Skip to contentSkip to conversation

net: mana: Batch ringing RX queue doorbell on receiving packets

PATCH Superseded [email protected] 6 participants 15 messages Jun 26, 2023 – Jul 6, 2023

Subsystems (6)

v3 of 5 Superseded Latest is v5 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 net: mana: Batch ringing RX queue doorbell on receiving packets

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)")

---
Change log:
v2:
Check for comp_read > 0 as it might be negative on completion error. Set rq.wqe_cnt to 0 according to BNIC spec.

v3:
Add details in the commit on the reason of performance increase and test numbers. Add details in the commit on why rq.wqe_cnt should be set to 0 according to hardware spec. Add "Reviewed-by" from Haiyang and Dexuan.

 drivers/net/ethernet/microsoft/mana/gdma_main.c |  5 ++++-
 drivers/net/ethernet/microsoft/mana/mana_en.c   | 10 ++++++++--
 2 files changed, 12 insertions(+), 3 deletions(-)
drivers/net/ethernet/microsoft/mana/gdma_main.c +4 -1
index 8f3f78b68592..3765d3389a9a 100644
@@ -300,8 +300,11 @@ static void mana_gd_ring_doorbell(struct gdma_context *gc, u32 db_index,
 
 void mana_gd_wq_ring_doorbell(struct gdma_context *gc, struct gdma_queue *queue)
 {
+	/* Hardware Spec specifies that software client should set 0 for
+	 * wqe_cnt for Receive Queues. This value is not used in Send Queues.
+	 */
 	mana_gd_ring_doorbell(gc, queue->gdma_dev->doorbell, queue->type,
-			      queue->id, queue->head * GDMA_WQE_BU_SIZE, 1);
+			      queue->id, queue->head * GDMA_WQE_BU_SIZE, 0);
 }
 
 void mana_gd_ring_cq(struct gdma_queue *cq, u8 arm_bit)
drivers/net/ethernet/microsoft/mana/mana_en.c +8 -2
index cd4d5ceb9f2d..1d8abe63fcb8 100644
@@ -1383,8 +1383,8 @@ static void mana_post_pkt_rxq(struct mana_rxq *rxq)
 
 	recv_buf_oob = &rxq->rx_oobs[curr_index];
 
-	err = mana_gd_post_and_ring(rxq->gdma_rq, &recv_buf_oob->wqe_req,
-				    &recv_buf_oob->wqe_inf);
+	err = mana_gd_post_work_request(rxq->gdma_rq, &recv_buf_oob->wqe_req,
+					&recv_buf_oob->wqe_inf);
 	if (WARN_ON_ONCE(err))
 		return;
 
@@ -1654,6 +1654,12 @@ static void mana_poll_rx_cq(struct mana_cq *cq)
 		mana_process_rx_cqe(rxq, cq, &comp[i]);
 	}
 
+	if (comp_read > 0) {
+		struct gdma_context *gc = rxq->gdma_rq->gdma_dev->gdma_context;
+
+		mana_gd_wq_ring_doorbell(gc, rxq->gdma_rq);
+	}
+
 	if (rxq->xdp_flush)
 		xdp_do_flush();
 }
Signature · 1 line
2.34.1
maintainer
#1 lore raw
Quoted 16 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

Quoted 8 lines

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)")

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

In any case you must avoid empty lines in the tag area.

If you really intend targeting the -net tree, please repost fixing the above and explicitly specifying the target tree in the subj prefix.

thanks!

Paolo

maintainer
#2 lore raw
Quoted 5 lines · On Thu, 29 Jun 2023 10:42:34 +0200 Paolo Abeni wrote:
Quoted 5 lines

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

And please split this cleanup into a separate patch, it doesn't sound like it has to be done as part of the optimization.

maintainer
#4 lore raw
In reply to Paolo Abeni
Quoted 40 lines

-----Original Message-----
From: Paolo Abeni <[email protected]>
Sent: Thursday, June 29, 2023 4:43 AM
To: [email protected]; Jason Gunthorpe <[email protected]>; Leon Romanovsky <[email protected]>; Ajay Sharma <[email protected]>; Dexuan Cui <[email protected]>; KY Srinivasan <[email protected]>; Haiyang Zhang <[email protected]>; Wei Liu <[email protected]>; David S. Miller <[email protected]>; Eric Dumazet <[email protected]>; Jakub Kicinski <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; Long Li <[email protected]>; [email protected]
Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 22 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

Quoted 6 lines
Quoted 2 lines
Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure Network

Adapter (MANA)")

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

This web page shows the net-next is "open":
http://vger.kernel.org/~davem/net-next.html

Is this still the right place to check net-next status?

- Haiyang

maintainer
#5 lore raw
Quoted 4 lines · On Thu, 29 Jun 2023 16:53:42 +0000 Haiyang Zhang wrote:

This web page shows the net-next is "open":
http://vger.kernel.org/~davem/net-next.html

Is this still the right place to check net-next status?

We're working on fixing it. Unfortunately it's a private page and most of the netdev maintainers don't have access to changing it :(

#6 lore raw
In reply to Paolo Abeni
Quoted 26 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 22 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

Quoted 6 lines
Quoted 3 lines
Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure

Network Adapter (MANA)")

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

This issue is a blocker for usage on 200G physical link. I think it can be categorized as a fix.

Quoted 5 lines

In any case you must avoid empty lines in the tag area.

If you really intend targeting the -net tree, please repost fixing the above and explicitly specifying the target tree in the subj prefix.

Will do, thank you.

Long

Quoted 4 lines

thanks!

Paolo

#3 lore raw
In reply to Jakub Kicinski
Quoted 4 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

On Thu, 29 Jun 2023 10:42:34 +0200 Paolo Abeni wrote:

Quoted 8 lines
Quoted 5 lines
Quoted 5 lines

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although currently the hardware doesn't enforce the check, in the future releases it may do.

And please split this cleanup into a separate patch, it doesn't sound like it has to be done as part of the optimization.

Will do, thanks.

Long

maintainer
#7 lore raw
In reply to Long Li
Quoted 39 lines · Long Li
Quoted 39 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving
packets

Quoted 33 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted
to the received queue. Excessive MMIO writes result in CPU spending
more time waiting on LOCK instructions (atomic operations), resulting
in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted
all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on
a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single
connection.

While we are making changes in this code path, change the code for
ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although
currently the hardware doesn't enforce the check, in the future releases it may do.

Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure

Network Adapter (MANA)")

Quoted 6 lines
Quoted 3 lines

Uhmmm... this looks like a performance improvement to me, more suitable for
the net-next tree ?!? (Note that net-next is closed now).

This issue is a blocker for usage on 200G physical link. I think it can be categorized as a fix.

Let me ask the question the other way around: is there any specific reason to have this fix into 6.5 and all the way back to 5.13? Especially the latest bit (CC-ing stable) looks at least debatable.

Thanks,

Paolo

#8 lore raw
Quoted 40 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 36 lines · Long Li
Quoted 35 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 24 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although
currently the hardware doesn't enforce the check, in the future releases it may do.

Quoted 3 lines
Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure

Network Adapter (MANA)")

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

Quoted 6 lines
Quoted 2 lines

This issue is a blocker for usage on 200G physical link. I think it can be categorized as a fix.

Let me ask the question the other way around: is there any specific reason to have this fix into 6.5 and all the way back to 5.13?
Especially the latest bit (CC-ing stable) looks at least debatable.

There are many deployed Linux distributions with MANA driver on kernel 5.15 and kernel 6.1. (those kernels are longterm) They need this fix to achieve the performance target.

Thanks,

Long

maintainer
#9 lore raw
Quoted 40 lines · Long Li
Quoted 40 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 36 lines · Long Li
Quoted 35 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 24 lines · On Mon, 2023-06-26 at 16:57 -0700, [email protected] wrote:

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although
currently the hardware doesn't enforce the check, in the future releases it may do.

Quoted 3 lines
Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure

Network Adapter (MANA)")

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

Quoted 8 lines
Quoted 6 lines
Quoted 2 lines

This issue is a blocker for usage on 200G physical link. I think it can be categorized as a fix.

Let me ask the question the other way around: is there any specific reason to have this fix into 6.5 and all the way back to 5.13?
Especially the latest bit (CC-ing stable) looks at least debatable.

There are many deployed Linux distributions with MANA driver on kernel 5.15 and kernel 6.1. (those kernels are longterm) They need this fix to achieve the performance target.

Why can't they be upgraded to get that performance target, and all the other goodness that those kernels have? We don't normally backport new features, right?

thanks,

greg k-h

#10 lore raw
Quoted 52 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 8 lines · Long Li
Quoted 8 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 4 lines · Long Li
Quoted 4 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

On Mon, 2023-06-26 at 16:57 -0700, [email protected]

Quoted 39 lines · wrote:
Quoted 38 lines
Quoted 28 lines
Quoted 28 lines
Quoted 27 lines

From: Long Li <[email protected]>

It's inefficient to ring the doorbell page every time a WQE is posted to the received queue. Excessive MMIO writes result in CPU spending more time waiting on LOCK instructions (atomic operations), resulting in poor scaling performance.

Move the code for ringing doorbell page to where after we have posted all WQEs to the receive queue during a callback from napi_poll().

With this change, tests showed an improvement from 120G/s to 160G/s on a 200G physical link, with 16 or 32 hardware queues.

Tests showed no regression in network latency benchmarks on single connection.

While we are making changes in this code path, change the code for ringing doorbell to set the WQE_COUNT to 0 for Receive Queue. The hardware specification specifies that it should set to 0. Although
currently the hardware doesn't enforce the check, in the future releases it may do.

Fixes ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft

Azure Network Adapter (MANA)")

Quoted 6 lines
Quoted 3 lines

Uhmmm... this looks like a performance improvement to me, more suitable for the net-next tree ?!? (Note that net-next is closed now).

This issue is a blocker for usage on 200G physical link. I think it can be categorized as a fix.

Let me ask the question the other way around: is there any specific reason to have this fix into 6.5 and all the way back to 5.13? Especially the latest bit (CC-ing stable) looks at least debatable.

Quoted 7 lines
Quoted 1 line

There are many deployed Linux distributions with MANA driver on kernel

5.15 and kernel 6.1. (those kernels are longterm) They need this fix to achieve the performance target.

Why can't they be upgraded to get that performance target, and all the other goodness that those kernels have? We don't normally backport new features, right?

I think this should be considered as a fix, not a new feature.

MANA is designed to be 200GB full duplex at the start. Due to lack of hardware testing capability at early stage of the project, we could only test 100GB for the Linux driver. When hardware is fully capable of reaching designed spec, this bug in the Linux driver shows up.

Thanks,

Long

Quoted 4 lines

thanks,

greg k-h

maintainer
#11 lore raw
Quoted 7 lines · On Fri, 30 Jun 2023 20:42:28 +0000 Long Li wrote:
Quoted 6 lines
Quoted 2 lines

5.15 and kernel 6.1. (those kernels are longterm) They need this fix to achieve the performance target.

Why can't they be upgraded to get that performance target, and all the other goodness that those kernels have? We don't normally backport new features, right?

Quoted 6 lines

I think this should be considered as a fix, not a new feature.

MANA is designed to be 200GB full duplex at the start. Due to lack of hardware testing capability at early stage of the project, we could only test 100GB for the Linux driver. When hardware is fully capable of reaching designed spec, this bug in the Linux driver shows up.

That part we understand.

If I were you I'd try to convince Greg and Paolo that the change is small and significant for user experience. And answer Greg's question why upgrading the kernel past 6.1 is a challenge in your environment.

#12 lore raw
Quoted 18 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 7 lines · On Fri, 30 Jun 2023 20:42:28 +0000 Long Li wrote:
Quoted 6 lines
Quoted 2 lines

5.15 and kernel 6.1. (those kernels are longterm) They need this fix to achieve the performance target.

Why can't they be upgraded to get that performance target, and all the other goodness that those kernels have? We don't normally backport new features, right?

Quoted 6 lines

I think this should be considered as a fix, not a new feature.

MANA is designed to be 200GB full duplex at the start. Due to lack of hardware testing capability at early stage of the project, we could only test 100GB for the Linux driver. When hardware is fully capable of reaching designed spec, this bug in the Linux driver shows up.

Quoted 5 lines

That part we understand.

If I were you I'd try to convince Greg and Paolo that the change is small and significant for user experience. And answer Greg's question why upgrading the kernel past 6.1 is a challenge in your environment.

I was under the impression that this patch was considered to be a feature, not a bug fix. I was trying to justify that the "Fixes:" tag was needed.

I apologize for misunderstanding this.

Without this fix, it's not possible to run a typical workload designed for 200Gb physical link speed.

We see a large number of customers and Linux distributions committed on 5.15 and 6.1 kernels. They planned the product cycles and certification processes around these longterm kernel versions. It's difficult for them to upgrade to newer kernel versions.

Thanks,

Long

maintainer
#13 lore raw
Quoted 67 lines · Long Li
Quoted 67 lines
Quoted 67 lines
Quoted 67 lines
Quoted 45 lines
Quoted 45 lines
Quoted 45 lines
Quoted 45 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue
doorbell
on receiving
packets

Quoted 20 lines · On Fri, 30 Jun 2023 20:42:28 +0000 Long Li wrote:
Quoted 20 lines
Quoted 20 lines
Quoted 20 lines
Quoted 15 lines

5.15 and kernel 6.1. (those
kernels are longterm)
They need
this
fix to achieve the performance
target.

Why can't they be upgraded to get that
performance
target, and
all
the other goodness that those kernels
have? We don't
normally
backport new features, right?

I think this should be considered as a fix, not
a new
feature.

Quoted 15 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

MANA is designed to be 200GB full duplex at the
start. Due
to
lack of
hardware testing capability at early stage of
the project,
we

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

could
only test 100GB for the Linux driver. When
hardware is
fully
capable
of reaching designed spec, this bug in the
Linux driver
shows up.

That part we understand.

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

If I were you I'd try to convince Greg and Paolo that the
change is
small and
significant for user experience. And answer Greg's question why
upgrading the
kernel past 6.1 is a challenge in your environment.

I was under the impression that this patch was considered to be a
feature,
not a bug fix. I was trying to justify that the "Fixes:" tag was
needed.

I apologize for misunderstanding this.

Without this fix, it's not possible to run a typical workload designed for 200Gb
physical link speed.

Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines

We see a large number of customers and Linux distributions committed
on 5.15
and 6.1 kernels. They planned the product cycles and certification
processes
around these longterm kernel versions. It's difficult for them to
upgrade to newer
kernel versions.

I think there are some misunderstanding WRT distros and stable kernels. (Commercial) distros will backport the patch as needed, regardless such patch landing in the 5.15 upstream tree or not. Individual users running their own vanilla 5.15 kernel can't expect performance improvement landing there.

All in all I feel undecided. I would endorse this change going trough net-next (without the stable tag). I would feel less torn with this change targeting -net without the stable tag. Targeting -net with the stable tag sounds a bit too much to me.

Cheers,
Paolo

#14 lore raw
Quoted 56 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 38 lines · Long Li
Quoted 38 lines
Quoted 38 lines
Quoted 38 lines
Quoted 25 lines
Quoted 25 lines
Quoted 25 lines
Quoted 25 lines

Subject: Re: [Patch v3] net: mana: Batch ringing RX queue doorbell on receiving packets

Quoted 20 lines · On Fri, 30 Jun 2023 20:42:28 +0000 Long Li wrote:
Quoted 20 lines
Quoted 20 lines
Quoted 20 lines

5.15 and kernel 6.1. (those
kernels are longterm) They need
this fix to achieve the
performance target.

Why can't they be upgraded to get that
performance target, and all the other
goodness that those kernels have? We
don't normally backport new features,
right?

I think this should be considered as a fix, not
a new feature.

MANA is designed to be 200GB full duplex at the
start. Due to lack of hardware testing
capability at early stage of the project, we
could only test 100GB for the Linux driver. When
hardware is fully capable of reaching designed
spec, this bug in the Linux driver shows up.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

That part we understand.

If I were you I'd try to convince Greg and Paolo that the change is small and significant for user experience. And answer Greg's question why upgrading the kernel past 6.1 is a challenge in your environment.

I was under the impression that this patch was considered to be a feature, not a bug fix. I was trying to justify that the "Fixes:" tag was needed.

I apologize for misunderstanding this.

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

Without this fix, it's not possible to run a typical workload designed for 200Gb physical link speed.

We see a large number of customers and Linux distributions committed on 5.15 and 6.1 kernels. They planned the product cycles and certification processes around these longterm kernel versions. It's difficult for them to upgrade to newer kernel versions.

I think there are some misunderstanding WRT distros and stable kernels. (Commercial) distros will backport the patch as needed, regardless such patch landing in the 5.15 upstream tree or not. Individual users running their own vanilla 5.15 kernel can't expect performance improvement landing there.

Quoted 7 lines

All in all I feel undecided. I would endorse this change going trough net-next (without the stable tag). I would feel less torn with this change targeting -net without the stable tag. Targeting -net with the stable tag sounds a bit too much to me.

Cheers,
Paolo

I'm sending this patch to net-next without stable tag.

Thanks,

Long