qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Richard Henderson <richard.henderson@linaro.org>, qemu-devel@nongnu.org
Cc: pbonzini@redhat.com, alex.bennee@linaro.org, stefanha@redhat.com
Subject: Re: [Qemu-devel] [RFC 1/3] exec: Adjust notdirty tracing
Date: Wed, 18 Sep 2019 10:34:31 +0200	[thread overview]
Message-ID: <38fead47-5d6e-5f67-c5b0-93273b22c0a7@redhat.com> (raw)
In-Reply-To: <20190918052641.21300-2-richard.henderson@linaro.org>

On 18.09.19 07:26, Richard Henderson wrote:
> The memory_region_tb_read tracepoint is unreachable, since notdirty
> is supposed to apply only to reads.  The memory_region_tb_write
> tracepoint is mis-named, because notdirty is not only used for TB
> invalidation.  It is also used for e.g. VGA RAM updates.
> 
> Replace memory_region_tb_write with memory_notdirty_write, and
> place it in memory_notdirty_write_prepare where it can catch all
> of the instances.  Add memory_notdirty_dirty to log when we no
> longer intercept writes to a page.
> 
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>  exec.c       | 3 +++
>  memory.c     | 4 ----
>  trace-events | 4 ++--
>  3 files changed, 5 insertions(+), 6 deletions(-)
> 
> diff --git a/exec.c b/exec.c
> index 8b998974f8..9babe57615 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2755,6 +2755,8 @@ void memory_notdirty_write_prepare(NotDirtyInfo *ndi,
>      ndi->size = size;
>      ndi->pages = NULL;
>  
> +    trace_memory_notdirty_write(mem_vaddr, ram_addr, size);
> +
>      assert(tcg_enabled());
>      if (!cpu_physical_memory_get_dirty_flag(ram_addr, DIRTY_MEMORY_CODE)) {
>          ndi->pages = page_collection_lock(ram_addr, ram_addr + size);
> @@ -2779,6 +2781,7 @@ void memory_notdirty_write_complete(NotDirtyInfo *ndi)
>      /* we remove the notdirty callback only if the code has been
>         flushed */
>      if (!cpu_physical_memory_is_clean(ndi->ram_addr)) {
> +        trace_memory_notdirty_dirty(ndi->mem_vaddr);
>          tlb_set_dirty(ndi->cpu, ndi->mem_vaddr);
>      }
>  }
> diff --git a/memory.c b/memory.c
> index b9dd6b94ca..57c44c97db 100644
> --- a/memory.c
> +++ b/memory.c
> @@ -438,7 +438,6 @@ static MemTxResult  memory_region_read_accessor(MemoryRegion *mr,
>          /* Accesses to code which has previously been translated into a TB show
>           * up in the MMIO path, as accesses to the io_mem_notdirty
>           * MemoryRegion. */
> -        trace_memory_region_tb_read(get_cpu_index(), addr, tmp, size);
>      } else if (TRACE_MEMORY_REGION_OPS_READ_ENABLED) {
>          hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
>          trace_memory_region_ops_read(get_cpu_index(), mr, abs_addr, tmp, size);
> @@ -465,7 +464,6 @@ static MemTxResult memory_region_read_with_attrs_accessor(MemoryRegion *mr,
>          /* Accesses to code which has previously been translated into a TB show
>           * up in the MMIO path, as accesses to the io_mem_notdirty
>           * MemoryRegion. */
> -        trace_memory_region_tb_read(get_cpu_index(), addr, tmp, size);
>      } else if (TRACE_MEMORY_REGION_OPS_READ_ENABLED) {
>          hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
>          trace_memory_region_ops_read(get_cpu_index(), mr, abs_addr, tmp, size);
> @@ -490,7 +488,6 @@ static MemTxResult memory_region_write_accessor(MemoryRegion *mr,
>          /* Accesses to code which has previously been translated into a TB show
>           * up in the MMIO path, as accesses to the io_mem_notdirty
>           * MemoryRegion. */
> -        trace_memory_region_tb_write(get_cpu_index(), addr, tmp, size);
>      } else if (TRACE_MEMORY_REGION_OPS_WRITE_ENABLED) {
>          hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
>          trace_memory_region_ops_write(get_cpu_index(), mr, abs_addr, tmp, size);
> @@ -515,7 +512,6 @@ static MemTxResult memory_region_write_with_attrs_accessor(MemoryRegion *mr,
>          /* Accesses to code which has previously been translated into a TB show
>           * up in the MMIO path, as accesses to the io_mem_notdirty
>           * MemoryRegion. */
> -        trace_memory_region_tb_write(get_cpu_index(), addr, tmp, size);
>      } else if (TRACE_MEMORY_REGION_OPS_WRITE_ENABLED) {
>          hwaddr abs_addr = memory_region_to_absolute_addr(mr, addr);
>          trace_memory_region_ops_write(get_cpu_index(), mr, abs_addr, tmp, size);
> diff --git a/trace-events b/trace-events
> index 823a4ae64e..5c9a1631e7 100644
> --- a/trace-events
> +++ b/trace-events
> @@ -52,14 +52,14 @@ dma_map_wait(void *dbs) "dbs=%p"
>  find_ram_offset(uint64_t size, uint64_t offset) "size: 0x%" PRIx64 " @ 0x%" PRIx64
>  find_ram_offset_loop(uint64_t size, uint64_t candidate, uint64_t offset, uint64_t next, uint64_t mingap) "trying size: 0x%" PRIx64 " @ 0x%" PRIx64 ", offset: 0x%" PRIx64" next: 0x%" PRIx64 " mingap: 0x%" PRIx64
>  ram_block_discard_range(const char *rbname, void *hva, size_t length, bool need_madvise, bool need_fallocate, int ret) "%s@%p + 0x%zx: madvise: %d fallocate: %d ret: %d"
> +memory_notdirty_write(uint64_t vaddr, uint64_t ram_addr, unsigned size) "0x%" PRIx64 " ram_addr 0x%" PRIx64 " size %u"
> +memory_notdirty_dirty(uint64_t vaddr) "0x%" PRIx64

My only suggestion would be to give slightly better names like

memory_notdirty_write_access()
memory_notdirty_set_dirty()

-- 

Thanks,

David / dhildenb


  reply	other threads:[~2019-09-18  8:35 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-18  5:26 [Qemu-devel] [RFC 0/3] Move notdirty handling to cputlb Richard Henderson
2019-09-18  5:26 ` [Qemu-devel] [RFC 1/3] exec: Adjust notdirty tracing Richard Henderson
2019-09-18  8:34   ` David Hildenbrand [this message]
2019-09-18  5:26 ` [Qemu-devel] [RFC 2/3] cputlb: Move NOTDIRTY handling from I/O path to TLB path Richard Henderson
2019-09-18  5:26 ` [Qemu-devel] [RFC 3/3] cputlb: Remove ATOMIC_MMU_DECLS Richard Henderson
2019-09-18  8:13 ` [Qemu-devel] [RFC 0/3] Move notdirty handling to cputlb Paolo Bonzini
2019-09-18  8:30 ` David Hildenbrand

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=38fead47-5d6e-5f67-c5b0-93273b22c0a7@redhat.com \
    --to=david@redhat.com \
    --cc=alex.bennee@linaro.org \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).