mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Emilio G. Cota" <cota@braap.org>
To: Greg KH <greg@kroah.com>
Cc: Martyn Welch <martyn.welch@ge.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Juan David Gonzalez Cobas <david.cobas@gmail.com>
Subject: [PATCH 04/30] staging/vme: allow non-dynamic allocation of bus numbers
Date: Fri, 22 Oct 2010 02:36:26 -0400	[thread overview]
Message-ID: <1287729412-24356-5-git-send-email-cota@braap.org> (raw)
In-Reply-To: <1287729412-24356-1-git-send-email-cota@braap.org>

From: Emilio G. Cota <cota@braap.org>

In a configuration with several bridges, each bridge is
assigned a certain bus number depending on the order in which
vme_register_bridge is called. This can complicate multi-bridge
installations because the eventual bus numbers will depend
on the order the bridges were loaded.

The appended allows bridges to register with a bus number of
their choice, while keeping the previous 'first come, first
served' behaviour as the default.

Signed-off-by: Emilio G. Cota <cota@braap.org>
---
 drivers/staging/vme/bridges/vme_ca91cx42.c |    1 +
 drivers/staging/vme/bridges/vme_tsi148.c   |    1 +
 drivers/staging/vme/vme.c                  |   46 +++++++++++++++++++++-------
 drivers/staging/vme/vme.h                  |    1 +
 4 files changed, 38 insertions(+), 11 deletions(-)

diff --git a/drivers/staging/vme/bridges/vme_ca91cx42.c b/drivers/staging/vme/bridges/vme_ca91cx42.c
index 1f2089f..14e10a6 100644
--- a/drivers/staging/vme/bridges/vme_ca91cx42.c
+++ b/drivers/staging/vme/bridges/vme_ca91cx42.c
@@ -1600,6 +1600,7 @@ static int ca91cx42_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	ca91cx42_bridge->driver_priv = ca91cx42_device;
+	ca91cx42_bridge->num = -1;
 
 	/* Enable the device */
 	retval = pci_enable_device(pdev);
diff --git a/drivers/staging/vme/bridges/vme_tsi148.c b/drivers/staging/vme/bridges/vme_tsi148.c
index 1cba1fa..e341122 100644
--- a/drivers/staging/vme/bridges/vme_tsi148.c
+++ b/drivers/staging/vme/bridges/vme_tsi148.c
@@ -2247,6 +2247,7 @@ static int tsi148_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 
 	tsi148_bridge->driver_priv = tsi148_device;
+	tsi148_bridge->num = -1;
 
 	/* Enable the device */
 	retval = pci_enable_device(pdev);
diff --git a/drivers/staging/vme/vme.c b/drivers/staging/vme/vme.c
index 47e9d5c..9fa0609 100644
--- a/drivers/staging/vme/vme.c
+++ b/drivers/staging/vme/vme.c
@@ -1307,20 +1307,42 @@ EXPORT_SYMBOL(vme_slot_get);
 
 /* - Bridge Registration --------------------------------------------------- */
 
-static int vme_alloc_bus_num(void)
+/* call with vme_bus_num_mtx held */
+static int __vme_alloc_bus_num(int *bus)
 {
-	int i;
-
-	mutex_lock(&vme_bus_num_mtx);
-	for (i = 0; i < sizeof(vme_bus_numbers) * 8; i++) {
-		if (((vme_bus_numbers >> i) & 0x1) == 0) {
-			vme_bus_numbers |= (0x1 << i);
-			break;
+	int index;
+
+	if (*bus == -1) {
+		/* try to find a free bus number */
+		for (index = 0; index < VME_MAX_BRIDGES; index++) {
+			if (~vme_bus_numbers & 1 << index) {
+				*bus = index;
+				break;
+			}
+		}
+		if (index == VME_MAX_BRIDGES) {
+			pr_warn("vme: No bus numbers left\n");
+			return -ENODEV;
+		}
+	} else {
+		/* check if the given bus number is already in use */
+		if (vme_bus_numbers & (1 << *bus)) {
+			pr_warn("vme: bus number %d already in use\n", *bus);
+			return -EBUSY;
 		}
 	}
-	mutex_unlock(&vme_bus_num_mtx);
+	vme_bus_numbers |= 1 << *bus;
+	return 0;
+}
+
+static int vme_alloc_bus_num(int *bus)
+{
+	int ret;
 
-	return i;
+	mutex_lock(&vme_bus_num_mtx);
+	ret = __vme_alloc_bus_num(bus);
+	mutex_unlock(&vme_bus_num_mtx);
+	return ret;
 }
 
 static void vme_free_bus_num(int bus)
@@ -1336,7 +1358,9 @@ int vme_register_bridge(struct vme_bridge *bridge)
 	int retval;
 	int i;
 
-	bridge->num = vme_alloc_bus_num();
+	retval = vme_alloc_bus_num(&bridge->num);
+	if (retval)
+		return retval;
 
 	/* This creates 32 vme "slot" devices. This equates to a slot for each
 	 * ID available in a system conforming to the ANSI/VITA 1-1994
diff --git a/drivers/staging/vme/vme.h b/drivers/staging/vme/vme.h
index 48768ca..5cdfa52 100644
--- a/drivers/staging/vme/vme.h
+++ b/drivers/staging/vme/vme.h
@@ -88,6 +88,7 @@ struct vme_resource {
 
 extern struct bus_type vme_bus_type;
 
+#define VME_MAX_BRIDGES		32
 #define VME_SLOT_CURRENT	-1
 #define VME_SLOT_ALL		-2
 
-- 
1.7.1


  parent reply	other threads:[~2010-10-22  7:23 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-22  6:36 [-next] staging/vme: various fixes + new driver model for VME Emilio G. Cota
2010-10-22  6:36 ` [PATCH 01/30] staging/vme: style: convert '&(foo)' to '&foo' Emilio G. Cota
2010-10-22  6:36 ` [PATCH 02/30] staging/vme_user: return the appropriate error code when module_init fails Emilio G. Cota
2010-10-22  6:36 ` [PATCH 03/30] staging/vme_user: remove unreachable line Emilio G. Cota
2010-10-22  6:36 ` Emilio G. Cota [this message]
2010-10-22  6:36 ` [PATCH 05/30] staging/vme: fix bogus clearing of the bus number in vme_free_bus_num Emilio G. Cota
2010-10-22  6:36 ` [PATCH 06/30] staging/vme/tsi148: use list_for_each_safe when deleting resources in .remove Emilio G. Cota
2010-10-22  6:36 ` [PATCH 07/30] staging/vme/tsi148: remove double freeing of the IRQ " Emilio G. Cota
2010-10-22  6:36 ` [PATCH 08/30] staging/vme/tsi148: fix warning in free_irq Emilio G. Cota
2010-10-22  6:36 ` [PATCH 09/30] staging/vme: fill in struct device's .release, even if it's a NOOP Emilio G. Cota
2010-10-22  6:36 ` [PATCH 10/30] staging/vme/tsi148: remove unreachable line Emilio G. Cota
2010-10-22  6:36 ` [PATCH 11/30] staging/vme/tsi148: declare static functions as such Emilio G. Cota
2010-10-22  6:36 ` [PATCH 12/30] staging/vme/ca91cx42: " Emilio G. Cota
2010-10-22  6:36 ` [PATCH 13/30] staging/vme_user: declare private variables as static Emilio G. Cota
2010-10-22  6:36 ` [PATCH 14/30] staging/vme_user: use an unsigned int for counting the number of kparams Emilio G. Cota
2010-10-22  6:36 ` [PATCH 15/30] staging/vme_user: remove __iomem marking from kern_buf and derivates Emilio G. Cota
2010-10-22  6:36 ` [PATCH 16/30] staging/vme_user: mark user-space buffers with __user Emilio G. Cota
2010-10-22  6:36 ` [PATCH 17/30] staging/vme: mark struct vme_master_resource's base address pointer as __iomem Emilio G. Cota
2010-10-22  6:36 ` [PATCH 18/30] staging/vme/tsi148: mark the registers' " Emilio G. Cota
2010-10-22  6:36 ` [PATCH 19/30] staging/vme/ca91cx42: " Emilio G. Cota
2010-10-22  6:36 ` [PATCH 20/30] staging/vme: trivial: rename vme_bus_num_mtx to vme_buses_lock Emilio G. Cota
2010-10-22  6:36 ` [PATCH 21/30] staging/vme: keep a list of registered buses (bridges) Emilio G. Cota
2010-10-22  6:36 ` [PATCH 22/30] staging/vme/vme_user: use __dev{init,exit} for .probe and .remove Emilio G. Cota
2010-10-22  6:36 ` [PATCH 23/30] staging/vme_user: fix usage of the slave resources after they've been freed Emilio G. Cota
2010-10-22  6:36 ` [PATCH 24/30] staging/vme_user: remove unnecessary call to vme_slave_set Emilio G. Cota
2010-10-22  6:36 ` [PATCH 25/30] staging/vme_user: add missing calls to vme_master_free calls in .remove Emilio G. Cota
2010-10-22  6:36 ` [PATCH 26/30] staging/vme: move all contents of vme_bridge.h to vme.h Emilio G. Cota
2010-10-22  6:36 ` [PATCH 27/30] staging/vme: rework the bus model Emilio G. Cota
2010-10-22 23:27   ` Emilio G. Cota
2010-10-25 11:24     ` Martyn Welch
2010-10-26  1:02       ` Emilio G. Cota
2010-10-27  9:16         ` Martyn Welch
2010-10-27 15:40           ` Emilio G. Cota
2010-10-22  6:36 ` [PATCH 28/30] staging/vme: convert vme_* users to vme_*_ng Emilio G. Cota
2010-10-22  6:36 ` [PATCH 29/30] staging/vme: remove unused vme_* functions and related code Emilio G. Cota
2010-10-22  6:36 ` [PATCH 30/30] staging/vme: remove _ng suffixes Emilio G. Cota
2010-10-22  9:26 ` [-next] staging/vme: various fixes + new driver model for VME Martyn Welch
2010-10-22 23:03   ` Emilio G. Cota
2010-10-22 10:08 ` Martyn Welch
2010-10-22 23:04   ` Emilio G. Cota
2010-10-22 13:56 ` Greg KH
2010-10-22 22:56   ` Emilio G. Cota
2010-10-26  1:10 [re-send][-next] " Emilio G. Cota
2010-10-26  1:10 ` [PATCH 04/30] staging/vme: allow non-dynamic allocation of bus numbers Emilio G. Cota
2010-10-27  9:41   ` Martyn Welch
2010-10-27 15:13     ` Emilio G. Cota
2010-11-04  3:40       ` Emilio G. Cota
2010-11-05 16:29         ` Martyn Welch

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=1287729412-24356-5-git-send-email-cota@braap.org \
    --to=cota@braap.org \
    --cc=david.cobas@gmail.com \
    --cc=greg@kroah.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=martyn.welch@ge.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

all inboxes | Powered by JetHome®