All of lore.kernel.org
 help / color / mirror / Atom feed
From: sujithpshankar@gmail.com (Sujith Pandel)
Subject: [PATCH] NVMe:Expose model attribute in sysfs
Date: Fri, 4 Sep 2015 10:24:48 +0530	[thread overview]
Message-ID: <20150904045439.GA10044@localhost.localdomain> (raw)
In-Reply-To: <alpine.LNX.2.00.1509011717550.23840@localhost.lm.intel.com>

Hi Keith,
On Tue, Sep 01, 2015@12:25:48PM -0500, Keith Busch wrote:
> >> But if we're going to have more than one sysfs entry, we can manage
> >> this easier using attribute groups instead. There are lots of examples
> >> in the kernel for this, like scsi_sysfs.c or blk-mq-sysfs.c
> >
> > Since there are only two sysfs attributes here as of now,
> > I will be going with your former suggestion of using a new goto label.
> 
> Aww, that just forces the person who finds a need to add a third entry
> to implement a better design, or maybe he/she will use the same reasoning
> to shift that problem to the next person.
> 
> The "zero, one, infinity rule" puts 2 in the same catagory as infinity.
 
:) Appreciate the way you and David are making me (a kernel newbie) interested to contribute more on this one.
Hoping that this turns out to be my first kernel-patch!

Does this approach look okay?
If any new attributes are required, it can be added in the existing array or else,
have to create a new group and call sysfs_create_group()/sysfs_remove_group() on it.

Please let me know your thoughts on this.

--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -3151,6 +3151,44 @@ static ssize_t nvme_sysfs_reset(struct device *dev,
 }
 static DEVICE_ATTR(reset_controller, S_IWUSR, NULL, nvme_sysfs_reset);

+#define nvme_string_attr(name)                                             \
+static ssize_t  name##_show(struct device *dev,                            \
+                struct device_attribute *attr, char *buf)                  \
+{                                                                          \
+        struct nvme_dev *ndev = dev_get_drvdata(dev);                      \
+        return sprintf(buf, "%.*s\n", (int)sizeof(ndev->name), ndev->name);\
+}                                                                          \
+static DEVICE_ATTR(name, S_IRUGO, name##_show, NULL);
+
+/*
+ * Device string attributes.
+ * Can add serial and firmware_rev attributes here.
+ */
+nvme_string_attr(model);
+
+static struct attribute *nvme_dev_attrs[] = {
+        &dev_attr_reset_controller.attr,
+        &dev_attr_model.attr,
+        NULL
+};
+
+static struct attribute_group nvme_dev_attrs_group = {
+        .attrs = nvme_dev_attrs,
+};
+
+void nvme_remove_sysfs_files(struct device *dev)
+{
+        sysfs_remove_group(&dev->kobj, &nvme_dev_attrs_group);
+}
+
+/*
+ * Create sysfs entries for device attributes.
+ */
+int nvme_create_sysfs_files(struct device *dev)
+{
+        return sysfs_create_group(&dev->kobj, &nvme_dev_attrs_group);
+}
+
 static void nvme_async_probe(struct work_struct *work);
 static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
@@ -3197,7 +3235,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
        get_device(dev->device);
        dev_set_drvdata(dev->device, dev);

-       result = device_create_file(dev->device, &dev_attr_reset_controller);
+       result = nvme_create_sysfs_files(dev->device);
        if (result)
                goto put_dev;

@@ -3208,6 +3246,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
        return 0;

  put_dev:
+       nvme_remove_sysfs_files(dev->device);
        device_destroy(nvme_class, MKDEV(nvme_char_major, dev->instance));
        put_device(dev->device);
  release_pools:
@@ -3259,7 +3298,7 @@ static void nvme_remove(struct pci_dev *pdev)
        flush_work(&dev->probe_work);
        flush_work(&dev->reset_work);
        flush_work(&dev->scan_work);
-       device_remove_file(dev->device, &dev_attr_reset_controller);
+       nvme_remove_sysfs_files(dev->device);
        nvme_dev_remove(dev);
        nvme_dev_shutdown(dev);
        nvme_dev_remove_admin(dev);


Regards,
Sujith

  reply	other threads:[~2015-09-04  4:54 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-25  4:54 [PATCH] NVMe:Expose model attribute in sysfs Sujith Pandel
2015-08-25 15:38 ` Matthew Wilcox
2015-09-01 16:38   ` Sujith Pandel
2015-08-25 17:14 ` Keith Busch
2015-09-01 16:55   ` Sujith Pandel
2015-09-01 17:25     ` Keith Busch
2015-09-04  4:54       ` Sujith Pandel [this message]
2015-09-08 21:29         ` Keith Busch
2015-09-09  4:46           ` Sujith Pandel

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=20150904045439.GA10044@localhost.localdomain \
    --to=sujithpshankar@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.