The prefetch window depending on the hardware platform. The current prefetch
policy may not be applicable to all platforms. In most cases, the number of
packets received by Rx burst is small (64 is used in most performance reports).
In L3fwd, the maximum value cannot exceed 512. Therefore, prefetch
On Mon, 23 Dec 2024 19:02:34 +0800
Junlong Wang wrote:
> V5:
> - Simplify the notify_data part in the zxdh_notify_queue function.
> - Replace rte_zmalloc with rte_calloc in the rss_reta_update function.
> - Remove unnecessary check in mtu_set function.
>
> V4:
> - resolved ci compile iss
On Mon, 23 Dec 2024 19:02:34 +0800
Junlong Wang wrote:
> V5:
> - Simplify the notify_data part in the zxdh_notify_queue function.
> - Replace rte_zmalloc with rte_calloc in the rss_reta_update function.
> - Remove unnecessary check in mtu_set function.
>
> V4:
> - resolved ci compile iss
On Mon, 23 Dec 2024 10:46:59 +
Shani Peretz wrote:
> Sorry if it wasn't clear, in mlx5 there is limited number of counters, so if
> there are X counters and 2X queues, there won't be enough counters to get
> statistics for all queues at the same time.
> In this case we thought the best appr
This patch makes use of the new FD entry cleanup callback
to close the VDUSE reconnect eventfd after its removal
from the FD set.
Signed-off-by: Maxime Coquelin
---
lib/vhost/vduse.c | 12 +---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/lib/vhost/vduse.c b/lib/vhost/vd
This patch leverages the new FD entry cleanup callback
to properly cleanup the vhost-user socket on disconnection
from the frontend.
Fixes: 0e38b42bf61c ("vhost: manage FD with epoll")
Cc: sta...@dpdk.org
Signed-off-by: Maxime Coquelin
---
lib/vhost/socket.c | 44 +--
This patch adds an optional cleanup callback to
FD entries in the FD manager, which is called by the
FD event thread at FD entry removal time (when the read
or write callback has requested the FD removal).
Doing this, we avoid closing the file descriptor before it
is removed from the epoll FD set.
The vhost FD manager provides a way for the read/write
callbacks to request removal of their associated FD from
the epoll FD set. Problem is that it is missing a cleanup
callback, so the read/write callback requesting the removal
have to perform cleanups before the FD is removed from the
FD set. It
Hi,
I think using rte_malloc(rte_calloc) here would be a better choice because:
- As Andre mentioned allocation happens outside of the performance
measurement section
- Due to relatively large size of the memory allocated for
ip_batch/next_hops some CPUs may experience TLB cache pressure, wh
Hello Gowri,
On Tue, Dec 24, 2024 at 8:39 AM Gowrishankar Muthukrishnan
wrote:
>
> Common virtio log include file.
That's really a short commitlog..
What are you trying to achieve?
The net/virtio and crypto/virtio drivers had dedicated logtypes so
far, which seems preferable.
I don't see a case
10 matches
Mail list logo
