*/
#include "qemu/osdep.h"
-#include "qemu-common.h"
+#include <sys/resource.h>
+#include <sys/sysctl.h>
+
+#include "qemu/help-texts.h"
#include "qemu/units.h"
#include "qemu/accel.h"
-#include "sysemu/tcg.h"
#include "qemu-version.h"
#include <machine/trap.h>
#include "qemu/help_option.h"
#include "qemu/module.h"
#include "exec/exec-all.h"
-#include "tcg/tcg.h"
+#include "tcg/startup.h"
#include "qemu/timer.h"
#include "qemu/envlist.h"
#include "qemu/cutils.h"
#include "exec/log.h"
#include "trace/control.h"
+#include "crypto/init.h"
+#include "qemu/guest-random.h"
+#include "gdbstub/user.h"
#include "host-os.h"
#include "target_arch_cpu.h"
-#include <sys/sysctl.h>
-
-int singlestep;
-unsigned long mmap_min_addr;
+static bool opt_one_insn_per_tb;
uintptr_t guest_base;
bool have_guest_base;
+/*
+ * When running 32-on-64 we should make sure we can fit all of the possible
+ * guest address space into a contiguous chunk of virtual host memory.
+ *
+ * This way we will never overlap with our own libraries or binaries or stack
+ * or anything else that QEMU maps.
+ *
+ * Many cpus reserve the high bit (or more than one for some 64-bit cpus)
+ * of the address for the kernel. Some cpus rely on this and user space
+ * uses the high bit(s) for pointer tagging and the like. For them, we
+ * must preserve the expected address space.
+ */
+#ifndef MAX_RESERVED_VA
+# if HOST_LONG_BITS > TARGET_VIRT_ADDR_SPACE_BITS
+# if TARGET_VIRT_ADDR_SPACE_BITS == 32 && \
+ (TARGET_LONG_BITS == 32 || defined(TARGET_ABI32))
+# define MAX_RESERVED_VA 0xfffffffful
+# else
+# define MAX_RESERVED_VA ((1ul << TARGET_VIRT_ADDR_SPACE_BITS) - 1)
+# endif
+# else
+# define MAX_RESERVED_VA 0
+# endif
+#endif
+
+/*
+ * That said, reserving *too* much vm space via mmap can run into problems
+ * with rlimits, oom due to page table creation, etc. We will still try it,
+ * if directed by the command-line option, but not by default.
+ */
+#if HOST_LONG_BITS == 64 && TARGET_VIRT_ADDR_SPACE_BITS <= 32
+unsigned long reserved_va = MAX_RESERVED_VA;
+#else
unsigned long reserved_va;
+#endif
-static const char *interp_prefix = CONFIG_QEMU_INTERP_PREFIX;
+const char *interp_prefix = CONFIG_QEMU_INTERP_PREFIX;
const char *qemu_uname_release;
-enum BSDType bsd_type;
char qemu_proc_pathname[PATH_MAX]; /* full path to exeutable */
-/*
- * XXX: on x86 MAP_GROWSDOWN only works if ESP <= address + 32, so
- * we allocate a bigger stack. Need a better solution, for example
- * by remapping the process stack directly at the right place
- */
-unsigned long x86_stack_size = 512 * 1024;
-
-void gemu_log(const char *fmt, ...)
-{
- va_list ap;
+unsigned long target_maxtsiz = TARGET_MAXTSIZ; /* max text size */
+unsigned long target_dfldsiz = TARGET_DFLDSIZ; /* initial data size limit */
+unsigned long target_maxdsiz = TARGET_MAXDSIZ; /* max data size */
+unsigned long target_dflssiz = TARGET_DFLSSIZ; /* initial data size limit */
+unsigned long target_maxssiz = TARGET_MAXSSIZ; /* max stack size */
+unsigned long target_sgrowsiz = TARGET_SGROWSIZ; /* amount to grow stack */
- va_start(ap, fmt);
- vfprintf(stderr, fmt, ap);
- va_end(ap);
-}
+/* Helper routines for implementing atomic operations. */
void fork_start(void)
{
+ start_exclusive();
+ cpu_list_lock();
+ mmap_fork_start();
}
void fork_end(int child)
{
if (child) {
+ CPUState *cpu, *next_cpu;
+ /*
+ * Child processes created by fork() only have a single thread. Discard
+ * information about the parent threads.
+ */
+ CPU_FOREACH_SAFE(cpu, next_cpu) {
+ if (cpu != thread_cpu) {
+ QTAILQ_REMOVE_RCU(&cpus_queue, cpu, node);
+ }
+ }
+ mmap_fork_end(child);
+ /*
+ * qemu_init_cpu_list() takes care of reinitializing the exclusive
+ * state, so we don't need to end_exclusive() here.
+ */
+ qemu_init_cpu_list();
gdbserver_fork(thread_cpu);
+ } else {
+ mmap_fork_end(child);
+ cpu_list_unlock();
+ end_exclusive();
}
}
"-E var=value sets/modifies targets environment variable(s)\n"
"-U var unsets targets environment variable(s)\n"
"-B address set guest_base address to address\n"
- "-bsd type select emulated BSD type FreeBSD/NetBSD/OpenBSD (default)\n"
"\n"
"Debug options:\n"
"-d item1[,...] enable logging of specified items\n"
" (use '-d help' for a list of log items)\n"
"-D logfile write logs to 'logfile' (default stderr)\n"
- "-p pagesize set the host page size to 'pagesize'\n"
- "-singlestep always run in singlestep mode\n"
+ "-one-insn-per-tb run with one guest instruction per emulated TB\n"
"-strace log system calls\n"
"-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
" specify tracing options\n"
,
TARGET_NAME,
interp_prefix,
- x86_stack_size);
+ target_dflssiz);
exit(1);
}
__thread CPUState *thread_cpu;
+void stop_all_tasks(void)
+{
+ /*
+ * We trust when using NPTL (pthreads) start_exclusive() handles thread
+ * stopping correctly.
+ */
+ start_exclusive();
+}
+
bool qemu_cpu_is_self(CPUState *cpu)
{
return thread_cpu == cpu;
}
/* Assumes contents are already zeroed. */
-void init_task_state(TaskState *ts)
+static void init_task_state(TaskState *ts)
+{
+ ts->sigaltstack_used = (struct target_sigaltstack) {
+ .ss_sp = 0,
+ .ss_size = 0,
+ .ss_flags = TARGET_SS_DISABLE,
+ };
+}
+
+void gemu_log(const char *fmt, ...)
+{
+ va_list ap;
+
+ va_start(ap, fmt);
+ vfprintf(stderr, fmt, ap);
+ va_end(ap);
+}
+
+static void
+adjust_ssize(void)
{
- int i;
+ struct rlimit rl;
- ts->used = 1;
- ts->first_free = ts->sigqueue_table;
- for (i = 0; i < MAX_SIGQUEUE_SIZE - 1; i++) {
- ts->sigqueue_table[i].next = &ts->sigqueue_table[i + 1];
+ if (getrlimit(RLIMIT_STACK, &rl) != 0) {
+ return;
}
- ts->sigqueue_table[i].next = NULL;
+
+ target_maxssiz = MIN(target_maxssiz, rl.rlim_max);
+ target_dflssiz = MIN(MAX(target_dflssiz, rl.rlim_cur), target_maxssiz);
+
+ rl.rlim_max = target_maxssiz;
+ rl.rlim_cur = target_dflssiz;
+ setrlimit(RLIMIT_STACK, &rl);
}
static void save_proc_pathname(char *argv0)
const char *cpu_type;
const char *log_file = NULL;
const char *log_mask = NULL;
+ const char *seed_optarg = NULL;
struct target_pt_regs regs1, *regs = ®s1;
struct image_info info1, *info = &info1;
struct bsd_binprm bprm;
const char *gdbstub = NULL;
char **target_environ, **wrk;
envlist_t *envlist = NULL;
- bsd_type = HOST_DEFAULT_BSD_TYPE;
+ char *argv0 = NULL;
+
+ adjust_ssize();
if (argc <= 1) {
usage();
envlist = envlist_create();
- /* add current environment into the list */
+ /*
+ * add current environment into the list
+ * envlist_setenv adds to the front of the list; to preserve environ
+ * order add from back to front
+ */
for (wrk = environ; *wrk != NULL; wrk++) {
+ continue;
+ }
+ while (wrk != environ) {
+ wrk--;
(void) envlist_setenv(envlist, *wrk);
}
}
} else if (!strcmp(r, "s")) {
r = argv[optind++];
- rv = qemu_strtoul(r, &r, 0, &x86_stack_size);
- if (rv < 0 || x86_stack_size <= 0) {
+ rv = qemu_strtoul(r, &r, 0, &target_dflssiz);
+ if (rv < 0 || target_dflssiz <= 0) {
usage();
}
if (*r == 'M') {
- x86_stack_size *= MiB;
+ target_dflssiz *= 1024 * 1024;
} else if (*r == 'k' || *r == 'K') {
- x86_stack_size *= KiB;
+ target_dflssiz *= 1024;
+ }
+ if (target_dflssiz > target_maxssiz) {
+ usage();
}
} else if (!strcmp(r, "L")) {
interp_prefix = argv[optind++];
} else if (!strcmp(r, "cpu")) {
cpu_model = argv[optind++];
if (is_help_option(cpu_model)) {
- /* XXX: implement xxx_cpu_list for targets that still miss it */
-#if defined(cpu_list)
- cpu_list();
-#endif
+ list_cpus();
exit(1);
}
} else if (!strcmp(r, "B")) {
have_guest_base = true;
} else if (!strcmp(r, "drop-ld-preload")) {
(void) envlist_unsetenv(envlist, "LD_PRELOAD");
- } else if (!strcmp(r, "bsd")) {
- if (!strcasecmp(argv[optind], "freebsd")) {
- bsd_type = target_freebsd;
- } else if (!strcasecmp(argv[optind], "netbsd")) {
- bsd_type = target_netbsd;
- } else if (!strcasecmp(argv[optind], "openbsd")) {
- bsd_type = target_openbsd;
- } else {
- usage();
- }
- optind++;
- } else if (!strcmp(r, "singlestep")) {
- singlestep = 1;
+ } else if (!strcmp(r, "seed")) {
+ seed_optarg = optarg;
+ } else if (!strcmp(r, "one-insn-per-tb")) {
+ opt_one_insn_per_tb = true;
} else if (!strcmp(r, "strace")) {
do_strace = 1;
} else if (!strcmp(r, "trace")) {
trace_opt_parse(optarg);
+ } else if (!strcmp(r, "0")) {
+ argv0 = argv[optind++];
} else {
usage();
}
}
/* init debug */
- qemu_log_needs_buffers();
- qemu_set_log_filename(log_file, &error_fatal);
- if (log_mask) {
- int mask;
-
- mask = qemu_str_to_log_mask(log_mask);
- if (!mask) {
- qemu_print_log_usage(stdout);
- exit(1);
+ {
+ int mask = 0;
+ if (log_mask) {
+ mask = qemu_str_to_log_mask(log_mask);
+ if (!mask) {
+ qemu_print_log_usage(stdout);
+ exit(1);
+ }
}
- qemu_set_log(mask);
+ qemu_set_log_filename_flags(log_file, mask, &error_fatal);
}
if (optind >= argc) {
usage();
}
filename = argv[optind];
+ if (argv0) {
+ argv[optind] = argv0;
+ }
if (!trace_init_backends()) {
exit(1);
/* init tcg before creating CPUs and to get qemu_host_page_size */
{
- AccelClass *ac = ACCEL_GET_CLASS(current_accel());
+ AccelState *accel = current_accel();
+ AccelClass *ac = ACCEL_GET_CLASS(accel);
accel_init_interfaces(ac);
+ object_property_set_bool(OBJECT(accel), "one-insn-per-tb",
+ opt_one_insn_per_tb, &error_abort);
ac->init_machine(NULL);
}
cpu = cpu_create(cpu_type);
- env = cpu->env_ptr;
+ env = cpu_env(cpu);
cpu_reset(cpu);
thread_cpu = cpu;
target_environ = envlist_to_environ(envlist, NULL);
envlist_free(envlist);
+ {
+ Error *err = NULL;
+ if (seed_optarg != NULL) {
+ qemu_guest_random_seed_main(seed_optarg, &err);
+ } else {
+ qcrypto_init(&err);
+ }
+ if (err) {
+ error_reportf_err(err, "cannot initialize crypto: ");
+ exit(1);
+ }
+ }
+
/*
* Now that page sizes are configured we can do
* proper page alignment for guest_base.
*/
- guest_base = HOST_PAGE_ALIGN(guest_base);
+ if (have_guest_base) {
+ if (guest_base & ~qemu_host_page_mask) {
+ error_report("Selected guest base not host page aligned");
+ exit(1);
+ }
+ }
+
+ /*
+ * If reserving host virtual address space, do so now.
+ * Combined with '-B', ensure that the chosen range is free.
+ */
+ if (reserved_va) {
+ void *p;
+
+ if (have_guest_base) {
+ p = mmap((void *)guest_base, reserved_va + 1, PROT_NONE,
+ MAP_ANON | MAP_PRIVATE | MAP_FIXED | MAP_EXCL, -1, 0);
+ } else {
+ p = mmap(NULL, reserved_va + 1, PROT_NONE,
+ MAP_ANON | MAP_PRIVATE, -1, 0);
+ }
+ if (p == MAP_FAILED) {
+ const char *err = strerror(errno);
+ char *sz = size_to_str(reserved_va + 1);
+
+ if (have_guest_base) {
+ error_report("Cannot allocate %s bytes at -B %p for guest "
+ "address space: %s", sz, (void *)guest_base, err);
+ } else {
+ error_report("Cannot allocate %s bytes for guest "
+ "address space: %s", sz, err);
+ }
+ exit(1);
+ }
+ guest_base = (uintptr_t)p;
+ have_guest_base = true;
+
+ /* Ensure that mmap_next_start is within range. */
+ if (reserved_va <= mmap_next_start) {
+ mmap_next_start = (reserved_va / 4 * 3)
+ & TARGET_PAGE_MASK & qemu_host_page_mask;
+ }
+ }
if (loader_exec(filename, argv + optind, target_environ, regs, info,
&bprm) != 0) {
g_free(target_environ);
if (qemu_loglevel_mask(CPU_LOG_PAGE)) {
- qemu_log("guest_base %p\n", (void *)guest_base);
- log_page_dump("binary load");
-
- qemu_log("start_brk 0x" TARGET_ABI_FMT_lx "\n", info->start_brk);
- qemu_log("end_code 0x" TARGET_ABI_FMT_lx "\n", info->end_code);
- qemu_log("start_code 0x" TARGET_ABI_FMT_lx "\n",
- info->start_code);
- qemu_log("start_data 0x" TARGET_ABI_FMT_lx "\n",
- info->start_data);
- qemu_log("end_data 0x" TARGET_ABI_FMT_lx "\n", info->end_data);
- qemu_log("start_stack 0x" TARGET_ABI_FMT_lx "\n",
- info->start_stack);
- qemu_log("brk 0x" TARGET_ABI_FMT_lx "\n", info->brk);
- qemu_log("entry 0x" TARGET_ABI_FMT_lx "\n", info->entry);
+ FILE *f = qemu_log_trylock();
+ if (f) {
+ fprintf(f, "guest_base %p\n", (void *)guest_base);
+ fprintf(f, "page layout changed following binary load\n");
+ page_dump(f);
+
+ fprintf(f, "end_code 0x" TARGET_ABI_FMT_lx "\n",
+ info->end_code);
+ fprintf(f, "start_code 0x" TARGET_ABI_FMT_lx "\n",
+ info->start_code);
+ fprintf(f, "start_data 0x" TARGET_ABI_FMT_lx "\n",
+ info->start_data);
+ fprintf(f, "end_data 0x" TARGET_ABI_FMT_lx "\n",
+ info->end_data);
+ fprintf(f, "start_stack 0x" TARGET_ABI_FMT_lx "\n",
+ info->start_stack);
+ fprintf(f, "brk 0x" TARGET_ABI_FMT_lx "\n", info->brk);
+ fprintf(f, "entry 0x" TARGET_ABI_FMT_lx "\n", info->entry);
+
+ qemu_log_unlock(f);
+ }
}
/* build Task State */
* generating the prologue until now so that the prologue can take
* the real value of GUEST_BASE into account.
*/
- tcg_prologue_init(tcg_ctx);
+ tcg_prologue_init();
target_cpu_init(env, regs);