qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v2 1/6] tcg: Ensure safe tb_jmp_cache lookup out


From: Sergey Fedorov
Subject: Re: [Qemu-devel] [PATCH v2 1/6] tcg: Ensure safe tb_jmp_cache lookup out of 'tb_lock'
Date: Fri, 8 Jul 2016 17:51:41 +0300
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.8.0

On 05/07/16 19:18, Alex Bennée wrote:
> From: Sergey Fedorov <address@hidden>
>
> First, ensure atomicity of CPU's 'tb_jmp_cache' access by:
>  * using atomic_read() to look up a TB when not holding 'tb_lock';
>  * using atomic_write() to remove a TB from each CPU's local cache on
>    TB invalidation.
>
> Second, add some memory barriers to ensure we don't put the TB being
> invalidated back to CPU's 'tb_jmp_cache'. If we fail to look up a TB in
> CPU's local cache because it is being invalidated by some other thread
> then it must not be found in the shared TB hash table. Otherwise we'd
> put it back to CPU's local cache.
>
> Note that this patch does *not* make CPU's TLB invalidation safe if it
> is done from some other thread while the CPU is in its execution loop.
>
> Signed-off-by: Sergey Fedorov <address@hidden>
> Signed-off-by: Sergey Fedorov <address@hidden>
> [AJB: fixed missing atomic set, tweak title]
> Signed-off-by: Alex Bennée <address@hidden>
> Reviewed-by: Richard Henderson <address@hidden>
> Reviewed-by: Emilio G. Cota <address@hidden>
>
> ---
> v1 (AJB):
>   - tweak title
>   - fixed missing set of tb_jmp_cache
> v2
>   - fix spelling s/con't/can't/
>   - add atomic_read while clearing tb_jmp_cache
>   - add r-b tags
> ---
>  cpu-exec.c      | 9 +++++++--
>  translate-all.c | 9 +++++++--
>  2 files changed, 14 insertions(+), 4 deletions(-)
>
> diff --git a/cpu-exec.c b/cpu-exec.c
> index b840e1d..10ce1cb 100644
> --- a/cpu-exec.c
> +++ b/cpu-exec.c
> @@ -285,6 +285,11 @@ static TranslationBlock *tb_find_slow(CPUState *cpu,
>  {
>      TranslationBlock *tb;
>  
> +    /* Ensure that we won't find a TB in the shared hash table
> +     * if it is being invalidated by some other thread.
> +     * Otherwise we'd put it back to CPU's local cache.
> +     * Pairs with smp_wmb() in tb_phys_invalidate(). */
> +    smp_rmb();
>      tb = tb_find_physical(cpu, pc, cs_base, flags);
>      if (tb) {
>          goto found;
> @@ -315,7 +320,7 @@ static TranslationBlock *tb_find_slow(CPUState *cpu,
>  
>  found:
>      /* we add the TB in the virtual pc hash table */
> -    cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)] = tb;
> +    atomic_set(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)], tb);
>      return tb;
>  }
>  
> @@ -333,7 +338,7 @@ static inline TranslationBlock *tb_find_fast(CPUState 
> *cpu,
>         is executed. */
>      cpu_get_tb_cpu_state(env, &pc, &cs_base, &flags);
>      tb_lock();
> -    tb = cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)];
> +    tb = atomic_read(&cpu->tb_jmp_cache[tb_jmp_cache_hash_func(pc)]);
>      if (unlikely(!tb || tb->pc != pc || tb->cs_base != cs_base ||
>                   tb->flags != flags)) {
>          tb = tb_find_slow(cpu, pc, cs_base, flags);
> diff --git a/translate-all.c b/translate-all.c
> index eaa95e4..96efe48 100644
> --- a/translate-all.c
> +++ b/translate-all.c
> @@ -1004,11 +1004,16 @@ void tb_phys_invalidate(TranslationBlock *tb, 
> tb_page_addr_t page_addr)
>          invalidate_page_bitmap(p);
>      }
>  
> +    /* Ensure that we won't find the TB in the shared hash table
> +     * if we can't see it in CPU's local cache.
> +     * Pairs with smp_rmb() in tb_find_slow(). */
> +    smp_wmb();
> +
>      /* remove the TB from the hash list */
>      h = tb_jmp_cache_hash_func(tb->pc);
>      CPU_FOREACH(cpu) {
> -        if (cpu->tb_jmp_cache[h] == tb) {
> -            cpu->tb_jmp_cache[h] = NULL;
> +        if (atomic_read(&cpu->tb_jmp_cache[h]) == tb) {

I think this can be a simple read because we can't race here with
setting 'tb_jmp_cache' in tb_find_slow().

Kind regards,
Sergey

> +            atomic_set(&cpu->tb_jmp_cache[h], NULL);
>          }
>      }
>  




reply via email to

[Prev in Thread] Current Thread [Next in Thread]