MIME-Version: 1.0
In-Reply-To: <20161116190504.0B2BCCFC@viggo.jf.intel.com>
References: <20161116190504.0B2BCCFC@viggo.jf.intel.com>
From: Dan Williams <dan.j.williams@intel.com>
Date: Wed, 16 Nov 2016 11:39:41 -0800
Message-ID: <CAPcyv4j-uVyvgfELab9Jjy6yr3cgvSGnHDfsvNn1+-tzwp48eg@mail.gmail.com>
Subject: Re: [RFC][PATCH] proc: mm: export PTE sizes directly in smaps
To: Dave Hansen <dave@sr71.net>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Christoph Hellwig <hch@lst.de>,
        Andrew Morton <akpm@linux-foundation.org>,
        Anshuman Khandual <khandual@linux.vnet.ibm.com>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2958
Lines: 78

On Wed, Nov 16, 2016 at 11:05 AM, Dave Hansen <dave@sr71.net> wrote:
>
> /proc/$pid/smaps has a number of fields that are intended to imply the
> kinds of PTEs used to map memory.  "AnonHugePages" obviously tells you
> how many PMDs are being used.  "MMUPageSize" along with the "Hugetlb"
> fields tells you how many PTEs you have for a huge page.
>
> The current mechanisms work fine when we have one or two page sizes.
> But, they start to get a bit muddled when we mix page sizes inside
> one VMA.  For instance, the DAX folks were proposing adding a set of
> fields like:
>
>         DevicePages:
>         DeviceHugePages:
>         DeviceGiganticPages:
>         DeviceGinormousPages:
>
> to unmuddle things when page sizes get mixed.  That's fine, but
> it does require userspace know the mapping from our various
> arbitrary names to hardware page sizes on each architecture and
> kernel configuration.
>
> What folks really want is to know how much memory is mapped with
> each page size.  How about we just do *that*?
>
> Patch attached.  Seems harmless enough.  Probably doesn't compile
> everywhere.  Makes smaps look like this:
>
> Private_Hugetlb:       0 kB
> Swap:                  0 kB
> SwapPss:               0 kB
> KernelPageSize:        4 kB
> MMUPageSize:           4 kB
> Locked:                0 kB
> Ptes:               4096 kB @        4 kB     4096 kB @     2048 kB
> VmFlags: rd wr mr mw me ac
>
> 1. I'd like to thank Dan Williams for showing me a mirror as I
>    complained about the bozo that introduced 'AnonHugePages'.
>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com>
>
> ---
>
>  b/fs/proc/task_mmu.c |   38 ++++++++++++++++++++++++++++++++++++--
>  1 file changed, 36 insertions(+), 2 deletions(-)
>
> diff -puN fs/proc/task_mmu.c~smaps-pte-sizes fs/proc/task_mmu.c
> --- a/fs/proc/task_mmu.c~smaps-pte-sizes        2016-11-16 08:47:35.609910046 -0800
> +++ b/fs/proc/task_mmu.c        2016-11-16 10:40:40.529937746 -0800
> @@ -445,6 +445,9 @@ struct mem_size_stats {
>         unsigned long swap;
>         unsigned long shared_hugetlb;
>         unsigned long private_hugetlb;
> +       unsigned long rss_pte;
> +       unsigned long rss_pmd;
> +       unsigned long rss_pud;
>         u64 pss;
>         u64 swap_pss;
>         bool check_shmem_swap;
> @@ -519,6 +522,7 @@ static void smaps_pte_entry(pte_t *pte,
>
>         if (pte_present(*pte)) {
>                 page = vm_normal_page(vma, addr, *pte);
> +               mss->rss_pte += PAGE_SIZE;
>         } else if (is_swap_pte(*pte)) {
>                 swp_entry_t swpent = pte_to_swp_entry(*pte);
>
> @@ -578,6 +582,7 @@ static void smaps_pmd_entry(pmd_t *pmd,
>                 /* pass */;
>         else
>                 VM_BUG_ON_PAGE(1, page);
> +       mss->rss_pmd += PAGE_SIZE;

Should this be "HPAGE_PMD_SIZE"?