Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1211557

Re: [PATCH 0/3] mm/vmalloc: Cache the /proc/meminfo vmalloc statistics

Show all headers | View raw

Linus wrote:
> I don't think any of this can be called "correct", in that the
> unlocked accesses to the cached state are clearly racy, but I think
> it's very much "acceptable".

I'd think you could easily fix that with a seqlock-like system.

What makes it so simple is that you can always fall back to
calc_vmalloc_info if there's any problem, rather than looping or blocking.

The basic idea is that you have a seqlock counter, but if either of
the two lsbits are set, the cached information is stale.

Basically, you need a seqlock and a spinlock.  The seqlock does
most of the work, and the spinlock ensures that there's only one
updater of the cache.

vmap_unlock() does set_bit(0, &seq->sequence).  This marks the information
as stale.

get_vmalloc_info reads the seqlock.  There are two case:
- If the two lsbits are 10, the cached information is valid.
  Copy it out, re-check the seqlock, and loop if the sequence
  number changes.
- In any other case, the cached information is
  not valid.
  - Try to obtain the spinlock.  Do not block if it's unavailable.
    - If unavailable, do not block.
    - If the lock is acquired:
      - Set the sequence to (sequence | 3) + 1 (we're the only writer)
      - This bumps the sequence number and leaves the lsbits at 00 (invalid)
      - Memory barrier TBD.  Do the RCU ops in calc_vmalloc_info do it for us?
  - Call calc_vmalloc_info
  - If we obtained the spinlock earlier:
    - Copy our vmi to cached_info
    - smp_wmb()
    - set_bit(1, &seq->sequence).  This marks the information as valid,
      as long as bit 0 is still clear.
    - Release the spinlock.

Basically, bit 0 says "vmalloc info has changed", and bit 1 says
"vmalloc cache has been updated".  This clears bit 0 before
starting the update so that an update during calc_vmalloc_info
will force a new update.

So the three case are basically:
00 - calc_vmalloc_info() in progress
01 - vmap_unlock() during calc_vmalloc_info()
10 - cached_info is valid
11 - vmap_unlock has invalidated cached_info, awaiting refresh

Logically, the sequence number should be initialized to ...01,
but the code above handles 00 okay.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Next in thread | Find similar | Unroll thread

Thread

Re: [PATCH 0/3] mm/vmalloc: Cache the /proc/meminfo vmalloc statistics "George Spelvin" <linux@horizon.com> - 2015-08-23 06:50 +0200
  Re: [PATCH 0/3] mm/vmalloc: Cache the /proc/meminfo vmalloc  statistics Ingo Molnar <mingo@kernel.org> - 2015-08-23 08:10 +0200
    Re: [PATCH 0/3] mm/vmalloc: Cache the /proc/meminfo vmalloc statistics "George Spelvin" <linux@horizon.com> - 2015-08-23 08:50 +0200
      [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-23 10:20 +0200
        Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Rasmus Villemoes <linux@rasmusvillemoes.dk> - 2015-08-23 23:00 +0200
          Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-24 09:00 +0200
            Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Rasmus Villemoes <linux@rasmusvillemoes.dk> - 2015-08-24 10:40 +0200
        Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Rasmus Villemoes <linux@rasmusvillemoes.dk> - 2015-08-24 00:00 +0200
          Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-24 09:10 +0200
          Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Linus Torvalds <torvalds@linux-foundation.org> - 2015-08-25 18:40 +0200
            Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Linus Torvalds <torvalds@linux-foundation.org> - 2015-08-25 19:10 +0200
        Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info "George Spelvin" <linux@horizon.com> - 2015-08-24 03:10 +0200
          [PATCH 3/3 v4] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-24 09:40 +0200
            Re: [PATCH 3/3 v4] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-24 09:50 +0200
              [PATCH 3/3 v5] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-24 10:00 +0200
                Re: [PATCH 3/3 v5] mm/vmalloc: Cache the vmalloc memory info "George Spelvin" <linux@horizon.com> - 2015-08-24 15:00 +0200
                [PATCH 3/3 v6] mm/vmalloc: Cache the vmalloc memory info Ingo Molnar <mingo@kernel.org> - 2015-08-25 12:00 +0200
                Re: [PATCH 3/3 v6] mm/vmalloc: Cache the vmalloc memory info "George Spelvin" <linux@horizon.com> - 2015-08-25 12:40 +0200
                Re: [PATCH 3/3 v6] mm/vmalloc: Cache the vmalloc memory info Peter Zijlstra <peterz@infradead.org> - 2015-08-25 15:10 +0200
                Re: [PATCH 3/3 v6] mm/vmalloc: Cache the vmalloc memory info Rasmus Villemoes <linux@rasmusvillemoes.dk> - 2015-08-25 16:30 +0200
                Re: [PATCH 3/3 v6] mm/vmalloc: Cache the vmalloc memory info "George Spelvin" <linux@horizon.com> - 2015-08-25 17:20 +0200
            Re: [PATCH 3/3 v4] mm/vmalloc: Cache the vmalloc memory info "John Stoffel" <john@stoffel.org> - 2015-08-24 15:50 +0200
              Re: [PATCH 3/3 v4] mm/vmalloc: Cache the vmalloc memory info "George Spelvin" <linux@horizon.com> - 2015-08-24 17:20 +0200
                Re: [PATCH 3/3 v4] mm/vmalloc: Cache the vmalloc memory info "John Stoffel" <john@stoffel.org> - 2015-08-24 18:00 +0200
        Re: [PATCH 3/3 v3] mm/vmalloc: Cache the vmalloc memory info Peter Zijlstra <peterz@infradead.org> - 2015-08-25 14:50 +0200

csiph-web