DPDK patches and discussions
 help / color / Atom feed
* Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
       [not found] <VI1PR08MB5376173EB8A8A58BC93CD2398F150@VI1PR08MB5376.eurprd08.prod.outlook.com>
@ 2020-02-14  8:04 ` Gavin Hu
  0 siblings, 0 replies; 6+ messages in thread
From: Gavin Hu @ 2020-02-14  8:04 UTC (permalink / raw)
  To: dev, jerinj
  Cc: Honnappa Nagarahalli, david.marchand, Joyce Kong, nd, Phil Yang,
	Ruifeng Wang, Steve Capper, thomas, xiaolong.ye, nd

Hi Jerin,

Could you help review this patch?  We discussed a lot on this topic and I think we agreed already to relax the barrier only on the fast path.

Best Regards,
Gavin


> -----Original Message-----
> From: Gavin Hu <Gavin.Hu@arm.com>
> Sent: Friday, February 14, 2020 3:58 PM
> To: Gavin Hu <Gavin.Hu@arm.com>
> Subject: Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx
> fastpath
> 
> To keep ordering of mixed accesses, rte_cio is sufficient.
> The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
> 
> [1] http://inbox.dpdk.org/dev/CALBAE1M-ezVWCjqCZDBw+MMDEC4O9
> qf0Kpn89EMdGDajepKoZQ@mail.gmail.com
> 
> Fixes: 4861cde46116 ("i40e: new poll mode driver")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Gavin Hu <gavin.hu@arm.com>
> ---
> V4:
> - add the Fixes tag and CC stable <Xiaolong Ye>
> V3:
> - optimize the barriers in the fast path only, leave as it is for the
>   barriers in the slow path and control path <jerin>
> - drop the virtio patches from the list as they are in the control path
> - it makes more sense to relax the barrier in the fast path, at the PMD level.
>   relaxing the fundamental rte_io_x barriers APIs requires scrutinizations for
>   each PMDs which use the barriers directly or indirectly.
> V2:
> - remove virtio_pci_read/write64 APIs definitions, they are not needed and
> generate compiling errors like " error: unused function 'virtio_pci_write64' [-
> Werror,-Wunused-function]"
> - update the reference link to kernel source code
> ---
>  drivers/net/i40e/i40e_rxtx.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
> index fd1ae80da..8c0f7cc67 100644
> --- a/drivers/net/i40e/i40e_rxtx.c
> +++ b/drivers/net/i40e/i40e_rxtx.c
> @@ -1248,7 +1248,8 @@ i40e_xmit_pkts(void *tx_queue, struct rte_mbuf
> **tx_pkts, uint16_t nb_pkts)
>     (unsigned) txq->port_id, (unsigned) txq->queue_id,
>     (unsigned) tx_id, (unsigned) nb_tx);
> 
> -I40E_PCI_REG_WRITE(txq->qtx_tail, tx_id);
> +rte_cio_wmb();
> +I40E_PCI_REG_WRITE_RELAXED(txq->qtx_tail, tx_id);
>  txq->tx_tail = tx_id;
> 
>  return nb_tx;
> --
> 2.17.1
> 


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
  2020-02-16 16:38       ` Ye Xiaolong
@ 2020-02-16 17:36         ` Thomas Monjalon
  0 siblings, 0 replies; 6+ messages in thread
From: Thomas Monjalon @ 2020-02-16 17:36 UTC (permalink / raw)
  To: Ye Xiaolong
  Cc: ferruh.yigit, Gavin Hu, dev, nd, david.marchand, jerinj,
	Honnappa.Nagarahalli, ruifeng.wang, phil.yang, joyce.kong,
	steve.capper, stable

16/02/2020 17:38, Ye Xiaolong:
> Hi, Thomas
> 
> On 02/16, Thomas Monjalon wrote:
> >15/02/2020 16:16, Ye Xiaolong:
> >> s/relaxed/relax
> >> 
> >> On 02/12, Gavin Hu wrote:
> >> >To keep ordering of mixed accesses, rte_cio is sufficient.
> >> >The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
> >[...]
> >> 
> >> Applied to dpdk-next-net-intel with Jerin's Reviewed-by tag, Thanks.
> >
> >I assume it is too much risky doing such optimization post-rc3.
> 
> Yes, this iss a valid concern, I agree to postpone it to next release.
> 
> >
> >Ferruh, Xiaolong, you don't plan anymore pull from dpdk-next-net-intel
> >in 20.02?
> 
> There are still some bug fixing work going on in PRC, so I assume there 
> should be some fix patches after RC3, they are still allowed to be merged
> to 20.02, if the fix is relatively small in terms of lines of code and scope,
> right?

Right



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
  2020-02-16  9:51     ` Thomas Monjalon
@ 2020-02-16 16:38       ` Ye Xiaolong
  2020-02-16 17:36         ` Thomas Monjalon
  0 siblings, 1 reply; 6+ messages in thread
From: Ye Xiaolong @ 2020-02-16 16:38 UTC (permalink / raw)
  To: Thomas Monjalon
  Cc: ferruh.yigit, Gavin Hu, dev, nd, david.marchand, jerinj,
	Honnappa.Nagarahalli, ruifeng.wang, phil.yang, joyce.kong,
	steve.capper, stable

Hi, Thomas

On 02/16, Thomas Monjalon wrote:
>15/02/2020 16:16, Ye Xiaolong:
>> s/relaxed/relax
>> 
>> On 02/12, Gavin Hu wrote:
>> >To keep ordering of mixed accesses, rte_cio is sufficient.
>> >The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
>[...]
>> 
>> Applied to dpdk-next-net-intel with Jerin's Reviewed-by tag, Thanks.
>
>I assume it is too much risky doing such optimization post-rc3.

Yes, this iss a valid concern, I agree to postpone it to next release.

>
>Ferruh, Xiaolong, you don't plan anymore pull from dpdk-next-net-intel
>in 20.02?

There are still some bug fixing work going on in PRC, so I assume there 
should be some fix patches after RC3, they are still allowed to be merged
to 20.02, if the fix is relatively small in terms of lines of code and scope,
right?


Thanks,
Xiaolong

>
>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
  2020-02-15 15:16   ` Ye Xiaolong
@ 2020-02-16  9:51     ` Thomas Monjalon
  2020-02-16 16:38       ` Ye Xiaolong
  0 siblings, 1 reply; 6+ messages in thread
From: Thomas Monjalon @ 2020-02-16  9:51 UTC (permalink / raw)
  To: Ye Xiaolong, ferruh.yigit
  Cc: Gavin Hu, dev, nd, david.marchand, jerinj, Honnappa.Nagarahalli,
	ruifeng.wang, phil.yang, joyce.kong, steve.capper, stable

15/02/2020 16:16, Ye Xiaolong:
> s/relaxed/relax
> 
> On 02/12, Gavin Hu wrote:
> >To keep ordering of mixed accesses, rte_cio is sufficient.
> >The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
[...]
> 
> Applied to dpdk-next-net-intel with Jerin's Reviewed-by tag, Thanks.

I assume it is too much risky doing such optimization post-rc3.

Ferruh, Xiaolong, you don't plan anymore pull from dpdk-next-net-intel
in 20.02?



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
  2020-02-12  5:56 ` [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath Gavin Hu
@ 2020-02-15 15:16   ` Ye Xiaolong
  2020-02-16  9:51     ` Thomas Monjalon
  0 siblings, 1 reply; 6+ messages in thread
From: Ye Xiaolong @ 2020-02-15 15:16 UTC (permalink / raw)
  To: Gavin Hu
  Cc: dev, nd, david.marchand, thomas, jerinj, Honnappa.Nagarahalli,
	ruifeng.wang, phil.yang, joyce.kong, steve.capper, stable

s/relaxed/relax

On 02/12, Gavin Hu wrote:
>To keep ordering of mixed accesses, rte_cio is sufficient.
>The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]
>
>[1] http://inbox.dpdk.org/dev/CALBAE1M-ezVWCjqCZDBw+MMDEC4O9
>qf0Kpn89EMdGDajepKoZQ@mail.gmail.com
>
>Fixes: 4861cde46116 ("i40e: new poll mode driver")
>Cc: stable@dpdk.org
>
>Signed-off-by: Gavin Hu <gavin.hu@arm.com>
>---
>V4:
>- add the Fixes tag and CC stable <Xiaolong Ye>
>V3:
>- optimize the barriers in the fast path only, leave as it is for the
>  barriers in the slow path and control path <jerin>
>- drop the virtio patches from the list as they are in the control path
>- it makes more sense to relax the barrier in the fast path, at the PMD level.
>  relaxing the fundamental rte_io_x barriers APIs requires scrutinizations for
>  each PMDs which use the barriers directly or indirectly.
>V2:
>- remove virtio_pci_read/write64 APIs definitions, they are not needed and generate compiling errors like " error: unused function 'virtio_pci_write64' [-Werror,-Wunused-function]"
>- update the reference link to kernel source code
>---
> drivers/net/i40e/i40e_rxtx.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
>diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
>index fd1ae80da..8c0f7cc67 100644
>--- a/drivers/net/i40e/i40e_rxtx.c
>+++ b/drivers/net/i40e/i40e_rxtx.c
>@@ -1248,7 +1248,8 @@ i40e_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
> 		   (unsigned) txq->port_id, (unsigned) txq->queue_id,
> 		   (unsigned) tx_id, (unsigned) nb_tx);
> 
>-	I40E_PCI_REG_WRITE(txq->qtx_tail, tx_id);
>+	rte_cio_wmb();
>+	I40E_PCI_REG_WRITE_RELAXED(txq->qtx_tail, tx_id);
> 	txq->tx_tail = tx_id;
> 
> 	return nb_tx;
>-- 
>2.17.1
>

Applied to dpdk-next-net-intel with Jerin's Reviewed-by tag, Thanks.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath
  2019-10-22 15:27 [dpdk-dev] [PATCH v1 0/3] relax io barrier for aarch64 and use smp barriers for virtual pci memory Gavin Hu
@ 2020-02-12  5:56 ` Gavin Hu
  2020-02-15 15:16   ` Ye Xiaolong
  0 siblings, 1 reply; 6+ messages in thread
From: Gavin Hu @ 2020-02-12  5:56 UTC (permalink / raw)
  To: dev
  Cc: nd, david.marchand, thomas, jerinj, xiaolong.ye,
	Honnappa.Nagarahalli, ruifeng.wang, phil.yang, joyce.kong,
	steve.capper, stable

To keep ordering of mixed accesses, rte_cio is sufficient.
The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1]

[1] http://inbox.dpdk.org/dev/CALBAE1M-ezVWCjqCZDBw+MMDEC4O9
qf0Kpn89EMdGDajepKoZQ@mail.gmail.com

Fixes: 4861cde46116 ("i40e: new poll mode driver")
Cc: stable@dpdk.org

Signed-off-by: Gavin Hu <gavin.hu@arm.com>
---
V4:
- add the Fixes tag and CC stable <Xiaolong Ye>
V3:
- optimize the barriers in the fast path only, leave as it is for the
  barriers in the slow path and control path <jerin>
- drop the virtio patches from the list as they are in the control path
- it makes more sense to relax the barrier in the fast path, at the PMD level.
  relaxing the fundamental rte_io_x barriers APIs requires scrutinizations for
  each PMDs which use the barriers directly or indirectly.
V2:
- remove virtio_pci_read/write64 APIs definitions, they are not needed and generate compiling errors like " error: unused function 'virtio_pci_write64' [-Werror,-Wunused-function]"
- update the reference link to kernel source code
---
 drivers/net/i40e/i40e_rxtx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
index fd1ae80da..8c0f7cc67 100644
--- a/drivers/net/i40e/i40e_rxtx.c
+++ b/drivers/net/i40e/i40e_rxtx.c
@@ -1248,7 +1248,8 @@ i40e_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
 		   (unsigned) txq->port_id, (unsigned) txq->queue_id,
 		   (unsigned) tx_id, (unsigned) nb_tx);
 
-	I40E_PCI_REG_WRITE(txq->qtx_tail, tx_id);
+	rte_cio_wmb();
+	I40E_PCI_REG_WRITE_RELAXED(txq->qtx_tail, tx_id);
 	txq->tx_tail = tx_id;
 
 	return nb_tx;
-- 
2.17.1


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, back to index

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <VI1PR08MB5376173EB8A8A58BC93CD2398F150@VI1PR08MB5376.eurprd08.prod.outlook.com>
2020-02-14  8:04 ` [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath Gavin Hu
2019-10-22 15:27 [dpdk-dev] [PATCH v1 0/3] relax io barrier for aarch64 and use smp barriers for virtual pci memory Gavin Hu
2020-02-12  5:56 ` [dpdk-dev] [PATCH v4] net/i40e: relaxed barrier in the tx fastpath Gavin Hu
2020-02-15 15:16   ` Ye Xiaolong
2020-02-16  9:51     ` Thomas Monjalon
2020-02-16 16:38       ` Ye Xiaolong
2020-02-16 17:36         ` Thomas Monjalon

DPDK patches and discussions

Archives are clonable:
	git clone --mirror http://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ http://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev


Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/ public-inbox