* [PATCH net] net: mvneta: set real interrupt per packet for tx_done
@ 2016-07-06 2:18 ` Marcin Wojtas
0 siblings, 0 replies; 7+ messages in thread
From: Marcin Wojtas @ 2016-07-06 2:18 UTC (permalink / raw)
To: linux-arm-kernel
From: Dmitri Epshtein <dima@marvell.com>
Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to
set coalescing threshold to a value guaranteeing interrupt generation
per each sent packet, so that buffers can be released with no delay.
In fact setting threshold to '1' was wrong, because it causes interrupt
every two packets. According to the documentation a reason behind it is
following - interrupt occurs once sent buffers counter reaches a value,
which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This
behavior was confirmed during tests. Also when testing the SoC working
as a NAS device, better performance was observed with int-per-packet,
as it strongly depends on the fact that all transmitted packets are
released immediately.
This commit enables NETA controller work in interrupt per sent packet mode
by setting coalescing threshold to 0.
Signed-off-by: Dmitri Epshtein <dima@marvell.com>
Signed-off-by: Marcin Wojtas <mw@semihalf.com>
Cc: <stable@vger.kernel.org> # v3.10+
Fixes aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay")
---
drivers/net/ethernet/marvell/mvneta.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index cf04c97..0ad2fa3 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -244,7 +244,7 @@
/* Various constants */
/* Coalescing */
-#define MVNETA_TXDONE_COAL_PKTS 1
+#define MVNETA_TXDONE_COAL_PKTS 0 /* interrupt per packet */
#define MVNETA_RX_COAL_PKTS 32
#define MVNETA_RX_COAL_USEC 100
--
1.8.3.1
^ permalink raw reply related [flat|nested] 7+ messages in thread* [PATCH net] net: mvneta: set real interrupt per packet for tx_done @ 2016-07-06 2:18 ` Marcin Wojtas 0 siblings, 0 replies; 7+ messages in thread From: Marcin Wojtas @ 2016-07-06 2:18 UTC (permalink / raw) To: linux-kernel, linux-arm-kernel, netdev Cc: davem, linux, sebastian.hesselbarth, andrew, jason, thomas.petazzoni, gregory.clement, nadavh, alior, nitroshift, mw, jaz, Dmitri Epshtein, stable From: Dmitri Epshtein <dima@marvell.com> Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to set coalescing threshold to a value guaranteeing interrupt generation per each sent packet, so that buffers can be released with no delay. In fact setting threshold to '1' was wrong, because it causes interrupt every two packets. According to the documentation a reason behind it is following - interrupt occurs once sent buffers counter reaches a value, which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This behavior was confirmed during tests. Also when testing the SoC working as a NAS device, better performance was observed with int-per-packet, as it strongly depends on the fact that all transmitted packets are released immediately. This commit enables NETA controller work in interrupt per sent packet mode by setting coalescing threshold to 0. Signed-off-by: Dmitri Epshtein <dima@marvell.com> Signed-off-by: Marcin Wojtas <mw@semihalf.com> Cc: <stable@vger.kernel.org> # v3.10+ Fixes aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") --- drivers/net/ethernet/marvell/mvneta.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index cf04c97..0ad2fa3 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -244,7 +244,7 @@ /* Various constants */ /* Coalescing */ -#define MVNETA_TXDONE_COAL_PKTS 1 +#define MVNETA_TXDONE_COAL_PKTS 0 /* interrupt per packet */ #define MVNETA_RX_COAL_PKTS 32 #define MVNETA_RX_COAL_USEC 100 -- 1.8.3.1 ^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH net] net: mvneta: set real interrupt per packet for tx_done 2016-07-06 2:18 ` Marcin Wojtas (?) @ 2016-07-06 5:04 ` Willy Tarreau -1 siblings, 0 replies; 7+ messages in thread From: Willy Tarreau @ 2016-07-06 5:04 UTC (permalink / raw) To: linux-arm-kernel Hi, On Wed, Jul 06, 2016 at 04:18:58AM +0200, Marcin Wojtas wrote: > From: Dmitri Epshtein <dima@marvell.com> > > Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to > set coalescing threshold to a value guaranteeing interrupt generation > per each sent packet, so that buffers can be released with no delay. > > In fact setting threshold to '1' was wrong, because it causes interrupt > every two packets. According to the documentation a reason behind it is > following - interrupt occurs once sent buffers counter reaches a value, > which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This > behavior was confirmed during tests. Also when testing the SoC working > as a NAS device, better performance was observed with int-per-packet, > as it strongly depends on the fact that all transmitted packets are > released immediately. > > This commit enables NETA controller work in interrupt per sent packet mode > by setting coalescing threshold to 0. We had a discussion about this in January 2015 and I thought I sent a patch to change it but I can't find it, so I think it was only proposed to some users for testing. I also remember that on more recent kernels by then (>=3.13) we observed a slightly better performance with this value set to zero. Acked-by: Willy Tarreau <w@1wt.eu> Willy ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH net] net: mvneta: set real interrupt per packet for tx_done @ 2016-07-06 5:04 ` Willy Tarreau 0 siblings, 0 replies; 7+ messages in thread From: Willy Tarreau @ 2016-07-06 5:04 UTC (permalink / raw) To: Marcin Wojtas Cc: thomas.petazzoni, andrew, linux, jason, netdev, linux-kernel, stable, Dmitri Epshtein, nadavh, alior, jaz, gregory.clement, nitroshift, davem, linux-arm-kernel, sebastian.hesselbarth Hi, On Wed, Jul 06, 2016 at 04:18:58AM +0200, Marcin Wojtas wrote: > From: Dmitri Epshtein <dima@marvell.com> > > Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to > set coalescing threshold to a value guaranteeing interrupt generation > per each sent packet, so that buffers can be released with no delay. > > In fact setting threshold to '1' was wrong, because it causes interrupt > every two packets. According to the documentation a reason behind it is > following - interrupt occurs once sent buffers counter reaches a value, > which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This > behavior was confirmed during tests. Also when testing the SoC working > as a NAS device, better performance was observed with int-per-packet, > as it strongly depends on the fact that all transmitted packets are > released immediately. > > This commit enables NETA controller work in interrupt per sent packet mode > by setting coalescing threshold to 0. We had a discussion about this in January 2015 and I thought I sent a patch to change it but I can't find it, so I think it was only proposed to some users for testing. I also remember that on more recent kernels by then (>=3.13) we observed a slightly better performance with this value set to zero. Acked-by: Willy Tarreau <w@1wt.eu> Willy ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH net] net: mvneta: set real interrupt per packet for tx_done @ 2016-07-06 5:04 ` Willy Tarreau 0 siblings, 0 replies; 7+ messages in thread From: Willy Tarreau @ 2016-07-06 5:04 UTC (permalink / raw) To: Marcin Wojtas Cc: linux-kernel, linux-arm-kernel, netdev, davem, linux, sebastian.hesselbarth, andrew, jason, thomas.petazzoni, gregory.clement, nadavh, alior, nitroshift, jaz, Dmitri Epshtein, stable Hi, On Wed, Jul 06, 2016 at 04:18:58AM +0200, Marcin Wojtas wrote: > From: Dmitri Epshtein <dima@marvell.com> > > Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to > set coalescing threshold to a value guaranteeing interrupt generation > per each sent packet, so that buffers can be released with no delay. > > In fact setting threshold to '1' was wrong, because it causes interrupt > every two packets. According to the documentation a reason behind it is > following - interrupt occurs once sent buffers counter reaches a value, > which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This > behavior was confirmed during tests. Also when testing the SoC working > as a NAS device, better performance was observed with int-per-packet, > as it strongly depends on the fact that all transmitted packets are > released immediately. > > This commit enables NETA controller work in interrupt per sent packet mode > by setting coalescing threshold to 0. We had a discussion about this in January 2015 and I thought I sent a patch to change it but I can't find it, so I think it was only proposed to some users for testing. I also remember that on more recent kernels by then (>=3.13) we observed a slightly better performance with this value set to zero. Acked-by: Willy Tarreau <w@1wt.eu> Willy ^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH net] net: mvneta: set real interrupt per packet for tx_done 2016-07-06 2:18 ` Marcin Wojtas @ 2016-07-09 3:45 ` David Miller -1 siblings, 0 replies; 7+ messages in thread From: David Miller @ 2016-07-09 3:45 UTC (permalink / raw) To: linux-arm-kernel From: Marcin Wojtas <mw@semihalf.com> Date: Wed, 6 Jul 2016 04:18:58 +0200 > From: Dmitri Epshtein <dima@marvell.com> > > Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to > set coalescing threshold to a value guaranteeing interrupt generation > per each sent packet, so that buffers can be released with no delay. > > In fact setting threshold to '1' was wrong, because it causes interrupt > every two packets. According to the documentation a reason behind it is > following - interrupt occurs once sent buffers counter reaches a value, > which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This > behavior was confirmed during tests. Also when testing the SoC working > as a NAS device, better performance was observed with int-per-packet, > as it strongly depends on the fact that all transmitted packets are > released immediately. > > This commit enables NETA controller work in interrupt per sent packet mode > by setting coalescing threshold to 0. > > Signed-off-by: Dmitri Epshtein <dima@marvell.com> > Signed-off-by: Marcin Wojtas <mw@semihalf.com> > Cc: <stable@vger.kernel.org> # v3.10+ > Fixes aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") Applied, thanks. ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH net] net: mvneta: set real interrupt per packet for tx_done @ 2016-07-09 3:45 ` David Miller 0 siblings, 0 replies; 7+ messages in thread From: David Miller @ 2016-07-09 3:45 UTC (permalink / raw) To: mw Cc: linux-kernel, linux-arm-kernel, netdev, linux, sebastian.hesselbarth, andrew, jason, thomas.petazzoni, gregory.clement, nadavh, alior, nitroshift, jaz, dima, stable From: Marcin Wojtas <mw@semihalf.com> Date: Wed, 6 Jul 2016 04:18:58 +0200 > From: Dmitri Epshtein <dima@marvell.com> > > Commit aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") intended to > set coalescing threshold to a value guaranteeing interrupt generation > per each sent packet, so that buffers can be released with no delay. > > In fact setting threshold to '1' was wrong, because it causes interrupt > every two packets. According to the documentation a reason behind it is > following - interrupt occurs once sent buffers counter reaches a value, > which is higher than one specified in MVNETA_TXQ_SIZE_REG(q). This > behavior was confirmed during tests. Also when testing the SoC working > as a NAS device, better performance was observed with int-per-packet, > as it strongly depends on the fact that all transmitted packets are > released immediately. > > This commit enables NETA controller work in interrupt per sent packet mode > by setting coalescing threshold to 0. > > Signed-off-by: Dmitri Epshtein <dima@marvell.com> > Signed-off-by: Marcin Wojtas <mw@semihalf.com> > Cc: <stable@vger.kernel.org> # v3.10+ > Fixes aebea2ba0f74 ("net: mvneta: fix Tx interrupt delay") Applied, thanks. ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2016-07-09 3:45 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2016-07-06 2:18 [PATCH net] net: mvneta: set real interrupt per packet for tx_done Marcin Wojtas 2016-07-06 2:18 ` Marcin Wojtas 2016-07-06 5:04 ` Willy Tarreau 2016-07-06 5:04 ` Willy Tarreau 2016-07-06 5:04 ` Willy Tarreau 2016-07-09 3:45 ` David Miller 2016-07-09 3:45 ` David Miller
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.