On 2020-06-11, Petr Mladek wrote:
> All this relies on the fact the the full barrier is called in
> data_push_tail() and data_push_tail() is called right above.
> But there are two situations where the barrier is not called.
> It is when:
>
> 1. desc.text_blk_lpos.next already is behind data_rin
On 2020-06-11, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +/*
>> + * Given a data ring (text or dict), put the associated descriptor of
>> each
>> + * data block from @lpos_begin until @lpos_end into the reusable state.
>> + *
>>>
On 2020-06-10, John Ogness wrote:
>>> +static bool data_make_reusable(struct printk_ringbuffer *rb,
>>> + struct prb_data_ring *data_ring,
>>> + unsigned long lpos_begin,
>>> + unsigned long lpos_end,
>>> +
On Wed 2020-06-10 12:24:01, John Ogness wrote:
> On 2020-06-10, Petr Mladek wrote:
> >> >> --- /dev/null
> >> >> +++ b/kernel/printk/printk_ringbuffer.c
> >> >> +/*
> >> >> + * Given a data ring (text or dict), put the associated descriptor of
> >> >> each
> >> >> + * data block from @lpos_begin
On Tue 2020-06-09 17:56:03, John Ogness wrote:
> On 2020-06-09, Petr Mladek wrote:
> >> --- /dev/null
> >> +++ b/kernel/printk/printk_ringbuffer.c
> >> +/*
> >> + * Advance the desc ring tail. This function advances the tail by one
> >> + * descriptor, thus invalidating the oldest descriptor. Befo
On 2020-06-10, Petr Mladek wrote:
>> +static bool data_make_reusable(struct printk_ringbuffer *rb,
>> + struct prb_data_ring *data_ring,
>> + unsigned long lpos_begin,
>> + unsigned long lpos_end,
>> +
On 2020-06-10, Petr Mladek wrote:
--- /dev/null
+++ b/kernel/printk/printk_ringbuffer.c
+/*
+ * Given a data ring (text or dict), put the associated descriptor of each
+ * data block from @lpos_begin until @lpos_end into the reusable state.
+ *
+ * If there is an
On 2020-06-10, Petr Mladek wrote:
>> >> --- /dev/null
>> >> +++ b/kernel/printk/printk_ringbuffer.c
>> >> +/*
>> >> + * Given a data ring (text or dict), put the associated descriptor of
>> >> each
>> >> + * data block from @lpos_begin until @lpos_end into the reusable state.
>> >> + *
+ * I
On Tue 2020-06-09 16:48:30, John Ogness wrote:
> On 2020-06-09, Petr Mladek wrote:
> >> --- /dev/null
> >> +++ b/kernel/printk/printk_ringbuffer.c
> >> +/*
> >> + * Given a data ring (text or dict), put the associated descriptor of each
> >> + * data block from @lpos_begin until @lpos_end into the
On Tue 2020-06-09 16:18:35, John Ogness wrote:
> On 2020-06-09, Petr Mladek wrote:
> >> --- /dev/null
> >> +++ b/kernel/printk/printk_ringbuffer.c
> >> +/*
> >> + * Given a data ring (text or dict), put the associated descriptor of each
> >> + * data block from @lpos_begin until @lpos_end into the
On 2020-06-09, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +/*
>> + * Advance the data ring tail to at least @lpos. This function puts
>> + * descriptors into the reusable state if the tail is pushed beyond
>> + * their associated data block.
>> + */
>> +stati
On 2020-06-09, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +/*
>> + * Advance the desc ring tail. This function advances the tail by one
>> + * descriptor, thus invalidating the oldest descriptor. Before advancing
>> + * the tail, the tail descriptor is made r
On 2020-06-09, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +static bool data_push_tail(struct printk_ringbuffer *rb,
>> + struct prb_data_ring *data_ring,
>> + unsigned long lpos)
>> +{
>> ...
>> +
>> +/*
On 2020-06-09, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +/*
>> + * Given a data ring (text or dict), put the associated descriptor of each
>> + * data block from @lpos_begin until @lpos_end into the reusable state.
>> + *
>> + * If there is any problem maki
On Fri 2020-05-01 11:46:09, John Ogness wrote:
> Introduce a multi-reader multi-writer lockless ringbuffer for storing
> the kernel log messages. Readers and writers may use their API from
> any context (including scheduler and NMI). This ringbuffer will make
> it possible to decouple printk() call
On 2020-06-09, Petr Mladek wrote:
>> --- /dev/null
>> +++ b/kernel/printk/printk_ringbuffer.c
>> +/*
>> + * Given a data ring (text or dict), put the associated descriptor of each
>> + * data block from @lpos_begin until @lpos_end into the reusable state.
>> + *
>> + * If there is any problem maki
On Fri 2020-05-01 11:46:09, John Ogness wrote:
> Introduce a multi-reader multi-writer lockless ringbuffer for storing
> the kernel log messages. Readers and writers may use their API from
> any context (including scheduler and NMI). This ringbuffer will make
> it possible to decouple printk() call
On Fri 2020-05-01 11:46:09, John Ogness wrote:
> Introduce a multi-reader multi-writer lockless ringbuffer for storing
> the kernel log messages. Readers and writers may use their API from
> any context (including scheduler and NMI). This ringbuffer will make
> it possible to decouple printk() call
On Fri 2020-05-01 11:46:09, John Ogness wrote:
> Introduce a multi-reader multi-writer lockless ringbuffer for storing
> the kernel log messages. Readers and writers may use their API from
> any context (including scheduler and NMI). This ringbuffer will make
> it possible to decouple printk() call
On Fri 2020-05-01 11:46:09, John Ogness wrote:
> Introduce a multi-reader multi-writer lockless ringbuffer for storing
> the kernel log messages. Readers and writers may use their API from
> any context (including scheduler and NMI). This ringbuffer will make
> it possible to decouple printk() call
On 2020-05-18, Linus Torvalds wrote:
>> smp_mb(); /* LMM(data_push_tail:C) */
>>
>> + if (atomic_long_try_cmpxchg_relaxed(&data_ring->tail_lpos,
>> + &tail_lpos,
>> + next_lpos)) { /* LMM(data_push_tail:D) */
On Mon, May 18, 2020 at 6:04 AM John Ogness wrote:
>
> The cmpxchg() needs to be moved out of the while condition so that a
> continue can be used as intended. Patch below.
This seems pointless and wrong (patch edited to remove the '-' lines
so that you see the end result):
> smp
Introduce a multi-reader multi-writer lockless ringbuffer for storing
the kernel log messages. Readers and writers may use their API from
any context (including scheduler and NMI). This ringbuffer will make
it possible to decouple printk() callers from any context, locking,
or console constraints.
23 matches
Mail list logo
