netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: <alucerop@amd.com>
To: <linux-cxl@vger.kernel.org>, <netdev@vger.kernel.org>,
	<dan.j.williams@intel.com>, <edward.cree@amd.com>,
	<davem@davemloft.net>, <kuba@kernel.org>, <pabeni@redhat.com>,
	<edumazet@google.com>, <dave.jiang@intel.com>
Cc: Alejandro Lucero <alucerop@amd.com>
Subject: [PATCH v10 16/26] cxl: define a driver interface for DPA allocation
Date: Wed, 5 Feb 2025 15:19:40 +0000	[thread overview]
Message-ID: <20250205151950.25268-17-alucerop@amd.com> (raw)
In-Reply-To: <20250205151950.25268-1-alucerop@amd.com>

From: Alejandro Lucero <alucerop@amd.com>

Region creation involves finding available DPA (device-physical-address)
capacity to map into HPA (host-physical-address) space. Define an API,
cxl_request_dpa(), that tries to allocate the DPA memory the driver
requires to operate. The memory requested should not be bigger than the
max available HPA obtained previously with cxl_get_hpa_freespace.

Based on https://lore.kernel.org/linux-cxl/168592158743.1948938.7622563891193802610.stgit@dwillia2-xfh.jf.intel.com/

Signed-off-by: Alejandro Lucero <alucerop@amd.com>
---
 drivers/cxl/core/hdm.c | 83 ++++++++++++++++++++++++++++++++++++++++++
 include/cxl/cxl.h      |  4 ++
 2 files changed, 87 insertions(+)

diff --git a/drivers/cxl/core/hdm.c b/drivers/cxl/core/hdm.c
index af025da81fa2..cec2c7dcaf3a 100644
--- a/drivers/cxl/core/hdm.c
+++ b/drivers/cxl/core/hdm.c
@@ -3,6 +3,7 @@
 #include <linux/seq_file.h>
 #include <linux/device.h>
 #include <linux/delay.h>
+#include <cxl/cxl.h>
 
 #include "cxlmem.h"
 #include "core.h"
@@ -587,6 +588,7 @@ int cxl_dpa_free(struct cxl_endpoint_decoder *cxled)
 	up_write(&cxl_dpa_rwsem);
 	return rc;
 }
+EXPORT_SYMBOL_NS_GPL(cxl_dpa_free, "CXL");
 
 int cxl_dpa_set_part(struct cxl_endpoint_decoder *cxled,
 		     enum cxl_partition_mode mode)
@@ -701,6 +703,87 @@ int cxl_dpa_alloc(struct cxl_endpoint_decoder *cxled, unsigned long long size)
 	return devm_add_action_or_reset(&port->dev, cxl_dpa_release, cxled);
 }
 
+static int find_free_decoder(struct device *dev, const void *data)
+{
+	struct cxl_endpoint_decoder *cxled;
+	struct cxl_port *port;
+
+	if (!is_endpoint_decoder(dev))
+		return 0;
+
+	cxled = to_cxl_endpoint_decoder(dev);
+	port = cxled_to_port(cxled);
+
+	if (cxled->cxld.id != port->hdm_end + 1)
+		return 0;
+
+	return 1;
+}
+
+/**
+ * cxl_request_dpa - search and reserve DPA given input constraints
+ * @cxlmd: memdev with an endpoint port with available decoders
+ * @is_ram: DPA operation mode (ram vs pmem)
+ * @min: the minimum amount of capacity the call needs
+ *
+ * Given that a region needs to allocate from limited HPA capacity it
+ * may be the case that a device has more mappable DPA capacity than
+ * available HPA. So, the expectation is that @min is a driver known
+ * value for how much capacity is needed, and @max is the limit of
+ * how much HPA space is available for a new region.
+ *
+ * Returns a pinned cxl_decoder with at least @min bytes of capacity
+ * reserved, or an error pointer. The caller is also expected to own the
+ * lifetime of the memdev registration associated with the endpoint to
+ * pin the decoder registered as well.
+ */
+struct cxl_endpoint_decoder *cxl_request_dpa(struct cxl_memdev *cxlmd,
+					     bool is_ram,
+					     resource_size_t alloc)
+{
+	struct cxl_port *endpoint = cxlmd->endpoint;
+	struct cxl_endpoint_decoder *cxled;
+	enum cxl_partition_mode mode;
+	struct device *cxled_dev;
+	int rc;
+
+	if (!IS_ALIGNED(alloc, SZ_256M))
+		return ERR_PTR(-EINVAL);
+
+	down_read(&cxl_dpa_rwsem);
+	cxled_dev = device_find_child(&endpoint->dev, NULL, find_free_decoder);
+	up_read(&cxl_dpa_rwsem);
+
+	if (!cxled_dev)
+		return ERR_PTR(-ENXIO);
+
+	cxled = to_cxl_endpoint_decoder(cxled_dev);
+
+	if (!cxled) {
+		rc = -ENODEV;
+		goto err;
+	}
+
+	if (is_ram)
+		mode = CXL_PARTMODE_RAM;
+	else
+		mode = CXL_PARTMODE_PMEM;
+
+	rc = cxl_dpa_set_part(cxled, mode);
+	if (rc)
+		goto err;
+
+	rc = cxl_dpa_alloc(cxled, alloc);
+	if (rc)
+		goto err;
+
+	return cxled;
+err:
+	put_device(cxled_dev);
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL_NS_GPL(cxl_request_dpa, "CXL");
+
 static void cxld_set_interleave(struct cxl_decoder *cxld, u32 *ctrl)
 {
 	u16 eig;
diff --git a/include/cxl/cxl.h b/include/cxl/cxl.h
index 3b72dc7ce8cf..3fa390b10089 100644
--- a/include/cxl/cxl.h
+++ b/include/cxl/cxl.h
@@ -90,4 +90,8 @@ struct cxl_root_decoder *cxl_get_hpa_freespace(struct cxl_memdev *cxlmd,
 					       unsigned long flags,
 					       resource_size_t *max);
 void cxl_put_root_decoder(struct cxl_root_decoder *cxlrd);
+struct cxl_endpoint_decoder *cxl_request_dpa(struct cxl_memdev *cxlmd,
+					     bool is_ram,
+					     resource_size_t alloc);
+int cxl_dpa_free(struct cxl_endpoint_decoder *cxled);
 #endif
-- 
2.17.1


  parent reply	other threads:[~2025-02-05 15:20 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-05 15:19 [PATCH v10 00/26] cxl: add type2 device basic support alucerop
2025-02-05 15:19 ` [PATCH v10 01/26] cxl: make memdev creation type agnostic alucerop
2025-02-06 19:37   ` Dan Williams
2025-02-17 12:32     ` Alejandro Lucero Palau
2025-02-19  2:29       ` Dan Williams
2025-02-20 18:17         ` Alejandro Lucero Palau
2025-02-17 13:05     ` Alejandro Lucero Palau
2025-02-13  3:57   ` Alison Schofield
2025-02-17 12:49     ` Alejandro Lucero Palau
2025-02-17 13:06     ` Alejandro Lucero Palau
2025-02-14 17:02   ` Jonathan Cameron
2025-02-17 13:08     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 02/26] sfc: add basic cxl initialization alucerop
2025-02-06  1:37   ` Edward Cree
2025-02-07 12:48   ` Simon Horman
2025-02-17 13:10     ` Alejandro Lucero Palau
2025-02-07 13:03   ` Simon Horman
2025-02-17 13:11     ` Alejandro Lucero Palau
2025-02-18 13:32       ` Simon Horman
2025-02-05 15:19 ` [PATCH v10 03/26] cxl: move pci generic code alucerop
2025-02-05 21:33   ` Ira Weiny
2025-02-06 17:49     ` Alejandro Lucero Palau
2025-02-14 17:11       ` Jonathan Cameron
2025-02-17 13:13         ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 04/26] cxl: move register/capability check to driver alucerop
2025-02-07 12:52   ` Simon Horman
2025-02-17 13:17     ` Alejandro Lucero Palau
2025-02-14 17:21   ` Jonathan Cameron
2025-02-17 13:18     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 05/26] cxl: add function for type2 cxl regs setup alucerop
2025-02-05 21:35   ` Ira Weiny
2025-02-06 17:50     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 06/26] sfc: use cxl api for regs setup and checking alucerop
2025-02-05 21:31   ` Ira Weiny
2025-02-06 17:47     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 07/26] cxl: add support for setting media ready by an accel driver alucerop
2025-02-05 21:42   ` Ira Weiny
2025-02-06 17:58     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 08/26] sfc: set cxl media ready alucerop
2025-02-05 15:19 ` [PATCH v10 09/26] cxl: support device identification without mailbox alucerop
2025-02-05 21:45   ` Ira Weiny
2025-02-06 18:10     ` Alejandro Lucero Palau
2025-02-06 19:23       ` Ira Weiny
2025-02-17 13:41         ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 10/26] cxl: modify dpa setup process for supporting type2 alucerop
2025-02-05 15:19 ` [PATCH v10 11/26] sfc: initialize dpa resources alucerop
2025-02-05 15:19 ` [PATCH v10 12/26] cxl: prepare memdev creation for type2 alucerop
2025-02-05 15:19 ` [PATCH v10 13/26] sfc: create type2 cxl memdev alucerop
2025-02-05 15:19 ` [PATCH v10 14/26] cxl: define a driver interface for HPA free space enumeration alucerop
2025-02-07 12:55   ` Simon Horman
2025-02-17 13:44     ` Alejandro Lucero Palau
2025-02-13  4:08   ` Alison Schofield
2025-02-17 13:49     ` Alejandro Lucero Palau
2025-02-05 15:19 ` [PATCH v10 15/26] sfc: obtain root decoder with enough HPA free space alucerop
2025-02-05 22:47   ` Ira Weiny
2025-02-17 13:54     ` Alejandro Lucero Palau
2025-02-18  0:03       ` Ira Weiny
2025-02-05 15:19 ` alucerop [this message]
2025-02-06 19:11   ` [PATCH v10 16/26] cxl: define a driver interface for DPA allocation kernel test robot
2025-02-07 13:46   ` Simon Horman
2025-02-17 14:08     ` Alejandro Lucero Palau
2025-02-18 13:34       ` Simon Horman
2025-02-18 14:09         ` Simon Horman
2025-02-05 15:19 ` [PATCH v10 17/26] sfc: get endpoint decoder alucerop
2025-02-05 15:19 ` [PATCH v10 18/26] cxl: make region type based on endpoint type alucerop
2025-02-05 15:19 ` [PATCH v10 19/26] cxl/region: factor out interleave ways setup alucerop
2025-02-05 15:19 ` [PATCH v10 20/26] cxl/region: factor out interleave granularity setup alucerop
2025-02-05 15:19 ` [PATCH v10 21/26] cxl: allow region creation by type2 drivers alucerop
2025-02-06 20:06   ` kernel test robot
2025-02-07 13:23   ` Simon Horman
2025-02-05 15:19 ` [PATCH v10 22/26] cxl: add region flag for precluding a device memory to be used for dax alucerop
2025-02-05 15:19 ` [PATCH v10 23/26] sfc: create cxl region alucerop
2025-02-05 15:19 ` [PATCH v10 24/26] cxl: add function for obtaining region range alucerop
2025-02-05 15:19 ` [PATCH v10 25/26] sfc: update MCDI protocol headers alucerop
2025-02-05 15:19 ` [PATCH v10 26/26] sfc: support pio mapping based on cxl alucerop
2025-02-13  1:51 ` [PATCH v10 00/26] cxl: add type2 device basic support Alison Schofield

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=20250205151950.25268-17-alucerop@amd.com \
    --to=alucerop@amd.com \
    --cc=dan.j.williams@intel.com \
    --cc=dave.jiang@intel.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=edward.cree@amd.com \
    --cc=kuba@kernel.org \
    --cc=linux-cxl@vger.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;
as well as URLs for NNTP newsgroup(s).