linux-block.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Josh Triplett <josh@joshtriplett.org>
To: Jens Axboe <axboe@kernel.dk>, Josef Bacik <josef@toxicpanda.com>,
	linux-block@vger.kernel.org, nbd@other.debian.org
Subject: [PATCH] nbd: Support Kconfig configuration of nbds_max and max_part
Date: Fri, 9 Oct 2020 14:18:37 -0700	[thread overview]
Message-ID: <f0a162a4c7073e3ed47d981b86d845d3bb9aa955.1602278122.git.josh@joshtriplett.org> (raw)

This allows configuring them at compile time, not just via the kernel
command line or module parameters. This doesn't change the default
defaults, just makes them configurable.

Signed-off-by: Josh Triplett <josh@joshtriplett.org>
---
I'd like to be able to simplify the kernel command line I'm currently
using, by removing the nbd module parameters from it; this change would
make it possible to configure them when building the kernel, instead.

 drivers/block/Kconfig | 28 ++++++++++++++++++++++++++++
 drivers/block/nbd.c   | 10 ++++++----
 2 files changed, 34 insertions(+), 4 deletions(-)

diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig
index ecceaaa1a66f..7df9a9609aee 100644
--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -273,6 +273,34 @@ config BLK_DEV_NBD
 
 	  If unsure, say N.
 
+config BLK_DEV_NBD_DEFAULT_NBDS_MAX
+	int "Default maximum number of NBD devices"
+	depends on BLK_DEV_NBD
+	default 16
+	help
+	  The NBD driver can provide a set of unconfigured NBD devices (nbd0,
+	  nbd1, ...) by default, so that userspace can open and configure these
+	  devices via the ioctl interface. If you know that your userspace uses
+	  exclusively the new netlink interface, you can set this to 0 to
+	  reduce the amount of time needed for the NBD driver to initialize.
+	  You can also set this parameter at runtime using the nbds_max module
+	  parameter.
+
+config BLK_DEV_NBD_DEFAULT_MAX_PART
+	int "Default maximum number of NBD device partitions"
+	depends on BLK_DEV_NBD
+	default 16
+	help
+	  Once an NBD device is set up and opened, the kernel can probe it for
+	  partitions, and set up corresponding devices for each partition
+	  (nbd0p1, nbd0p2, ...). By default, the kernel probes for up to 16
+	  partitions per device. This also adds a bit of time to NBD device
+	  initialization. You can set this to a lower number if you know you'll
+	  never use more than that many partitions on an NBD device. If you
+	  exclusively use unpartitioned NBD devices, you can set this to 0 to
+	  skip partition probing entirely.  You can also set this parameter at
+	  runtime using the max_part module parameter.
+
 config BLK_DEV_SKD
 	tristate "STEC S1120 Block Driver"
 	depends on PCI
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index edf8b632e3d2..42851956c9ce 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -145,8 +145,8 @@ static struct dentry *nbd_dbg_dir;
 
 #define NBD_DEF_BLKSIZE 1024
 
-static unsigned int nbds_max = 16;
-static int max_part = 16;
+static unsigned int nbds_max = CONFIG_BLK_DEV_NBD_DEFAULT_NBDS_MAX;
+static int max_part = CONFIG_BLK_DEV_NBD_DEFAULT_MAX_PART;
 static int part_shift;
 
 static int nbd_dev_dbg_init(struct nbd_device *nbd);
@@ -2456,6 +2456,8 @@ MODULE_DESCRIPTION("Network Block Device");
 MODULE_LICENSE("GPL");
 
 module_param(nbds_max, int, 0444);
-MODULE_PARM_DESC(nbds_max, "number of network block devices to initialize (default: 16)");
+MODULE_PARM_DESC(nbds_max, "number of network block devices to initialize (default: "
+	__stringify(CONFIG_BLK_DEV_NBD_DEFAULT_NBDS_MAX) ")");
 module_param(max_part, int, 0444);
-MODULE_PARM_DESC(max_part, "number of partitions per device (default: 16)");
+MODULE_PARM_DESC(max_part, "number of partitions per device (default: "
+	__stringify(CONFIG_BLK_DEV_NBD_DEFAULT_MAX_PART) ")");
-- 
2.28.0


                 reply	other threads:[~2020-10-09 21:28 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=f0a162a4c7073e3ed47d981b86d845d3bb9aa955.1602278122.git.josh@joshtriplett.org \
    --to=josh@joshtriplett.org \
    --cc=axboe@kernel.dk \
    --cc=josef@toxicpanda.com \
    --cc=linux-block@vger.kernel.org \
    --cc=nbd@other.debian.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;
as well as URLs for NNTP newsgroup(s).