tcg: remove tb_lock

Use mmap_lock in user-mode to protect TCG state and the page descriptors.
In !user-mode, each vCPU has its own TCG state, so no locks needed.
Per-page locks are used to protect the page descriptors.

Per-TB locks are used in both modes to protect TB jumps.

Some notes:

- tb_lock is removed from notdirty_mem_write by passing a
  locked page_collection to tb_invalidate_phys_page_fast.

- tcg_tb_lookup/remove/insert/etc have their own internal lock(s),
  so there is no need to further serialize access to them.

- do_tb_flush is run in a safe async context, meaning no other
  vCPU threads are running. Therefore acquiring mmap_lock there
  is just to please tools such as thread sanitizer.

- Not visible in the diff, but tb_invalidate_phys_page already
  has an assert_memory_lock.

- cpu_io_recompile is !user-only, so no mmap_lock there.

- Added mmap_unlock()'s before all siglongjmp's that could
  be called in user-mode while mmap_lock is held.
  + Added an assert for !have_mmap_lock() after returning from
    the longjmp in cpu_exec, just like we do in cpu_exec_step_atomic.

Performance numbers before/after:

Host: AMD Opteron(tm) Processor 6376

                 ubuntu 17.04 ppc64 bootup+shutdown time

  700 +-+--+----+------+------------+-----------+------------*--+-+
      |    +    +      +            +           +           *B    |
      |         before ***B***                            ** *    |
      |tb lock removal ###D###                         ***        |
  600 +-+                                           ***         +-+
      |                                           **         #    |
      |                                        *B*          #D    |
      |                                     *** *         ##      |
  500 +-+                                ***           ###      +-+
      |                             * ***           ###           |
      |                            *B*          # ##              |
      |                          ** *          #D#                |
  400 +-+                      **            ##                 +-+
      |                      **           ###                     |
      |                    **           ##                        |
      |                  **         # ##                          |
  300 +-+  *           B*          #D#                          +-+
      |    B         ***        ###                               |
      |    *       **       ####                                  |
      |     *   ***      ###                                      |
  200 +-+   B  *B     #D#                                       +-+
      |     #B* *   ## #                                          |
      |     #*    ##                                              |
      |    + D##D#     +            +           +            +    |
  100 +-+--+----+------+------------+-----------+------------+--+-+
           1    8      16      Guest CPUs       48           64
  png: https://imgur.com/HwmBHXe

              debian jessie aarch64 bootup+shutdown time

  90 +-+--+-----+-----+------------+------------+------------+--+-+
     |    +     +     +            +            +            +    |
     |         before ***B***                                B    |
  80 +tb lock removal ###D###                              **D  +-+
     |                                                   **###    |
     |                                                 **##       |
  70 +-+                                             ** #       +-+
     |                                             ** ##          |
     |                                           **  #            |
  60 +-+                                       *B  ##           +-+
     |                                       **  ##               |
     |                                    ***  #D                 |
  50 +-+                               ***   ##                 +-+
     |                             * **   ###                     |
     |                           **B*  ###                        |
  40 +-+                     ****  # ##                         +-+
     |                   ****     #D#                             |
     |             ***B**      ###                                |
  30 +-+    B***B**        ####                                 +-+
     |    B *   *     # ###                                       |
     |     B       ###D#                                          |
  20 +-+   D  ##D##                                             +-+
     |      D#                                                    |
     |    +     +     +            +            +            +    |
  10 +-+--+-----+-----+------------+------------+------------+--+-+
          1     8     16      Guest CPUs        48           64
  png: https://imgur.com/iGpGFtv

The gains are high for 4-8 CPUs. Beyond that point, however, unrelated
lock contention significantly hurts scalability.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

1 files changed, 11 insertions, 15 deletions

diff --git a/exec.c b/exec.c
index ebadc0e302..28f9bdcbf9 100644
--- a/exec.c
+++ b/exec.c
@@ -1031,9 +1031,7 @@ const char *parse_cpu_model(const char *cpu_model)
 static void breakpoint_invalidate(CPUState *cpu, target_ulong pc)
 {
     mmap_lock();
-    tb_lock();
     tb_invalidate_phys_page_range(pc, pc + 1, 0);
-    tb_unlock();
     mmap_unlock();
 }
 #else
@@ -2644,21 +2642,21 @@ void memory_notdirty_write_prepare(NotDirtyInfo *ndi,
     ndi->ram_addr = ram_addr;
     ndi->mem_vaddr = mem_vaddr;
     ndi->size = size;
-    ndi->locked = false;
+    ndi->pages = NULL;
 
     assert(tcg_enabled());
     if (!cpu_physical_memory_get_dirty_flag(ram_addr, DIRTY_MEMORY_CODE)) {
-        ndi->locked = true;
-        tb_lock();
-        tb_invalidate_phys_page_fast(ram_addr, size);
+        ndi->pages = page_collection_lock(ram_addr, ram_addr + size);
+        tb_invalidate_phys_page_fast(ndi->pages, ram_addr, size);
     }
 }
 
 /* Called within RCU critical section. */
 void memory_notdirty_write_complete(NotDirtyInfo *ndi)
 {
-    if (ndi->locked) {
-        tb_unlock();
+    if (ndi->pages) {
+        page_collection_unlock(ndi->pages);
+        ndi->pages = NULL;
     }
 
     /* Set both VGA and migration bits for simplicity and to remove
@@ -2745,18 +2743,16 @@ static void check_watchpoint(int offset, int len, MemTxAttrs attrs, int flags)
                 }
                 cpu->watchpoint_hit = wp;
 
-                /* Both tb_lock and iothread_mutex will be reset when
-                 * cpu_loop_exit or cpu_loop_exit_noexc longjmp
-                 * back into the cpu_exec main loop.
-                 */
-                tb_lock();
+                mmap_lock();
                 tb_check_watchpoint(cpu);
                 if (wp->flags & BP_STOP_BEFORE_ACCESS) {
                     cpu->exception_index = EXCP_DEBUG;
+                    mmap_unlock();
                     cpu_loop_exit(cpu);
                 } else {
                     /* Force execution of one insn next time.  */
                     cpu->cflags_next_tb = 1 | curr_cflags();
+                    mmap_unlock();
                     cpu_loop_exit_noexc(cpu);
                 }
             }
@@ -3147,9 +3143,9 @@ static void invalidate_and_set_dirty(MemoryRegion *mr, hwaddr addr,
     }
     if (dirty_log_mask & (1 << DIRTY_MEMORY_CODE)) {
         assert(tcg_enabled());
-        tb_lock();
+        mmap_lock();
         tb_invalidate_phys_range(addr, addr + length);
-        tb_unlock();
+        mmap_unlock();
         dirty_log_mask &= ~(1 << DIRTY_MEMORY_CODE);
     }
     cpu_physical_memory_set_dirty_range(addr, length, dirty_log_mask);