMP is enabled in that case.
+ */
+#define SIGIO_BIT 0
+#define SIGIO_MASK (1 << SIGIO_BIT)
+
+#define SIGALRM_BIT 1
+#define SIGALRM_MASK (1 << SIGALRM_BIT)
+
+#define SIGCHLD_BIT 2
+#define SIGCHLD_MASK (1 << SIGCHLD_BIT)
+
+__thread int signals_enabled;
+#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
+static int signals_blocked, signals_blocked_pending;
+#endif
+static __thread unsigned int signals_pending;
+static __thread unsigned int signals_active;
+
+static void sig_handler(int sig, struct siginfo *si, mcontext_t *mc)
+{
+	int enabled = signals_enabled;
+
+#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
+	if ((signals_blocked ||
+	     __atomic_load_n(&signals_blocked_pending, __ATOMIC_SEQ_CST)) &&
+	    (sig == SIGIO)) {
+		/* increment so unblock will do another round */
+		__atomic_add_fetch(&signals_blocked_pending, 1,
+				   __ATOMIC_SEQ_CST);
+		return;
+	}
+#endif
+
+	if (!enabled && (sig == SIGIO)) {
+		/*
+		 * In TT_MODE_EXTERNAL, need to still call time-travel
+		 * handlers. This will mark signals_pending by itself
+		 * (only if necessary.)
+		 * Note we won't get here if signals are hard-blocked
+		 * (which is handled above), in that case the hard-
+		 * unblock will handle things.
+		 */
+		if (time_travel_mode == TT_MODE_EXTERNAL)
+			sigio_run_timetravel_handlers();
+		else
+			signals_pending |= SIGIO_MASK;
+		return;
+	}
+
+	if (!enabled && (sig == SIGCHLD)) {
+		signals_pending |= SIGCHLD_MASK;
+		return;
+	}
+
+	block_signals_trace();
+
+	sig_handler_common(sig, si, mc);
+
+	um_set_signals_trace(enabled);
+}
+
+static void timer_real_alarm_handler(mcontext_t *mc)
+{
+	struct uml_pt_regs regs;
+
+	if (mc != NULL)
+		get_regs_from_mc(&regs, mc);
+	else
+		memset(&regs, 0, sizeof(regs));
+	timer_handler(SIGALRM, NULL, &regs);
+}
+
+static void timer_alarm_handler(int sig, struct siginfo *unused_si, mcontext_t *mc)
+{
+	int enabled;
+
+	enabled = signals_enabled;
+	if (!signals_enabled) {
+		signals_pending |= SIGALRM_MASK;
+		return;
+	}
+
+	block_signals_trace();
+
+	signals_active |= SIGALRM_MASK;
+
+	timer_real_alarm_handler(mc);
+
+	signals_active &= ~SIGALRM_MASK;
+
+	um_set_signals_trace(enabled);
+}
+
+void deliver_alarm(void) {
+    timer_alarm_handler(SIGALRM, NULL, NULL);
+}
+
+void timer_set_signal_handler(void)
+{
+	set_handler(SIGALRM);
+}
+
+int timer_alarm_pending(void)
+{
+	return !!(signals_pending & SIGALRM_MASK);
+}
+
+void set_sigstack(void *sig_stack, int size)
+{
+	stack_t stack = {
+		.ss_flags = 0,
+		.ss_sp = sig_stack,
+		.ss_size = size
+	};
+
+	if (sigaltstack(&stack, NULL) != 0)
+		panic("enabling signal stack failed, errno = %d\n", errno);
+}
+
+static void sigusr1_handler(int sig, struct siginfo *unused_si, mcontext_t *mc)
+{
+	uml_pm_wake();
+}
+
+void register_pm_wake_signal(void)
+{
+	set_handler(SIGUSR1);
+}
+
+static void (*handlers[_NSIG])(int sig, struct siginfo *si, mcontext_t *mc) = {
+	[SIGSEGV] = sig_handler,
+	[SIGBUS] = sig_handler,
+	[SIGILL] = sig_handler,
+	[SIGFPE] = sig_handler,
+	[SIGTRAP] = sig_handler,
+
+	[SIGIO] = sig_handler,
+	[SIGWINCH] = sig_handler,
+	/* SIGCHLD is only actually registered in seccomp mode. */
+	[SIGCHLD] = sig_handler,
+	[SIGALRM] = timer_alarm_handler,
+
+	[SIGUSR1] = sigusr1_handler,
+};
+
+static void hard_handler(int sig, siginfo_t *si, void *p)
+{
+	ucontext_t *uc = p;
+	mcontext_t *mc = &uc->uc_mcontext;
+	int save_errno = errno;
+
+	(*handlers[sig])(sig, (struct siginfo *)si, mc);
+
+	errno = save_errno;
+}
+
+void set_handler(int sig)
+{
+	struct sigaction action;
+	int flags = SA_SIGINFO | SA_ONSTACK;
+	sigset_t sig_mask;
+
+	action.sa_sigaction = hard_handler;
+
+	/* block irq ones */
+	sigemptyset(&action.sa_mask);
+	sigaddset(&action.sa_mask, SIGIO);
+	sigaddset(&action.sa_mask, SIGWINCH);
+	sigaddset(&action.sa_mask, SIGALRM);
+	if (using_seccomp)
+		sigaddset(&action.sa_mask, SIGCHLD);
+
+	if (sig == SIGSEGV)
+		flags |= SA_NODEFER;
+
+	if (sigismember(&action.sa_mask, sig))
+		flags |= SA_RESTART; /* if it's an irq signal */
+
+	action.sa_flags = flags;
+	action.sa_restorer = NULL;
+	if (sigaction(sig, &action, NULL) < 0)
+		panic("sigaction failed - errno = %d\n", errno);
+
+	sigemptyset(&sig_mask);
+	sigaddset(&sig_mask, sig);
+	if (sigprocmask(SIG_UNBLOCK, &sig_mask, NULL) < 0)
+		panic("sigprocmask failed - errno = %d\n", errno);
+}
+
+void send_sigio_to_self(void)
+{
+	kill(os_getpid(), SIGIO);
+}
+
+int change_sig(int signal, int on)
+{
+	sigset_t sigset;
+
+	sigemptyset(&sigset);
+	sigaddset(&sigset, signal);
+	if (sigprocmask(on ? SIG_UNBLOCK : SIG_BLOCK, &sigset, NULL) < 0)
+		return -errno;
+
+	return 0;
+}
+
+static inline void __block_signals(void)
+{
+	if (!signals_enabled)
+		return;
+
+	os_local_ipi_disable();
+	barrier();
+	signals_enabled = 0;
+}
+
+static inline void __unblock_signals(void)
+{
+	if (signals_enabled)
+		return;
+
+	signals_enabled = 1;
+	barrier();
+	os_local_ipi_enable();
+}
+
+void block_signals(void)
+{
+	__block_signals();
+	/*
+	 * This must return with signals disabled, so this barrier
+	 * ensures that writes are flushed out before the return.
+	 * This might matter if gcc figures out how to inline this and
+	 * decides to shuffle this code into the caller.
+	 */
+	barrier();
+}
+
+void unblock_signals(void)
+{
+	int save_pending;
+
+	if (signals_enabled == 1)
+		return;
+
+	__unblock_signals();
+
+#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
+	deliver_time_travel_irqs();
+#endif
+
+	/*
+	 * We loop because the IRQ handler returns with interrupts off.  So,
+	 * interrupts may have arrived and we need to re-enable them and
+	 * recheck signals_pending.
+	 */
+	while (1) {
+		/*
+		 * Save and reset save_pending after enabling signals.  This
+		 * way, signals_pending won't be changed while we're reading it.
+		 *
+		 * Setting signals_enabled and reading signals_pending must
+		 * happen in this order, so have the barrier here.
+		 */
+		barrier();
+
+		save_pending = signals_pending;
+		if (save_pending == 0)
+			return;
+
+		signals_pending = 0;
+
+		/*
+		 * We have pending interrupts, so disable signals, as the
+		 * handlers expect them off when they are called.  They will
+		 * be enabled again above. We need to trace this, as we're
+		 * expected to be enabling interrupts already, but any more
+		 * tracing that happens inside the handlers we call for the
+		 * pending signals will mess up the tracing state.
+		 */
+		__block_signals();
+		um_trace_signals_off();
+
+		/*
+		 * Deal with SIGIO first because the alarm handler might
+		 * schedule, leaving the pending SIGIO stranded until we come
+		 * back here.
+		 *
+		 * SIGIO's handler doesn't use siginfo or mcontext,
+		 * so they can be NULL.
+		 */
+		if (save_pending & SIGIO_MASK)
+			sig_handler_common(SIGIO, NULL, NULL);
+
+		if (save_pending & SIGCHLD_MASK) {
+			struct uml_pt_regs regs = {};
+
+			sigchld_handler(SIGCHLD, NULL, &regs, NULL);
+		}
+
+		/* Do not reenter the handler */
+
+		if ((save_pending & SIGALRM_MASK) && (!(signals_active & SIGALRM_MASK)))
+			timer_real_alarm_handler(NULL);
+
+		/* Rerun the loop only if there is still pending SIGIO and not in TIMER handler */
+
+		if (!(signals_pending & SIGIO_MASK) && (signals_active & SIGALRM_MASK))
+			return;
+
+		/* Re-enable signals and trace that we're doing so. */
+		um_trace_signals_on();
+		__unblock_signals();
+	}
+}
+
+int um_get_signals(void)
+{
+	return signals_enabled;
+}
+
+int um_set_signals(int enable)
+{
+	int ret;
+	if (signals_enabled == enable)
+		return enable;
+
+	ret = signals_enabled;
+	if (enable)
+		unblock_signals();
+	else block_signals();
+
+	return ret;
+}
+
+int um_set_signals_trace(int enable)
+{
+	int ret;
+	if (signals_enabled == enable)
+		return enable;
+
+	ret = signals_enabled;
+	if (enable)
+		unblock_signals_trace();
+	else
+		block_signals_trace();
+
+	return ret;
+}
+
+#if IS_ENABLED(CONFIG_UML_TIME_TRAVEL_SUPPORT)
+void mark_sigio_pending(void)
+{
+	/*
+	 * It would seem that this should be atomic so
+	 * it isn't a read-modify-write with a signal
+	 * that could happen in the middle, losing the
+	 * value set by the signal.
+	 *
+	 * However, this function is only called when in
+	 * time-travel=ext simulation mode, in which case
+	 * the only signal ever pending is SIGIO, which
+	 * is blocked while this can be called, and the
+	 * timer signal (SIGALRM) cannot happen.
+	 */
+	signals_pending |= SIGIO_MASK;
+}
+
+void block_signals_hard(void)
+{
+	signals_blocked++;
+	barrier();
+}
+
+void unblock_signals_hard(void)
+{
+	static bool unblocking;
+
+	if (!signals_blocked)
+		panic("unblocking signals while not blocked");
+
+	if (--signals_blocked)
+		return;
+	/*
+	 * Must be set to 0 before we check pending so the
+	 * SIGIO handler will run as normal unless we're still
+	 * going to process signals_blocked_pending.
+	 */
+	barrier();
+
+	/*
+	 * Note that block_signals_hard()/unblock_signals_hard() can be called
+	 * within the unblock_signals()/sigio_run_timetravel_handlers() below.
+	 * This would still be prone to race conditions since it's actually a
+	 * call _within_ e.g. vu_req_read_message(), where we observed this
+	 * issue, which loops. Thus, if the inner call handles the recorded
+	 * pending signals, we can get out of the inner call with the real
+	 * signal hander no longer blocked, and still have a race. Thus don't
+	 * handle unblocking in the inner call, if it happens, but only in
+	 * the outermost call - 'unblocking' serves as an ownership for the
+	 * signals_blocked_pending decrement.
+	 */
+	if (unblocking)
+		return;
+	unblocking = true;
+
+	while (__atomic_load_n(&signals_blocked_pending, __ATOMIC_SEQ_CST)) {
+		if (signals_enabled) {
+			/* signals are enabled so we can touch this */
+			signals_pending |= SIGIO_MASK;
+			/*
+			 * this is a bit inefficient, but that's
+			 * not really important
+			 */
+			block_signals();
+			unblock_signals();
+		} else {
+			/*
+			 * we need to run time-travel handlers even
+			 * if not enabled
+			 */
+			sigio_run_timetravel_handlers();
+		}
+
+		/*
+		 * The decrement of signals_blocked_pending must be atomic so
+		 * that the signal handler will either happen before or after
+		 * the decrement, not during a read-modify-write:
+		 *  - If it happens before, it can increment it and we'll
+		 *    decrement it and do another round in the loop.
+		 *  - If it happens after it'll see 0 for both signals_blocked
+		 *    and signals_blocked_pending and thus run the handler as
+		 *    usual (subject to signals_enabled, but that's unrelated.)
+		 *
+		 * Note that a call to unblock_signals_hard() within the calls
+		 * to unblock_signals() or sigio_run_timetravel_handlers() above
+		 * will do nothing due to the 'unblocking' state, so this cannot
+		 * underflow as the only one decrementing will be the outermost
+		 * one.
+		 */
+		if (__atomic_sub_fetch(&signals_blocked_pending, 1,
+				       __ATOMIC_SEQ_CST) < 0)
+			panic("signals_blocked_pending underflow");
+	}
+
+	unblocking = false;
+}
+#endif

linux-7.2.3/arch/um/os-Linux/skas/Makefile

@@ -0,0 +1,10 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (C) 2002 - 2007 Jeff Dike (jdike@{linux.intel,addtoit}.com)
+#
+
+obj-y := mem.o process.o
+
+USER_OBJS := $(obj-y)
+
+include $(srctree)/arch/um/scripts/Makefile.rules

linux-7.2.3/arch/um/os-Linux/skas/mem.c

@@ -0,0 +1,284 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2021 Benjamin Berg <benjamin@sipsolutions.net>
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stddef.h>
+#include <unistd.h>
+#include <errno.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <init.h>
+#include <as-layout.h>
+#include <mm_id.h>
+#include <os.h>
+#include <ptrace_user.h>
+#include <registers.h>
+#include <skas.h>
+#include <sysdep/ptrace.h>
+#include <sysdep/stub.h>
+#include "../internal.h"
+
+extern char __syscall_stub_start[];
+
+void syscall_stub_dump_error(struct mm_id *mm_idp)
+{
+	struct stub_data *proc_data = (void *)mm_idp->stack;
+	struct stub_syscall *sc;
+
+	if (proc_data->syscall_data_len < 0 ||
+	    proc_data->syscall_data_len >= ARRAY_SIZE(proc_data->syscall_data))
+		panic("Syscall data was corrupted by stub (len is: %d, expected maximum: %d)!",
+			proc_data->syscall_data_len,
+			mm_idp->syscall_data_len);
+
+	sc = &proc_data->syscall_data[proc_data->syscall_data_len];
+
+	printk(UM_KERN_ERR "%s : length = %d, last offset = %d",
+		__func__, mm_idp->syscall_data_len,
+		proc_data->syscall_data_len);
+	printk(UM_KERN_ERR "%s : stub syscall type %d failed, return value = 0x%lx\n",
+		__func__, sc->syscall, proc_data->err);
+
+	print_hex_dump(UM_KERN_ERR, "    syscall data: ", 0,
+		       16, 4, sc, sizeof(*sc), 0);
+
+	if (using_seccomp) {
+		printk(UM_KERN_ERR "%s: FD map num: %d", __func__,
+		       mm_idp->syscall_fd_num);
+		print_hex_dump(UM_KERN_ERR,
+				"    FD map: ", 0, 16,
+				sizeof(mm_idp->syscall_fd_map[0]),
+				mm_idp->syscall_fd_map,
+				sizeof(mm_idp->syscall_fd_map), 0);
+	}
+}
+
+static inline unsigned long *check_init_stack(struct mm_id * mm_idp,
+					      unsigned long *stack)
+{
+	if (stack == NULL) {
+		stack = (unsigned long *) mm_idp->stack + 2;
+		*stack = 0;
+	}
+	return stack;
+}
+
+static unsigned long syscall_regs[MAX_REG_NR];
+
+static int __init init_syscall_regs(void)
+{
+	get_safe_registers(syscall_regs, NULL);
+
+	syscall_regs[REGS_IP_INDEX] = STUB_CODE +
+		((unsigned long) stub_syscall_handler -
+		 (unsigned long) __syscall_stub_start);
+	syscall_regs[REGS_SP_INDEX] = STUB_DATA +
+		offsetof(struct stub_data, sigstack) +
+		sizeof(((struct stub_data *) 0)->sigstack) -
+		sizeof(void *);
+
+	return 0;
+}
+
+__initcall(init_syscall_regs);
+
+static inline long do_syscall_stub(struct mm_id *mm_idp)
+{
+	struct stub_data *proc_data = (void *)mm_idp->stack;
+	int n, i;
+	int err, pid = mm_idp->pid;
+
+	/* Inform process how much we have filled in. */
+	proc_data->syscall_data_len = mm_idp->syscall_data_len;
+
+	if (using_seccomp) {
+		proc_data->restart_wait = 1;
+		wait_stub_done_seccomp(mm_idp, 0, 1);
+	} else {
+		n = ptrace_setregs(pid, syscall_regs);
+		if (n < 0) {
+			printk(UM_KERN_ERR "Registers -\n");
+			for (i = 0; i < MAX_REG_NR; i++)
+				printk(UM_KERN_ERR "\t%d\t0x%lx\n", i, syscall_regs[i]);
+			panic("%s : PTRACE_SETREGS failed, errno = %d\n",
+			      __func__, -n);
+		}
+
+		err = ptrace(PTRACE_CONT, pid, 0, 0);
+		if (err)
+			panic("Failed to continue stub, pid = %d, errno = %d\n",
+			      pid, errno);
+
+		wait_stub_done(pid);
+	}
+
+	/*
+	 * proc_data->err will be negative if there was an (unexpected) error.
+	 * In that case, syscall_data_len points to the last executed syscall,
+	 * otherwise it will be zero (but we do not need to rely on that).
+	 */
+	if (proc_data->err < 0) {
+		syscall_stub_dump_error(mm_idp);
+
+		/* Store error code in case someone tries to add more syscalls */
+		mm_idp->syscall_data_len = proc_data->err;
+	} else {
+		mm_idp->syscall_data_len = 0;
+	}
+
+	if (using_seccomp)
+		mm_idp->syscall_fd_num = 0;
+
+	return mm_idp->syscall_data_len;
+}
+
+int syscall_stub_flush(struct mm_id *mm_idp)
+{
+	int res;
+
+	if (mm_idp->syscall_data_len == 0)
+		return 0;
+
+	/* If an error happened already, report it and reset the state. */
+	if (mm_idp->syscall_data_len < 0) {
+		res = mm_idp->syscall_data_len;
+		mm_idp->syscall_data_len = 0;
+		return res;
+	}
+
+	res = do_syscall_stub(mm_idp);
+	mm_idp->syscall_data_len = 0;
+
+	return res;
+}
+
+struct stub_syscall *syscall_stub_alloc(struct mm_id *mm_idp)
+{
+	struct stub_syscall *sc;
+	struct stub_data *proc_data = (struct stub_data *) mm_idp->stack;
+
+	if (mm_idp->syscall_data_len > 0 &&
+	    mm_idp->syscall_data_len == ARRAY_SIZE(proc_data->syscall_data))
+		do_syscall_stub(mm_idp);
+
+	if (mm_idp->syscall_data_len < 0) {
+		/* Return dummy to retain error state. */
+		sc = &proc_data->syscall_data[0];
+	} else {
+		sc = &proc_data->syscall_data[mm_idp->syscall_data_len];
+		mm_idp->syscall_data_len += 1;
+	}
+	memset(sc, 0, sizeof(*sc));
+
+	return sc;
+}
+
+static struct stub_syscall *syscall_stub_get_previous(struct mm_id *mm_idp,
+						      int syscall_type,
+						      unsigned long virt)
+{
+	if (mm_idp->syscall_data_len > 0) {
+		struct stub_data *proc_data = (void *) mm_idp->stack;
+		struct stub_syscall *sc;
+
+		sc = &proc_data->syscall_data[mm_idp->syscall_data_len - 1];
+
+		if (sc->syscall == syscall_type &&
+		    sc->mem.addr + sc->mem.length == virt)
+			return sc;
+	}
+
+	return NULL;
+}
+
+static int get_stub_fd(struct mm_id *mm_idp, int fd)
+{
+	int i;
+
+	/* Find an FD slot (or flush and use first) */
+	if (!using_seccomp)
+		return fd;
+
+	/* Already crashed, value does not matter */
+	if (mm_idp->syscall_data_len < 0)
+		return 0;
+
+	/* Find existing FD in map if we can allocate another syscall */
+	if (mm_idp->syscall_data_len <
+	    ARRAY_SIZE(((struct stub_data *)NULL)->syscall_data)) {
+		for (i = 0; i < mm_idp->syscall_fd_num; i++) {
+			if (mm_idp->syscall_fd_map[i] == fd)
+				return i;
+		}
+
+		if (mm_idp->syscall_fd_num < STUB_MAX_FDS) {
+			i = mm_idp->syscall_fd_num;
+			mm_idp->syscall_fd_map[i] = fd;
+
+			mm_idp->syscall_fd_num++;
+
+			return i;
+		}
+	}
+
+	/* FD map full or no syscall space available, continue after flush */
+	do_syscall_stub(mm_idp);
+	mm_idp->syscall_fd_map[0] = fd;
+	mm_idp->syscall_fd_num = 1;
+
+	return 0;
+}
+
+int map(struct mm_id *mm_idp, unsigned long virt, unsigned long len, int prot,
+	int phys_fd, unsigned long long offset)
+{
+	struct stub_syscall *sc;
+
+	/* Compress with previous syscall if that is possible */
+	sc = syscall_stub_get_previous(mm_idp, STUB_SYSCALL_MMAP, virt);
+	if (sc && sc->mem.prot == prot &&
+	    sc->mem.offset == MMAP_OFFSET(offset - sc->mem.length)) {
+		int prev_fd = sc->mem.fd;
+
+		if (using_seccomp)
+			prev_fd = mm_idp->syscall_fd_map[sc->mem.fd];
+
+		if (phys_fd == prev_fd) {
+			sc->mem.length += len;
+			return 0;
+		}
+	}
+
+	phys_fd = get_stub_fd(mm_idp, phys_fd);
+
+	sc = syscall_stub_alloc(mm_idp);
+	sc->syscall = STUB_SYSCALL_MMAP;
+	sc->mem.addr = virt;
+	sc->mem.length = len;
+	sc->mem.prot = prot;
+	sc->mem.fd = phys_fd;
+	sc->mem.offset = MMAP_OFFSET(offset);
+
+	return 0;
+}
+
+int unmap(struct mm_id *mm_idp, unsigned long addr, unsigned long len)
+{
+	struct stub_syscall *sc;
+
+	/* Compress with previous syscall if that is possible */
+	sc = syscall_stub_get_previous(mm_idp, STUB_SYSCALL_MUNMAP, addr);
+	if (sc) {
+		sc->mem.length += len;
+		return 0;
+	}
+
+	sc = syscall_stub_alloc(mm_idp);
+	sc->syscall = STUB_SYSCALL_MUNMAP;
+	sc->mem.addr = addr;
+	sc->mem.length = len;
+
+	return 0;
+}

linux-7.2.3/arch/um/os-Linux/skas/process.c

@@ -0,0 +1,908 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2021 Benjamin Berg <benjamin@sipsolutions.net>
+ * Copyright (C) 2015 Thomas Meyer (thomas@m3y3r.de)
+ * Copyright (C) 2002- 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdlib.h>
+#include <stdbool.h>
+#include <unistd.h>
+#include <sched.h>
+#include <errno.h>
+#include <string.h>
+#include <fcntl.h>
+#include <mem_user.h>
+#include <sys/mman.h>
+#include <sys/wait.h>
+#include <sys/stat.h>
+#include <sys/socket.h>
+#include <asm/unistd.h>
+#include <as-layout.h>
+#include <init.h>
+#include <kern_util.h>
+#include <mem.h>
+#include <os.h>
+#include <ptrace_user.h>
+#include <registers.h>
+#include <skas.h>
+#include <sysdep/stub.h>
+#include <sysdep/mcontext.h>
+#include <linux/futex.h>
+#include <linux/threads.h>
+#include <timetravel.h>
+#include <asm-generic/rwonce.h>
+#include "../internal.h"
+
+int is_skas_winch(int pid, int fd, void *data)
+{
+	return pid == getpgrp();
+}
+
+static const char *ptrace_reg_name(int idx)
+{
+#define R(n) case HOST_##n: return #n
+
+	switch (idx) {
+#ifdef __x86_64__
+	R(BX);
+	R(CX);
+	R(DI);
+	R(SI);
+	R(DX);
+	R(BP);
+	R(AX);
+	R(R8);
+	R(R9);
+	R(R10);
+	R(R11);
+	R(R12);
+	R(R13);
+	R(R14);
+	R(R15);
+	R(ORIG_AX);
+	R(CS);
+	R(SS);
+	R(EFLAGS);
+#elif defined(__i386__)
+	R(IP);
+	R(SP);
+	R(EFLAGS);
+	R(AX);
+	R(BX);
+	R(CX);
+	R(DX);
+	R(SI);
+	R(DI);
+	R(BP);
+	R(CS);
+	R(SS);
+	R(DS);
+	R(FS);
+	R(ES);
+	R(GS);
+	R(ORIG_AX);
+#endif
+	}
+	return "";
+}
+
+static int ptrace_dump_regs(int pid)
+{
+	unsigned long regs[MAX_REG_NR];
+	int i;
+
+	if (ptrace(PTRACE_GETREGS, pid, 0, regs) < 0)
+		return -errno;
+
+	printk(UM_KERN_ERR "Stub registers -\n");
+	for (i = 0; i < ARRAY_SIZE(regs); i++) {
+		const char *regname = ptrace_reg_name(i);
+
+		printk(UM_KERN_ERR "\t%s\t(%2d): %lx\n", regname, i, regs[i]);
+	}
+
+	return 0;
+}
+
+/*
+ * Signals that are OK to receive in the stub - we'll just continue it.
+ * SIGWINCH will happen when UML is inside a detached screen.
+ */
+#define STUB_SIG_MASK ((1 << SIGALRM) | (1 << SIGWINCH))
+
+/* Signals that the stub will finish with - anything else is an error */
+#define STUB_DONE_MASK (1 << SIGTRAP)
+
+void wait_stub_done(int pid)
+{
+	int n, status, err;
+
+	while (1) {
+		CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED | __WALL));
+		if ((n < 0) || !WIFSTOPPED(status))
+			goto bad_wait;
+
+		if (((1 << WSTOPSIG(status)) & STUB_SIG_MASK) == 0)
+			break;
+
+		err = ptrace(PTRACE_CONT, pid, 0, 0);
+		if (err) {
+			printk(UM_KERN_ERR "%s : continue failed, errno = %d\n",
+			       __func__, errno);
+			fatal_sigsegv();
+		}
+	}
+
+	if (((1 << WSTOPSIG(status)) & STUB_DONE_MASK) != 0)
+		return;
+
+bad_wait:
+	err = ptrace_dump_regs(pid);
+	if (err)
+		printk(UM_KERN_ERR "Failed to get registers from stub, errno = %d\n",
+		       -err);
+	printk(UM_KERN_ERR "%s : failed to wait for SIGTRAP, pid = %d, n = %d, errno = %d, status = 0x%x\n",
+	       __func__, pid, n, errno, status);
+	fatal_sigsegv();
+}
+
+void wait_stub_done_seccomp(struct mm_id *mm_idp, int running, int wait_sigsys)
+{
+	struct stub_data *data = (void *)mm_idp->stack;
+	int ret;
+
+	do {
+		const char byte = 0;
+		struct iovec iov = {
+			.iov_base = (void *)&byte,
+			.iov_len = sizeof(byte),
+		};
+		union {
+			char data[CMSG_SPACE(sizeof(mm_idp->syscall_fd_map))];
+			struct cmsghdr align;
+		} ctrl;
+		struct msghdr msgh = {
+			.msg_iov = &iov,
+			.msg_iovlen = 1,
+		};
+
+		if (!running) {
+			if (mm_idp->syscall_fd_num) {
+				unsigned int fds_size =
+					sizeof(int) * mm_idp->syscall_fd_num;
+				struct cmsghdr *cmsg;
+
+				msgh.msg_control = ctrl.data;
+				msgh.msg_controllen = CMSG_SPACE(fds_size);
+				cmsg = CMSG_FIRSTHDR(&msgh);
+				cmsg->cmsg_level = SOL_SOCKET;
+				cmsg->cmsg_type = SCM_RIGHTS;
+				cmsg->cmsg_len = CMSG_LEN(fds_size);
+				memcpy(CMSG_DATA(cmsg), mm_idp->syscall_fd_map,
+				       fds_size);
+
+				CATCH_EINTR(syscall(__NR_sendmsg, mm_idp->sock,
+						&msgh, 0));
+			}
+
+			data->signal = 0;
+			data->futex = FUTEX_IN_CHILD;
+			CATCH_EINTR(syscall(__NR_futex, &data->futex,
+					    FUTEX_WAKE, 1, NULL, NULL, 0));
+		}
+
+		do {
+			/*
+			 * We need to check whether the child is still alive
+			 * before and after the FUTEX_WAIT call. Before, in
+			 * case it just died but we still updated data->futex
+			 * to FUTEX_IN_CHILD. And after, in case it died while
+			 * we were waiting (and SIGCHLD woke us up, see the
+			 * IRQ handler in mmu.c).
+			 *
+			 * Either way, if PID is negative, then we have no
+			 * choice but to kill the task.
+			 */
+			if (__READ_ONCE(mm_idp->pid) < 0)
+				goto out_kill;
+
+			ret = syscall(__NR_futex, &data->futex,
+				      FUTEX_WAIT, FUTEX_IN_CHILD,
+				      NULL, NULL, 0);
+			if (ret < 0 && errno != EINTR && errno != EAGAIN) {
+				printk(UM_KERN_ERR "%s : FUTEX_WAIT failed, errno = %d\n",
+				       __func__, errno);
+				goto out_kill;
+			}
+		} while (data->futex == FUTEX_IN_CHILD);
+
+		if (__READ_ONCE(mm_idp->pid) < 0)
+			goto out_kill;
+
+		running = 0;
+
+		/* We may receive a SIGALRM before SIGSYS, iterate again. */
+	} while (wait_sigsys && data->signal == SIGALRM);
+
+	if (data->mctx_offset > sizeof(data->sigstack) - sizeof(mcontext_t)) {
+		printk(UM_KERN_ERR "%s : invalid mcontext offset", __func__);
+		goto out_kill;
+	}
+
+	if (wait_sigsys && data->signal != SIGSYS) {
+		printk(UM_KERN_ERR "%s : expected SIGSYS but got %d",
+		       __func__, data->signal);
+		goto out_kill;
+	}
+
+	return;
+
+out_kill:
+	printk(UM_KERN_ERR "%s : failed to wait for stub, pid = %d, errno = %d\n",
+	       __func__, mm_idp->pid, errno);
+	/* This is not true inside start_userspace */
+	if (current_mm_id() == mm_idp)
+		fatal_sigsegv();
+}
+
+extern unsigned long current_stub_stack(void);
+
+static void get_skas_faultinfo(int pid, struct faultinfo *fi)
+{
+	int err;
+
+	err = ptrace(PTRACE_CONT, pid, 0, SIGSEGV);
+	if (err) {
+		printk(UM_KERN_ERR "Failed to continue stub, pid = %d, "
+		       "errno = %d\n", pid, errno);
+		fatal_sigsegv();
+	}
+	wait_stub_done(pid);
+
+	/*
+	 * faultinfo is prepared by the stub_segv_handler at start of
+	 * the stub stack page. We just have to copy it.
+	 */
+	memcpy(fi, (void *)current_stub_stack(), sizeof(*fi));
+}
+
+static void handle_trap(struct uml_pt_regs *regs)
+{
+	if ((UPT_IP(regs) >= STUB_START) && (UPT_IP(regs) < STUB_END))
+		fatal_sigsegv();
+
+	handle_syscall(regs);
+}
+
+extern char __syscall_stub_start[];
+
+static int stub_exe_fd;
+
+struct tramp_data {
+	struct stub_data *stub_data;
+	/* 0 is inherited, 1 is the kernel side */
+	int sockpair[2];
+};
+
+#ifndef CLOSE_RANGE_CLOEXEC
+#define CLOSE_RANGE_CLOEXEC	(1U << 2)
+#endif
+
+static int userspace_tramp(void *data)
+{
+	struct tramp_data *tramp_data = data;
+	char *const argv[] = { "uml-userspace", NULL };
+	unsigned long long offset;
+	struct stub_init_data init_data = {
+		.seccomp = using_seccomp,
+		.stub_start = STUB_START,
+	};
+	int ret;
+
+	if (using_seccomp) {
+		init_data.signal_handler = STUB_CODE +
+					   (unsigned long) stub_signal_interrupt -
+					   (unsigned long) __syscall_stub_start;
+		init_data.signal_restorer = STUB_CODE +
+					   (unsigned long) stub_signal_restorer -
+					   (unsigned long) __syscall_stub_start;
+	} else {
+		init_data.signal_handler = STUB_CODE +
+					   (unsigned long) stub_segv_handler -
+					   (unsigned long) __syscall_stub_start;
+		init_data.signal_restorer = 0;
+	}
+
+	init_data.stub_code_fd = phys_mapping(uml_to_phys(__syscall_stub_start),
+					      &offset);
+	init_data.stub_code_offset = MMAP_OFFSET(offset);
+
+	init_data.stub_data_fd = phys_mapping(uml_to_phys(tramp_data->stub_data),
+					      &offset);
+	init_data.stub_data_offset = MMAP_OFFSET(offset);
+
+	/*
+	 * Avoid leaking unneeded FDs to the stub by setting CLOEXEC on all FDs
+	 * and then unsetting it on all memory related FDs.
+	 * This is not strictly necessary from a safety perspective.
+	 */
+	syscall(__NR_close_range, 0, ~0U, CLOSE_RANGE_CLOEXEC);
+
+	fcntl(init_data.stub_data_fd, F_SETFD, 0);
+
+	/* dup2 signaling FD/socket to STDIN */
+	if (dup2(tramp_data->sockpair[0], 0) < 0)
+		exit(3);
+	close(tramp_data->sockpair[0]);
+
+	/* Write init_data and close write side */
+	ret = write(tramp_data->sockpair[1], &init_data, sizeof(init_data));
+	close(tramp_data->sockpair[1]);
+
+	if (ret != sizeof(init_data))
+		exit(4);
+
+	/* Raw execveat for compatibility with older libc versions */
+	syscall(__NR_execveat, stub_exe_fd, (unsigned long)"",
+		(unsigned long)argv, NULL, AT_EMPTY_PATH);
+
+	exit(5);
+}
+
+extern char stub_exe_start[];
+extern char stub_exe_end[];
+
+extern char *tempdir;
+
+#define STUB_EXE_NAME_TEMPLATE "/uml-userspace-XXXXXX"
+
+#ifndef MFD_EXEC
+#define MFD_EXEC 0x0010U
+#endif
+
+static int __init init_stub_exe_fd(void)
+{
+	size_t written = 0;
+	char *tmpfile = NULL;
+
+	stub_exe_fd = memfd_create("uml-userspace",
+				   MFD_EXEC | MFD_CLOEXEC | MFD_ALLOW_SEALING);
+
+	if (stub_exe_fd < 0) {
+		printk(UM_KERN_INFO "Could not create executable memfd, using temporary file!");
+
+		tmpfile = malloc(strlen(tempdir) +
+				  strlen(STUB_EXE_NAME_TEMPLATE) + 1);
+		if (tmpfile == NULL)
+			panic("Failed to allocate memory for stub binary name");
+
+		strcpy(tmpfile, tempdir);
+		strcat(tmpfile, STUB_EXE_NAME_TEMPLATE);
+
+		stub_exe_fd = mkstemp(tmpfile);
+		if (stub_exe_fd < 0)
+			panic("Could not create temporary file for stub binary: %d",
+			      -errno);
+	}
+
+	while (written < stub_exe_end - stub_exe_start) {
+		ssize_t res = write(stub_exe_fd, stub_exe_start + written,
+				    stub_exe_end - stub_exe_start - written);
+		if (res < 0) {
+			if (errno == EINTR)
+				continue;
+
+			if (tmpfile)
+				unlink(tmpfile);
+			panic("Failed write stub binary: %d", -errno);
+		}
+
+		written += res;
+	}
+
+	if (!tmpfile) {
+		fcntl(stub_exe_fd, F_ADD_SEALS,
+		      F_SEAL_WRITE | F_SEAL_SHRINK | F_SEAL_GROW | F_SEAL_SEAL);
+	} else {
+		if (fchmod(stub_exe_fd, 00500) < 0) {
+			unlink(tmpfile);
+			panic("Could not make stub binary executable: %d",
+			      -errno);
+		}
+
+		close(stub_exe_fd);
+		stub_exe_fd = open(tmpfile, O_RDONLY | O_CLOEXEC | O_NOFOLLOW);
+		if (stub_exe_fd < 0) {
+			unlink(tmpfile);
+			panic("Could not reopen stub binary: %d", -errno);
+		}
+
+		unlink(tmpfile);
+		free(tmpfile);
+	}
+
+	return 0;
+}
+__initcall(init_stub_exe_fd);
+
+int using_seccomp;
+
+/**
+ * start_userspace() - prepare a new userspace process
+ * @mm_id: The corresponding struct mm_id
+ *
+ * Setups a new temporary stack page that is used while userspace_tramp() runs
+ * Clones the kernel process into a new userspace process, with FDs only.
+ *
+ * Return: When positive: the process id of the new userspace process,
+ *         when negative: an error number.
+ * FIXME: can PIDs become negative?!
+ */
+int start_userspace(struct mm_id *mm_id)
+{
+	struct stub_data *proc_data = (void *)mm_id->stack;
+	struct tramp_data tramp_data = {
+		.stub_data = proc_data,
+	};
+	void *stack;
+	unsigned long sp;
+	int status, n, err;
+
+	/* setup a temporary stack page */
+	stack = mmap(NULL, UM_KERN_PAGE_SIZE,
+		     PROT_READ | PROT_WRITE | PROT_EXEC,
+		     MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+	if (stack == MAP_FAILED) {
+		err = -errno;
+		printk(UM_KERN_ERR "%s : mmap failed, errno = %d\n",
+		       __func__, errno);
+		return err;
+	}
+
+	/* set stack pointer to the end of the stack page, so it can grow downwards */
+	sp = (unsigned long)stack + UM_KERN_PAGE_SIZE;
+
+	/* socket pair for init data and SECCOMP FD passing (no CLOEXEC here) */
+	if (socketpair(AF_UNIX, SOCK_STREAM, 0, tramp_data.sockpair)) {
+		err = -errno;
+		printk(UM_KERN_ERR "%s : socketpair failed, errno = %d\n",
+		       __func__, errno);
+		return err;
+	}
+
+	if (using_seccomp)
+		proc_data->futex = FUTEX_IN_CHILD;
+
+	mm_id->pid = clone(userspace_tramp, (void *) sp,
+		    CLONE_VFORK | CLONE_VM | SIGCHLD,
+		    (void *)&tramp_data);
+	if (mm_id->pid < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "%s : clone failed, errno = %d\n",
+		       __func__, errno);
+		goto out_close;
+	}
+
+	if (using_seccomp) {
+		wait_stub_done_seccomp(mm_id, 1, 1);
+	} else {
+		do {
+			CATCH_EINTR(n = waitpid(mm_id->pid, &status,
+						WUNTRACED | __WALL));
+			if (n < 0) {
+				err = -errno;
+				printk(UM_KERN_ERR "%s : wait failed, errno = %d\n",
+				       __func__, errno);
+				goto out_kill;
+			}
+		} while (WIFSTOPPED(status) && (WSTOPSIG(status) == SIGALRM));
+
+		if (!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) {
+			err = -EINVAL;
+			printk(UM_KERN_ERR "%s : expected SIGSTOP, got status = %d\n",
+			       __func__, status);
+			goto out_kill;
+		}
+
+		if (ptrace(PTRACE_SETOPTIONS, mm_id->pid, NULL,
+			   (void *) PTRACE_O_TRACESYSGOOD) < 0) {
+			err = -errno;
+			printk(UM_KERN_ERR "%s : PTRACE_SETOPTIONS failed, errno = %d\n",
+			       __func__, errno);
+			goto out_kill;
+		}
+	}
+
+	if (munmap(stack, UM_KERN_PAGE_SIZE) < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "%s : munmap failed, errno = %d\n",
+		       __func__, errno);
+		goto out_kill;
+	}
+
+	close(tramp_data.sockpair[0]);
+	if (using_seccomp)
+		mm_id->sock = tramp_data.sockpair[1];
+	else
+		close(tramp_data.sockpair[1]);
+
+	return 0;
+
+out_kill:
+	os_kill_ptraced_process(mm_id->pid, 1);
+out_close:
+	close(tramp_data.sockpair[0]);
+	close(tramp_data.sockpair[1]);
+
+	mm_id->pid = -1;
+
+	return err;
+}
+
+static int unscheduled_userspace_iterations;
+extern unsigned long tt_extra_sched_jiffies;
+
+void userspace(struct uml_pt_regs *regs)
+{
+	int err, status, op;
+	siginfo_t si_local;
+	siginfo_t *si;
+	int sig;
+
+	/* Handle any immediate reschedules or signals */
+	interrupt_end();
+
+	while (1) {
+		struct mm_id *mm_id = current_mm_id();
+
+		/*
+		 * At any given time, only one CPU thread can enter the
+		 * turnstile to operate on the same stub process, including
+		 * executing stub system calls (mmap and munmap).
+		 */
+		enter_turnstile(mm_id);
+
+		/*
+		 * When we are in time-travel mode, userspace can theoretically
+		 * do a *lot* of work without being scheduled. The problem with
+		 * this is that it will prevent kernel bookkeeping (primarily
+		 * the RCU) from running and this can for example cause OOM
+		 * situations.
+		 *
+		 * This code accounts a jiffie against the scheduling clock
+		 * after the defined userspace iterations in the same thread.
+		 * By doing so the situation is effectively prevented.
+		 */
+		if (time_travel_mode == TT_MODE_INFCPU ||
+		    time_travel_mode == TT_MODE_EXTERNAL) {
+#ifdef CONFIG_UML_MAX_USERSPACE_ITERATIONS
+			if (CONFIG_UML_MAX_USERSPACE_ITERATIONS &&
+			    unscheduled_userspace_iterations++ >
+			    CONFIG_UML_MAX_USERSPACE_ITERATIONS) {
+				tt_extra_sched_jiffies += 1;
+				unscheduled_userspace_iterations = 0;
+			}
+#endif
+		}
+
+		time_travel_print_bc_msg();
+
+		current_mm_sync();
+
+		if (using_seccomp) {
+			struct stub_data *proc_data = (void *) mm_id->stack;
+
+			err = set_stub_state(regs, proc_data, singlestepping());
+			if (err) {
+				printk(UM_KERN_ERR "%s - failed to set regs: %d",
+				       __func__, err);
+				fatal_sigsegv();
+			}
+
+			/* Must have been reset by the syscall caller */
+			if (proc_data->restart_wait != 0)
+				panic("Programming error: Flag to only run syscalls in child was not cleared!");
+
+			/* Mark pending syscalls for flushing */
+			proc_data->syscall_data_len = mm_id->syscall_data_len;
+
+			wait_stub_done_seccomp(mm_id, 0, 0);
+
+			sig = proc_data->signal;
+
+			if (sig == SIGTRAP && proc_data->err != 0) {
+				printk(UM_KERN_ERR "%s - Error flushing stub syscalls",
+				       __func__);
+				syscall_stub_dump_error(mm_id);
+				mm_id->syscall_data_len = proc_data->err;
+				fatal_sigsegv();
+			}
+
+			mm_id->syscall_data_len = 0;
+			mm_id->syscall_fd_num = 0;
+
+			err = get_stub_state(regs, proc_data, NULL);
+			if (err) {
+				printk(UM_KERN_ERR "%s - failed to get regs: %d",
+				       __func__, err);
+				fatal_sigsegv();
+			}
+
+			if (proc_data->si_offset > sizeof(proc_data->sigstack) - sizeof(*si))
+				panic("%s - Invalid siginfo offset from child", __func__);
+
+			si = &si_local;
+			memcpy(si, &proc_data->sigstack[proc_data->si_offset], sizeof(*si));
+
+			regs->is_user = 1;
+
+			/* Fill in ORIG_RAX and extract fault information */
+			PT_SYSCALL_NR(regs->gp) = si->si_syscall;
+			if (sig == SIGSEGV) {
+				mcontext_t *mcontext = (void *)&proc_data->sigstack[proc_data->mctx_offset];
+
+				GET_FAULTINFO_FROM_MC(regs->faultinfo, mcontext);
+			}
+		} else {
+			int pid = mm_id->pid;
+
+			/* Flush out any pending syscalls */
+			err = syscall_stub_flush(mm_id);
+			if (err) {
+				if (err == -ENOMEM)
+					report_enomem();
+
+				printk(UM_KERN_ERR "%s - Error flushing stub syscalls: %d",
+					__func__, -err);
+				fatal_sigsegv();
+			}
+
+			/*
+			 * This can legitimately fail if the process loads a
+			 * bogus value into a segment register.  It will
+			 * segfault and PTRACE_GETREGS will read that value
+			 * out of the process.  However, PTRACE_SETREGS will
+			 * fail.  In this case, there is nothing to do but
+			 * just kill the process.
+			 */
+			if (ptrace(PTRACE_SETREGS, pid, 0, regs->gp)) {
+				printk(UM_KERN_ERR "%s - ptrace set regs failed, errno = %d\n",
+				       __func__, errno);
+				fatal_sigsegv();
+			}
+
+			if (put_fp_registers(pid, regs->fp)) {
+				printk(UM_KERN_ERR "%s - ptrace set fp regs failed, errno = %d\n",
+				       __func__, errno);
+				fatal_sigsegv();
+			}
+
+			if (singlestepping())
+				op = PTRACE_SYSEMU_SINGLESTEP;
+			else
+				op = PTRACE_SYSEMU;
+
+			if (ptrace(op, pid, 0, 0)) {
+				printk(UM_KERN_ERR "%s - ptrace continue failed, op = %d, errno = %d\n",
+				       __func__, op, errno);
+				fatal_sigsegv();
+			}
+
+			CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED | __WALL));
+			if (err < 0) {
+				printk(UM_KERN_ERR "%s - wait failed, errno = %d\n",
+				       __func__, errno);
+				fatal_sigsegv();
+			}
+
+			regs->is_user = 1;
+			if (ptrace(PTRACE_GETREGS, pid, 0, regs->gp)) {
+				printk(UM_KERN_ERR "%s - PTRACE_GETREGS failed, errno = %d\n",
+				       __func__, errno);
+				fatal_sigsegv();
+			}
+
+			if (get_fp_registers(pid, regs->fp)) {
+				printk(UM_KERN_ERR "%s -  get_fp_registers failed, errno = %d\n",
+				       __func__, errno);
+				fatal_sigsegv();
+			}
+
+			if (WIFSTOPPED(status)) {
+				sig = WSTOPSIG(status);
+
+				/*
+				 * These signal handlers need the si argument
+				 * and SIGSEGV needs the faultinfo.
+				 * The SIGIO and SIGALARM handlers which constitute
+				 * the majority of invocations, do not use it.
+				 */
+				switch (sig) {
+				case SIGSEGV:
+					get_skas_faultinfo(pid,
+							   &regs->faultinfo);
+					fallthrough;
+				case SIGTRAP:
+				case SIGILL:
+				case SIGBUS:
+				case SIGFPE:
+				case SIGWINCH:
+					ptrace(PTRACE_GETSIGINFO, pid, 0,
+					       (struct siginfo *)&si_local);
+					si = &si_local;
+					break;
+				default:
+					si = NULL;
+					break;
+				}
+			} else {
+				sig = 0;
+			}
+		}
+
+		exit_turnstile(mm_id);
+
+		UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */
+
+		if (sig) {
+			switch (sig) {
+			case SIGSEGV:
+				if (using_seccomp || PTRACE_FULL_FAULTINFO)
+					(*sig_info[SIGSEGV])(SIGSEGV,
+							     (struct siginfo *)si,
+							     regs, NULL);
+				else
+					segv(regs->faultinfo, 0, 1, NULL, NULL);
+
+				break;
+			case SIGSYS:
+				handle_syscall(regs);
+				break;
+			case SIGTRAP + 0x80:
+				handle_trap(regs);
+				break;
+			case SIGTRAP:
+				relay_signal(SIGTRAP, (struct siginfo *)si, regs, NULL);
+				break;
+			case SIGALRM:
+				break;
+			case SIGIO:
+			case SIGILL:
+			case SIGBUS:
+			case SIGFPE:
+			case SIGWINCH:
+				block_signals_trace();
+				(*sig_info[sig])(sig, (struct siginfo *)si, regs, NULL);
+				unblock_signals_trace();
+				break;
+			default:
+				printk(UM_KERN_ERR "%s - child stopped with signal %d\n",
+				       __func__, sig);
+				fatal_sigsegv();
+			}
+			interrupt_end();
+
+			/* Avoid -ERESTARTSYS handling in host */
+			if (PT_SYSCALL_NR_OFFSET != PT_SYSCALL_RET_OFFSET)
+				PT_SYSCALL_NR(regs->gp) = -1;
+		}
+	}
+}
+
+void new_thread(void *stack, jmp_buf *buf, void (*handler)(void))
+{
+	(*buf)[0].JB_IP = (unsigned long) handler;
+	(*buf)[0].JB_SP = (unsigned long) stack + UM_THREAD_SIZE -
+		sizeof(void *);
+}
+
+#define INIT_JMP_NEW_THREAD 0
+#define INIT_JMP_CALLBACK 1
+#define INIT_JMP_HALT 2
+#define INIT_JMP_REBOOT 3
+
+void switch_threads(jmp_buf *me, jmp_buf *you)
+{
+	unscheduled_userspace_iterations = 0;
+
+	if (UML_SETJMP(me) == 0)
+		UML_LONGJMP(you, 1);
+}
+
+static jmp_buf initial_jmpbuf;
+
+static __thread void (*cb_proc)(void *arg);
+static __thread void *cb_arg;
+static __thread jmp_buf *cb_back;
+
+int start_idle_thread(void *stack, jmp_buf *switch_buf)
+{
+	int n;
+
+	set_handler(SIGWINCH);
+
+	/*
+	 * Can't use UML_SETJMP or UML_LONGJMP here because they save
+	 * and restore signals, with the possible side-effect of
+	 * trying to handle any signals which came when they were
+	 * blocked, which can't be done on this stack.
+	 * Signals must be blocked when jumping back here and restored
+	 * after returning to the jumper.
+	 */
+	n = setjmp(initial_jmpbuf);
+	switch (n) {
+	case INIT_JMP_NEW_THREAD:
+		(*switch_buf)[0].JB_IP = (unsigned long) uml_finishsetup;
+		(*switch_buf)[0].JB_SP = (unsigned long) stack +
+			UM_THREAD_SIZE - sizeof(void *);
+		break;
+	case INIT_JMP_CALLBACK:
+		(*cb_proc)(cb_arg);
+		longjmp(*cb_back, 1);
+		break;
+	case INIT_JMP_HALT:
+		kmalloc_ok = 0;
+		return 0;
+	case INIT_JMP_REBOOT:
+		kmalloc_ok = 0;
+		return 1;
+	default:
+		printk(UM_KERN_ERR "Bad sigsetjmp return in %s - %d\n",
+		       __func__, n);
+		fatal_sigsegv();
+	}
+	longjmp(*switch_buf, 1);
+
+	/* unreachable */
+	printk(UM_KERN_ERR "impossible long jump!");
+	fatal_sigsegv();
+	return 0;
+}
+
+void initial_thread_cb_skas(void (*proc)(void *), void *arg)
+{
+	jmp_buf here;
+
+	cb_proc = proc;
+	cb_arg = arg;
+	cb_back = &here;
+
+	initial_jmpbuf_lock();
+	if (UML_SETJMP(&here) == 0)
+		UML_LONGJMP(&initial_jmpbuf, INIT_JMP_CALLBACK);
+	initial_jmpbuf_unlock();
+
+	cb_proc = NULL;
+	cb_arg = NULL;
+	cb_back = NULL;
+}
+
+void halt_skas(void)
+{
+	initial_jmpbuf_lock();
+	UML_LONGJMP(&initial_jmpbuf, INIT_JMP_HALT);
+	/* unreachable */
+}
+
+static bool noreboot;
+
+static int __init noreboot_cmd_param(char *str, int *add)
+{
+	*add = 0;
+	noreboot = true;
+	return 0;
+}
+
+__uml_setup("noreboot", noreboot_cmd_param,
+"noreboot\n"
+"    Rather than rebooting, exit always, akin to QEMU's -no-reboot option.\n"
+"    This is useful if you're using CONFIG_PANIC_TIMEOUT in order to catch\n"
+"    crashes in CI\n\n");
+
+void reboot_skas(void)
+{
+	initial_jmpbuf_lock();
+	UML_LONGJMP(&initial_jmpbuf, noreboot ? INIT_JMP_HALT : INIT_JMP_REBOOT);
+	/* unreachable */
+}

linux-7.2.3/arch/um/os-Linux/smp.c

@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025 Ant Group
+ * Author: Tiwei Bie <tiwei.btw@antgroup.com>
+ */
+
+#include <errno.h>
+#include <pthread.h>
+#include <signal.h>
+#include <kern_util.h>
+#include <um_malloc.h>
+#include <init.h>
+#include <os.h>
+#include <smp.h>
+#include "internal.h"
+
+struct cpu_thread_data {
+	int cpu;
+	sigset_t sigset;
+};
+
+static __thread int __curr_cpu;
+
+int uml_curr_cpu(void)
+{
+	return __curr_cpu;
+}
+
+static pthread_t cpu_threads[CONFIG_NR_CPUS];
+
+static void *cpu_thread(void *arg)
+{
+	struct cpu_thread_data *data = arg;
+
+	__curr_cpu = data->cpu;
+
+	uml_start_secondary(data);
+
+	return NULL;
+}
+
+int os_start_cpu_thread(int cpu)
+{
+	struct cpu_thread_data *data;
+	sigset_t sigset, oset;
+	int err;
+
+	data = uml_kmalloc(sizeof(*data), UM_GFP_ATOMIC);
+	if (!data)
+		return -ENOMEM;
+
+	sigfillset(&sigset);
+	if (sigprocmask(SIG_SETMASK, &sigset, &oset) < 0) {
+		err = errno;
+		goto err;
+	}
+
+	data->cpu = cpu;
+	data->sigset = oset;
+
+	err = pthread_create(&cpu_threads[cpu], NULL, cpu_thread, data);
+	if (sigprocmask(SIG_SETMASK, &oset, NULL) < 0)
+		panic("Failed to restore the signal mask, errno = %d", errno);
+	if (err != 0)
+		goto err;
+
+	return 0;
+
+err:
+	kfree(data);
+	return -err;
+}
+
+void os_start_secondary(void *arg, jmp_buf *switch_buf)
+{
+	struct cpu_thread_data *data = arg;
+
+	sigaddset(&data->sigset, IPI_SIGNAL);
+	sigaddset(&data->sigset, SIGIO);
+
+	if (sigprocmask(SIG_SETMASK, &data->sigset, NULL) < 0)
+		panic("Failed to restore the signal mask, errno = %d", errno);
+
+	kfree(data);
+	longjmp(*switch_buf, 1);
+
+	/* unreachable */
+	printk(UM_KERN_ERR "impossible long jump!");
+	fatal_sigsegv();
+}
+
+int os_send_ipi(int cpu, int vector)
+{
+	union sigval value = { .sival_int = vector };
+
+	return pthread_sigqueue(cpu_threads[cpu], IPI_SIGNAL, value);
+}
+
+static void __local_ipi_set(int enable)
+{
+	sigset_t sigset;
+
+	sigemptyset(&sigset);
+	sigaddset(&sigset, IPI_SIGNAL);
+
+	if (sigprocmask(enable ? SIG_UNBLOCK : SIG_BLOCK, &sigset, NULL) < 0)
+		panic("%s: sigprocmask failed, errno = %d", __func__, errno);
+}
+
+void os_local_ipi_enable(void)
+{
+	__local_ipi_set(1);
+}
+
+void os_local_ipi_disable(void)
+{
+	__local_ipi_set(0);
+}
+
+static void ipi_sig_handler(int sig, siginfo_t *si, void *uc)
+{
+	int save_errno = errno;
+
+	signals_enabled = 0;
+	um_trace_signals_off();
+
+	uml_ipi_handler(si->si_value.sival_int);
+
+	um_trace_signals_on();
+	signals_enabled = 1;
+
+	errno = save_errno;
+}
+
+void __init os_init_smp(void)
+{
+	struct sigaction action = {
+		.sa_sigaction = ipi_sig_handler,
+		.sa_flags = SA_SIGINFO | SA_ONSTACK | SA_RESTART,
+	};
+
+	sigfillset(&action.sa_mask);
+
+	if (sigaction(IPI_SIGNAL, &action, NULL) < 0)
+		panic("%s: sigaction failed, errno = %d", __func__, errno);
+
+	cpu_threads[0] = pthread_self();
+}

linux-7.2.3/arch/um/os-Linux/start_up.c

@@ -0,0 +1,495 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2021 Benjamin Berg <benjamin@sipsolutions.net>
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdarg.h>
+#include <unistd.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <sched.h>
+#include <signal.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <sys/time.h>
+#include <sys/resource.h>
+#include <asm/ldt.h>
+#include <asm/unistd.h>
+#include <init.h>
+#include <os.h>
+#include <smp.h>
+#include <kern_util.h>
+#include <mem_user.h>
+#include <ptrace_user.h>
+#include <stdbool.h>
+#include <stub-data.h>
+#include <sys/prctl.h>
+#include <linux/seccomp.h>
+#include <linux/filter.h>
+#include <sysdep/mcontext.h>
+#include <sysdep/stub.h>
+#include <registers.h>
+#include <skas.h>
+#include "internal.h"
+
+static void ptrace_child(void)
+{
+	int ret;
+	/* Calling os_getpid because some libcs cached getpid incorrectly */
+	int pid = os_getpid(), ppid = getppid();
+	int sc_result;
+
+	if (change_sig(SIGWINCH, 0) < 0 ||
+	    ptrace(PTRACE_TRACEME, 0, 0, 0) < 0) {
+		perror("ptrace");
+		kill(pid, SIGKILL);
+	}
+	kill(pid, SIGSTOP);
+
+	/*
+	 * This syscall will be intercepted by the parent. Don't call more than
+	 * once, please.
+	 */
+	sc_result = os_getpid();
+
+	if (sc_result == pid)
+		/* Nothing modified by the parent, we are running normally. */
+		ret = 1;
+	else if (sc_result == ppid)
+		/*
+		 * Expected in check_ptrace and check_sysemu when they succeed
+		 * in modifying the stack frame
+		 */
+		ret = 0;
+	else
+		/* Serious trouble! This could be caused by a bug in host 2.6
+		 * SKAS3/2.6 patch before release -V6, together with a bug in
+		 * the UML code itself.
+		 */
+		ret = 2;
+
+	exit(ret);
+}
+
+static void fatal_perror(const char *str)
+{
+	perror(str);
+	exit(1);
+}
+
+static void fatal(char *fmt, ...)
+{
+	va_list list;
+
+	va_start(list, fmt);
+	vfprintf(stderr, fmt, list);
+	va_end(list);
+
+	exit(1);
+}
+
+static void non_fatal(char *fmt, ...)
+{
+	va_list list;
+
+	va_start(list, fmt);
+	vfprintf(stderr, fmt, list);
+	va_end(list);
+}
+
+static int start_ptraced_child(void)
+{
+	int pid, n, status;
+
+	fflush(stdout);
+
+	pid = fork();
+	if (pid == 0)
+		ptrace_child();
+	else if (pid < 0)
+		fatal_perror("start_ptraced_child : fork failed");
+
+	CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
+	if (n < 0)
+		fatal_perror("check_ptrace : waitpid failed");
+	if (!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP))
+		fatal("check_ptrace : expected SIGSTOP, got status = %d",
+		      status);
+
+	return pid;
+}
+
+static void stop_ptraced_child(int pid, int exitcode)
+{
+	int status, n;
+
+	if (ptrace(PTRACE_CONT, pid, 0, 0) < 0)
+		fatal_perror("stop_ptraced_child : ptrace failed");
+
+	CATCH_EINTR(n = waitpid(pid, &status, 0));
+	if (!WIFEXITED(status) || (WEXITSTATUS(status) != exitcode)) {
+		int exit_with = WEXITSTATUS(status);
+		fatal("stop_ptraced_child : child exited with exitcode %d, "
+		      "while expecting %d; status 0x%x\n", exit_with,
+		      exitcode, status);
+	}
+}
+
+static void __init check_sysemu(void)
+{
+	int pid, n, status, count=0;
+
+	os_info("Checking syscall emulation for ptrace...");
+	pid = start_ptraced_child();
+
+	if ((ptrace(PTRACE_SETOPTIONS, pid, 0,
+		   (void *) PTRACE_O_TRACESYSGOOD) < 0))
+		fatal_perror("check_sysemu: PTRACE_SETOPTIONS failed");
+
+	while (1) {
+		count++;
+		if (ptrace(PTRACE_SYSEMU_SINGLESTEP, pid, 0, 0) < 0)
+			goto fail;
+		CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
+		if (n < 0)
+			fatal_perror("check_sysemu: wait failed");
+
+		if (WIFSTOPPED(status) &&
+		    (WSTOPSIG(status) == (SIGTRAP|0x80))) {
+			if (!count) {
+				non_fatal("check_sysemu: SYSEMU_SINGLESTEP "
+					  "doesn't singlestep");
+				goto fail;
+			}
+			n = ptrace(PTRACE_POKEUSER, pid, PT_SYSCALL_RET_OFFSET,
+				   os_getpid());
+			if (n < 0)
+				fatal_perror("check_sysemu : failed to modify "
+					     "system call return");
+			break;
+		}
+		else if (WIFSTOPPED(status) && (WSTOPSIG(status) == SIGTRAP))
+			count++;
+		else {
+			non_fatal("check_sysemu: expected SIGTRAP or "
+				  "(SIGTRAP | 0x80), got status = %d\n",
+				  status);
+			goto fail;
+		}
+	}
+	stop_ptraced_child(pid, 0);
+
+	os_info("OK\n");
+
+	return;
+
+fail:
+	stop_ptraced_child(pid, 1);
+	fatal("missing\n");
+}
+
+static void __init check_ptrace(void)
+{
+	int pid, syscall, n, status;
+
+	os_info("Checking that ptrace can change system call numbers...");
+	pid = start_ptraced_child();
+
+	if ((ptrace(PTRACE_SETOPTIONS, pid, 0,
+		   (void *) PTRACE_O_TRACESYSGOOD) < 0))
+		fatal_perror("check_ptrace: PTRACE_SETOPTIONS failed");
+
+	while (1) {
+		if (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0)
+			fatal_perror("check_ptrace : ptrace failed");
+
+		CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
+		if (n < 0)
+			fatal_perror("check_ptrace : wait failed");
+
+		if (!WIFSTOPPED(status) ||
+		   (WSTOPSIG(status) != (SIGTRAP | 0x80)))
+			fatal("check_ptrace : expected (SIGTRAP|0x80), "
+			       "got status = %d", status);
+
+		syscall = ptrace(PTRACE_PEEKUSER, pid, PT_SYSCALL_NR_OFFSET,
+				 0);
+		if (syscall == __NR_getpid) {
+			n = ptrace(PTRACE_POKEUSER, pid, PT_SYSCALL_NR_OFFSET,
+				   __NR_getppid);
+			if (n < 0)
+				fatal_perror("check_ptrace : failed to modify "
+					     "system call");
+			break;
+		}
+	}
+	stop_ptraced_child(pid, 0);
+	os_info("OK\n");
+	check_sysemu();
+}
+
+extern unsigned long host_fp_size;
+extern unsigned long exec_regs[MAX_REG_NR];
+extern unsigned long *exec_fp_regs;
+
+__initdata static struct stub_data *seccomp_test_stub_data;
+
+static void __init sigsys_handler(int sig, siginfo_t *info, void *p)
+{
+	ucontext_t *uc = p;
+
+	/* Stow away the location of the mcontext in the stack */
+	seccomp_test_stub_data->mctx_offset = (unsigned long)&uc->uc_mcontext -
+					      (unsigned long)&seccomp_test_stub_data->sigstack[0];
+
+	/* Prevent libc from clearing memory (mctx_offset in particular) */
+	syscall(__NR_exit, 0);
+}
+
+static int __init seccomp_helper(void *data)
+{
+	static struct sock_filter filter[] = {
+		BPF_STMT(BPF_LD | BPF_W | BPF_ABS,
+			 offsetof(struct seccomp_data, nr)),
+		BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_clock_nanosleep, 1, 0),
+		BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_ALLOW),
+		BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_TRAP),
+	};
+	static struct sock_fprog prog = {
+		.len = ARRAY_SIZE(filter),
+		.filter = filter,
+	};
+	struct sigaction sa;
+
+	/* close_range is needed for the stub */
+	if (stub_syscall3(__NR_close_range, 1, ~0U, 0))
+		exit(1);
+
+	set_sigstack(seccomp_test_stub_data->sigstack,
+			sizeof(seccomp_test_stub_data->sigstack));
+
+	sa.sa_flags = SA_ONSTACK | SA_NODEFER | SA_SIGINFO;
+	sa.sa_sigaction = (void *) sigsys_handler;
+	sa.sa_restorer = NULL;
+	if (sigaction(SIGSYS, &sa, NULL) < 0)
+		exit(2);
+
+	prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
+	if (syscall(__NR_seccomp, SECCOMP_SET_MODE_FILTER,
+			SECCOMP_FILTER_FLAG_TSYNC, &prog) != 0)
+		exit(3);
+
+	sleep(0);
+
+	/* Never reached. */
+	_exit(4);
+}
+
+static bool __init init_seccomp(void)
+{
+	int pid;
+	int status;
+	int n;
+	unsigned long sp;
+
+	/*
+	 * We check that we can install a seccomp filter and then exit(0)
+	 * from a trapped syscall.
+	 *
+	 * Note that we cannot verify that no seccomp filter already exists
+	 * for a syscall that results in the process/thread to be killed.
+	 */
+
+	os_info("Checking that seccomp filters can be installed...");
+
+	seccomp_test_stub_data = mmap(0, sizeof(*seccomp_test_stub_data),
+				      PROT_READ | PROT_WRITE,
+				      MAP_SHARED | MAP_ANON, 0, 0);
+
+	/* Use the syscall data area as stack, we just need something */
+	sp = (unsigned long)&seccomp_test_stub_data->syscall_data +
+	     sizeof(seccomp_test_stub_data->syscall_data) -
+	     sizeof(void *);
+	pid = clone(seccomp_helper, (void *)sp, CLONE_VFORK | CLONE_VM, NULL);
+
+	if (pid < 0)
+		fatal_perror("check_seccomp : clone failed");
+
+	CATCH_EINTR(n = waitpid(pid, &status, __WCLONE));
+	if (n < 0)
+		fatal_perror("check_seccomp : waitpid failed");
+
+	if (WIFEXITED(status) && WEXITSTATUS(status) == 0) {
+		struct uml_pt_regs *regs;
+		unsigned long fp_size;
+		int r;
+
+		/* Fill in the host_fp_size from the mcontext. */
+		regs = calloc(1, sizeof(struct uml_pt_regs));
+		get_stub_state(regs, seccomp_test_stub_data, &fp_size);
+		host_fp_size = fp_size;
+		free(regs);
+
+		/* Repeat with the correct size */
+		regs = calloc(1, sizeof(struct uml_pt_regs) + host_fp_size);
+		r = get_stub_state(regs, seccomp_test_stub_data, NULL);
+
+		/* Store as the default startup registers */
+		exec_fp_regs = malloc(host_fp_size);
+		memcpy(exec_regs, regs->gp, sizeof(exec_regs));
+		memcpy(exec_fp_regs, regs->fp, host_fp_size);
+
+		munmap(seccomp_test_stub_data, sizeof(*seccomp_test_stub_data));
+
+		free(regs);
+
+		if (r) {
+			os_info("failed to fetch registers: %d\n", r);
+			return false;
+		}
+
+		os_info("OK\n");
+		return true;
+	}
+
+	if (WIFEXITED(status) && WEXITSTATUS(status) == 2)
+		os_info("missing\n");
+	else
+		os_info("error\n");
+
+	munmap(seccomp_test_stub_data, sizeof(*seccomp_test_stub_data));
+	return false;
+}
+
+
+static void __init check_coredump_limit(void)
+{
+	struct rlimit lim;
+	int err = getrlimit(RLIMIT_CORE, &lim);
+
+	if (err) {
+		perror("Getting core dump limit");
+		return;
+	}
+
+	os_info("Core dump limits :\n\tsoft - ");
+	if (lim.rlim_cur == RLIM_INFINITY)
+		os_info("NONE\n");
+	else
+		os_info("%llu\n", (unsigned long long)lim.rlim_cur);
+
+	os_info("\thard - ");
+	if (lim.rlim_max == RLIM_INFINITY)
+		os_info("NONE\n");
+	else
+		os_info("%llu\n", (unsigned long long)lim.rlim_max);
+}
+
+void  __init get_host_cpu_features(
+		void (*flags_helper_func)(char *line),
+		void (*cache_helper_func)(char *line))
+{
+	FILE *cpuinfo;
+	char *line = NULL;
+	size_t len = 0;
+	int done_parsing = 0;
+
+	cpuinfo = fopen("/proc/cpuinfo", "r");
+	if (cpuinfo == NULL) {
+		os_info("Failed to get host CPU features\n");
+	} else {
+		while ((getline(&line, &len, cpuinfo)) != -1) {
+			if (strstr(line, "flags")) {
+				flags_helper_func(line);
+				done_parsing++;
+			}
+			if (strstr(line, "cache_alignment")) {
+				cache_helper_func(line);
+				done_parsing++;
+			}
+			free(line);
+			line = NULL;
+			if (done_parsing > 1)
+				break;
+		}
+		fclose(cpuinfo);
+	}
+}
+
+static int seccomp_config __initdata;
+
+static int __init uml_seccomp_config(char *line, int *add)
+{
+	*add = 0;
+
+	if (strcmp(line, "off") == 0)
+		seccomp_config = 0;
+	else if (strcmp(line, "auto") == 0)
+		seccomp_config = 1;
+	else if (strcmp(line, "on") == 0)
+		seccomp_config = 2;
+	else
+		fatal("Invalid seccomp option '%s', expected on/auto/off\n",
+		      line);
+
+	return 0;
+}
+
+__uml_setup("seccomp=", uml_seccomp_config,
+"seccomp=<on/auto/off>\n"
+"    Configure whether or not SECCOMP is used. With SECCOMP, userspace\n"
+"    processes work collaboratively with the kernel instead of being\n"
+"    traced using ptrace. All syscalls from the application are caught and\n"
+"    redirected using a signal. This signal handler in turn is permitted to\n"
+"    do the selected set of syscalls to communicate with the UML kernel and\n"
+"    do the required memory management.\n"
+"\n"
+"    This method is overall faster than the ptrace based userspace, primarily\n"
+"    because it reduces the number of context switches for (minor) page faults.\n"
+"\n"
+"    However, the SECCOMP filter is not (yet) restrictive enough to prevent\n"
+"    userspace from reading and writing all physical memory. Userspace\n"
+"    processes could also trick the stub into disabling SIGALRM which\n"
+"    prevents it from being interrupted for scheduling purposes.\n"
+"\n"
+"    This is insecure and should only be used with a trusted userspace\n\n"
+);
+
+void __init os_early_checks(void)
+{
+	int pid;
+
+	/* Print out the core dump limits early */
+	check_coredump_limit();
+
+	/* Need to check this early because mmapping happens before the
+	 * kernel is running.
+	 */
+	check_tmpexec();
+
+	if (seccomp_config) {
+		if (init_seccomp()) {
+			using_seccomp = 1;
+			return;
+		}
+
+		if (seccomp_config == 2)
+			fatal("SECCOMP userspace requested but not functional!\n");
+	}
+
+	if (uml_ncpus > 1)
+		fatal("SMP is not supported with PTRACE userspace.\n");
+
+	using_seccomp = 0;
+	check_ptrace();
+
+	pid = start_ptraced_child();
+	if (init_pid_registers(pid))
+		fatal("Failed to initialize default registers");
+	stop_ptraced_child(pid, 1);
+}

linux-7.2.3/arch/um/os-Linux/time.c

@@ -0,0 +1,157 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2015 Anton Ivanov (aivanov@{brocade.com,kot-begemot.co.uk})
+ * Copyright (C) 2015 Thomas Meyer (thomas@m3y3r.de)
+ * Copyright (C) 2012-2014 Cisco Systems
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike{addtoit,linux.intel}.com)
+ */
+
+#include <stddef.h>
+#include <unistd.h>
+#include <errno.h>
+#include <signal.h>
+#include <time.h>
+#include <sys/signalfd.h>
+#include <sys/time.h>
+#include <kern_util.h>
+#include <os.h>
+#include <smp.h>
+#include <string.h>
+#include "internal.h"
+
+static timer_t event_high_res_timer[CONFIG_NR_CPUS] = { 0 };
+
+static inline long long timespec_to_ns(const struct timespec *ts)
+{
+	return ((long long) ts->tv_sec * UM_NSEC_PER_SEC) + ts->tv_nsec;
+}
+
+long long os_persistent_clock_emulation(void)
+{
+	struct timespec realtime_tp;
+
+	clock_gettime(CLOCK_REALTIME, &realtime_tp);
+	return timespec_to_ns(&realtime_tp);
+}
+
+#ifndef sigev_notify_thread_id
+#define sigev_notify_thread_id _sigev_un._tid
+#endif
+
+/**
+ * os_timer_create() - create an new posix (interval) timer
+ */
+int os_timer_create(void)
+{
+	int cpu = uml_curr_cpu();
+	timer_t *t = &event_high_res_timer[cpu];
+	struct sigevent sev = {
+		.sigev_notify = SIGEV_THREAD_ID,
+		.sigev_signo = SIGALRM,
+		.sigev_value.sival_ptr = t,
+		.sigev_notify_thread_id = gettid(),
+	};
+
+	if (timer_create(CLOCK_MONOTONIC, &sev, t) == -1)
+		return -1;
+
+	return 0;
+}
+
+int os_timer_set_interval(int cpu, unsigned long long nsecs)
+{
+	struct itimerspec its;
+
+	its.it_value.tv_sec = nsecs / UM_NSEC_PER_SEC;
+	its.it_value.tv_nsec = nsecs % UM_NSEC_PER_SEC;
+
+	its.it_interval.tv_sec = nsecs / UM_NSEC_PER_SEC;
+	its.it_interval.tv_nsec = nsecs % UM_NSEC_PER_SEC;
+
+	if (timer_settime(event_high_res_timer[cpu], 0, &its, NULL) == -1)
+		return -errno;
+
+	return 0;
+}
+
+int os_timer_one_shot(int cpu, unsigned long long nsecs)
+{
+	struct itimerspec its = {
+		.it_value.tv_sec = nsecs / UM_NSEC_PER_SEC,
+		.it_value.tv_nsec = nsecs % UM_NSEC_PER_SEC,
+
+		.it_interval.tv_sec = 0,
+		.it_interval.tv_nsec = 0, // we cheat here
+	};
+
+	timer_settime(event_high_res_timer[cpu], 0, &its, NULL);
+	return 0;
+}
+
+/**
+ * os_timer_disable() - disable the posix (interval) timer
+ * @cpu: the CPU for which the timer is to be disabled
+ */
+void os_timer_disable(int cpu)
+{
+	struct itimerspec its;
+
+	memset(&its, 0, sizeof(struct itimerspec));
+	timer_settime(event_high_res_timer[cpu], 0, &its, NULL);
+}
+
+long long os_nsecs(void)
+{
+	struct timespec ts;
+
+	clock_gettime(CLOCK_MONOTONIC,&ts);
+	return timespec_to_ns(&ts);
+}
+
+static __thread int wake_signals;
+
+void os_idle_prepare(void)
+{
+	sigset_t set;
+
+	sigemptyset(&set);
+	sigaddset(&set, SIGALRM);
+	sigaddset(&set, IPI_SIGNAL);
+
+	/*
+	 * We need to use signalfd rather than sigsuspend in idle sleep
+	 * because the IPI signal is a real-time signal that carries data,
+	 * and unlike handling SIGALRM, we cannot simply flag it in
+	 * signals_pending.
+	 */
+	wake_signals = signalfd(-1, &set, SFD_CLOEXEC);
+	if (wake_signals < 0)
+		panic("Failed to create signal FD, errno = %d", errno);
+}
+
+/**
+ * os_idle_sleep() - sleep until interrupted
+ */
+void os_idle_sleep(void)
+{
+	sigset_t set;
+
+	/*
+	 * Block SIGALRM while performing the need_resched check.
+	 * Note that, because IRQs are disabled, the IPI signal is
+	 * already blocked.
+	 */
+	sigemptyset(&set);
+	sigaddset(&set, SIGALRM);
+	sigprocmask(SIG_BLOCK, &set, NULL);
+
+	/*
+	 * Because disabling IRQs does not block SIGALRM, it is also
+	 * necessary to check for any pending timer alarms.
+	 */
+	if (!uml_need_resched() && !timer_alarm_pending())
+		os_poll(1, &wake_signals);
+
+	/* Restore the signal mask. */
+	sigprocmask(SIG_UNBLOCK, &set, NULL);
+}

linux-7.2.3/arch/um/os-Linux/tty.c

@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <kern_util.h>
+#include <os.h>
+
+struct grantpt_info {
+	int fd;
+	int res;
+	int err;
+};
+
+static void grantpt_cb(void *arg)
+{
+	struct grantpt_info *info = arg;
+
+	info->res = grantpt(info->fd);
+	info->err = errno;
+}
+
+int get_pty(void)
+{
+	struct grantpt_info info;
+	int fd, err;
+
+	fd = open("/dev/ptmx", O_RDWR);
+	if (fd < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "get_pty : Couldn't open /dev/ptmx - "
+		       "err = %d\n", errno);
+		return err;
+	}
+
+	info.fd = fd;
+	initial_thread_cb(grantpt_cb, &info);
+
+	if (info.res < 0) {
+		err = -info.err;
+		printk(UM_KERN_ERR "get_pty : Couldn't grant pty - "
+		       "errno = %d\n", -info.err);
+		goto out;
+	}
+
+	if (unlockpt(fd) < 0) {
+		err = -errno;
+		printk(UM_KERN_ERR "get_pty : Couldn't unlock pty - "
+		       "errno = %d\n", errno);
+		goto out;
+	}
+	return fd;
+out:
+	close(fd);
+	return err;
+}

linux-7.2.3/arch/um/os-Linux/umid.c

@@ -0,0 +1,409 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <dirent.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/stat.h>
+#include <init.h>
+#include <os.h>
+
+#define UML_DIR "~/.uml/"
+
+#define UMID_LEN 64
+
+/* Changed by set_umid, which is run early in boot */
+static char umid[UMID_LEN] = { 0 };
+
+/* Changed by set_uml_dir and make_uml_dir, which are run early in boot */
+static char *uml_dir = UML_DIR;
+
+static int __init make_uml_dir(void)
+{
+	char dir[512] = { '\0' };
+	int len, err;
+
+	if (*uml_dir == '~') {
+		char *home = getenv("HOME");
+
+		err = -ENOENT;
+		if (home == NULL) {
+			printk(UM_KERN_ERR
+				"%s: no value in environment for $HOME\n",
+				__func__);
+			goto err;
+		}
+		strscpy(dir, home);
+		uml_dir++;
+	}
+	strlcat(dir, uml_dir, sizeof(dir));
+	len = strlen(dir);
+	if (len > 0 && dir[len - 1] != '/')
+		strlcat(dir, "/", sizeof(dir));
+
+	err = -ENOMEM;
+	uml_dir = malloc(strlen(dir) + 1);
+	if (uml_dir == NULL) {
+		printk(UM_KERN_ERR "%s : malloc failed, errno = %d\n",
+			__func__, errno);
+		goto err;
+	}
+	strcpy(uml_dir, dir);
+
+	if ((mkdir(uml_dir, 0777) < 0) && (errno != EEXIST)) {
+		printk(UM_KERN_ERR "Failed to mkdir '%s': %s\n",
+			uml_dir, strerror(errno));
+		err = -errno;
+		goto err_free;
+	}
+	return 0;
+
+err_free:
+	free(uml_dir);
+err:
+	uml_dir = NULL;
+	return err;
+}
+
+/*
+ * Unlinks the files contained in @dir and then removes @dir.
+ * Doesn't handle directory trees, so it's not like rm -rf, but almost such. We
+ * ignore ENOENT errors for anything (they happen, strangely enough - possibly
+ * due to races between multiple dying UML threads).
+ */
+static int remove_files_and_dir(char *dir)
+{
+	DIR *directory;
+	struct dirent *ent;
+	int len;
+	char file[256];
+	int ret;
+
+	directory = opendir(dir);
+	if (directory == NULL) {
+		if (errno != ENOENT)
+			return -errno;
+		else
+			return 0;
+	}
+
+	while ((ent = readdir(directory)) != NULL) {
+		if (!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, ".."))
+			continue;
+		len = strlen(dir) + strlen("/") + strlen(ent->d_name) + 1;
+		if (len > sizeof(file)) {
+			ret = -E2BIG;
+			goto out;
+		}
+
+		sprintf(file, "%s/%s", dir, ent->d_name);
+		if (unlink(file) < 0 && errno != ENOENT) {
+			ret = -errno;
+			goto out;
+		}
+	}
+
+	if (rmdir(dir) < 0 && errno != ENOENT) {
+		ret = -errno;
+		goto out;
+	}
+
+	ret = 0;
+out:
+	closedir(directory);
+	return ret;
+}
+
+/*
+ * This says that there isn't already a user of the specified directory even if
+ * there are errors during the checking.  This is because if these errors
+ * happen, the directory is unusable by the pre-existing UML, so we might as
+ * well take it over.  This could happen either by
+ * 	the existing UML somehow corrupting its umid directory
+ * 	something other than UML sticking stuff in the directory
+ *	this boot racing with a shutdown of the other UML
+ * In any of these cases, the directory isn't useful for anything else.
+ *
+ * Boolean return: 1 if in use, 0 otherwise.
+ */
+static inline int is_umdir_used(char *dir)
+{
+	char pid[sizeof("nnnnnnnnn")], *end, *file;
+	int fd, p, n;
+	size_t filelen = strlen(dir) + sizeof("/pid") + 1;
+
+	file = malloc(filelen);
+	if (!file)
+		return -ENOMEM;
+
+	snprintf(file, filelen, "%s/pid", dir);
+
+	fd = open(file, O_RDONLY);
+	if (fd < 0) {
+		fd = -errno;
+		if (fd != -ENOENT) {
+			printk(UM_KERN_ERR "is_umdir_used : couldn't open pid "
+			       "file '%s', err = %d\n", file, -fd);
+		}
+		goto out;
+	}
+
+	n = read(fd, pid, sizeof(pid));
+	if (n < 0) {
+		printk(UM_KERN_ERR "is_umdir_used : couldn't read pid file "
+		       "'%s', err = %d\n", file, errno);
+		goto out_close;
+	} else if (n == 0) {
+		printk(UM_KERN_ERR "is_umdir_used : couldn't read pid file "
+		       "'%s', 0-byte read\n", file);
+		goto out_close;
+	}
+
+	p = strtoul(pid, &end, 0);
+	if (end == pid) {
+		printk(UM_KERN_ERR "is_umdir_used : couldn't parse pid file "
+		       "'%s', errno = %d\n", file, errno);
+		goto out_close;
+	}
+
+	if ((kill(p, 0) == 0) || (errno != ESRCH)) {
+		printk(UM_KERN_ERR "umid \"%s\" is already in use by pid %d\n",
+		       umid, p);
+		return 1;
+	}
+
+out_close:
+	close(fd);
+out:
+	free(file);
+	return 0;
+}
+
+/*
+ * Try to remove the directory @dir unless it's in use.
+ * Precondition: @dir exists.
+ * Returns 0 for success, < 0 for failure in removal or if the directory is in
+ * use.
+ */
+static int umdir_take_if_dead(char *dir)
+{
+	int ret;
+	if (is_umdir_used(dir))
+		return -EEXIST;
+
+	ret = remove_files_and_dir(dir);
+	if (ret) {
+		printk(UM_KERN_ERR "is_umdir_used - remove_files_and_dir "
+		       "failed with err = %d\n", ret);
+	}
+	return ret;
+}
+
+static void __init create_pid_file(void)
+{
+	char pid[sizeof("nnnnnnnnn")], *file;
+	int fd, n;
+
+	n = strlen(uml_dir) + UMID_LEN + sizeof("/pid");
+	file = malloc(n);
+	if (!file)
+		return;
+
+	if (umid_file_name("pid", file, n))
+		goto out;
+
+	fd = open(file, O_RDWR | O_CREAT | O_EXCL, 0644);
+	if (fd < 0) {
+		printk(UM_KERN_ERR "Open of machine pid file \"%s\" failed: "
+		       "%s\n", file, strerror(errno));
+		goto out;
+	}
+
+	snprintf(pid, sizeof(pid), "%d\n", getpid());
+	n = write(fd, pid, strlen(pid));
+	if (n != strlen(pid))
+		printk(UM_KERN_ERR "Write of pid file failed - err = %d\n",
+		       errno);
+
+	close(fd);
+out:
+	free(file);
+}
+
+int __init set_umid(char *name)
+{
+	if (strlen(name) > UMID_LEN - 1)
+		return -E2BIG;
+
+	strscpy(umid, name);
+
+	return 0;
+}
+
+/* Changed in make_umid, which is called during early boot */
+static int umid_setup = 0;
+
+static int __init make_umid(void)
+{
+	int fd, err;
+	char tmp[256];
+
+	if (umid_setup)
+		return 0;
+
+	make_uml_dir();
+
+	if (*umid == '\0') {
+		strscpy(tmp, uml_dir);
+		strlcat(tmp, "XXXXXX", sizeof(tmp));
+		fd = mkstemp(tmp);
+		if (fd < 0) {
+			printk(UM_KERN_ERR "make_umid - mkstemp(%s) failed: "
+			       "%s\n", tmp, strerror(errno));
+			err = -errno;
+			goto err;
+		}
+
+		close(fd);
+
+		set_umid(&tmp[strlen(uml_dir)]);
+
+		/*
+		 * There's a nice tiny little race between this unlink and
+		 * the mkdir below.  It'd be nice if there were a mkstemp
+		 * for directories.
+		 */
+		if (unlink(tmp)) {
+			err = -errno;
+			goto err;
+		}
+	}
+
+	snprintf(tmp, sizeof(tmp), "%s%s", uml_dir, umid);
+	err = mkdir(tmp, 0777);
+	if (err < 0) {
+		err = -errno;
+		if (err != -EEXIST)
+			goto err;
+
+		if (umdir_take_if_dead(tmp) < 0)
+			goto err;
+
+		err = mkdir(tmp, 0777);
+	}
+	if (err) {
+		err = -errno;
+		printk(UM_KERN_ERR "Failed to create '%s' - err = %d\n", umid,
+		       errno);
+		goto err;
+	}
+
+	umid_setup = 1;
+
+	create_pid_file();
+
+	err = 0;
+ err:
+	return err;
+}
+
+static int __init make_umid_init(void)
+{
+	if (!make_umid())
+		return 0;
+
+	/*
+	 * If initializing with the given umid failed, then try again with
+	 * a random one.
+	 */
+	printk(UM_KERN_ERR "Failed to initialize umid \"%s\", trying with a "
+	       "random umid\n", umid);
+	*umid = '\0';
+	make_umid();
+
+	return 0;
+}
+
+__initcall(make_umid_init);
+
+int __init umid_file_name(char *name, char *buf, int len)
+{
+	int n, err;
+
+	err = make_umid();
+	if (err)
+		return err;
+
+	n = snprintf(buf, len, "%s%s/%s", uml_dir, umid, name);
+	if (n >= len) {
+		printk(UM_KERN_ERR "umid_file_name : buffer too short\n");
+		return -E2BIG;
+	}
+
+	return 0;
+}
+
+char *get_umid(void)
+{
+	return umid;
+}
+
+static int __init set_uml_dir(char *name, int *add)
+{
+	*add = 0;
+
+	if (*name == '\0') {
+		os_warn("uml_dir can't be an empty string\n");
+		return 0;
+	}
+
+	if (name[strlen(name) - 1] == '/') {
+		uml_dir = name;
+		return 0;
+	}
+
+	uml_dir = malloc(strlen(name) + 2);
+	if (uml_dir == NULL) {
+		os_warn("Failed to malloc uml_dir - error = %d\n", errno);
+
+		/*
+		 * Return 0 here because do_initcalls doesn't look at
+		 * the return value.
+		 */
+		return 0;
+	}
+	sprintf(uml_dir, "%s/", name);
+
+	return 0;
+}
+
+__uml_setup("uml_dir=", set_uml_dir,
+"uml_dir=<directory>\n"
+"    The location to place the pid and umid files.\n\n"
+);
+
+static void remove_umid_dir(void)
+{
+	char *dir, err;
+
+	dir = malloc(strlen(uml_dir) + UMID_LEN + 1);
+	if (!dir)
+		return;
+
+	sprintf(dir, "%s%s", uml_dir, umid);
+	err = remove_files_and_dir(dir);
+	if (err)
+		os_warn("%s - remove_files_and_dir failed with err = %d\n",
+			__func__, err);
+
+	free(dir);
+}
+
+__uml_exitcall(remove_umid_dir);

linux-7.2.3/arch/um/os-Linux/user_syms.c

@@ -0,0 +1,37 @@
+// SPDX-License-Identifier: GPL-2.0
+#define __NO_FORTIFY
+#include <linux/types.h>
+#include <linux/module.h>
+
+/*
+ * This file exports some critical string functions and compiler
+ * built-in functions (where calls are emitted by the compiler
+ * itself that we cannot avoid even in kernel code) to modules.
+ *
+ * "_user.c" code that previously used exports here such as hostfs
+ * really should be considered part of the 'hypervisor' and define
+ * its own API boundary like hostfs does now; don't add exports to
+ * this file for such cases.
+ */
+
+/* If it's not defined, the export is included in lib/string.c.*/
+#ifdef __HAVE_ARCH_STRSTR
+#undef strstr
+EXPORT_SYMBOL(strstr);
+#endif
+
+#ifndef __x86_64__
+#undef memcpy
+extern void *memcpy(void *, const void *, size_t);
+EXPORT_SYMBOL(memcpy);
+extern void *memmove(void *, const void *, size_t);
+EXPORT_SYMBOL(memmove);
+#undef memset
+extern void *memset(void *, int, size_t);
+EXPORT_SYMBOL(memset);
+#endif
+
+#ifdef _FORTIFY_SOURCE
+extern int __sprintf_chk(char *str, int flag, size_t len, const char *format);
+EXPORT_SYMBOL(__sprintf_chk);
+#endif

linux-7.2.3/arch/um/os-Linux/util.c

@@ -0,0 +1,209 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
+ */
+
+#include <stdarg.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <errno.h>
+#include <signal.h>
+#include <string.h>
+#include <termios.h>
+#include <sys/wait.h>
+#include <sys/mman.h>
+#include <sys/utsname.h>
+#include <sys/random.h>
+#include <init.h>
+#include <os.h>
+
+void stack_protections(unsigned long address)
+{
+	if (mprotect((void *) address, UM_THREAD_SIZE, PROT_READ | PROT_WRITE) < 0)
+		panic("protecting stack failed, errno = %d", errno);
+}
+
+int raw(int fd)
+{
+	struct termios tt;
+	int err;
+
+	CATCH_EINTR(err = tcgetattr(fd, &tt));
+	if (err < 0)
+		return -errno;
+
+	cfmakeraw(&tt);
+
+	CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt));
+	if (err < 0)
+		return -errno;
+
+	/*
+	 * XXX tcsetattr could have applied only some changes
+	 * (and cfmakeraw() is a set of changes)
+	 */
+	return 0;
+}
+
+void setup_machinename(char *machine_out)
+{
+	struct utsname host;
+
+	uname(&host);
+#if IS_ENABLED(CONFIG_UML_X86)
+# if !IS_ENABLED(CONFIG_64BIT)
+	if (!strcmp(host.machine, "x86_64")) {
+		strcpy(machine_out, "i686");
+		return;
+	}
+# else
+	if (!strcmp(host.machine, "i686")) {
+		strcpy(machine_out, "x86_64");
+		return;
+	}
+# endif
+#endif
+	strcpy(machine_out, host.machine);
+}
+
+void setup_hostinfo(char *buf, int len)
+{
+	struct utsname host;
+
+	uname(&host);
+	snprintf(buf, len, "%s %s %s %s %s", host.sysname, host.nodename,
+		 host.release, host.version, host.machine);
+}
+
+/*
+ * We cannot use glibc's abort(). It makes use of tgkill() which
+ * has no effect within UML's kernel threads.
+ * After that glibc would execute an invalid instruction to kill
+ * the calling process and UML crashes with SIGSEGV.
+ */
+static inline void __attribute__ ((noreturn)) uml_abort(void)
+{
+	sigset_t sig;
+
+	fflush(NULL);
+
+	if (!sigemptyset(&sig) && !sigaddset(&sig, SIGABRT))
+		sigprocmask(SIG_UNBLOCK, &sig, 0);
+
+	for (;;)
+		if (kill(getpid(), SIGABRT) < 0)
+			exit(127);
+}
+
+ssize_t os_getrandom(void *buf, size_t len, unsigned int flags)
+{
+	return getrandom(buf, len, flags);
+}
+
+/*
+ * UML helper threads must not handle SIGWINCH/INT/TERM
+ */
+void os_fix_helper_signals(void)
+{
+	signal(SIGWINCH, SIG_IGN);
+	signal(SIGINT, SIG_DFL);
+	signal(SIGTERM, SIG_DFL);
+}
+
+void os_dump_core(void)
+{
+	int pid;
+
+	signal(SIGSEGV, SIG_DFL);
+
+	/*
+	 * We are about to SIGTERM this entire process group to ensure that
+	 * nothing is around to run after the kernel exits.  The
+	 * kernel wants to abort, not die through SIGTERM, so we
+	 * ignore it here.
+	 */
+
+	signal(SIGTERM, SIG_IGN);
+	kill(0, SIGTERM);
+	/*
+	 * Most of the other processes associated with this UML are
+	 * likely sTopped, so give them a SIGCONT so they see the
+	 * SIGTERM.
+	 */
+	kill(0, SIGCONT);
+
+	/*
+	 * Now, having sent signals to everyone but us, make sure they
+	 * die by ptrace.  Processes can survive what's been done to
+	 * them so far - the mechanism I understand is receiving a
+	 * SIGSEGV and segfaulting immediately upon return.  There is
+	 * always a SIGSEGV pending, and (I'm guessing) signals are
+	 * processed in numeric order so the SIGTERM (signal 15 vs
+	 * SIGSEGV being signal 11) is never handled.
+	 *
+	 * Run a waitpid loop until we get some kind of error.
+	 * Hopefully, it's ECHILD, but there's not a lot we can do if
+	 * it's something else.  Tell os_kill_ptraced_process not to
+	 * wait for the child to report its death because there's
+	 * nothing reasonable to do if that fails.
+	 */
+
+	while ((pid = waitpid(-1, NULL, WNOHANG | __WALL)) > 0)
+		os_kill_ptraced_process(pid, 0);
+
+	uml_abort();
+}
+
+void um_early_printk(const char *s, unsigned int n)
+{
+	printf("%.*s", n, s);
+}
+
+static int quiet_info;
+
+static int __init quiet_cmd_param(char *str, int *add)
+{
+	quiet_info = 1;
+	return 0;
+}
+
+__uml_setup("quiet", quiet_cmd_param,
+"quiet\n"
+"    Turns off information messages during boot.\n\n");
+
+/*
+ * The os_info/os_warn functions will be called by helper threads. These
+ * have a very limited stack size and using the libc formatting functions
+ * may overflow the stack.
+ * So pull in the kernel vscnprintf and use that instead with a fixed
+ * on-stack buffer.
+ */
+int vscnprintf(char *buf, size_t size, const char *fmt, va_list args);
+
+void os_info(const char *fmt, ...)
+{
+	char buf[256];
+	va_list list;
+	int len;
+
+	if (quiet_info)
+		return;
+
+	va_start(list, fmt);
+	len = vscnprintf(buf, sizeof(buf), fmt, list);
+	fwrite(buf, len, 1, stderr);
+	va_end(list);
+}
+
+void os_warn(const char *fmt, ...)
+{
+	char buf[256];
+	va_list list;
+	int len;
+
+	va_start(list, fmt);
+	len = vscnprintf(buf, sizeof(buf), fmt, list);
+	fwrite(buf, len, 1, stderr);
+	va_end(list);
+}