lkml.org 
[lkml]   [2017]   [Oct]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    SubjectRe: [PATCH v6 02/11] locking/rwsem: Implement a new locking scheme
    From
    Date
    On 10/11/2017 02:58 PM, Waiman Long wrote:
    > On 10/11/2017 02:40 PM, Peter Zijlstra wrote:
    >> On Wed, Oct 11, 2017 at 02:01:53PM -0400, Waiman Long wrote:
    >>> +/*
    >>> + * The definition of the atomic counter in the semaphore:
    >>> + *
    >>> + * Bit 0 - writer locked bit
    >>> + * Bit 1 - waiters present bit
    >>> + * Bits 2-7 - reserved
    >>> + * Bits 8-31 - 24-bit reader count
    >>> + *
    >>> + * atomic_fetch_add() is used to obtain reader lock, whereas atomic_cmpxchg()
    >>> + * will be used to obtain writer lock.
    >>> + */
    >>> +#define RWSEM_WRITER_LOCKED 0X00000001
    >>> +#define RWSEM_FLAG_WAITERS 0X00000002
    >>> +#define RWSEM_READER_BIAS 0x00000100
    >>> +#define RWSEM_READER_SHIFT 8
    >>> +#define RWSEM_READER_MASK (~((1U << RWSEM_READER_SHIFT) - 1))
    >>> +#define RWSEM_LOCK_MASK (RWSEM_WRITER_LOCKED|RWSEM_READER_MASK)
    >>> +#define RWSEM_READ_FAILED_MASK (RWSEM_WRITER_LOCKED|RWSEM_FLAG_WAITERS)
    >>> +
    >>> +#define RWSEM_COUNT_IS_LOCKED(c) ((c) & RWSEM_LOCK_MASK)
    >>> +
    >>> +/*
    >>> + * lock for reading
    >>> + */
    >>> +static inline void __down_read(struct rw_semaphore *sem)
    >>> +{
    >>> + if (unlikely(atomic_fetch_add_acquire(RWSEM_READER_BIAS, &sem->count)
    >>> + & RWSEM_READ_FAILED_MASK))
    >>> + rwsem_down_read_failed(sem);
    >>> +}
    >> So I implemented rwsem-mutex (also qrwlock based) that puts
    >>
    >> (unsigned long)current | RWSEM_WRITER
    >>
    >> in the atomic_long_t rw_semaphore::owner field. The down-side is that
    >> you can't do fetch_add based __down_read, because that would clobber the
    >> pointer. The up-side is that we have a stable owner pointer (which is
    >> what I needed for PI like things).
    > Without fetch_add for readers, it could lead to reduced performance for
    > reader heavy workloads.
    >
    > Are you trying to do a PI version of rwsem? It can work when the lock is
    > writer owned, but not when it is reader owned.

    I have actually been thinking about giving priority to RT or DL task by
    putting the task in front of the wait queue and assert the lock handoff
    bit, for example. There are extra reserved bits left that can be useful
    for adding these additional features. That will be later when I am done
    with the current patch.

    Cheers,
    Longman




    \
     
     \ /
      Last update: 2017-10-11 21:06    [W:3.878 / U:0.004 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site