Message-ID: <Z_7qhcoDwIz2GJW4@aschofie-mobl2.lan>
Date: Tue, 15 Apr 2025 16:23:49 -0700
From: Alison Schofield <alison.schofield@...el.com>
To: Robert Richter <rrichter@....com>
CC: Vishal Verma <vishal.l.verma@...el.com>, Ira Weiny <ira.weiny@...el.com>,
	Dan Williams <dan.j.williams@...el.com>, Jonathan Cameron
	<Jonathan.Cameron@...wei.com>, Dave Jiang <dave.jiang@...el.com>, "Davidlohr
 Bueso" <dave@...olabs.net>, <linux-cxl@...r.kernel.org>,
	<linux-kernel@...r.kernel.org>, Gregory Price <gourry@...rry.net>, "Fabio M.
 De Francesco" <fabio.m.de.francesco@...ux.intel.com>, Terry Bowman
	<terry.bowman@....com>
Subject: Re: [PATCH v4 00/14] cxl: Address translation support, part 1:
 Cleanups and refactoring

On Thu, Mar 06, 2025 at 05:44:34PM +0100, Robert Richter wrote:
> This series is the first part of adding support for CXL address
> translation. It contains cleanup and code refactoring in preparation
> of the actual implementation that will be sent in part 2. Cleanup and
> code refactoring have been split in a separate series to reduce the
> number of patches of the series. Even without address translation on
> top this rework improves esp. the region code, cleans it up,
> simplifies it and adds debugging messages to better analyze region
> creation failures:
> 
> Content of patches:
> 
>  * Patches 1: Remove else after return.
> 
>  * Patches 2-3: Cleanups and comments around cxl_hdm_decode_init().
> 
>  * Patches 4, 5: Adding and modifying helper functions.
> 
>  * Patches 6-10: Refactoring of endpoint decoder setup.
> 
>  * Patches 11-14: Adding and modifying debug messages.
> 

Hi Robert, 

cross-post from Discord here:

@Robert Richter @Dave Jiang (Intel)  I sync'd my work on top of Roberts
v4 Part One, and then walked thru and tagged most of the 14 patch set.
Can a bunch of those go onto cxl/next now?  I saw 'hard' no's on 5,6,10
and a couple of maybes 2,3,9 but even if you omit all those, we could
get 8 patches simmering on cxl/next now.

In the collab sync you mentioned working this set. After using and
reviewing today, and seeing I'm not the first person to suggest this,
I think we would benefit from merging part of this set while you 
refine the rest.

I believe they all have plenty of tags, so it's based on what you are
willing to let fly!

-- Alison


> v4:
>  * rebased onto cxl/next, commit 0a14566be090 ("cxl/Documentation:
>    Remove 'mixed' from sysfs mode doc"),
>  * added tags to SOB chain,
>  * reworked comments in cxl_hdm_decode_init() (dropped moving comment
>    and updated patch that modifies comments) (Jonathan),
>  * reworded patch description that removes duplicate call of
>    cxl_find_decoder_early() (Jonathan),
>  * moved some patches out of this rework and cleanup series (Dave,
>    Jonathan),
> 
> v3:
>  * added tags to SOB chain,
>  * fixed NULL pointer dereference in cxl_find_root_decoder() (Alison),
>  * updated subject line of patches that add kernel messages and
>    included example log messages (Alison),
> 
> v2:
>  * rebased onto cxl/next,
>  * added tags to SOB chain,
>  * move patches with cleanups and refactoring into this separate
>    series (Dave),
>  * added patch "cxl/acpi: Unify CFMWS memory log messages with SRAT
>    messages" to improve CFMWS log messages,
>  * renamed endpoint decoder functions to cxl_endpoint_decoder_*() (Li),
>  * reworded patch description that moves find_cxl_root() and reworks
>    cxl_find_root_decoder() (Terry),
>  * small changes to cxl_find_root_decoder()/
>    cxl_endpoint_decoder_initialize() (Jonanthan),
>  * updated comment in cxl_port_find_switch_decoder() (Ben),
>  * cxl_endpoint_decoder_initialize(): Simplify variable declaration
>    (Jonathan, Ben),
>  * cxl_find_decoder_early(): Added comment on function usage (Gregory),
>  * reordered patches and reworded some of the subject for a better
>    structure.
> 
> Robert Richter (14):
>   cxl: Remove else after return
>   cxl/pci: Moving code in cxl_hdm_decode_init()
>   cxl/pci: Add comments to cxl_hdm_decode_init()
>   cxl: Introduce parent_port_of() helper
>   cxl/region: Rename function to cxl_find_decoder_early()
>   cxl/region: Avoid duplicate call of cxl_find_decoder_early()
>   cxl/region: Move find_cxl_root() to cxl_add_to_region()
>   cxl/region: Factor out code to find the root decoder
>   cxl/region: Factor out code to find a root decoder's region
>   cxl/region: Add function to find a port's switch decoder by range
>   cxl/region: Add a dev_warn() on registration failure
>   cxl/region: Add a dev_err() on missing target list entries
>   cxl: Add a dev_dbg() when a decoder was added to a port
>   cxl/acpi: Unify CFMWS memory log messages with SRAT messages
> 
>  drivers/cxl/acpi.c        |  12 ++-
>  drivers/cxl/core/cdat.c   |   2 +-
>  drivers/cxl/core/hdm.c    |   3 +-
>  drivers/cxl/core/pci.c    |  50 +++++++----
>  drivers/cxl/core/port.c   |  15 ++--
>  drivers/cxl/core/region.c | 170 +++++++++++++++++++++++++-------------
>  drivers/cxl/cxl.h         |   7 +-
>  drivers/cxl/port.c        |  15 +---
>  8 files changed, 172 insertions(+), 102 deletions(-)
> 
> 
> base-commit: 0a14566be090ca51a32ebdd8a8e21678062dac08
> -- 
> 2.39.5
> 

Powered by blists - more mailing lists