qemu-e2k/util/oslib-win32.c

824 lines
21 KiB
C
Raw Normal View History

/*
* os-win32.c
*
* Copyright (c) 2003-2008 Fabrice Bellard
* Copyright (c) 2010-2016 Red Hat, Inc.
*
* QEMU library functions for win32 which are shared between QEMU and
* the QEMU tools.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*
* The implementation of g_poll (functions poll_rest, g_poll) at the end of
* this file are based on code from GNOME glib-2 and use a different license,
* see the license comment there.
*/
#include "qemu/osdep.h"
#include <windows.h>
#include "qemu-common.h"
2016-03-14 09:01:28 +01:00
#include "qapi/error.h"
#include "sysemu/sysemu.h"
#include "qemu/main-loop.h"
#include "trace.h"
#include "qemu/sockets.h"
#include "qemu/cutils.h"
/* this must come after including "trace.h" */
#include <shlobj.h>
void *qemu_oom_check(void *ptr)
{
if (ptr == NULL) {
fprintf(stderr, "Failed to allocate memory: %lu\n", GetLastError());
abort();
}
return ptr;
}
void *qemu_try_memalign(size_t alignment, size_t size)
{
void *ptr;
if (!size) {
abort();
}
ptr = VirtualAlloc(NULL, size, MEM_COMMIT, PAGE_READWRITE);
trace_qemu_memalign(alignment, size, ptr);
return ptr;
}
void *qemu_memalign(size_t alignment, size_t size)
{
return qemu_oom_check(qemu_try_memalign(alignment, size));
}
util/oslib-win32: indicate alignment for qemu_anon_ram_alloc() Let's set the alignment just like for the posix variant. This will implicitly set the alignment of the underlying memory region and therefore make memory_region_get_alignment(mr) return something > 0 for all memory backends applicable to PCDIMM/NVDIMM. The allocation granularity is ususally 64k, while the page size is 4k. The documentation of VirtualAlloc is not really comprehensible in case only MEM_COMMIT is specified without an address. We'll detect the actual values and then go for the bigger one. The expection is, that it will always be 64k aligned. (The assumption is that MEM_COMMIT does an implicit MEM_RESERVE, so the address will always be aligned to the allocation granularity. And the allocation granularity is always bigger than the page size). This will allow us to drop special handling in pc.c for memory_region_get_alignment(mr) == 0, as we can then assume that it is always set (and AFAICS >= getpagesize()). For pc in pc_memory_plug(), under Windows TARGET_PAGE_SIZE == getpagesize(), therefore alignment of DIMMs will not change, and therefore also not the guest physical memory layout. For spapr in spapr_memory_plug(), an alignment of 0 would have been used until now. As QEMU_ALIGN_UP will crash with the alignment being 0, this never worked, so we don't have to care about compatibility handling. Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: David Hildenbrand <david@redhat.com> Message-Id: <20180801133444.11269-3-david@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2018-08-01 15:34:42 +02:00
static int get_allocation_granularity(void)
{
SYSTEM_INFO system_info;
GetSystemInfo(&system_info);
return system_info.dwAllocationGranularity;
}
void *qemu_anon_ram_alloc(size_t size, uint64_t *align, bool shared)
{
void *ptr;
ptr = VirtualAlloc(NULL, size, MEM_COMMIT, PAGE_READWRITE);
trace_qemu_anon_ram_alloc(size, ptr);
util/oslib-win32: indicate alignment for qemu_anon_ram_alloc() Let's set the alignment just like for the posix variant. This will implicitly set the alignment of the underlying memory region and therefore make memory_region_get_alignment(mr) return something > 0 for all memory backends applicable to PCDIMM/NVDIMM. The allocation granularity is ususally 64k, while the page size is 4k. The documentation of VirtualAlloc is not really comprehensible in case only MEM_COMMIT is specified without an address. We'll detect the actual values and then go for the bigger one. The expection is, that it will always be 64k aligned. (The assumption is that MEM_COMMIT does an implicit MEM_RESERVE, so the address will always be aligned to the allocation granularity. And the allocation granularity is always bigger than the page size). This will allow us to drop special handling in pc.c for memory_region_get_alignment(mr) == 0, as we can then assume that it is always set (and AFAICS >= getpagesize()). For pc in pc_memory_plug(), under Windows TARGET_PAGE_SIZE == getpagesize(), therefore alignment of DIMMs will not change, and therefore also not the guest physical memory layout. For spapr in spapr_memory_plug(), an alignment of 0 would have been used until now. As QEMU_ALIGN_UP will crash with the alignment being 0, this never worked, so we don't have to care about compatibility handling. Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: David Hildenbrand <david@redhat.com> Message-Id: <20180801133444.11269-3-david@redhat.com> Reviewed-by: Igor Mammedov <imammedo@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2018-08-01 15:34:42 +02:00
if (ptr && align) {
*align = MAX(get_allocation_granularity(), getpagesize());
}
return ptr;
}
void qemu_vfree(void *ptr)
{
trace_qemu_vfree(ptr);
if (ptr) {
VirtualFree(ptr, 0, MEM_RELEASE);
}
}
void qemu_anon_ram_free(void *ptr, size_t size)
{
trace_qemu_anon_ram_free(ptr, size);
if (ptr) {
VirtualFree(ptr, 0, MEM_RELEASE);
}
}
oslib-win32: only provide localtime_r/gmtime_r if missing The oslib-win32 file currently provides a localtime_r and gmtime_r replacement unconditionally. Some versions of Mingw-w64 would provide crude macros for localtime_r/gmtime_r which QEMU takes care to disable. Latest versions of Mingw-w64 now provide actual functions for localtime_r/gmtime_r, but with a twist that you have to include unistd.h or pthread.h before including time.h. By luck some files in QEMU have such an include order, resulting in compile errors: CC util/osdep.o In file included from include/qemu-common.h:48:0, from util/osdep.c:48: include/sysemu/os-win32.h:77:12: error: redundant redeclaration of 'gmtime_r' [-Werror=redundant-decls] struct tm *gmtime_r(const time_t *timep, struct tm *result); ^ In file included from include/qemu-common.h:35:0, from util/osdep.c:48: /usr/i686-w64-mingw32/sys-root/mingw/include/time.h:272:107: note: previous definition of 'gmtime_r' was here In file included from include/qemu-common.h:48:0, from util/osdep.c:48: include/sysemu/os-win32.h:79:12: error: redundant redeclaration of 'localtime_r' [-Werror=redundant-decls] struct tm *localtime_r(const time_t *timep, struct tm *result); ^ In file included from include/qemu-common.h:35:0, from util/osdep.c:48: /usr/i686-w64-mingw32/sys-root/mingw/include/time.h:269:107: note: previous definition of 'localtime_r' was here This change adds a configure test to see if localtime_r exits, and only enables the QEMU impl if missing. We also re-arrange qemu-common.h try attempt to guarantee that all source files get unistd.h before time.h and thus see the localtime_r/gmtime_r defs. [sw: Use "official" spellings for Mingw-w64, MinGW in comments.] [sw: Terminate sentences with a dot in comments.] Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Stefan Weil <sw@weilnetz.de>
2015-09-22 16:13:26 +02:00
#ifndef CONFIG_LOCALTIME_R
/* FIXME: add proper locking */
struct tm *gmtime_r(const time_t *timep, struct tm *result)
{
struct tm *p = gmtime(timep);
memset(result, 0, sizeof(*result));
if (p) {
*result = *p;
p = result;
}
return p;
}
/* FIXME: add proper locking */
struct tm *localtime_r(const time_t *timep, struct tm *result)
{
struct tm *p = localtime(timep);
memset(result, 0, sizeof(*result));
if (p) {
*result = *p;
p = result;
}
return p;
}
oslib-win32: only provide localtime_r/gmtime_r if missing The oslib-win32 file currently provides a localtime_r and gmtime_r replacement unconditionally. Some versions of Mingw-w64 would provide crude macros for localtime_r/gmtime_r which QEMU takes care to disable. Latest versions of Mingw-w64 now provide actual functions for localtime_r/gmtime_r, but with a twist that you have to include unistd.h or pthread.h before including time.h. By luck some files in QEMU have such an include order, resulting in compile errors: CC util/osdep.o In file included from include/qemu-common.h:48:0, from util/osdep.c:48: include/sysemu/os-win32.h:77:12: error: redundant redeclaration of 'gmtime_r' [-Werror=redundant-decls] struct tm *gmtime_r(const time_t *timep, struct tm *result); ^ In file included from include/qemu-common.h:35:0, from util/osdep.c:48: /usr/i686-w64-mingw32/sys-root/mingw/include/time.h:272:107: note: previous definition of 'gmtime_r' was here In file included from include/qemu-common.h:48:0, from util/osdep.c:48: include/sysemu/os-win32.h:79:12: error: redundant redeclaration of 'localtime_r' [-Werror=redundant-decls] struct tm *localtime_r(const time_t *timep, struct tm *result); ^ In file included from include/qemu-common.h:35:0, from util/osdep.c:48: /usr/i686-w64-mingw32/sys-root/mingw/include/time.h:269:107: note: previous definition of 'localtime_r' was here This change adds a configure test to see if localtime_r exits, and only enables the QEMU impl if missing. We also re-arrange qemu-common.h try attempt to guarantee that all source files get unistd.h before time.h and thus see the localtime_r/gmtime_r defs. [sw: Use "official" spellings for Mingw-w64, MinGW in comments.] [sw: Terminate sentences with a dot in comments.] Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Stefan Weil <sw@weilnetz.de>
2015-09-22 16:13:26 +02:00
#endif /* CONFIG_LOCALTIME_R */
void qemu_set_block(int fd)
{
unsigned long opt = 0;
WSAEventSelect(fd, NULL, 0);
ioctlsocket(fd, FIONBIO, &opt);
}
void qemu_set_nonblock(int fd)
{
unsigned long opt = 1;
ioctlsocket(fd, FIONBIO, &opt);
qemu_fd_register(fd);
}
int socket_set_fast_reuse(int fd)
{
/* Enabling the reuse of an endpoint that was used by a socket still in
* TIME_WAIT state is usually performed by setting SO_REUSEADDR. On Windows
* fast reuse is the default and SO_REUSEADDR does strange things. So we
* don't have to do anything here. More info can be found at:
* http://msdn.microsoft.com/en-us/library/windows/desktop/ms740621.aspx */
return 0;
}
static int socket_error(void)
{
switch (WSAGetLastError()) {
case 0:
return 0;
case WSAEINTR:
return EINTR;
case WSAEINVAL:
return EINVAL;
case WSA_INVALID_HANDLE:
return EBADF;
case WSA_NOT_ENOUGH_MEMORY:
return ENOMEM;
case WSA_INVALID_PARAMETER:
return EINVAL;
case WSAENAMETOOLONG:
return ENAMETOOLONG;
case WSAENOTEMPTY:
return ENOTEMPTY;
case WSAEWOULDBLOCK:
/* not using EWOULDBLOCK as we don't want code to have
* to check both EWOULDBLOCK and EAGAIN */
return EAGAIN;
case WSAEINPROGRESS:
return EINPROGRESS;
case WSAEALREADY:
return EALREADY;
case WSAENOTSOCK:
return ENOTSOCK;
case WSAEDESTADDRREQ:
return EDESTADDRREQ;
case WSAEMSGSIZE:
return EMSGSIZE;
case WSAEPROTOTYPE:
return EPROTOTYPE;
case WSAENOPROTOOPT:
return ENOPROTOOPT;
case WSAEPROTONOSUPPORT:
return EPROTONOSUPPORT;
case WSAEOPNOTSUPP:
return EOPNOTSUPP;
case WSAEAFNOSUPPORT:
return EAFNOSUPPORT;
case WSAEADDRINUSE:
return EADDRINUSE;
case WSAEADDRNOTAVAIL:
return EADDRNOTAVAIL;
case WSAENETDOWN:
return ENETDOWN;
case WSAENETUNREACH:
return ENETUNREACH;
case WSAENETRESET:
return ENETRESET;
case WSAECONNABORTED:
return ECONNABORTED;
case WSAECONNRESET:
return ECONNRESET;
case WSAENOBUFS:
return ENOBUFS;
case WSAEISCONN:
return EISCONN;
case WSAENOTCONN:
return ENOTCONN;
case WSAETIMEDOUT:
return ETIMEDOUT;
case WSAECONNREFUSED:
return ECONNREFUSED;
case WSAELOOP:
return ELOOP;
case WSAEHOSTUNREACH:
return EHOSTUNREACH;
default:
return EIO;
}
}
int inet_aton(const char *cp, struct in_addr *ia)
{
uint32_t addr = inet_addr(cp);
if (addr == 0xffffffff) {
return 0;
}
ia->s_addr = addr;
return 1;
}
void qemu_set_cloexec(int fd)
{
}
/* Offset between 1/1/1601 and 1/1/1970 in 100 nanosec units */
#define _W32_FT_OFFSET (116444736000000000ULL)
int qemu_gettimeofday(qemu_timeval *tp)
{
union {
unsigned long long ns100; /*time since 1 Jan 1601 in 100ns units */
FILETIME ft;
} _now;
if(tp) {
GetSystemTimeAsFileTime (&_now.ft);
tp->tv_usec=(long)((_now.ns100 / 10ULL) % 1000000ULL );
tp->tv_sec= (long)((_now.ns100 - _W32_FT_OFFSET) / 10000000ULL);
}
/* Always return 0 as per Open Group Base Specifications Issue 6.
Do not set errno on error. */
return 0;
}
int qemu_get_thread_id(void)
{
return GetCurrentThreadId();
}
char *
qemu_get_local_state_pathname(const char *relative_pathname)
{
HRESULT result;
char base_path[MAX_PATH+1] = "";
result = SHGetFolderPath(NULL, CSIDL_COMMON_APPDATA, NULL,
/* SHGFP_TYPE_CURRENT */ 0, base_path);
if (result != S_OK) {
/* misconfigured environment */
g_critical("CSIDL_COMMON_APPDATA unavailable: %ld", (long)result);
abort();
}
return g_strdup_printf("%s" G_DIR_SEPARATOR_S "%s", base_path,
relative_pathname);
}
void qemu_set_tty_echo(int fd, bool echo)
{
HANDLE handle = (HANDLE)_get_osfhandle(fd);
DWORD dwMode = 0;
if (handle == INVALID_HANDLE_VALUE) {
return;
}
GetConsoleMode(handle, &dwMode);
if (echo) {
SetConsoleMode(handle, dwMode | ENABLE_ECHO_INPUT | ENABLE_LINE_INPUT);
} else {
SetConsoleMode(handle,
dwMode & ~(ENABLE_ECHO_INPUT | ENABLE_LINE_INPUT));
}
}
static char exec_dir[PATH_MAX];
void qemu_init_exec_dir(const char *argv0)
{
char *p;
char buf[MAX_PATH];
DWORD len;
len = GetModuleFileName(NULL, buf, sizeof(buf) - 1);
if (len == 0) {
return;
}
buf[len] = 0;
p = buf + len - 1;
while (p != buf && *p != '\\') {
p--;
}
*p = 0;
if (access(buf, R_OK) == 0) {
pstrcpy(exec_dir, sizeof(exec_dir), buf);
}
}
char *qemu_get_exec_dir(void)
{
return g_strdup(exec_dir);
}
#if !GLIB_CHECK_VERSION(2, 50, 0)
/*
* The original implementation of g_poll from glib has a problem on Windows
* when using timeouts < 10 ms.
*
* Whenever g_poll is called with timeout < 10 ms, it does a quick poll instead
* of wait. This causes significant performance degradation of QEMU.
*
* The following code is a copy of the original code from glib/gpoll.c
* (glib commit 20f4d1820b8d4d0fc4447188e33efffd6d4a88d8 from 2014-02-19).
* Some debug code was removed and the code was reformatted.
* All other code modifications are marked with 'QEMU'.
*/
/*
* gpoll.c: poll(2) abstraction
* Copyright 1998 Owen Taylor
* Copyright 2008 Red Hat, Inc.
*
* This 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 of the License, or (at your option) any later version.
*
* This 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 this library; if not, see <http://www.gnu.org/licenses/>.
*/
static int poll_rest(gboolean poll_msgs, HANDLE *handles, gint nhandles,
GPollFD *fds, guint nfds, gint timeout)
{
DWORD ready;
GPollFD *f;
int recursed_result;
if (poll_msgs) {
/* Wait for either messages or handles
* -> Use MsgWaitForMultipleObjectsEx
*/
ready = MsgWaitForMultipleObjectsEx(nhandles, handles, timeout,
QS_ALLINPUT, MWMO_ALERTABLE);
if (ready == WAIT_FAILED) {
gchar *emsg = g_win32_error_message(GetLastError());
g_warning("MsgWaitForMultipleObjectsEx failed: %s", emsg);
g_free(emsg);
}
} else if (nhandles == 0) {
/* No handles to wait for, just the timeout */
if (timeout == INFINITE) {
ready = WAIT_FAILED;
} else {
SleepEx(timeout, TRUE);
ready = WAIT_TIMEOUT;
}
} else {
/* Wait for just handles
* -> Use WaitForMultipleObjectsEx
*/
ready =
WaitForMultipleObjectsEx(nhandles, handles, FALSE, timeout, TRUE);
if (ready == WAIT_FAILED) {
gchar *emsg = g_win32_error_message(GetLastError());
g_warning("WaitForMultipleObjectsEx failed: %s", emsg);
g_free(emsg);
}
}
if (ready == WAIT_FAILED) {
return -1;
} else if (ready == WAIT_TIMEOUT || ready == WAIT_IO_COMPLETION) {
return 0;
} else if (poll_msgs && ready == WAIT_OBJECT_0 + nhandles) {
for (f = fds; f < &fds[nfds]; ++f) {
if (f->fd == G_WIN32_MSG_HANDLE && f->events & G_IO_IN) {
f->revents |= G_IO_IN;
}
}
/* If we have a timeout, or no handles to poll, be satisfied
* with just noticing we have messages waiting.
*/
if (timeout != 0 || nhandles == 0) {
return 1;
}
/* If no timeout and handles to poll, recurse to poll them,
* too.
*/
recursed_result = poll_rest(FALSE, handles, nhandles, fds, nfds, 0);
return (recursed_result == -1) ? -1 : 1 + recursed_result;
} else if (/* QEMU: removed the following unneeded statement which causes
* a compiler warning: ready >= WAIT_OBJECT_0 && */
ready < WAIT_OBJECT_0 + nhandles) {
for (f = fds; f < &fds[nfds]; ++f) {
if ((HANDLE) f->fd == handles[ready - WAIT_OBJECT_0]) {
f->revents = f->events;
}
}
/* If no timeout and polling several handles, recurse to poll
* the rest of them.
*/
if (timeout == 0 && nhandles > 1) {
/* Remove the handle that fired */
int i;
for (i = ready - WAIT_OBJECT_0 + 1; i < nhandles; i++) {
handles[i-1] = handles[i];
}
nhandles--;
recursed_result = poll_rest(FALSE, handles, nhandles, fds, nfds, 0);
return (recursed_result == -1) ? -1 : 1 + recursed_result;
}
return 1;
}
return 0;
}
gint g_poll(GPollFD *fds, guint nfds, gint timeout)
{
HANDLE handles[MAXIMUM_WAIT_OBJECTS];
gboolean poll_msgs = FALSE;
GPollFD *f;
gint nhandles = 0;
int retval;
for (f = fds; f < &fds[nfds]; ++f) {
if (f->fd == G_WIN32_MSG_HANDLE && (f->events & G_IO_IN)) {
poll_msgs = TRUE;
} else if (f->fd > 0) {
/* Don't add the same handle several times into the array, as
* docs say that is not allowed, even if it actually does seem
* to work.
*/
gint i;
for (i = 0; i < nhandles; i++) {
if (handles[i] == (HANDLE) f->fd) {
break;
}
}
if (i == nhandles) {
if (nhandles == MAXIMUM_WAIT_OBJECTS) {
g_warning("Too many handles to wait for!\n");
break;
} else {
handles[nhandles++] = (HANDLE) f->fd;
}
}
}
}
for (f = fds; f < &fds[nfds]; ++f) {
f->revents = 0;
}
if (timeout == -1) {
timeout = INFINITE;
}
/* Polling for several things? */
if (nhandles > 1 || (nhandles > 0 && poll_msgs)) {
/* First check if one or several of them are immediately
* available
*/
retval = poll_rest(poll_msgs, handles, nhandles, fds, nfds, 0);
/* If not, and we have a significant timeout, poll again with
* timeout then. Note that this will return indication for only
* one event, or only for messages. We ignore timeouts less than
* ten milliseconds as they are mostly pointless on Windows, the
* MsgWaitForMultipleObjectsEx() call will timeout right away
* anyway.
*
* Modification for QEMU: replaced timeout >= 10 by timeout > 0.
*/
if (retval == 0 && (timeout == INFINITE || timeout > 0)) {
retval = poll_rest(poll_msgs, handles, nhandles,
fds, nfds, timeout);
}
} else {
/* Just polling for one thing, so no need to check first if
* available immediately
*/
retval = poll_rest(poll_msgs, handles, nhandles, fds, nfds, timeout);
}
if (retval == -1) {
for (f = fds; f < &fds[nfds]; ++f) {
f->revents = 0;
}
}
return retval;
}
#endif
int getpagesize(void)
{
SYSTEM_INFO system_info;
GetSystemInfo(&system_info);
return system_info.dwPageSize;
}
mem-prealloc: reduce large guest start-up and migration time. Using "-mem-prealloc" option for a large guest leads to higher guest start-up and migration time. This is because with "-mem-prealloc" option qemu tries to map every guest page (create address translations), and make sure the pages are available during runtime. virsh/libvirt by default, seems to use "-mem-prealloc" option in case the guest is configured to use huge pages. The patch tries to map all guest pages simultaneously by spawning multiple threads. Currently limiting the change to QEMU library functions on POSIX compliant host only, as we are not sure if the problem exists on win32. Below are some stats with "-mem-prealloc" option for guest configured to use huge pages. ------------------------------------------------------------------------ Idle Guest | Start-up time | Migration time ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - single threaded (existing code) ------------------------------------------------------------------------ 64 Core - 4TB | 54m11.796s | 75m43.843s 64 Core - 1TB | 8m56.576s | 14m29.049s 64 Core - 256GB | 2m11.245s | 3m26.598s ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - map guest pages using 8 threads ------------------------------------------------------------------------ 64 Core - 4TB | 5m1.027s | 34m10.565s 64 Core - 1TB | 1m10.366s | 8m28.188s 64 Core - 256GB | 0m19.040s | 2m10.148s ----------------------------------------------------------------------- Guest stats with 2M HugePage usage - map guest pages using 16 threads ----------------------------------------------------------------------- 64 Core - 4TB | 1m58.970s | 31m43.400s 64 Core - 1TB | 0m39.885s | 7m55.289s 64 Core - 256GB | 0m11.960s | 2m0.135s ----------------------------------------------------------------------- Changed in v2: - modify number of memset threads spawned to min(smp_cpus, 16). - removed 64GB memory restriction for spawning memset threads. Changed in v3: - limit number of threads spawned based on min(sysconf(_SC_NPROCESSORS_ONLN), 16, smp_cpus) - implement memset thread specific siglongjmp in SIGBUS signal_handler. Changed in v4 - remove sigsetjmp/siglongjmp and SIGBUS unblock/block for main thread as main thread no longer touches any pages. - simplify code my returning memset_thread_failed status from touch_all_pages. Signed-off-by: Jitendra Kolhe <jitendra.kolhe@hpe.com> Message-Id: <1487907103-32350-1-git-send-email-jitendra.kolhe@hpe.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-02-24 04:31:43 +01:00
void os_mem_prealloc(int fd, char *area, size_t memory, int smp_cpus,
Error **errp)
{
int i;
size_t pagesize = qemu_real_host_page_size;
memory = (memory + pagesize - 1) & -pagesize;
for (i = 0; i < memory / pagesize; i++) {
memset(area + pagesize * i, 0, 1);
}
}
char *qemu_get_pid_name(pid_t pid)
{
/* XXX Implement me */
abort();
}
pid_t qemu_fork(Error **errp)
{
errno = ENOSYS;
error_setg_errno(errp, errno,
"cannot fork child process");
return -1;
}
osdep: add wrappers for socket functions The windows socket functions look identical to the normal POSIX sockets functions, but instead of setting errno, the caller needs to call WSAGetLastError(). QEMU has tried to deal with this incompatibility by defining a socket_error() method that callers must use that abstracts the difference between WSAGetLastError() and errno. This approach is somewhat error prone though - many callers of the sockets functions are just using errno directly because it is easy to forget the need use a QEMU specific wrapper. It is not always immediately obvious that a particular function will in fact call into Windows sockets functions, so the dev may not even realize they need to use socket_error(). This introduces an alternative approach to portability inspired by the way GNULIB fixes portability problems. We use a macro to redefine the original socket function names to refer to a QEMU wrapper function. The wrapper function calls the original Win32 sockets method and then sets errno from the WSAGetLastError() value. Thus all code can simply call the normal POSIX sockets APIs are have standard errno reporting on error, even on Windows. This makes the socket_error() method obsolete. We also bring closesocket & ioctlsocket into this approach. Even though they are non-standard Win32 names, we can't wrap the normal close/ioctl methods since there's no reliable way to distinguish between a file descriptor and HANDLE in Win32. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2016-03-07 21:25:19 +01:00
#undef connect
int qemu_connect_wrap(int sockfd, const struct sockaddr *addr,
socklen_t addrlen)
{
int ret;
ret = connect(sockfd, addr, addrlen);
if (ret < 0) {
if (WSAGetLastError() == WSAEWOULDBLOCK) {
errno = EINPROGRESS;
} else {
errno = socket_error();
}
osdep: add wrappers for socket functions The windows socket functions look identical to the normal POSIX sockets functions, but instead of setting errno, the caller needs to call WSAGetLastError(). QEMU has tried to deal with this incompatibility by defining a socket_error() method that callers must use that abstracts the difference between WSAGetLastError() and errno. This approach is somewhat error prone though - many callers of the sockets functions are just using errno directly because it is easy to forget the need use a QEMU specific wrapper. It is not always immediately obvious that a particular function will in fact call into Windows sockets functions, so the dev may not even realize they need to use socket_error(). This introduces an alternative approach to portability inspired by the way GNULIB fixes portability problems. We use a macro to redefine the original socket function names to refer to a QEMU wrapper function. The wrapper function calls the original Win32 sockets method and then sets errno from the WSAGetLastError() value. Thus all code can simply call the normal POSIX sockets APIs are have standard errno reporting on error, even on Windows. This makes the socket_error() method obsolete. We also bring closesocket & ioctlsocket into this approach. Even though they are non-standard Win32 names, we can't wrap the normal close/ioctl methods since there's no reliable way to distinguish between a file descriptor and HANDLE in Win32. Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2016-03-07 21:25:19 +01:00
}
return ret;
}
#undef listen
int qemu_listen_wrap(int sockfd, int backlog)
{
int ret;
ret = listen(sockfd, backlog);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef bind
int qemu_bind_wrap(int sockfd, const struct sockaddr *addr,
socklen_t addrlen)
{
int ret;
ret = bind(sockfd, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef socket
int qemu_socket_wrap(int domain, int type, int protocol)
{
int ret;
ret = socket(domain, type, protocol);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef accept
int qemu_accept_wrap(int sockfd, struct sockaddr *addr,
socklen_t *addrlen)
{
int ret;
ret = accept(sockfd, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef shutdown
int qemu_shutdown_wrap(int sockfd, int how)
{
int ret;
ret = shutdown(sockfd, how);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef ioctlsocket
int qemu_ioctlsocket_wrap(int fd, int req, void *val)
{
int ret;
ret = ioctlsocket(fd, req, val);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef closesocket
int qemu_closesocket_wrap(int fd)
{
int ret;
ret = closesocket(fd);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef getsockopt
int qemu_getsockopt_wrap(int sockfd, int level, int optname,
void *optval, socklen_t *optlen)
{
int ret;
ret = getsockopt(sockfd, level, optname, optval, optlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef setsockopt
int qemu_setsockopt_wrap(int sockfd, int level, int optname,
const void *optval, socklen_t optlen)
{
int ret;
ret = setsockopt(sockfd, level, optname, optval, optlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef getpeername
int qemu_getpeername_wrap(int sockfd, struct sockaddr *addr,
socklen_t *addrlen)
{
int ret;
ret = getpeername(sockfd, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef getsockname
int qemu_getsockname_wrap(int sockfd, struct sockaddr *addr,
socklen_t *addrlen)
{
int ret;
ret = getsockname(sockfd, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef send
ssize_t qemu_send_wrap(int sockfd, const void *buf, size_t len, int flags)
{
int ret;
ret = send(sockfd, buf, len, flags);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef sendto
ssize_t qemu_sendto_wrap(int sockfd, const void *buf, size_t len, int flags,
const struct sockaddr *addr, socklen_t addrlen)
{
int ret;
ret = sendto(sockfd, buf, len, flags, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef recv
ssize_t qemu_recv_wrap(int sockfd, void *buf, size_t len, int flags)
{
int ret;
ret = recv(sockfd, buf, len, flags);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
#undef recvfrom
ssize_t qemu_recvfrom_wrap(int sockfd, void *buf, size_t len, int flags,
struct sockaddr *addr, socklen_t *addrlen)
{
int ret;
ret = recvfrom(sockfd, buf, len, flags, addr, addrlen);
if (ret < 0) {
errno = socket_error();
}
return ret;
}
util: add qemu_write_pidfile() There are variants of qemu_create_pidfile() in qemu-pr-helper and qemu-ga. Let's have a common implementation in libqemuutil. The code is initially based from pr-helper write_pidfile(), with various improvements and suggestions from Daniel Berrangé: QEMU will leave the pidfile existing on disk when it exits which initially made me think it avoids the deletion race. The app managing QEMU, however, may well delete the pidfile after it has seen QEMU exit, and even if the app locks the pidfile before deleting it, there is still a race. eg consider the following sequence QEMU 1 libvirtd QEMU 2 1. lock(pidfile) 2. exit() 3. open(pidfile) 4. lock(pidfile) 5. open(pidfile) 6. unlink(pidfile) 7. close(pidfile) 8. lock(pidfile) IOW, at step 8 the new QEMU has successfully acquired the lock, but the pidfile no longer exists on disk because it was deleted after the original QEMU exited. While we could just say no external app should ever delete the pidfile, I don't think that is satisfactory as people don't read docs, and admins don't like stale pidfiles being left around on disk. To make this robust, I think we might want to copy libvirt's approach to pidfile acquisition which runs in a loop and checks that the file on disk /after/ acquiring the lock matches the file that was locked. Then we could in fact safely let QEMU delete its own pidfiles on clean exit.. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20180831145314.14736-2-marcandre.lureau@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2018-08-31 16:53:12 +02:00
bool qemu_write_pidfile(const char *filename, Error **errp)
{
char buffer[128];
int len;
HANDLE file;
OVERLAPPED overlap;
BOOL ret;
memset(&overlap, 0, sizeof(overlap));
file = CreateFile(filename, GENERIC_WRITE, FILE_SHARE_READ, NULL,
OPEN_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL);
if (file == INVALID_HANDLE_VALUE) {
error_setg(errp, "Failed to create PID file");
return false;
}
len = snprintf(buffer, sizeof(buffer), FMT_pid "\n", (pid_t)getpid());
ret = WriteFile(file, (LPCVOID)buffer, (DWORD)len,
NULL, &overlap);
CloseHandle(file);
if (ret == 0) {
error_setg(errp, "Failed to write PID file");
return false;
}
return true;
}
char *qemu_get_host_name(Error **errp)
{
wchar_t tmp[MAX_COMPUTERNAME_LENGTH + 1];
DWORD size = G_N_ELEMENTS(tmp);
if (GetComputerNameW(tmp, &size) == 0) {
error_setg_win32(errp, GetLastError(), "failed close handle");
return NULL;
}
return g_utf16_to_utf8(tmp, size, NULL, NULL, NULL);
}