* [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic
@ 2019-05-09 4:39 Nikhil Rao
2019-05-09 4:39 ` Nikhil Rao
` (2 more replies)
0 siblings, 3 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-09 4:39 UTC (permalink / raw)
To: jerinj; +Cc: dev, matias.elo, Nikhil Rao
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8d178be..fd3761b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic
2019-05-09 4:39 [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic Nikhil Rao
@ 2019-05-09 4:39 ` Nikhil Rao
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
2019-05-10 8:08 ` [dpdk-dev] [PATCH v3] " Nikhil Rao
2 siblings, 0 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-09 4:39 UTC (permalink / raw)
To: jerinj; +Cc: dev, matias.elo, Nikhil Rao
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8d178be..fd3761b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 4:39 [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic Nikhil Rao
2019-05-09 4:39 ` Nikhil Rao
@ 2019-05-09 4:50 ` Nikhil Rao
2019-05-09 4:50 ` Nikhil Rao
` (2 more replies)
2019-05-10 8:08 ` [dpdk-dev] [PATCH v3] " Nikhil Rao
2 siblings, 3 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-09 4:50 UTC (permalink / raw)
To: jerinj; +Cc: dev, matias.elo, Nikhil Rao
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Bugzilla ID: 277
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
v2:
* added Bugzilla ID
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8d178be..fd3761b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
@ 2019-05-09 4:50 ` Nikhil Rao
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:22 ` Mattias Rönnblom
2 siblings, 0 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-09 4:50 UTC (permalink / raw)
To: jerinj; +Cc: dev, matias.elo, Nikhil Rao
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Bugzilla ID: 277
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
v2:
* added Bugzilla ID
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8d178be..fd3761b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
2019-05-09 4:50 ` Nikhil Rao
@ 2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:17 ` Thomas Monjalon
` (2 more replies)
2019-05-09 19:22 ` Mattias Rönnblom
2 siblings, 3 replies; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-09 19:17 UTC (permalink / raw)
To: Nikhil Rao; +Cc: dev, jerinj, matias.elo
09/05/2019 06:50, Nikhil Rao:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count)
I don't this code, but it looks you are removing a test
which was probably done on purpose.
By the way, you should make your test explicit with "!= 0".
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:17 ` Thomas Monjalon
@ 2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:23 ` Mattias Rönnblom
2019-05-10 8:19 ` Rao, Nikhil
2 siblings, 0 replies; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-09 19:17 UTC (permalink / raw)
To: Nikhil Rao; +Cc: dev, jerinj, matias.elo
09/05/2019 06:50, Nikhil Rao:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count)
I don't this code, but it looks you are removing a test
which was probably done on purpose.
By the way, you should make your test explicit with "!= 0".
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:17 ` Thomas Monjalon
@ 2019-05-09 19:23 ` Mattias Rönnblom
2019-05-09 19:23 ` Mattias Rönnblom
2019-05-10 8:19 ` Rao, Nikhil
2 siblings, 1 reply; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-09 19:23 UTC (permalink / raw)
To: Thomas Monjalon, Nikhil Rao; +Cc: dev, jerinj, matias.elo
On 2019-05-09 21:17, Thomas Monjalon wrote:
> 09/05/2019 06:50, Nikhil Rao:
>> The Rx adapter flushes events only if it has BATCH_SIZE
>> events buffered where BATCH_SIZE is set to 32, e.g., if a
>> single packet is sent, it is never passed to
>> eventdev. Fix this issue by adding an event buffer flush
>> either when a Rx queue is found to be empty or the adapter service
>> function has processed the max number of packets for an invocation.
>>
>> Bugzilla ID: 277
>>
>> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>>
>> Reported-by: Matias Elo <matias.elo@nokia.com>
>>
>> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
>> ---
>> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
>> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
>> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
>> - if (buf->count >= BATCH_SIZE)
>> + if (buf->count)
>
> I don't this code, but it looks you are removing a test
> which was probably done on purpose.
> By the way, you should make your test explicit with "!= 0".
>
"> 0". count is unsigned.
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:23 ` Mattias Rönnblom
@ 2019-05-09 19:23 ` Mattias Rönnblom
0 siblings, 0 replies; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-09 19:23 UTC (permalink / raw)
To: Thomas Monjalon, Nikhil Rao; +Cc: dev, jerinj, matias.elo
On 2019-05-09 21:17, Thomas Monjalon wrote:
> 09/05/2019 06:50, Nikhil Rao:
>> The Rx adapter flushes events only if it has BATCH_SIZE
>> events buffered where BATCH_SIZE is set to 32, e.g., if a
>> single packet is sent, it is never passed to
>> eventdev. Fix this issue by adding an event buffer flush
>> either when a Rx queue is found to be empty or the adapter service
>> function has processed the max number of packets for an invocation.
>>
>> Bugzilla ID: 277
>>
>> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>>
>> Reported-by: Matias Elo <matias.elo@nokia.com>
>>
>> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
>> ---
>> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
>> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
>> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
>> - if (buf->count >= BATCH_SIZE)
>> + if (buf->count)
>
> I don't this code, but it looks you are removing a test
> which was probably done on purpose.
> By the way, you should make your test explicit with "!= 0".
>
"> 0". count is unsigned.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:23 ` Mattias Rönnblom
@ 2019-05-10 8:19 ` Rao, Nikhil
2019-05-10 8:19 ` Rao, Nikhil
2 siblings, 1 reply; 24+ messages in thread
From: Rao, Nikhil @ 2019-05-10 8:19 UTC (permalink / raw)
To: Thomas Monjalon; +Cc: dev, jerinj, matias.elo
> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas@monjalon.net]
> Sent: Friday, May 10, 2019 12:48 AM
> To: Rao, Nikhil <nikhil.rao@intel.com>
> Cc: dev@dpdk.org; jerinj@marvell.com; matias.elo@nokia.com
> Subject: Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
>
> 09/05/2019 06:50, Nikhil Rao:
> > The Rx adapter flushes events only if it has BATCH_SIZE events
> > buffered where BATCH_SIZE is set to 32, e.g., if a single packet is
> > sent, it is never passed to eventdev. Fix this issue by adding an
> > event buffer flush either when a Rx queue is found to be empty or the
> > adapter service function has processed the max number of packets for
> > an invocation.
> >
> > Bugzilla ID: 277
> >
> > Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
> >
> > Reported-by: Matias Elo <matias.elo@nokia.com>
> >
> > Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> > ---
> > --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> > - if (buf->count >= BATCH_SIZE)
> > + if (buf->count)
>
> I don't this code, but it looks you are removing a test which was probably
> done on purpose.
The previous test (against BATCH_SIZE) resulted in events permanently sitting
in the buffer if the number of events in the buffer was less than BATCH_SIZE.
> By the way, you should make your test explicit with "!= 0".
>
Done.
Thanks,
Nikhil
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-10 8:19 ` Rao, Nikhil
@ 2019-05-10 8:19 ` Rao, Nikhil
0 siblings, 0 replies; 24+ messages in thread
From: Rao, Nikhil @ 2019-05-10 8:19 UTC (permalink / raw)
To: Thomas Monjalon; +Cc: dev, jerinj, matias.elo
> -----Original Message-----
> From: Thomas Monjalon [mailto:thomas@monjalon.net]
> Sent: Friday, May 10, 2019 12:48 AM
> To: Rao, Nikhil <nikhil.rao@intel.com>
> Cc: dev@dpdk.org; jerinj@marvell.com; matias.elo@nokia.com
> Subject: Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
>
> 09/05/2019 06:50, Nikhil Rao:
> > The Rx adapter flushes events only if it has BATCH_SIZE events
> > buffered where BATCH_SIZE is set to 32, e.g., if a single packet is
> > sent, it is never passed to eventdev. Fix this issue by adding an
> > event buffer flush either when a Rx queue is found to be empty or the
> > adapter service function has processed the max number of packets for
> > an invocation.
> >
> > Bugzilla ID: 277
> >
> > Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
> >
> > Reported-by: Matias Elo <matias.elo@nokia.com>
> >
> > Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> > ---
> > --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> > - if (buf->count >= BATCH_SIZE)
> > + if (buf->count)
>
> I don't this code, but it looks you are removing a test which was probably
> done on purpose.
The previous test (against BATCH_SIZE) resulted in events permanently sitting
in the buffer if the number of events in the buffer was less than BATCH_SIZE.
> By the way, you should make your test explicit with "!= 0".
>
Done.
Thanks,
Nikhil
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
2019-05-09 4:50 ` Nikhil Rao
2019-05-09 19:17 ` Thomas Monjalon
@ 2019-05-09 19:22 ` Mattias Rönnblom
2019-05-09 19:22 ` Mattias Rönnblom
2019-05-10 8:20 ` Rao, Nikhil
2 siblings, 2 replies; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-09 19:22 UTC (permalink / raw)
To: Nikhil Rao, jerinj; +Cc: dev, matias.elo
On 2019-05-09 06:50, Nikhil Rao wrote:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> v2:
> * added Bugzilla ID
>
> diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> index 8d178be..fd3761b 100644
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> break;
> }
>
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count)
> rxa_flush_event_buffer(rx_adapter);
>
> return nb_rx;
>
After this change, is there any point in having the event_enqueue_buffer
being a part of the rx_adapter struct? It could be stack allocated. In
the process, you could eliminate the rte_event copying that goes on and
the logic around that. Just make mbufs ETH_EVENT_BUFFER_SIZE in size,
and retry until the NIC queue is empty or you accumulated
ETH_EVENT_BUFFER_SIZE events. Then you ship it off to the event device.
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:22 ` Mattias Rönnblom
@ 2019-05-09 19:22 ` Mattias Rönnblom
2019-05-10 8:20 ` Rao, Nikhil
1 sibling, 0 replies; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-09 19:22 UTC (permalink / raw)
To: Nikhil Rao, jerinj; +Cc: dev, matias.elo
On 2019-05-09 06:50, Nikhil Rao wrote:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> v2:
> * added Bugzilla ID
>
> diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> index 8d178be..fd3761b 100644
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> break;
> }
>
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count)
> rxa_flush_event_buffer(rx_adapter);
>
> return nb_rx;
>
After this change, is there any point in having the event_enqueue_buffer
being a part of the rx_adapter struct? It could be stack allocated. In
the process, you could eliminate the rte_event copying that goes on and
the logic around that. Just make mbufs ETH_EVENT_BUFFER_SIZE in size,
and retry until the NIC queue is empty or you accumulated
ETH_EVENT_BUFFER_SIZE events. Then you ship it off to the event device.
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-09 19:22 ` Mattias Rönnblom
2019-05-09 19:22 ` Mattias Rönnblom
@ 2019-05-10 8:20 ` Rao, Nikhil
2019-05-10 8:20 ` Rao, Nikhil
1 sibling, 1 reply; 24+ messages in thread
From: Rao, Nikhil @ 2019-05-10 8:20 UTC (permalink / raw)
To: Mattias Rönnblom, jerinj; +Cc: dev, matias.elo
> -----Original Message-----
> From: Mattias Rönnblom [mailto:mattias.ronnblom@ericsson.com]
> Sent: Friday, May 10, 2019 12:52 AM
> To: Rao, Nikhil <nikhil.rao@intel.com>; jerinj@marvell.com
> Cc: dev@dpdk.org; matias.elo@nokia.com
> Subject: Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
>
> On 2019-05-09 06:50, Nikhil Rao wrote:
> > The Rx adapter flushes events only if it has BATCH_SIZE events
> > buffered where BATCH_SIZE is set to 32, e.g., if a single packet is
> > sent, it is never passed to eventdev. Fix this issue by adding an
> > event buffer flush either when a Rx queue is found to be empty or the
> > adapter service function has processed the max number of packets for
> > an invocation.
> >
> > Bugzilla ID: 277
> >
> > Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
> >
> > Reported-by: Matias Elo <matias.elo@nokia.com>
> >
> > Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> > ---
> > lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > v2:
> > * added Bugzilla ID
> >
> > diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > index 8d178be..fd3761b 100644
> > --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> > break;
> > }
> >
> > - if (buf->count >= BATCH_SIZE)
> > + if (buf->count)
> > rxa_flush_event_buffer(rx_adapter);
> >
> > return nb_rx;
> >
>
> After this change, is there any point in having the event_enqueue_buffer
> being a part of the rx_adapter struct? It could be stack allocated. In the
> process, you could eliminate the rte_event copying that goes on and the
> logic around that. Just make mbufs ETH_EVENT_BUFFER_SIZE in size, and
> retry until the NIC queue is empty or you accumulated
> ETH_EVENT_BUFFER_SIZE events. Then you ship it off to the event device.
You are right, I will post a separate (from the bug fix) patch for this.
Thanks,
Nikhil
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
2019-05-10 8:20 ` Rao, Nikhil
@ 2019-05-10 8:20 ` Rao, Nikhil
0 siblings, 0 replies; 24+ messages in thread
From: Rao, Nikhil @ 2019-05-10 8:20 UTC (permalink / raw)
To: Mattias Rönnblom, jerinj; +Cc: dev, matias.elo
> -----Original Message-----
> From: Mattias Rönnblom [mailto:mattias.ronnblom@ericsson.com]
> Sent: Friday, May 10, 2019 12:52 AM
> To: Rao, Nikhil <nikhil.rao@intel.com>; jerinj@marvell.com
> Cc: dev@dpdk.org; matias.elo@nokia.com
> Subject: Re: [dpdk-dev] [PATCH v2] eventdev: fix Rx adapter event flush logic
>
> On 2019-05-09 06:50, Nikhil Rao wrote:
> > The Rx adapter flushes events only if it has BATCH_SIZE events
> > buffered where BATCH_SIZE is set to 32, e.g., if a single packet is
> > sent, it is never passed to eventdev. Fix this issue by adding an
> > event buffer flush either when a Rx queue is found to be empty or the
> > adapter service function has processed the max number of packets for
> > an invocation.
> >
> > Bugzilla ID: 277
> >
> > Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
> >
> > Reported-by: Matias Elo <matias.elo@nokia.com>
> >
> > Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> > ---
> > lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > v2:
> > * added Bugzilla ID
> >
> > diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > index 8d178be..fd3761b 100644
> > --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> > @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> > break;
> > }
> >
> > - if (buf->count >= BATCH_SIZE)
> > + if (buf->count)
> > rxa_flush_event_buffer(rx_adapter);
> >
> > return nb_rx;
> >
>
> After this change, is there any point in having the event_enqueue_buffer
> being a part of the rx_adapter struct? It could be stack allocated. In the
> process, you could eliminate the rte_event copying that goes on and the
> logic around that. Just make mbufs ETH_EVENT_BUFFER_SIZE in size, and
> retry until the NIC queue is empty or you accumulated
> ETH_EVENT_BUFFER_SIZE events. Then you ship it off to the event device.
You are right, I will post a separate (from the bug fix) patch for this.
Thanks,
Nikhil
^ permalink raw reply [flat|nested] 24+ messages in thread
* [dpdk-dev] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-09 4:39 [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic Nikhil Rao
2019-05-09 4:39 ` Nikhil Rao
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
@ 2019-05-10 8:08 ` Nikhil Rao
2019-05-10 8:08 ` Nikhil Rao
2019-05-10 13:30 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2 siblings, 2 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-10 8:08 UTC (permalink / raw)
To: thomas; +Cc: jerinj, dev, matias.elo, Nikhil Rao, stable
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Bugzilla ID: 277
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
CC: stable@dpdk.org
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
v2:
* add Bugzilla ID
v3:
* add stable CC
* replace boolean check with check for > 0
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8b6e69d..b60674b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count > 0)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* [dpdk-dev] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 8:08 ` [dpdk-dev] [PATCH v3] " Nikhil Rao
@ 2019-05-10 8:08 ` Nikhil Rao
2019-05-10 13:30 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
1 sibling, 0 replies; 24+ messages in thread
From: Nikhil Rao @ 2019-05-10 8:08 UTC (permalink / raw)
To: thomas; +Cc: jerinj, dev, matias.elo, Nikhil Rao, stable
The Rx adapter flushes events only if it has BATCH_SIZE
events buffered where BATCH_SIZE is set to 32, e.g., if a
single packet is sent, it is never passed to
eventdev. Fix this issue by adding an event buffer flush
either when a Rx queue is found to be empty or the adapter service
function has processed the max number of packets for an invocation.
Bugzilla ID: 277
Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
Reported-by: Matias Elo <matias.elo@nokia.com>
CC: stable@dpdk.org
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
---
lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
v2:
* add Bugzilla ID
v3:
* add stable CC
* replace boolean check with check for > 0
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8b6e69d..b60674b 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
break;
}
- if (buf->count >= BATCH_SIZE)
+ if (buf->count > 0)
rxa_flush_event_buffer(rx_adapter);
return nb_rx;
--
1.8.3.1
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 8:08 ` [dpdk-dev] [PATCH v3] " Nikhil Rao
2019-05-10 8:08 ` Nikhil Rao
@ 2019-05-10 13:30 ` Thomas Monjalon
2019-05-10 13:30 ` Thomas Monjalon
2019-05-10 16:13 ` Mattias Rönnblom
1 sibling, 2 replies; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-10 13:30 UTC (permalink / raw)
To: dev; +Cc: stable, Nikhil Rao, jerinj, matias.elo
Any review please?
10/05/2019 10:08, Nikhil Rao:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> CC: stable@dpdk.org
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> v2:
> * add Bugzilla ID
>
> v3:
> * add stable CC
> * replace boolean check with check for > 0
>
> diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> index 8b6e69d..b60674b 100644
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> break;
> }
>
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count > 0)
> rxa_flush_event_buffer(rx_adapter);
>
> return nb_rx;
>
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 13:30 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
@ 2019-05-10 13:30 ` Thomas Monjalon
2019-05-10 16:13 ` Mattias Rönnblom
1 sibling, 0 replies; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-10 13:30 UTC (permalink / raw)
To: dev; +Cc: stable, Nikhil Rao, jerinj, matias.elo
Any review please?
10/05/2019 10:08, Nikhil Rao:
> The Rx adapter flushes events only if it has BATCH_SIZE
> events buffered where BATCH_SIZE is set to 32, e.g., if a
> single packet is sent, it is never passed to
> eventdev. Fix this issue by adding an event buffer flush
> either when a Rx queue is found to be empty or the adapter service
> function has processed the max number of packets for an invocation.
>
> Bugzilla ID: 277
>
> Fixes: 6b83f5935543 ("eventdev: add event buffer flush in Rx adapter")
>
> Reported-by: Matias Elo <matias.elo@nokia.com>
>
> CC: stable@dpdk.org
>
> Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
> ---
> lib/librte_eventdev/rte_event_eth_rx_adapter.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> v2:
> * add Bugzilla ID
>
> v3:
> * add stable CC
> * replace boolean check with check for > 0
>
> diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> index 8b6e69d..b60674b 100644
> --- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> +++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
> @@ -872,7 +872,7 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
> break;
> }
>
> - if (buf->count >= BATCH_SIZE)
> + if (buf->count > 0)
> rxa_flush_event_buffer(rx_adapter);
>
> return nb_rx;
>
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 13:30 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2019-05-10 13:30 ` Thomas Monjalon
@ 2019-05-10 16:13 ` Mattias Rönnblom
2019-05-10 16:13 ` Mattias Rönnblom
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
1 sibling, 2 replies; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-10 16:13 UTC (permalink / raw)
To: Thomas Monjalon, dev; +Cc: stable, Nikhil Rao, jerinj, matias.elo
On 2019-05-10 15:30, Thomas Monjalon wrote:
> Any review please?
>
Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Mattias Elo reported "Thanks, I’ve tested this patch and can confirm
that it fixes the problem." for the (nearly identical) v2 of this patch.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 16:13 ` Mattias Rönnblom
@ 2019-05-10 16:13 ` Mattias Rönnblom
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
1 sibling, 0 replies; 24+ messages in thread
From: Mattias Rönnblom @ 2019-05-10 16:13 UTC (permalink / raw)
To: Thomas Monjalon, dev; +Cc: stable, Nikhil Rao, jerinj, matias.elo
On 2019-05-10 15:30, Thomas Monjalon wrote:
> Any review please?
>
Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
Mattias Elo reported "Thanks, I’ve tested this patch and can confirm
that it fixes the problem." for the (nearly identical) v2 of this patch.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-10 16:13 ` Mattias Rönnblom
2019-05-10 16:13 ` Mattias Rönnblom
@ 2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 14:00 ` Thomas Monjalon
1 sibling, 2 replies; 24+ messages in thread
From: Elo, Matias (Nokia - FI/Espoo) @ 2019-05-13 4:20 UTC (permalink / raw)
To: Thomas Monjalon
Cc: dev, stable, Nikhil Rao, Jerin Jacob Kollanukkaran,
Mattias Rönnblom
> On 2019-05-10 15:30, Thomas Monjalon wrote:
>> Any review please?
>
> Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
>
> Mattias Elo reported "Thanks, I’ve tested this patch and can confirm that it fixes the problem." for the (nearly identical) v2 of this patch.
I’ve now tested also the v3:
Tested-by: Matias Elo <matias.elo@nokia.com>
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
@ 2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 14:00 ` Thomas Monjalon
1 sibling, 0 replies; 24+ messages in thread
From: Elo, Matias (Nokia - FI/Espoo) @ 2019-05-13 4:20 UTC (permalink / raw)
To: Thomas Monjalon
Cc: dev, stable, Nikhil Rao, Jerin Jacob Kollanukkaran,
Mattias Rönnblom
> On 2019-05-10 15:30, Thomas Monjalon wrote:
>> Any review please?
>
> Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
>
> Mattias Elo reported "Thanks, I’ve tested this patch and can confirm that it fixes the problem." for the (nearly identical) v2 of this patch.
I’ve now tested also the v3:
Tested-by: Matias Elo <matias.elo@nokia.com>
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
@ 2019-05-13 14:00 ` Thomas Monjalon
2019-05-13 14:00 ` Thomas Monjalon
1 sibling, 1 reply; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-13 14:00 UTC (permalink / raw)
To: Nikhil Rao
Cc: dev, Elo, Matias (Nokia - FI/Espoo),
stable, Jerin Jacob Kollanukkaran, Mattias Rönnblom
13/05/2019 06:20, Elo, Matias (Nokia - FI/Espoo):
>
> > On 2019-05-10 15:30, Thomas Monjalon wrote:
> >> Any review please?
> >
> > Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> >
> > Mattias Elo reported "Thanks, I’ve tested this patch and can confirm that it fixes the problem." for the (nearly identical) v2 of this patch.
>
> I’ve now tested also the v3:
>
> Tested-by: Matias Elo <matias.elo@nokia.com>
Applied, thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [dpdk-dev] [dpdk-stable] [PATCH v3] eventdev: fix Rx adapter event flush logic
2019-05-13 14:00 ` Thomas Monjalon
@ 2019-05-13 14:00 ` Thomas Monjalon
0 siblings, 0 replies; 24+ messages in thread
From: Thomas Monjalon @ 2019-05-13 14:00 UTC (permalink / raw)
To: Nikhil Rao
Cc: dev, Elo, Matias (Nokia - FI/Espoo),
stable, Jerin Jacob Kollanukkaran, Mattias Rönnblom
13/05/2019 06:20, Elo, Matias (Nokia - FI/Espoo):
>
> > On 2019-05-10 15:30, Thomas Monjalon wrote:
> >> Any review please?
> >
> > Reviewed-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
> >
> > Mattias Elo reported "Thanks, I’ve tested this patch and can confirm that it fixes the problem." for the (nearly identical) v2 of this patch.
>
> I’ve now tested also the v3:
>
> Tested-by: Matias Elo <matias.elo@nokia.com>
Applied, thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
end of thread, other threads:[~2019-05-13 14:01 UTC | newest]
Thread overview: 24+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-09 4:39 [dpdk-dev] [PATCH] eventdev: fix Rx adapter event flush logic Nikhil Rao
2019-05-09 4:39 ` Nikhil Rao
2019-05-09 4:50 ` [dpdk-dev] [PATCH v2] " Nikhil Rao
2019-05-09 4:50 ` Nikhil Rao
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:17 ` Thomas Monjalon
2019-05-09 19:23 ` Mattias Rönnblom
2019-05-09 19:23 ` Mattias Rönnblom
2019-05-10 8:19 ` Rao, Nikhil
2019-05-10 8:19 ` Rao, Nikhil
2019-05-09 19:22 ` Mattias Rönnblom
2019-05-09 19:22 ` Mattias Rönnblom
2019-05-10 8:20 ` Rao, Nikhil
2019-05-10 8:20 ` Rao, Nikhil
2019-05-10 8:08 ` [dpdk-dev] [PATCH v3] " Nikhil Rao
2019-05-10 8:08 ` Nikhil Rao
2019-05-10 13:30 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2019-05-10 13:30 ` Thomas Monjalon
2019-05-10 16:13 ` Mattias Rönnblom
2019-05-10 16:13 ` Mattias Rönnblom
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 4:20 ` Elo, Matias (Nokia - FI/Espoo)
2019-05-13 14:00 ` Thomas Monjalon
2019-05-13 14:00 ` Thomas Monjalon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).