From: Robert Richter <rrichter@amd.com>
To: Alison Schofield <alison.schofield@intel.com>
Cc: Vishal Verma <vishal.l.verma@intel.com>,
Ira Weiny <ira.weiny@intel.com>,
Dan Williams <dan.j.williams@intel.com>,
Jonathan Cameron <Jonathan.Cameron@huawei.com>,
Dave Jiang <dave.jiang@intel.com>,
Davidlohr Bueso <dave@stgolabs.net>,
linux-cxl@vger.kernel.org, linux-kernel@vger.kernel.org,
Gregory Price <gourry@gourry.net>,
"Fabio M. De Francesco" <fabio.m.de.francesco@linux.intel.com>,
Terry Bowman <terry.bowman@amd.com>,
Joshua Hahn <joshua.hahnjy@gmail.com>
Subject: Re: [PATCH v4 12/14] cxl: Simplify cxl_rd_ops allocation and handling
Date: Tue, 11 Nov 2025 13:07:28 +0100 [thread overview]
Message-ID: <aRMnAIwdMnTTK1Tp@rric.localdomain> (raw)
In-Reply-To: <aQqF-xCnL2Ugh3hp@aschofie-mobl2.lan>
Alison,
On 04.11.25 15:02:19, Alison Schofield wrote:
> On Mon, Nov 03, 2025 at 07:47:53PM +0100, Robert Richter wrote:
> > A root decoder's callback handlers are collected in struct cxl_rd_ops.
> > The structure is dynamically allocated, though it contains only a few
> > pointers in it. This also requires to check two pointes to check for
> > the existance of a callback.
> >
> > Simplify the allocation, release and handler check by embedding the
> > ops statical in struct cxl_root_decoder.
> >
> > Implementation is equivalent to how struct cxl_root_ops handles the
> > callbacks.
>
> The allocation was intentionally dynamic because the root decoder ops
> only existed for CFMWS's defined w XOR Arithmetic.
>
> From the commit msg:
> >> To avoid maintaining a static ops instance populated with mostly NULL
> >> pointers, allocate the ops structure dynamically only when a platform
> >> requires overrides (e.g. XOR interleave decoding).
> >> The setup can be extended as additional callbacks are added.
> See: 524b2b76f365 ("cxl: Move hpa_to_spa callback to a new root decoder ops structure")
>
> Has the usage changed?
No, code did not change. In another patch I had to change struct
cxl_root and the cxl_root_ops allocation there. The pattern here is
the same and the code would benefit from the same change to simplify
it.
Dynamic allocation does not save much here (only a single pointer
entry) but requires resource allocation and a 2-level pointer check
which must be run even if the pointers are NULL. If there is at least
one pointer set (which becomes more likely with an increased number of
ops) you need to allocate the whole struct anyway and dynamic alloc
does net help then. The diffstat also shows the simplification:
3 files changed, 8 insertions(+), 22 deletions(-)
-Robert
next prev parent reply other threads:[~2025-11-11 12:07 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-03 18:47 [PATCH v4 00/14] cxl: ACPI PRM Address Translation Support and AMD Zen5 enablement Robert Richter
2025-11-03 18:47 ` [PATCH v4 01/14] cxl/region: Store root decoder in struct cxl_region Robert Richter
2025-11-11 14:45 ` Jonathan Cameron
2025-11-14 9:38 ` Robert Richter
2025-11-03 18:47 ` [PATCH v4 02/14] cxl/region: Store HPA range " Robert Richter
2025-11-11 11:25 ` Robert Richter
2025-11-03 18:47 ` [PATCH v4 03/14] cxl/region: Rename misleading variable name @hpa to @hpa_range Robert Richter
2025-11-03 21:36 ` Dave Jiang
2025-11-11 14:41 ` Jonathan Cameron
2025-11-12 16:23 ` Dave Jiang
2025-11-03 18:47 ` [PATCH v4 04/14] cxl/region: Add @hpa_range argument to function cxl_calc_interleave_pos() Robert Richter
2025-11-03 21:52 ` Dave Jiang
2025-11-04 3:04 ` Alison Schofield
2025-11-11 11:28 ` Robert Richter
2025-11-04 16:52 ` kernel test robot
2025-11-03 18:47 ` [PATCH v4 05/14] cxl: Simplify cxl_root_ops allocation and handling Robert Richter
2025-11-03 21:53 ` Dave Jiang
2025-11-04 23:02 ` Dave Jiang
2025-11-07 15:45 ` Robert Richter
2025-11-07 15:50 ` Dave Jiang
2025-11-11 14:52 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 06/14] cxl/region: Separate region parameter setup and region construction Robert Richter
2025-11-03 22:05 ` Dave Jiang
2025-11-07 15:59 ` Robert Richter
2025-11-11 14:59 ` Jonathan Cameron
2025-11-11 15:02 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 07/14] cxl/region: Use region data to get the root decoder Robert Richter
2025-11-03 22:30 ` Dave Jiang
2025-11-11 15:14 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 08/14] cxl: Introduce callback for HPA address ranges translation Robert Richter
2025-11-03 23:09 ` Dave Jiang
2025-11-11 15:15 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 09/14] cxl/acpi: Prepare use of EFI runtime services Robert Richter
2025-11-03 23:34 ` Dave Jiang
2025-11-11 15:17 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 10/14] cxl: Enable AMD Zen5 address translation using ACPI PRMT Robert Richter
2025-11-04 1:00 ` Dave Jiang
2025-11-11 9:23 ` Robert Richter
2025-11-04 9:33 ` kernel test robot
2025-11-04 12:49 ` Robert Richter
2025-11-04 23:35 ` kernel test robot
2025-11-11 15:30 ` Jonathan Cameron
2025-11-13 11:24 ` Robert Richter
2025-11-03 18:47 ` [PATCH v4 11/14] cxl/atl: Lock decoders that need address translation Robert Richter
2025-11-04 17:13 ` Dave Jiang
2025-11-11 12:54 ` Robert Richter
2025-11-12 16:34 ` Dave Jiang
2025-11-13 20:05 ` Robert Richter
2025-11-13 20:36 ` Dave Jiang
2025-11-14 7:34 ` Robert Richter
2025-11-14 15:21 ` Dave Jiang
2025-11-11 15:31 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 12/14] cxl: Simplify cxl_rd_ops allocation and handling Robert Richter
2025-11-04 17:26 ` Dave Jiang
2025-11-04 23:02 ` Alison Schofield
2025-11-11 12:07 ` Robert Richter [this message]
2025-11-11 15:34 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 13/14] cxl/acpi: Group xor arithmetric setup code in a single block Robert Richter
2025-11-11 15:35 ` Jonathan Cameron
2025-11-03 18:47 ` [PATCH v4 14/14] cxl/region: Remove local variable @inc in cxl_port_setup_targets() Robert Richter
2025-11-11 15:36 ` Jonathan Cameron
2025-11-13 20:10 ` Robert Richter
2025-11-04 16:17 ` [PATCH v4 00/14] cxl: ACPI PRM Address Translation Support and AMD Zen5 enablement Alison Schofield
2025-11-17 15:34 ` Robert Richter
2025-11-17 17:23 ` Gregory Price
2025-11-11 14:01 ` Gregory Price
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=aRMnAIwdMnTTK1Tp@rric.localdomain \
--to=rrichter@amd.com \
--cc=Jonathan.Cameron@huawei.com \
--cc=alison.schofield@intel.com \
--cc=dan.j.williams@intel.com \
--cc=dave.jiang@intel.com \
--cc=dave@stgolabs.net \
--cc=fabio.m.de.francesco@linux.intel.com \
--cc=gourry@gourry.net \
--cc=ira.weiny@intel.com \
--cc=joshua.hahnjy@gmail.com \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=terry.bowman@amd.com \
--cc=vishal.l.verma@intel.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