[PATCH v8 0/3] MR741: ntdll: Implement NtFlushProcessWriteBuffers.

[Torge Matthies (＠tmatthies)]

Based on [a patch](https://www.winehq.org/mailman3/hyperkitty/list/[e-mail]/messag...) by Jinoh Kang (@iamahuman) from February 2022.

I removed the need for the event object and implemented fast paths for Linux.  
On macOS 10.14+ `thread_get_register_pointer_values` is used on every thread of the process.  
On Linux 4.14+ `membarrier(MEMBARRIER_CMD_GLOBAL_EXPEDITED, ...)` is used.  
On x86 Linux <= 4.13 `madvise(..., MADV_DONTNEED)` is used, which sends IPIs to all cores causing them to do a memory barrier.  
On non-x86 Linux <= 4.2 and on other platforms the fallback path using APCs is used.

--
  v8: ntdll: Add thread_get_register_pointer_values-based implementation of NtFlushProcessWriteBuffers.
      ntdll: Add sys_membarrier-based implementation of NtFlushProcessWriteBuffers.
      ntdll: Add MADV_DONTNEED-based implementation of NtFlushProcessWriteBuffers.

https://gitlab.winehq.org/wine/wine/-/merge_requests/741

[PATCH v8 1/3] ntdll: Add MADV_DONTNEED-based implementation of NtFlushProcessWriteBuffers.

[Torge Matthies]

From: Torge Matthies

Credits to Avi Kivity (scylladb) and Aliaksei Kandratsenka (gperftools) for this trick, see [1].

[1] https://github.com/scylladb/seastar/commit/77a58e4dc020233f66fccb8d9e8f7a8b7...
---
 dlls/ntdll/unix/virtual.c | 42 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

diff --git a/dlls/ntdll/unix/virtual.c b/dlls/ntdll/unix/virtual.c
index 8087a12785c..b5d7ba39d99 100644
--- a/dlls/ntdll/unix/virtual.c
+++ b/dlls/ntdll/unix/virtual.c
@@ -215,6 +215,11 @@ struct range_entry
 static struct range_entry *free_ranges;
 static struct range_entry *free_ranges_end;

+#if defined(__linux__) && (defined(__i386__) || defined(__x86_64__))
+static void *dontneed_page;
+static pthread_mutex_t dontneed_page_mutex = PTHREAD_MUTEX_INITIALIZER;
+#endif
+

static inline BOOL is_beyond_limit( const void *addr, size_t size, const void *limit )
 {
@@ -5174,13 +5179,48 @@ NTSTATUS WINAPI NtFlushInstructionCache( HANDLE handle, const void *addr, SIZE_T
 }

+#if defined(__linux__) && (defined(__i386__) || defined(__x86_64__))
+static int try_madvise( void )
+{
+    int ret = 0;
+    char *mem;
+
+    pthread_mutex_lock(&dontneed_page_mutex);
+    /* Credits to Avi Kivity (scylladb) and Aliaksei Kandratsenka (gperftools) for this trick,
+     * see https://github.com/scylladb/seastar/commit/77a58e4dc020233f66fccb8d9e8f7a8b7... */
+    mem = dontneed_page;
+    if (!mem)
+    {
+        mem = anon_mmap_alloc( page_size, PROT_READ | PROT_WRITE );
+        if (mem == MAP_FAILED)
+            goto failed;
+        if (mlock( mem, page_size ))
+        {
+            munmap( mem, page_size );
+            goto failed;
+        }
+        dontneed_page = mem;
+    }
+    *mem = 3;
+    ret = !madvise( mem, page_size, MADV_DONTNEED );
+failed:
+    pthread_mutex_unlock(&dontneed_page_mutex);
+    return ret;
+}
+#else
+static int try_madvise( void ) { return 0; }
+#endif
+
+
 /**********************************************************************
  *           NtFlushProcessWriteBuffers  (NTDLL.@)
  */
 void WINAPI NtFlushProcessWriteBuffers(void)
 {
     static int once = 0;
-    if (!once++) FIXME( "stub\n" );
+    if (try_madvise())
+        return;
+    if (!once++) FIXME( "no implementation available on this platform\n" );
 }

[PATCH v8 2/3] ntdll: Add sys_membarrier-based implementation of NtFlushProcessWriteBuffers.

[Torge Matthies]

From: Torge Matthies

Uses the MEMBARRIER_CMD_PRIVATE_EXPEDITED membarrier command introduced in Linux 4.14.
---
 dlls/ntdll/unix/virtual.c | 49 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 48 insertions(+), 1 deletion(-)

diff --git a/dlls/ntdll/unix/virtual.c b/dlls/ntdll/unix/virtual.c
index b5d7ba39d99..10dd5a997fe 100644
--- a/dlls/ntdll/unix/virtual.c
+++ b/dlls/ntdll/unix/virtual.c
@@ -39,6 +39,9 @@
 #ifdef HAVE_SYS_SYSINFO_H
 # include <sys/sysinfo.h>
 #endif
+#ifdef HAVE_SYS_SYSCALL_H
+# include <sys/syscall.h>
+#endif
 #ifdef HAVE_SYS_SYSCTL_H
 # include <sys/sysctl.h>
 #endif
@@ -215,10 +218,16 @@ struct range_entry
 static struct range_entry *free_ranges;
 static struct range_entry *free_ranges_end;

-#if defined(__linux__) && (defined(__i386__) || defined(__x86_64__))
+#ifdef __linux__
+#ifdef __NR_membarrier
+static BOOL membarrier_exp_available;
+static pthread_once_t membarrier_init_once = PTHREAD_ONCE_INIT;
+#endif
+#if defined(__i386__) || defined(__x86_64__)
 static void *dontneed_page;
 static pthread_mutex_t dontneed_page_mutex = PTHREAD_MUTEX_INITIALIZER;
 #endif
+#endif

static inline BOOL is_beyond_limit( const void *addr, size_t size, const void *limit )
@@ -5179,6 +5188,42 @@ NTSTATUS WINAPI NtFlushInstructionCache( HANDLE handle, const void *addr, SIZE_T
 }

+#if defined(__linux__) && defined(__NR_membarrier)
+#define MEMBARRIER_CMD_QUERY                        0x00
+#define MEMBARRIER_CMD_PRIVATE_EXPEDITED            0x08
+#define MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED   0x10
+
+
+static int membarrier( int cmd, unsigned int flags, int cpu_id )
+{
+    return syscall( __NR_membarrier, cmd, flags, cpu_id );
+}
+
+
+static void membarrier_init( void )
+{
+    static const int exp_required_cmds =
+        MEMBARRIER_CMD_PRIVATE_EXPEDITED | MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED;
+    int available_cmds = membarrier( MEMBARRIER_CMD_QUERY, 0, 0 );
+    if (available_cmds == -1)
+        return;
+    if ((available_cmds & exp_required_cmds) == exp_required_cmds)
+        membarrier_exp_available = !membarrier( MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED, 0, 0 );
+}
+
+
+static int try_exp_membarrier( void )
+{
+    pthread_once(&membarrier_init_once, membarrier_init);
+    if (!membarrier_exp_available)
+        return 0;
+    return !membarrier( MEMBARRIER_CMD_PRIVATE_EXPEDITED, 0, 0 );
+}
+#else
+static int try_exp_membarrier( void ) { return 0; }
+#endif
+
+
 #if defined(__linux__) && (defined(__i386__) || defined(__x86_64__))
 static int try_madvise( void )
 {
@@ -5218,6 +5263,8 @@ static int try_madvise( void ) { return 0; }
 void WINAPI NtFlushProcessWriteBuffers(void)
 {
     static int once = 0;
+    if (try_exp_membarrier())
+        return;
     if (try_madvise())
         return;
     if (!once++) FIXME( "no implementation available on this platform\n" );

[PATCH v8 3/3] ntdll: Add thread_get_register_pointer_values-based implementation of NtFlushProcessWriteBuffers.

[Torge Matthies]

From: Torge Matthies

---
 dlls/ntdll/unix/virtual.c | 62 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)

diff --git a/dlls/ntdll/unix/virtual.c b/dlls/ntdll/unix/virtual.c
index 10dd5a997fe..f2c5bd92de9 100644
--- a/dlls/ntdll/unix/virtual.c
+++ b/dlls/ntdll/unix/virtual.c
@@ -65,6 +65,9 @@
 #if defined(__APPLE__)
 # include <mach/mach_init.h>
 # include <mach/mach_vm.h>
+# include <mach/task.h>
+# include <mach/thread_state.h>
+# include <mach/vm_map.h>
 #endif

#include "ntstatus.h"
@@ -218,6 +221,11 @@ struct range_entry
 static struct range_entry *free_ranges;
 static struct range_entry *free_ranges_end;

+#ifdef __APPLE__
+static kern_return_t (*p_thread_get_register_pointer_values)( thread_t, uintptr_t*, size_t*, uintptr_t* );
+static pthread_once_t tgrpvs_init_once = PTHREAD_ONCE_INIT;
+#endif
+
 #ifdef __linux__
 #ifdef __NR_membarrier
 static BOOL membarrier_exp_available;
@@ -5188,6 +5196,58 @@ NTSTATUS WINAPI NtFlushInstructionCache( HANDLE handle, const void *addr, SIZE_T
 }

+#ifdef __APPLE__
+
+static void tgrpvs_init( void )
+{
+    p_thread_get_register_pointer_values = dlsym( RTLD_DEFAULT, "thread_get_register_pointer_values" );
+}
+
+static int try_mach_tgrpvs( void )
+{
+    /* Taken from https://github.com/dotnet/runtime/blob/7be37908e5a1cbb83b1062768c1649827eeac... */
+    mach_msg_type_number_t count, i;
+    thread_act_array_t threads;
+    kern_return_t kret;
+    int ret = 0;
+
+    pthread_once(&tgrpvs_init_once, tgrpvs_init);
+    if (!p_thread_get_register_pointer_values)
+        return 0;
+
+    kret = task_threads( mach_task_self(), &threads, &count );
+    if (kret)
+        return 0;
+
+    for (i = 0; i < count; i++)
+    {
+        uintptr_t reg_values[128];
+        size_t reg_count = ARRAY_SIZE( reg_values );
+        uintptr_t sp;
+
+        /* This function always fails when querying Rosetta's exception handling thread, so we will only handle
+           KERN_INSUFFICIENT_BUFFER_SIZE as an error, like .NET core does. */
+        kret = p_thread_get_register_pointer_values( threads[i], &sp, &reg_count, reg_values );
+        if (kret == KERN_INSUFFICIENT_BUFFER_SIZE)
+            goto fail;
+
+        kret = mach_port_deallocate( mach_task_self(), threads[i] );
+        if (kret)
+            goto fail;
+    }
+    ret = 1;
+fail:
+    for (; i < count; i++)
+        mach_port_deallocate( mach_task_self(), threads[i] );
+    vm_deallocate( mach_task_self(), (vm_address_t)threads, count * sizeof(threads[0]) );
+    return ret;
+}
+
+#else
+static int try_mach_tgrpvs( void ) { return 0; }
+#endif
+
+
 #if defined(__linux__) && defined(__NR_membarrier)
 #define MEMBARRIER_CMD_QUERY                        0x00
 #define MEMBARRIER_CMD_PRIVATE_EXPEDITED            0x08
@@ -5263,6 +5323,8 @@ static int try_madvise( void ) { return 0; }
 void WINAPI NtFlushProcessWriteBuffers(void)
 {
     static int once = 0;
+    if (try_mach_tgrpvs())
+        return;
     if (try_exp_membarrier())
         return;
     if (try_madvise())

[Jinoh Kang (＠iamahuman)]

Jinoh Kang (@iamahuman) commented about dlls/ntdll/unix/virtual.c:
> }
>
> +#if defined(__linux__) && (defined(__i386__) || defined(__x86_64__))

Why restrict to x86?

[Jinoh Kang (＠iamahuman)]

Jinoh Kang (@iamahuman) commented about dlls/ntdll/unix/virtual.c:
> -
>    if (mem == MAP_FAILED)
> -
>        goto failed;
> -
>    if (mlock( mem, page_size ))
> -
>    {
> -
>        munmap( mem, page_size );
> -
>        goto failed;
> -
>    }
> -
>    dontneed_page = mem;
> - }
> - *mem = 3;
> - ret = !madvise( mem, page_size, MADV_DONTNEED );
>
> +failed:
> - pthread_mutex_unlock(&dontneed_page_mutex);
> - return ret;
>
> +}
> +#else

Condition comments? (Also below)