Hi Michael / Xuan Zhuo,
On Wednesday, 2024-10-16 at 08:55:21 +01, Darren Kenny wrote:
> Hi Michael,
>
> On Monday, 2024-10-14 at 00:57:41 -04, Michael S. Tsirkin wrote:
>> On Mon, Oct 14, 2024 at 11:12:29AM +0800, Xuan Zhuo wrote:
>>> In the last linux version, we disabled this feature to fix the
On Mon, Oct 14, 2024 at 11:12 AM Xuan Zhuo wrote:
>
> Currently, the virtio core will perform a dma operation for each
> buffer. Although, the same page may be operated multiple times.
>
> In premapped mod, we can perform only one dma operation for the pages of
> the alloc frag. This is beneficial
On Mon, Oct 14, 2024 at 11:12 AM Xuan Zhuo wrote:
>
> Now, if we want to judge the rx work mode, we have to use such codes:
>
> 1. merge mode: vi->mergeable_rx_bufs
> 2. big mode: vi->big_packets && !vi->mergeable_rx_bufs
> 3. small: !vi->big_packets && !vi->mergeable_rx_bufs
>
> This is inc
On Mon, Oct 14, 2024 at 11:12 AM Xuan Zhuo wrote:
>
> Now, the premapped mode can be enabled unconditionally.
>
> So we can remove the failover code for merge and small mode.
>
> Signed-off-by: Xuan Zhuo
Let's be more verbose here. For example, the virtnet_rq_xxx() helper
would be only used if t
On Mon, Oct 14, 2024 at 11:12 AM Xuan Zhuo wrote:
>
> When the frag just got a page, then may lead to regression on VM.
> Specially if the sysctl net.core.high_order_alloc_disable value is 1,
> then the frag always get a page when do refill.
>
> Which could see reliable crashes or scp failure (scp
