public inbox for linux-cxl@vger.kernel.org
 help / color / mirror / Atom feed
From: Dave Jiang <dave.jiang@intel.com>
To: Jonathan Cameron <Jonathan.Cameron@Huawei.com>
Cc: linux-cxl@vger.kernel.org, dan.j.williams@intel.com,
	ira.weiny@intel.com, vishal.l.verma@intel.com,
	alison.schofield@intel.com, dave@stgolabs.net, jgg@nvidia.com,
	shiju.jose@huawei.com
Subject: Re: [RFC PATCH 04/13] cxl: Add Get Supported Features command for kernel usage
Date: Fri, 27 Sep 2024 09:22:01 -0700	[thread overview]
Message-ID: <806fbf8c-c3c3-41ed-b4b7-cb0a74d408df@intel.com> (raw)
In-Reply-To: <20240726185010.0000295c@Huawei.com>



On 7/26/24 10:50 AM, Jonathan Cameron wrote:
> On Thu, 18 Jul 2024 14:32:22 -0700
> Dave Jiang <dave.jiang@intel.com> wrote:
> 
>> CXL spec r3.1 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.
>>
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> 
> Already a fairly well reviewed implementation of this in 
> 
> https://lore.kernel.org/linux-cxl/20240726160556.2079-5-shiju.jose@huawei.com/
> 
> Though it wasn't focused on providing userspace access to it and doesn't handle
> large feature lists which would be eventually needed.
> 
> I'd have preferred to see review on that rather than another version
> of the same functionality.
> 
> Note the QEMU testing used for that set will be useable here too
> and is now upstream.
> 
> A few comments inline.
> 
> Jonathan
> 
> 
>> +int cxl_get_supported_features(struct cxl_mailbox *cxl_mbox)
>> +{
>> +	int remain_feats, max_size, max_feats, start, rc;
>> +	int feat_size = sizeof(struct cxl_feat_entry);
>> +	struct cxl_mbox_get_sup_feats_out *mbox_out;
>> +	struct cxl_mbox_get_sup_feats_in mbox_in;
>> +	int hdr_size = sizeof(*mbox_out);
>> +	struct cxl_mbox_cmd mbox_cmd;
>> +	struct cxl_mem_command *cmd;
>> +	void *ptr;
>> +
>> +	/* Get supported features is optional, need to check */
>> +	cmd = cxl_mem_find_command(CXL_MBOX_OP_GET_SUPPORTED_FEATURES);
>> +	if (!cmd)
>> +		return -EOPNOTSUPP;
>> +	if (!test_bit(cmd->info.id, cxl_mbox->enabled_cmds))
>> +		return -EOPNOTSUPP;
>> +
>> +	rc = cxl_get_supported_features_count(cxl_mbox);
>> +	if (rc)
>> +		return rc;
>> +
>> +	struct cxl_feat_entry *entries __free(kvfree) =
>> +		kvmalloc(cxl_mbox->num_features * feat_size, GFP_KERNEL);
>> +
>> +	if (!entries)
>> +		return -ENOMEM;
>> +
>> +	cxl_mbox->entries = no_free_ptr(entries);
> 
> Hmm. Not srue I'd bother with __free(kvfree) for one line that doesn't
> do the free.
> 
>> +	rc = devm_add_action_or_reset(cxl_mbox->host, cxl_free_features,
>> +				      cxl_mbox->entries);
>> +	if (rc)
>> +		return rc;
>> +
>> +	max_size = cxl_mbox->payload_size - hdr_size;
>> +	/* max feat entries that can fit in mailbox max payload size */
>> +	max_feats = max_size / feat_size;
>> +	ptr = &cxl_mbox->entries[0];
>> +
>> +	mbox_out = kvmalloc(cxl_mbox->payload_size, GFP_KERNEL);
> Freed?
> 
>> +	if (!mbox_out)
>> +		return -ENOMEM;
>> +
>> +	start = 0;
>> +	remain_feats = cxl_mbox->num_features;
>> +	do {
>> +		int retrieved, alloc_size, copy_feats;
>> +
>> +		if (remain_feats > max_feats) {
>> +			alloc_size = sizeof(*mbox_out) + max_feats * feat_size;
>> +			remain_feats = remain_feats - max_feats;
>> +			copy_feats = max_feats;
>> +		} else {
>> +			alloc_size = sizeof(*mbox_out) + remain_feats * feat_size;
>> +			copy_feats = remain_feats;
>> +			remain_feats = 0;
>> +		}
>> +
>> +		memset(&mbox_in, 0, sizeof(mbox_in));
>> +		mbox_in.count = alloc_size;
>> +		mbox_in.start_idx = start;
>> +		memset(mbox_out, 0, alloc_size);
>> +		mbox_cmd = (struct cxl_mbox_cmd) {
>> +			.opcode = CXL_MBOX_OP_GET_SUPPORTED_FEATURES,
>> +			.size_in = sizeof(mbox_in),
>> +			.payload_in = &mbox_in,
>> +			.size_out = alloc_size,
>> +			.payload_out = mbox_out,
>> +			.min_out = hdr_size,
>> +		};
>> +		rc = cxl_internal_send_cmd(cxl_mbox, &mbox_cmd);
>> +		if (rc < 0)
>> +			return rc;
>> +
>> +		if (mbox_cmd.size_out <= hdr_size) {
>> +			rc = -ENXIO;
>> +			goto err;
>> +		}
>> +
>> +		/*
>> +		 * Make sure retrieved out buffer is multiple of feature
>> +		 * entries.
>> +		 */
>> +		retrieved = mbox_cmd.size_out - hdr_size;
>> +		if (retrieved % sizeof(struct cxl_feat_entry)) {
>> +			rc = -ENXIO;
>> +			goto err;
>> +		}
>> +
>> +		/*
>> +		 * If the reported output entries * defined entry size !=
>> +		 * retrieved output bytes, then the output package is incorrect.
>> +		 */
>> +		if (mbox_out->num_entries * feat_size != retrieved) {
>> +			rc = -ENXIO;
>> +			goto err;
>> +		}
>> +
>> +		memcpy(ptr, mbox_out->ents, retrieved);
>> +		ptr += retrieved;
>> +		/*
>> +		 * If the number of output entries is less than expected, add the
>> +		 * remaining entries to the next batch.
>> +		 */
>> +		remain_feats += copy_feats - mbox_out->num_entries;
>> +		start += mbox_out->num_entries;
>> +	} while (remain_feats);
>> +
>> +	return 0;
>> +
>> +err:
>> +	cxl_mbox->num_features = 0;
>> +	return rc;
>> +}
>> +EXPORT_SYMBOL_NS_GPL(cxl_get_supported_features, CXL);
>> +
>>  /**
>>   * cxl_enumerate_cmds() - Enumerate commands for a device.
>>   * @mds: The driver data for the operation
>> diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h
>> index efdf833f2c51..4d3690aa2f3b 100644
>> --- a/drivers/cxl/cxlmem.h
>> +++ b/drivers/cxl/cxlmem.h
> 
>> +struct cxl_mbox_get_sup_feats_out {
>> +	__le16 num_entries;
>> +	__le16 supported_feats;
>> +	__le32 reserved;
>> +	struct cxl_feat_entry ents[] __counted_by(le32_to_cpu(supported_feats));
> 
> I didn't know __counted_by worked with function calls. Interesting.

Actually I need to use __counted_by_le() instead as someone pointed out via kbot reporting.

DJ

> 
>> +} __packed;
>> +
> 
>>  #endif /* __CXL_MEM_H__ */
>> diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
>> index 7e26da706921..6a00238446f9 100644
>> --- a/drivers/cxl/pci.c
>> +++ b/drivers/cxl/pci.c
>> @@ -872,6 +872,10 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>>  	if (rc)
>>  		return rc;
>>  
>> +	rc = cxl_get_supported_features(&cxlds->cxl_mbox);
> 
> I would separate no features from an error in trying to enumerate them.
> 
>> +	if (rc)
>> +		dev_dbg(&pdev->dev, "No features enumerated.\n");
>> +
>>  	rc = cxl_set_timestamp(mds);
>>  	if (rc)
>>  		return rc;
>> diff --git a/include/linux/cxl/mailbox.h b/include/linux/cxl/mailbox.h
>> index 2380b22d7a12..570864239b8f 100644
>> --- a/include/linux/cxl/mailbox.h
>> +++ b/include/linux/cxl/mailbox.h
>> @@ -53,6 +53,7 @@ struct cxl_mbox_cmd {
>>   * @mbox_mutex: mutex protects device mailbox and firmware
>>   * @mbox_wait: rcuwait for mailbox
>>   * @mbox_send: @dev specific transport for transmitting mailbox commands
>> + * @features: number of supported features
> 
> missing docs for entries.
> 
>>   */
>>  struct cxl_mailbox {
>>  	struct device *host;
>> @@ -63,6 +64,8 @@ struct cxl_mailbox {
>>  	struct mutex mbox_mutex; /* lock to protect mailbox context */
>>  	struct rcuwait mbox_wait;
>>  	int (*mbox_send)(struct cxl_mailbox *cxl_mbox, struct cxl_mbox_cmd *cmd);
>> +	int num_features;
>> +	struct cxl_feat_entry *entries;
>>  };
>>  
>>  #endif
>> diff --git a/include/uapi/linux/cxl_mem.h b/include/uapi/linux/cxl_mem.h
>> index c6c0fe27495d..bd2535962f70 100644
>> --- a/include/uapi/linux/cxl_mem.h
>> +++ b/include/uapi/linux/cxl_mem.h
>> @@ -50,6 +50,7 @@
>>  	___C(GET_LOG_CAPS, "Get Log Capabilities"),			  \
>>  	___C(CLEAR_LOG, "Clear Log"),					  \
>>  	___C(GET_SUP_LOG_SUBLIST, "Get Supported Logs Sub-List"),	  \
>> +	___C(GET_SUPPORTED_FEATURES, "Get Supported Features"),		  \
>>  	___C(MAX, "invalid / last command")
>>  
>>  #define ___C(a, b) CXL_MEM_COMMAND_ID_##a
> 


  reply	other threads:[~2024-09-27 16:22 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-18 21:32 [RFC PATCH 0/13] fwctl/cxl: Add CXL feature commands support via fwctl Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 01/13] cxl: Move mailbox related bits to the same context Dave Jiang
2024-07-19  6:31   ` Alejandro Lucero Palau
2024-07-19 15:47     ` Dave Jiang
2024-07-19 16:07       ` Alejandro Lucero Palau
2024-07-26 17:28   ` Jonathan Cameron
2024-07-18 21:32 ` [RFC PATCH 02/13] cxl: Fix comment regarding cxl_query_cmd() return data Dave Jiang
2024-07-26 17:29   ` Jonathan Cameron
2024-09-24 23:41     ` Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 03/13] cxl: Refactor user ioctl command path from mds to mailbox Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 04/13] cxl: Add Get Supported Features command for kernel usage Dave Jiang
2024-07-26 17:50   ` Jonathan Cameron
2024-09-27 16:22     ` Dave Jiang [this message]
2024-08-21 16:05   ` Shiju Jose
2024-08-21 18:10     ` Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 05/13] cxl/test: Add Get Supported Features mailbox command support Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 06/13] cxl: Add Get Feature " Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 07/13] cxl: Add Set " Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 08/13] fwctl/cxl: Add driver for CXL mailbox for handling CXL features commands Dave Jiang
2024-07-22 15:31   ` Jason Gunthorpe
2024-07-22 21:42     ` Dave Jiang
2024-07-26 18:02   ` Jonathan Cameron
2024-09-24 23:44     ` Dave Jiang
2024-09-26 17:37       ` Jason Gunthorpe
2024-09-26 20:26         ` Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 09/13] fwctl/cxl: Add support for get driver information Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 10/13] fwctl/cxl: Add support for fwctl RPC command to enable CXL feature commands Dave Jiang
2024-07-22 15:29   ` Jason Gunthorpe
2024-07-22 22:52     ` Dave Jiang
2024-07-29 11:29   ` Jonathan Cameron
2024-11-13 15:41     ` Dave Jiang
2024-11-19 11:41       ` Jonathan Cameron
2024-07-18 21:32 ` [RFC PATCH 11/13] fwctl/cxl: Add query commands software command for ->fw_rpc() Dave Jiang
2024-07-22 15:24   ` Jason Gunthorpe
2024-07-22 23:23     ` Dave Jiang
2024-08-08 12:56       ` Jason Gunthorpe
2024-07-29 11:48   ` Jonathan Cameron
2024-07-18 21:32 ` [RFC PATCH 12/13] cxl/test: Add Get Feature support to cxl_test Dave Jiang
2024-07-18 21:32 ` [RFC PATCH 13/13] cxl/test: Add Set " Dave Jiang
2024-07-19  6:23 ` [RFC PATCH 0/13] fwctl/cxl: Add CXL feature commands support via fwctl Alejandro Lucero Palau
2024-07-19 15:48   ` Dave Jiang
2024-07-22 15:32 ` Jason Gunthorpe
2024-07-29 12:05 ` Jonathan Cameron
2024-08-06 16:44   ` Dave Jiang
2024-11-13  0:17   ` Dave Jiang
2024-11-19 11:43     ` Jonathan Cameron
2024-11-19 15:58       ` Dave Jiang

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=806fbf8c-c3c3-41ed-b4b7-cb0a74d408df@intel.com \
    --to=dave.jiang@intel.com \
    --cc=Jonathan.Cameron@Huawei.com \
    --cc=alison.schofield@intel.com \
    --cc=dan.j.williams@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