search for: tsq

...l S. Tsirkin wrote: >>>>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>>>> similar to wifi. It's worth considering whether something similar to >>>>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>>>> benefitial. >>>>>> I've played a similar patch several days before. The tricky part is the mode >>>>>> switching between napi and no napi. We should make sure when the packet is >>>>>&gt...

...l S. Tsirkin wrote: >>>>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>>>> similar to wifi. It's worth considering whether something similar to >>>>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>>>> benefitial. >>>>>> I've played a similar patch several days before. The tricky part is the mode >>>>>> switching between napi and no napi. We should make sure when the packet is >>>>>&gt...

...the solution converges, sometimes not.   Since I¢m planning to do this 5000 times, is there a way where I can make sure the solution always converge?   size <- 25 n <- rbinom(1, size, 0.6) data <- log(rlnorm(n, -0.075, sqrt(.15))) xbar <- mean(data) sigma2 <- var(data) s <- n*xbar tsq <- sum(data^2) t_init <- xbar-2*sqrt(sigma2/n)   l2 <- function(x){         x1 <- x[1]         x2 <- x[2]         (size - n)*log(1 - exp(t_init - x1 - x2/2)) + n*(t_init -x1 - x2/2) - (n/2)*log(2*pi) - (n/2)*log(x2) - tsq/(2*x2) + x1*s/x2 - n*(x1^2)/(2*x2)   }     gr.l2 <- funct...

...>>>>>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>>>>>> similar to wifi. It's worth considering whether something similar to >>>>>>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>>>>>> benefitial. >>>>>>>> I've played a similar patch several days before. The tricky part is the mode >>>>>>>> switching between napi and no napi. We should make sure when the packet is...

...>>>>>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>>>>>> similar to wifi. It's worth considering whether something similar to >>>>>>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>>>>>> benefitial. >>>>>>>> I've played a similar patch several days before. The tricky part is the mode >>>>>>>> switching between napi and no napi. We should make sure when the packet is...

...2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>> similar to wifi. It's worth considering whether something similar to >>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>> benefitial. >>>> I've played a similar patch several days before. The tricky part is the mode >>>> switching between napi and no napi. We should make sure when the packet is >>>> sent and trakced by BQL,? it s...

...2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>> similar to wifi. It's worth considering whether something similar to >>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>> benefitial. >>>> I've played a similar patch several days before. The tricky part is the mode >>>> switching between napi and no napi. We should make sure when the packet is >>>> sent and trakced by BQL,? it s...

...Wang wrote: >> On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>> similar to wifi. It's worth considering whether something similar to >>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>> benefitial. >> >> I've played a similar patch several days before. The tricky part is the mode >> switching between napi and no napi. We should make sure when the packet is >> sent and trakced by BQL,? it should be consumed by BQ...

...Wang wrote: >> On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>> similar to wifi. It's worth considering whether something similar to >>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>> benefitial. >> >> I've played a similar patch several days before. The tricky part is the mode >> switching between napi and no napi. We should make sure when the packet is >> sent and trakced by BQL,? it should be consumed by BQ...

On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: > When use_napi is set, let's enable BQLs. Note: some of the issues are > similar to wifi. It's worth considering whether something similar to > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > benefitial. I've played a similar patch several days before. The tricky part is the mode switching between napi and no napi. We should make sure when the packet is sent and trakced by BQL,? it should be consumed by BQL as well. I did it by tracking it thro...

On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: > When use_napi is set, let's enable BQLs. Note: some of the issues are > similar to wifi. It's worth considering whether something similar to > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > benefitial. I've played a similar patch several days before. The tricky part is the mode switching between napi and no napi. We should make sure when the packet is sent and trakced by BQL,? it should be consumed by BQL as well. I did it by tracking it thro...

Now that we have brought the virtio overhead way down with a fast packed ring implementation, we seem to be actually observing TCP drops indicative of bufferbloat. So let's try to enable TSQ. Note: it isn't clear that the default pacing is great for the virt usecase. It's worth trying to play with sk_pacing_shift_update to see what happens. For this reason, and for a more important one that I didn't have time to test it yet, sending as an RFC. Michael S. Tsirkin (2): v...

...; > > > > > When use_napi is set, let's enable BQLs. Note: some of the issues are > > > > > > > > similar to wifi. It's worth considering whether something similar to > > > > > > > > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > > > > > > > > benefitial. > > > > > > > I've played a similar patch several days before. The tricky part is the mode > > > > > > > switching between napi and no napi. We should make sure when the pa...

...Jason Wang wrote: > > On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: > > When use_napi is set, let's enable BQLs. Note: some of the issues are > > similar to wifi. It's worth considering whether something similar to > > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > > benefitial. > > > I've played a similar patch several days before. The tricky part is the mode > switching between napi and no napi. We should make sure when the packet is > sent and trakced by BQL,? it should be consumed by BQL as well. I...

...; > When use_napi is set, let's enable BQLs. Note: some of the issues are > > > > > > > > > > similar to wifi. It's worth considering whether something similar to > > > > > > > > > > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > > > > > > > > > > benefitial. > > > > > > > > > I've played a similar patch several days before. The tricky part is the mode > > > > > > > > > switching between napi and no napi. W...

...o obvious regression on throughput were > >>> noticed. More cpu utilization were noticed in few cases. > >>> - For guest transmission. Very huge improvement on througput for > >>>small > >>> packet transmission were noticed. This is expected since TSQ and > >>>other > >>> optimization for small packet transmission work after tx interrupt. > >>>But > >>> will use more cpu for large packets. > >>> - For TCP_RR, regression (10% on transaction rate and cpu > >>>utilization)...

...o obvious regression on throughput were > >>> noticed. More cpu utilization were noticed in few cases. > >>> - For guest transmission. Very huge improvement on througput for > >>>small > >>> packet transmission were noticed. This is expected since TSQ and > >>>other > >>> optimization for small packet transmission work after tx interrupt. > >>>But > >>> will use more cpu for large packets. > >>> - For TCP_RR, regression (10% on transaction rate and cpu > >>>utilization)...

...2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>>>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>>>> similar to wifi. It's worth considering whether something similar to >>>>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>>>> benefitial. >>>> I've played a similar patch several days before. The tricky part is the mode >>>> switching between napi and no napi. We should make sure when the packet is >>>> sent and trakced by BQL,? it s...

...On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: > > > > When use_napi is set, let's enable BQLs. Note: some of the issues are > > > > similar to wifi. It's worth considering whether something similar to > > > > commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be > > > > benefitial. > > > > > > I've played a similar patch several days before. The tricky part is the mode > > > switching between napi and no napi. We should make sure when the packet is > > > sent and trakced by B...

...Wang wrote: >> On 2018/12/6 ??6:54, Michael S. Tsirkin wrote: >>> When use_napi is set, let's enable BQLs. Note: some of the issues are >>> similar to wifi. It's worth considering whether something similar to >>> commit 36148c2bbfbe ("mac80211: Adjust TSQ pacing shift") might be >>> benefitial. >> >> I've played a similar patch several days before. The tricky part is the mode >> switching between napi and no napi. We should make sure when the packet is >> sent and trakced by BQL,? it should be consumed by BQ...