Linux CXL
 help / color / mirror / Atom feed
From: Dave Jiang <dave.jiang@intel.com>
To: alucerop@amd.com, linux-cxl@vger.kernel.org, netdev@vger.kernel.org
Cc: davem@davemloft.net, kuba@kernel.org, pabeni@redhat.com,
	edumazet@google.com, ecree.xilinx@gmail.com, icheng@nvidia.com,
	rafael@kernel.org
Subject: Re: [PATCH v2 2/4] cxl/region: Add region reference in memdev attach
Date: Thu, 1 Oct 2026 14:38:16 -0700	[thread overview]
Message-ID: <6bc33514-8bfb-44d6-8fde-28f45dff5eb9@intel.com> (raw)
In-Reply-To: <20261001132023.17032-3-alucerop@amd.com>



On 10/1/26 6:20 AM, alucerop@amd.com wrote:
> From: Alejandro Lucero <alucerop@amd.com>
> 
> Use a new field in cxl_attach_region struct for easily link it with the
> region the memdev is attached to.
> 
> This facilitates device links creation where such a region is the supplier
> with non-PF0 physical functions wanting to use the CXL region being the
> consumers.
> 
> Signed-off-by: Alejandro Lucero <alucerop@amd.com>
> ---
>  drivers/cxl/core/region.c | 1 +
>  drivers/cxl/cxlmem.h      | 2 ++
>  2 files changed, 3 insertions(+)
> 
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index 27e63e6dab7c..78ca7ebc3e55 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -4132,6 +4132,7 @@ int cxl_memdev_attach_region(struct cxl_memdev *cxlmd)
>  	if (rc)
>  		return rc;
>  
> +	attach->cxlr = cxlr;
>  	attach->hpa_range = (struct range) {
>  		.start = cxlr->params.res->start,
>  		.end = cxlr->params.res->end,
> diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h
> index c401e3a1af06..c598561b8e5f 100644
> --- a/drivers/cxl/cxlmem.h
> +++ b/drivers/cxl/cxlmem.h
> @@ -104,6 +104,7 @@ struct cxl_memdev_attach {
>  /**
>   * struct cxl_attach_region - coordinate mapping a region at memdev registration
>   * @attach: common core attachment descriptor
> + * @cxlr: cxl region the memdev is attached to.
>   * @hpa_range: physical address range of the region
>   *
>   * For the common simple case of a CXL device with private (non-general purpose
> @@ -112,6 +113,7 @@ struct cxl_memdev_attach {
>   */
>  struct cxl_attach_region {
>  	struct cxl_memdev_attach attach;
> +	struct cxl_region *cxlr;
>  	struct range hpa_range;
>  };
>  

attach->cxlr is never cleared when the region goes away. Unbinding the endpoint port runs endpoint_unregister_region(), which unregisters the region and drops its reference. PF0 stays bound until the detach work runs.

A non-PF0 probe in that window still finds the memdev and calls device_link_add() on the freed region.

How about something like this?

diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index 27e63e6dab7c..38ca73f12b84 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -4076,6 +4076,23 @@ static int first_mapped_decoder(struct device *dev, const void *data)
 	return 0;
 }
 
+/*
+ * Invalidate @attach before the region goes away so that
+ * cxl_get_range_and_link() can not pick up a stale region.
+ */
+static void endpoint_detach_attach_region(void *_attach)
+{
+	struct cxl_attach_region *attach = _attach;
+	struct cxl_region *cxlr;
+
+	scoped_guard(rwsem_write, &cxl_rwsem.region) {
+		cxlr = attach->cxlr;
+		WRITE_ONCE(attach->cxlr, NULL);
+		attach->hpa_range = DEFINE_RANGE(0, -1);
+	}
+	endpoint_unregister_region(cxlr);
+}
+
 /*
  * Runs in cxl_mem_probe context after successful endpoint probe, assumes the
  * simple case of single mapped decoder per memdev.
@@ -4127,15 +4144,23 @@ int cxl_memdev_attach_region(struct cxl_memdev *cxlmd)
 
 	/* Only teardown regions that pass validation, ignore the rest */
 	get_device(&cxlr->dev);
-	rc = devm_add_action_or_reset(&endpoint->dev,
-				      endpoint_unregister_region, cxlr);
-	if (rc)
+	/*
+	 * Not devm_add_action_or_reset(): the reset path would take
+	 * cxl_rwsem.region for write while it is held for read here. The
+	 * endpoint lock keeps the action from running before @attach is set.
+	 */
+	rc = devm_add_action(&endpoint->dev, endpoint_detach_attach_region,
+			     attach);
+	if (rc) {
+		put_device(&cxlr->dev);
 		return rc;
+	}
 
 	attach->hpa_range = (struct range) {
 		.start = cxlr->params.res->start,
 		.end = cxlr->params.res->end,
 	};
+	WRITE_ONCE(attach->cxlr, cxlr);
 	return 0;
 }
 EXPORT_SYMBOL_FOR_MODULES(cxl_memdev_attach_region, "cxl_mem");
diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h
index c401e3a1af06..7cd3a69cd5f5 100644
--- a/drivers/cxl/cxlmem.h
+++ b/drivers/cxl/cxlmem.h
@@ -104,6 +104,8 @@ struct cxl_memdev_attach {
 /**
  * struct cxl_attach_region - coordinate mapping a region at memdev registration
  * @attach: common core attachment descriptor
+ * @cxlr: cxl region the memdev is attached to, cleared under cxl_rwsem.region
+ *	before the region is unregistered
  * @hpa_range: physical address range of the region
  *
  * For the common simple case of a CXL device with private (non-general purpose
@@ -112,6 +114,7 @@ struct cxl_memdev_attach {
  */
 struct cxl_attach_region {
 	struct cxl_memdev_attach attach;
+	struct cxl_region *cxlr;
 	struct range hpa_range;
 };
 

  reply	other threads:[~2026-10-01 21:38 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-01 13:20 [PATCH v2 0/4] Type2 multipf support alucerop
2026-10-01 13:20 ` [PATCH v2 1/4] driver core: Check for supplier requiring PM at link creation alucerop
2026-10-01 20:31   ` Dave Jiang
2026-10-02  4:32     ` Lucero Palau, Alejandro
2026-10-02 15:31       ` Dave Jiang
2026-10-02 12:02   ` sashiko-bot
2026-10-01 13:20 ` [PATCH v2 2/4] cxl/region: Add region reference in memdev attach alucerop
2026-10-01 21:38   ` Dave Jiang [this message]
2026-10-02  4:41     ` Lucero Palau, Alejandro
2026-10-02 15:52       ` Dave Jiang
2026-10-08 13:50         ` Lucero Palau, Alejandro
2026-10-08 16:18           ` Dave Jiang
2026-10-08 18:07             ` Lucero Palau, Alejandro
2026-10-08 21:05               ` Dave Jiang
2026-10-09  6:58                 ` Lucero Palau, Alejandro
2026-10-09 16:57                   ` Dave Jiang
2026-10-02 12:02   ` sashiko-bot
2026-10-01 13:20 ` [PATCH v2 3/4] cxl/memdev: Add support for multi PF devices alucerop
2026-10-01 22:11   ` Dave Jiang
2026-10-01 22:41     ` Dave Jiang
2026-10-02  4:50     ` Lucero Palau, Alejandro
2026-10-02 15:55       ` Dave Jiang
2026-10-02 12:02   ` sashiko-bot
2026-10-01 13:20 ` [PATCH v2 4/4] sfc: add multipf support alucerop
2026-10-01 22:32   ` Dave Jiang
2026-10-02  5:33     ` Lucero Palau, Alejandro
2026-10-02 12:02   ` sashiko-bot

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=6bc33514-8bfb-44d6-8fde-28f45dff5eb9@intel.com \
    --to=dave.jiang@intel.com \
    --cc=alucerop@amd.com \
    --cc=davem@davemloft.net \
    --cc=ecree.xilinx@gmail.com \
    --cc=edumazet@google.com \
    --cc=icheng@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=linux-cxl@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=rafael@kernel.org \
    /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