Thereby making it per-TCGContext. Once we remove tb_lock, this will
avoid an atomic increment every time a TB is invalidated.

Signed-off-by: Emilio G. Cota <cota@braap.org>
---
 accel/tcg/translate-all.c |  5 +++--
 include/exec/tb-context.h |  1 -
 tcg/tcg.c                 | 14 ++++++++++++++
 tcg/tcg.h                 |  3 +++
 4 files changed, 20 insertions(+), 3 deletions(-)

diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index 3a51d49..20ad3fc 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -1072,7 +1072,8 @@ void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr)
     /* suppress any remaining jumps to this TB */
     tb_jmp_unlink(tb);
 
-    tb_ctx.tb_phys_invalidate_count++;
+    atomic_set(&tcg_ctx->tb_phys_invalidate_count,
+               tcg_ctx->tb_phys_invalidate_count + 1);
 }
 
 #ifdef CONFIG_SOFTMMU
@@ -1862,7 +1863,7 @@ void dump_exec_info(FILE *f, fprintf_function cpu_fprintf)
     cpu_fprintf(f, "\nStatistics:\n");
     cpu_fprintf(f, "TB flush count      %u\n",
                 atomic_read(&tb_ctx.tb_flush_count));
-    cpu_fprintf(f, "TB invalidate count %d\n", tb_ctx.tb_phys_invalidate_count);
+    cpu_fprintf(f, "TB invalidate count %zu\n", tcg_tb_phys_invalidate_count());
     cpu_fprintf(f, "TLB flush count     %zu\n", tlb_flush_count());
     tcg_dump_info(f, cpu_fprintf);
 }
diff --git a/include/exec/tb-context.h b/include/exec/tb-context.h
index d8472c8..8c9b49c 100644
--- a/include/exec/tb-context.h
+++ b/include/exec/tb-context.h
@@ -37,7 +37,6 @@ struct TBContext {
 
     /* statistics */
     unsigned tb_flush_count;
-    int tb_phys_invalidate_count;
 };
 
 extern TBContext tb_ctx;
diff --git a/tcg/tcg.c b/tcg/tcg.c
index b471708..a7b596e 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -791,6 +791,20 @@ size_t tcg_code_capacity(void)
     return capacity;
 }
 
+size_t tcg_tb_phys_invalidate_count(void)
+{
+    unsigned int n_ctxs = atomic_read(&n_tcg_ctxs);
+    unsigned int i;
+    size_t total = 0;
+
+    for (i = 0; i < n_ctxs; i++) {
+        const TCGContext *s = atomic_read(&tcg_ctxs[i]);
+
+        total += atomic_read(&s->tb_phys_invalidate_count);
+    }
+    return total;
+}
+
 /* pool based memory allocation */
 void *tcg_malloc_internal(TCGContext *s, int size)
 {
diff --git a/tcg/tcg.h b/tcg/tcg.h
index 8bf29cc..9dd9448 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -694,6 +694,8 @@ struct TCGContext {
     /* Threshold to flush the translated code buffer.  */
     void *code_gen_highwater;
 
+    size_t tb_phys_invalidate_count;
+
     /* Track which vCPU triggers events */
     CPUState *cpu;                      /* *_trans */
 
@@ -852,6 +854,7 @@ size_t tcg_code_capacity(void);
 
 void tcg_tb_insert(TranslationBlock *tb);
 void tcg_tb_remove(TranslationBlock *tb);
+size_t tcg_tb_phys_invalidate_count(void);
 TranslationBlock *tcg_tb_lookup(uintptr_t tc_ptr);
 void tcg_tb_foreach(GTraverseFunc func, gpointer user_data);
 size_t tcg_nb_tbs(void);
-- 
2.7.4

On 02/26/2018 09:39 PM, Emilio G. Cota wrote:
> Thereby making it per-TCGContext. Once we remove tb_lock, this will
> avoid an atomic increment every time a TB is invalidated.
> 
> Signed-off-by: Emilio G. Cota <cota@braap.org>
> ---
>  accel/tcg/translate-all.c |  5 +++--
>  include/exec/tb-context.h |  1 -
>  tcg/tcg.c                 | 14 ++++++++++++++
>  tcg/tcg.h                 |  3 +++
>  4 files changed, 20 insertions(+), 3 deletions(-)

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~

Emilio G. Cota <cota@braap.org> writes:

> Thereby making it per-TCGContext. Once we remove tb_lock, this will
> avoid an atomic increment every time a TB is invalidated.
>
> Signed-off-by: Emilio G. Cota <cota@braap.org>
> ---
>  accel/tcg/translate-all.c |  5 +++--
>  include/exec/tb-context.h |  1 -
>  tcg/tcg.c                 | 14 ++++++++++++++
>  tcg/tcg.h                 |  3 +++
>  4 files changed, 20 insertions(+), 3 deletions(-)
>
> diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
> index 3a51d49..20ad3fc 100644
> --- a/accel/tcg/translate-all.c
> +++ b/accel/tcg/translate-all.c
> @@ -1072,7 +1072,8 @@ void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr)
>      /* suppress any remaining jumps to this TB */
>      tb_jmp_unlink(tb);
>
> -    tb_ctx.tb_phys_invalidate_count++;
> +    atomic_set(&tcg_ctx->tb_phys_invalidate_count,
> +               tcg_ctx->tb_phys_invalidate_count + 1);

We do have an atomic_inc helper for this or we need comment that we only
have atomic_reads() to worry about hence no races.

Otherwise:

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

>  }
>
>  #ifdef CONFIG_SOFTMMU
> @@ -1862,7 +1863,7 @@ void dump_exec_info(FILE *f, fprintf_function cpu_fprintf)
>      cpu_fprintf(f, "\nStatistics:\n");
>      cpu_fprintf(f, "TB flush count      %u\n",
>                  atomic_read(&tb_ctx.tb_flush_count));
> -    cpu_fprintf(f, "TB invalidate count %d\n", tb_ctx.tb_phys_invalidate_count);
> +    cpu_fprintf(f, "TB invalidate count %zu\n", tcg_tb_phys_invalidate_count());
>      cpu_fprintf(f, "TLB flush count     %zu\n", tlb_flush_count());
>      tcg_dump_info(f, cpu_fprintf);
>  }
> diff --git a/include/exec/tb-context.h b/include/exec/tb-context.h
> index d8472c8..8c9b49c 100644
> --- a/include/exec/tb-context.h
> +++ b/include/exec/tb-context.h
> @@ -37,7 +37,6 @@ struct TBContext {
>
>      /* statistics */
>      unsigned tb_flush_count;
> -    int tb_phys_invalidate_count;
>  };
>
>  extern TBContext tb_ctx;
> diff --git a/tcg/tcg.c b/tcg/tcg.c
> index b471708..a7b596e 100644
> --- a/tcg/tcg.c
> +++ b/tcg/tcg.c
> @@ -791,6 +791,20 @@ size_t tcg_code_capacity(void)
>      return capacity;
>  }
>
> +size_t tcg_tb_phys_invalidate_count(void)
> +{
> +    unsigned int n_ctxs = atomic_read(&n_tcg_ctxs);
> +    unsigned int i;
> +    size_t total = 0;
> +
> +    for (i = 0; i < n_ctxs; i++) {
> +        const TCGContext *s = atomic_read(&tcg_ctxs[i]);
> +
> +        total += atomic_read(&s->tb_phys_invalidate_count);
> +    }
> +    return total;
> +}
> +
>  /* pool based memory allocation */
>  void *tcg_malloc_internal(TCGContext *s, int size)
>  {
> diff --git a/tcg/tcg.h b/tcg/tcg.h
> index 8bf29cc..9dd9448 100644
> --- a/tcg/tcg.h
> +++ b/tcg/tcg.h
> @@ -694,6 +694,8 @@ struct TCGContext {
>      /* Threshold to flush the translated code buffer.  */
>      void *code_gen_highwater;
>
> +    size_t tb_phys_invalidate_count;
> +
>      /* Track which vCPU triggers events */
>      CPUState *cpu;                      /* *_trans */
>
> @@ -852,6 +854,7 @@ size_t tcg_code_capacity(void);
>
>  void tcg_tb_insert(TranslationBlock *tb);
>  void tcg_tb_remove(TranslationBlock *tb);
> +size_t tcg_tb_phys_invalidate_count(void);
>  TranslationBlock *tcg_tb_lookup(uintptr_t tc_ptr);
>  void tcg_tb_foreach(GTraverseFunc func, gpointer user_data);
>  size_t tcg_nb_tbs(void);


--
Alex Bennée

On Thu, Mar 29, 2018 at 11:06:07 +0100, Alex Bennée wrote:
> Emilio G. Cota <cota@braap.org> writes:
(snip)
> > diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
> > index 3a51d49..20ad3fc 100644
> > --- a/accel/tcg/translate-all.c
> > +++ b/accel/tcg/translate-all.c
> > @@ -1072,7 +1072,8 @@ void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr)
> >      /* suppress any remaining jumps to this TB */
> >      tb_jmp_unlink(tb);
> >
> > -    tb_ctx.tb_phys_invalidate_count++;
> > +    atomic_set(&tcg_ctx->tb_phys_invalidate_count,
> > +               tcg_ctx->tb_phys_invalidate_count + 1);
> 
> We do have an atomic_inc helper for this or we need comment that we only
> have atomic_reads() to worry about hence no races.
> 
> Otherwise:
> 
> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>

tcg_ctx is per-thread (or otherwise tb_lock is held here), so
yes we only worry here about concurrent atomic_reads. This
is common enough not to need a comment, me thinks.
[we have quite a few more instances of atomic_set(foo, foo + 1),
for instance when incrementing TCGProfile counts.]

Thanks,

		Emilio