Hi Michael ,
I have solved all the data race issue , it seems.
The data race culprit is the ring index which I have removed
in the latest datarace-free code.
The diff can be found at https://www.diffchecker.com/w0Pxp2mF
I plan to study the internal implementation of ptr_ring c coding
in the comi
Hi Michael,
> If I had to guess I'd say the way you play with indices is probably racy
> so you are producing an invalid index.
You are probably right.
I am suspecting item_recv_push_index and item_send_push_index in
https://gist.github.com/promach/7716ee8addcaa33fda140d74d1ad94d6#file-riffa_dri
On Fri, Feb 01, 2019 at 04:12:46PM +0800, fei phung wrote:
> > I am not sure what does assignment of pointers mean in this context.
> > ptr_ring is designed for a single producer and a single consumer. For
> > why it works see explanation about data dependencies in
> > Documentation/memory-barrier
> I am not sure what does assignment of pointers mean in this context.
> ptr_ring is designed for a single producer and a single consumer. For
> why it works see explanation about data dependencies in
> Documentation/memory-barriers.txt. You will have to be more specific
> about the data race tha
On Thu, Jan 31, 2019 at 01:16:31PM +0800, fei phung wrote:
> Hi,
>
> /*
> * Filename: circ_ring.c
> * Version: 1.0
> * Description: A circular buffer using API from
> * https://github.com/torvalds/linux/blob/master/include/linux/ptr_ring.h
> */
>
> ptr_ring's void** queue is just giving data
Hi,
/*
* Filename: circ_ring.c
* Version: 1.0
* Description: A circular buffer using API from
* https://github.com/torvalds/linux/blob/master/include/linux/ptr_ring.h
*/
ptr_ring's void** queue is just giving data race problem, running
consume() together with [assignment of pointers+produce(
Hi,
I am having data race from threadsanitizer report.
Could you see if there is still data race in the following code using
ptr_ring API ?
/*
* Filename: circ_ring.c
* Version: 1.0
* Description: A circular buffer using API from
* https://github.com/torvalds/linux/blob/master/include/linu
On Wed, Jan 16, 2019 at 06:48:41AM +, Cheng Fei Phung wrote:
> Hi,
>
> > https://gist.github.com/promach/65e9331d55a43a2815239430a28e29c6#file-circ_ring-c-L62
> > racy if there are multiple consumers.
> > just call ptr_ring_consume_any.
>
> If I modify pop_circ_queue() below to directly use p
Hi,
> https://gist.github.com/promach/65e9331d55a43a2815239430a28e29c6#file-circ_ring-c-L62
> racy if there are multiple consumers.
> just call ptr_ring_consume_any.
If I modify pop_circ_queue() below to directly use
ptr_ring_consume_any() without
ptr_ring_empty_any() , I am afraid I am running i
On Wed, Jan 16, 2019 at 01:10:31AM +0800, fei phung wrote:
> Hi,
>
> inline int pop_circ_queue(struct ptr_ring * buffer, struct item * item_pop)
> {
> if (!ptr_ring_empty_any(buffer)) // if (not empty)
> {
> DEBUG_MSG(KERN_INFO "Before pop, head = %u , tail =
> %u\n
Hi,
inline int pop_circ_queue(struct ptr_ring * buffer, struct item * item_pop)
{
if (!ptr_ring_empty_any(buffer)) // if (not empty)
{
DEBUG_MSG(KERN_INFO "Before pop, head = %u , tail =
%u\n", buffer->consumer_head, buffer->consumer_tail);
/* extra
On Tue, Jan 15, 2019 at 12:33:28PM +0800, fei phung wrote:
> Hi netdev mailing list and Michael,
>
> I am having problem getting proper ptr_ring operation where one
> ptr_ring entry (val1=2 for item_recv_pop) is missing from the void **
> queue
>
> Did I initialize the ring pointers (ptr_ring_ini
Hi netdev mailing list and Michael,
I am having problem getting proper ptr_ring operation where one
ptr_ring entry (val1=2 for item_recv_pop) is missing from the void **
queue
Did I initialize the ring pointers (ptr_ring_init()) correctly ?
See the following for more context:
https://i.imgur.co
13 matches
Mail list logo
