Linux CXL
 help / color / mirror / Atom feed
From: Dan Williams <dan.j.williams@intel.com>
To: Dave Jiang <dave.jiang@intel.com>, <linux-cxl@vger.kernel.org>
Cc: <dan.j.williams@intel.com>, <ira.weiny@intel.com>,
	<vishal.l.verma@intel.com>, <alison.schofield@intel.com>,
	<Jonathan.Cameron@huawei.com>, <dave@stgolabs.net>,
	<jgg@nvidia.com>, <shiju.jose@huawei.com>
Subject: Re: [PATCH v3 03/16] cxl: Add Get Supported Features command for kernel usage
Date: Tue, 4 Feb 2025 15:50:46 -0800	[thread overview]
Message-ID: <67a2a7d62e298_2d2c294d7@dwillia2-xfh.jf.intel.com.notmuch> (raw)
In-Reply-To: <20250204220430.4146187-4-dave.jiang@intel.com>

Dave Jiang wrote:
> CXL spec r3.2 8.2.9.6.1 Get Supported Features (Opcode 0500h)
> The command retrieve the list of supported device-specific features
> (identified by UUID) and general information about each Feature.
> 
> The driver will retrieve the feature entries in order to make checks and
> provide information for the Get Feature and Set Feature command. One of
> the main piece of information retrieved are the effects a Set Feature
> command would have for a particular feature.
> 
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> ---
> v3:
> - Use struct_size() instead of open coding the calculation. (Jonathan)
> ---
>  drivers/cxl/features.c | 138 +++++++++++++++++++++++++++++++++++++++++
>  include/cxl/features.h |  57 +++++++++++++++++
>  2 files changed, 195 insertions(+)
> 
> diff --git a/drivers/cxl/features.c b/drivers/cxl/features.c
> index 958e4828a58d..4d8c603121f0 100644
> --- a/drivers/cxl/features.c
> +++ b/drivers/cxl/features.c
> @@ -7,6 +7,139 @@
>  #include "cxlmem.h"
>  #include "features.h"
>  
> +static void cxl_free_feature_entries(void *entries)
> +{
> +	kvfree(entries);
> +}
> +
> +static int cxl_get_supported_features_count(struct cxl_mailbox *cxl_mbox)
> +{
[..]
> +
> +	cxlfs->num_features = count;
> +	cxlfs->entries = no_free_ptr(entries);
> +	return devm_add_action_or_reset(&cxlmd->dev, cxl_free_feature_entries,
> +					cxlfs->entries);

This is a memory leak. At this point cxlfs->entries is NULL having been
marked NULL by no_free_ptr().

It is awkward to mix a return code, a return buffer, and a return
features count all at the same time near a devm handoff. I would suggest
solving it by:

1/ skip calling devm_add_action_or_reset() inside of
   get_supported_features(), make the caller responsible for devm
   wrapping any allocations.

2/ Convey the error code and the object via an ERR_PTR() from
   get_supported_features()

3/ Move "num_features" into the allocation for @entries. Something like:

struct cxl_features_state {
     struct cxl_memdev *cxlmd;
     enum cxl_features_capability cap;
     struct cxl_feat_entries {
         int num_features;
         struct cxl_feat_entry ent[] __counted_by(num_features);
     } *entries;
};

...and then have get_supported_features() return a 'struct
cxl_feat_entries *' pointer.

> +}
> +
>  static void enumerate_feature_cmds(struct cxl_memdev *cxlmd,
>  				   struct cxl_features_state *cxlfs)
>  {
> @@ -77,12 +210,17 @@ DEFINE_FREE(cxl_free_cxlfs, struct cxl_features_state *, if (_T) cxl_cxlfs_free(
>   */
>  int devm_cxl_add_features(struct cxl_memdev *cxlmd)
>  {
> +	int rc;
> +
>  	struct cxl_features_state *cxlfs __free(cxl_free_cxlfs) =
>  		devm_cxlfs_allocate(cxlmd);

devm and scope-based should never appear in the same statement.

>  	if (!cxlfs)
>  		return -ENOMEM;
>  
>  	enumerate_feature_cmds(cxlmd, cxlfs);
> +	rc = get_supported_features(cxlmd, cxlfs);
> +	if (rc)
> +		return rc;
>  
>  	cxlmd->cxlfs = no_free_ptr(cxlfs);
>  
> diff --git a/include/cxl/features.h b/include/cxl/features.h
> index 8e1830f0ccba..c65760342f97 100644
> --- a/include/cxl/features.h
> +++ b/include/cxl/features.h
> @@ -3,6 +3,8 @@
>  #ifndef __CXL_FEATURES_H__
>  #define __CXL_FEATURES_H__
>  
> +#include <linux/uuid.h>
> +
>  struct cxl_mailbox;
>  
>  /* Index IDs for CXL mailbox Feature commands */
> @@ -25,11 +27,66 @@ enum cxl_features_capability {
>   * @cxlmd: Pointer to cxl mem device
>   * @cap: Feature commands capability
>   * @num_features: total Features supported by the device
> + * @entries: Feature detail entries fetched from the device
>   */
>  struct cxl_features_state {
>  	struct cxl_memdev *cxlmd;
>  	enum cxl_features_capability cap;
>  	int num_features;
> +	struct cxl_feat_entry *entries;

Per above, this organization results in some buggy gymanistics.

  reply	other threads:[~2025-02-04 23:50 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-04 22:03 [PATCH v3 00/16] cxl: Add CXL feature commands support via fwctl Dave Jiang
2025-02-04 22:03 ` [PATCH v3 01/16] cxl: Refactor user ioctl command path from mds to mailbox Dave Jiang
2025-02-05 17:41   ` Jonathan Cameron
2025-02-05 17:52     ` Dave Jiang
2025-02-07  5:47   ` Li Ming
2025-02-04 22:03 ` [PATCH v3 02/16] cxl: Enumerate feature commands Dave Jiang
2025-02-04 23:34   ` Dan Williams
2025-02-04 22:03 ` [PATCH v3 03/16] cxl: Add Get Supported Features command for kernel usage Dave Jiang
2025-02-04 23:50   ` Dan Williams [this message]
2025-02-07  5:42   ` Li Ming
2025-02-08  0:03     ` Dave Jiang
2025-02-04 22:03 ` [PATCH v3 04/16] cxl/test: Add Get Supported Features mailbox command support Dave Jiang
2025-02-07  5:51   ` Li Ming
2025-02-04 22:03 ` [PATCH v3 05/16] cxl/mbox: Add GET_FEATURE mailbox command Dave Jiang
2025-02-05  0:54   ` Dan Williams
2025-02-05 17:57     ` Jonathan Cameron
2025-02-05 23:12       ` Dan Williams
2025-02-06 11:03         ` Jonathan Cameron
2025-02-07 20:18           ` Dan Williams
2025-02-07  6:18   ` Li Ming
2025-02-04 22:03 ` [PATCH v3 06/16] cxl/mbox: Add SET_FEATURE " Dave Jiang
2025-02-04 22:03 ` [PATCH v3 07/16] cxl: Setup exclusive CXL features that are reserved for the kernel Dave Jiang
2025-02-04 22:03 ` [PATCH v3 08/16] cxl: Add FWCTL support to the CXL memdev driver Dave Jiang
2025-02-05  1:18   ` Dan Williams
2025-02-04 22:03 ` [PATCH v3 09/16] cxl: Add support for FWCTL get driver information callback Dave Jiang
2025-02-05  1:27   ` Dan Williams
2025-02-04 22:03 ` [PATCH v3 10/16] cxl: Move cxl feature command structs to user header Dave Jiang
2025-02-04 22:03 ` [PATCH v3 11/16] cxl: Add support for fwctl RPC command to enable CXL feature commands Dave Jiang
2025-02-05  1:41   ` Dan Williams
2025-02-04 22:03 ` [PATCH v3 12/16] cxl: Add support to handle user feature commands for get feature Dave Jiang
2025-02-04 22:03 ` [PATCH v3 13/16] cxl: Add support to handle user feature commands for set feature Dave Jiang
2025-02-04 22:03 ` [PATCH v3 14/16] cxl/test: Add Get Feature support to cxl_test Dave Jiang
2025-02-04 22:03 ` [PATCH v3 15/16] cxl/test: Add Set " Dave Jiang
2025-02-04 22:03 ` [PATCH v3 16/16] fwctl/cxl: Add documentation to FWCTL CXL Dave Jiang
2025-02-05  1:50   ` Dan Williams

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=67a2a7d62e298_2d2c294d7@dwillia2-xfh.jf.intel.com.notmuch \
    --to=dan.j.williams@intel.com \
    --cc=Jonathan.Cameron@huawei.com \
    --cc=alison.schofield@intel.com \
    --cc=dave.jiang@intel.com \
    --cc=dave@stgolabs.net \
    --cc=ira.weiny@intel.com \
    --cc=jgg@nvidia.com \
    --cc=linux-cxl@vger.kernel.org \
    --cc=shiju.jose@huawei.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