[Qemu-devel] [PATCH v3 4/6] s390x/tcg: Flush the TLB of all CPUs on SSKE and RRBE

David Hildenbrand posted 6 patches 6 years, 5 months ago
Maintainers: Richard Henderson <rth@twiddle.net>, Cornelia Huck <cohuck@redhat.com>, David Hildenbrand <david@redhat.com>
[Qemu-devel] [PATCH v3 4/6] s390x/tcg: Flush the TLB of all CPUs on SSKE and RRBE
Posted by David Hildenbrand 6 years, 5 months ago
Whenever we modify a storage key, we should flush the TLBs of all CPUs,
so the MMU fault handling code can properly consider the changed storage
key (to e.g., properly set the reference and change bit on the next
accesses).

These functions are barely used in modern Linux guests, so the performance
implications are neglectable for now.

This is a preparation for better reference and change bit handling for
TCG, which will require more MMU changes.

Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 target/s390x/mem_helper.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/target/s390x/mem_helper.c b/target/s390x/mem_helper.c
index 29d9eaa5b7..91ba2e03d9 100644
--- a/target/s390x/mem_helper.c
+++ b/target/s390x/mem_helper.c
@@ -1815,6 +1815,11 @@ void HELPER(sske)(CPUS390XState *env, uint64_t r1, uint64_t r2)
 
     key = (uint8_t) r1;
     skeyclass->set_skeys(ss, addr / TARGET_PAGE_SIZE, 1, &key);
+   /*
+    * As we can only flush by virtual address and not all the entries
+    * that point to a physical address we have to flush the whole TLB.
+    */
+    tlb_flush_all_cpus_synced(env_cpu(env));
 }
 
 /* reset reference bit extended */
@@ -1843,6 +1848,11 @@ uint32_t HELPER(rrbe)(CPUS390XState *env, uint64_t r2)
     if (skeyclass->set_skeys(ss, r2 / TARGET_PAGE_SIZE, 1, &key)) {
         return 0;
     }
+   /*
+    * As we can only flush by virtual address and not all the entries
+    * that point to a physical address we have to flush the whole TLB.
+    */
+    tlb_flush_all_cpus_synced(env_cpu(env));
 
     /*
      * cc
-- 
2.21.0


Re: [Qemu-devel] [PATCH v3 4/6] s390x/tcg: Flush the TLB of all CPUs on SSKE and RRBE
Posted by Alex Bennée 6 years, 5 months ago
David Hildenbrand <david@redhat.com> writes:

> Whenever we modify a storage key, we should flush the TLBs of all CPUs,
> so the MMU fault handling code can properly consider the changed storage
> key (to e.g., properly set the reference and change bit on the next
> accesses).
>
> These functions are barely used in modern Linux guests, so the performance
> implications are neglectable for now.
>
> This is a preparation for better reference and change bit handling for
> TCG, which will require more MMU changes.
>
> Reviewed-by: Cornelia Huck <cohuck@redhat.com>
> Signed-off-by: David Hildenbrand <david@redhat.com>

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

> ---
>  target/s390x/mem_helper.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
>
> diff --git a/target/s390x/mem_helper.c b/target/s390x/mem_helper.c
> index 29d9eaa5b7..91ba2e03d9 100644
> --- a/target/s390x/mem_helper.c
> +++ b/target/s390x/mem_helper.c
> @@ -1815,6 +1815,11 @@ void HELPER(sske)(CPUS390XState *env, uint64_t r1, uint64_t r2)
>
>      key = (uint8_t) r1;
>      skeyclass->set_skeys(ss, addr / TARGET_PAGE_SIZE, 1, &key);
> +   /*
> +    * As we can only flush by virtual address and not all the entries
> +    * that point to a physical address we have to flush the whole TLB.
> +    */
> +    tlb_flush_all_cpus_synced(env_cpu(env));
>  }
>
>  /* reset reference bit extended */
> @@ -1843,6 +1848,11 @@ uint32_t HELPER(rrbe)(CPUS390XState *env, uint64_t r2)
>      if (skeyclass->set_skeys(ss, r2 / TARGET_PAGE_SIZE, 1, &key)) {
>          return 0;
>      }
> +   /*
> +    * As we can only flush by virtual address and not all the entries
> +    * that point to a physical address we have to flush the whole TLB.
> +    */
> +    tlb_flush_all_cpus_synced(env_cpu(env));
>
>      /*
>       * cc


--
Alex Bennée