Netdev List
 help / color / mirror / Atom feed
From: Potnuri Bharat Teja <bharat@chelsio.com>
To: Simon Horman <horms@kernel.org>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"davem@davemloft.net" <davem@davemloft.net>,
	"kuba@kernel.org" <kuba@kernel.org>,
	"edumazet@google.com" <edumazet@google.com>,
	"pabeni@redhat.com" <pabeni@redhat.com>,
	"andrew+netdev@lunn.ch" <andrew+netdev@lunn.ch>,
	"Harshita V. Rajput" <harshitha.vr@chelsio.com>
Subject: Re: [PATCH net-next v5 REPOST 03/12] cxgb4: add T7 hardware management support
Date: Tue, 17 Nov 2026 20:15:07 +0530	[thread overview]
Message-ID: <avxoc4fRV6sxEuh/@chelsio.com> (raw)
In-Reply-To: <20261111155034.GF18859@horms.kernel.org>

On Wednesday, November 11/11/26, 2026 at 21:20:34 +0530, Simon Horman wrote:
> On Thu, Nov 05, 2026 at 08:15:40AM -0500, Potnuri Bharat Teja wrote:
> > Extend the cxgb4 hardware management layer to support T7 adapters.
> > 
> > Add T7 handling to hardware resource and topology discovery.  The TP
> > load balancer mode and the firmware supplied TP channel map are read
> > into adapter_params, and t4_get_mps_bg_map() and t4_get_tp_ch_map()
> > account for the T7 channel layout.  The load balancer mode distributes
> > the four TP channels across the active ports as four ports of one
> > channel, two ports of two channels, or one port of four channels.
> > 
> > t4_get_chip_type() moves from cxgb4_main.c into t4_hw.c so that chip
> > identification lives with the rest of the hardware layer, and learns
> > the T7 chip version.
> > 
> > Add T7 flash support.  The flash layout is described by a table
> > indexed by flash location rather than by fixed FLASH_*_START_SEC and
> > FLASH_*_NSECS constants, with t4_flash_location_start_sec(),
> > t4_flash_location_nsecs(), t4_flash_location_start() and
> > t4_flash_location_size() returning the bounds of a region.
> > t4_flash_cfg_addr() now returns int so that it can report a region
> > that does not exist on the running adapter.  t4_get_flash_params()
> > learns the flash parts used on T7 and a four byte address serial
> > flash interface, which is required for parts larger than 16MB.
> > 
> > Add the T7 register ranges to t4_get_regs(), the T7 PCIE memory window
> > registers, and T7 support to the CIM ingress and egress queue, CIM LA,
> > SGE context, MPS trace filter, PM statistics, scheduler and I2C
> > accessors.
> > 
> > T7 implements multiple uP cores.  The number of enabled cores is read
> > during device log initialisation and the per core device log
> > parameters are kept in an array, so that the device log and the CIM
> > queue accessors can address a specific core.
> > 
> > Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
> > ---
> >  .../net/ethernet/chelsio/cxgb4/cudbg_lib.c    |    4 +-
> >  drivers/net/ethernet/chelsio/cxgb4/cxgb4.h    |   89 +-
> >  .../ethernet/chelsio/cxgb4/cxgb4_debugfs.c    |    2 +-
> >  .../net/ethernet/chelsio/cxgb4/cxgb4_main.c   |   17 -
> >  drivers/net/ethernet/chelsio/cxgb4/t4_hw.c    | 2200 ++++++++++++++---
> >  drivers/net/ethernet/chelsio/cxgb4/t4_hw.h    |  139 +-
> 
> Hi Potnuri,
> 
> I've looked over the first three patches of this series, with the help
> of AI, and I'd like to provide some general feedback:
> 
> * These patches, and the overall patchset are very long.
>   I would really suggest splitting up these patches.
>   And I think that means also splitting up the patch-set,
>   into stages.
> 
Hi Simon,
Thanks for your detailed response and time.
I had to leave these first 2 patches huge as they are adding new HW register/structures. 
and they may be remain reltively huge even after splitting them. I may need your suggestion here. 

The rest of the patches are relatively small and I am working on splitting the rest of the patches 
to smaller ones where all possible.
>   I know that sounds like it's going to take longer.
>   But experience tells me that getting review of large
>   patchsets to the extent that a merge occurs tends
>   to be a very long road.
> 
> * Being long, these patches do many things.
>   But the preferred approach is one thing per patch.
>   This is much easier on my human brain.
>   And I don't believe I am alone here.

agree with you, however these changes are interdependent so not sure if they can be split into series.
> 
> * It seems to be a bit of a stretch for patches 1/12 and 2/12 to
>   claim there are no functional changes. And to be honest,
>   they are two long for me to dive into the details.
> 
1 and 2 only contain device specific definitions
> And, some specific feedback on this patch:
> 
> * It appears to introduce new kdoc warnings. Which I believe
>   are gating part of our CI. This is not desirable.
> 
Shall fix these and validate with kernel-doc prior to posing the next series.
Thanks.

> Warnings before: 262 after: 276 (add: 20 del: 6)
> 
> Warnings removed:
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:12909 No description found for return value of 't4_read_cim_ibq'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:12951 No description found for return value of 't4_read_cim_obq'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13550 No description found for return value of 't4_i2c_rd'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:12909 No description found for return value of 't4_read_cim_ibq'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:12951 No description found for return value of 't4_read_cim_obq'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13550 No description found for return value of 't4_i2c_rd'
> Per-file breakdown:
>      6 drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
> 
> New warnings added:
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:9269 No description found for return value of 't7_wait_sram_done'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:10396 No description found for return value of 't4_get_tp_port_chan'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13332 No description found for return value of 't4_get_chip_type'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13641 No description found for return value of 't4_sge_get_qpp'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14243 No description found for return value of 't4_read_cim_ibq_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14319 No description found for return value of 't4_read_cim_obq_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14373 No description found for return value of 't4_cim_read_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14429 No description found for return value of 't4_cim_write_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14484 No description found for return value of 't4_cim_read_la_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:15019 No description found for return value of 't4_i2c_io'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:9269 No description found for return value of 't7_wait_sram_done'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:10396 No description found for return value of 't4_get_tp_port_chan'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13332 No description found for return value of 't4_get_chip_type'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:13641 No description found for return value of 't4_sge_get_qpp'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14243 No description found for return value of 't4_read_cim_ibq_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14319 No description found for return value of 't4_read_cim_obq_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14373 No description found for return value of 't4_cim_read_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14429 No description found for return value of 't4_cim_write_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:14484 No description found for return value of 't4_cim_read_la_core'
> Warning: drivers/net/ethernet/chelsio/cxgb4/t4_hw.c:15019 No description found for return value of 't4_i2c_io'
> Per-file breakdown:
>     20 drivers/net/ethernet/chelsio/cxgb4/t4_hw.c
> 
> -- 
> pw-bot: changes-requested
> 
> 

  reply	other threads:[~2026-11-17 14:45 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-05 13:15 [PATCH net-next v5 REPOST 00/12] cxgb4: Add T7 adapter support Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 01/12] cxgb4: add T7 hardware definitions and firmware interfaces Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 02/12] cxgb4: rework the interrupt handling framework Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 03/12] cxgb4: add T7 hardware management support Potnuri Bharat Teja
2026-11-11 15:50   ` Simon Horman
2026-11-17 14:45     ` Potnuri Bharat Teja [this message]
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 04/12] cxgb4: move PCI device management into cxgb4_pci.c and rework driver helpers Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 05/12] cxgb4: add T7 support to the main driver Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 06/12] cxgb4: update PTP register access for T7 Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 07/12] cxgb4: extend ethtool support for T7 adapters Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 08/12] cxgb4: Add T7 support to the filter infrastructure Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 09/12] cxgb4: Add indirect register definitions for T7 Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 10/12] cxgb4: extend CUDBG support for T7 adapters Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 11/12] cxgb4: extend debugfs " Potnuri Bharat Teja
2026-11-05 13:15 ` [PATCH net-next v5 REPOST 12/12] cxgb4: select ULD Tx and control queues by TID on T7 Potnuri Bharat Teja

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=avxoc4fRV6sxEuh/@chelsio.com \
    --to=bharat@chelsio.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=harshitha.vr@chelsio.com \
    --cc=horms@kernel.org \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox