linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Yinghai Lu <yinghai@kernel.org>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Tony Luck <tony.luck@intel.com>,
	David Miller <davem@davemloft.net>, x86 <x86@kernel.org>,
	Dominik Brodowski <linux@dominikbrodowski.net>,
	Andrew Morton <akpm@linux-foundation.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arch@vger.kernel.org
Subject: Re: [PATCH -v12 01/15] resources: Split out __allocate_resource()
Date: Tue, 26 Jun 2012 13:33:40 -0700	[thread overview]
Message-ID: <CAE9FiQVBPoEinWC__9DF-oi8T-78YJRn95J=xXdL2GzNoKoCaQ@mail.gmail.com> (raw)
In-Reply-To: <CA+55aFzi7tHg=3k8Fmga8XmxFbn6tZLLV97vt67Xx_WtckQo6g@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 928 bytes --]

On Tue, Jun 26, 2012 at 12:01 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> On Tue, Jun 26, 2012 at 11:53 AM, Yinghai Lu <yinghai@kernel.org> wrote:
>> It will take bool lock, so we could use it in other functions that
>> hold the resource lock already.
>
> This is too damn ugly.
>
> These kinds of "conditionally take lock" things are always just bugs
> waiting to happen. Don't do it.
>
> Just make the rule be that the caller of the __allocate_resource
> helper has to hold the lock. Sure, that means that you need to then
> use split reallocate_resource() into a helper function (ie a static
> __reallocate_resource() that needs to have the lock taken by the
> caller too), but dammit, that's definitely the right thing to do
> anyway.
>
> These kinds of "bool lock" crap things have to die. They are *wrong*.
> They are a sign of bad locking rules.

You are right, please check updated one.

Thanks

Yinghai

[-- Attachment #2: probe_resource_1.patch --]
[-- Type: application/octet-stream, Size: 3516 bytes --]

Subject: [PATCH] resources: Split out __allocate_resource()

It will out hold lock, so we could use it in other functions that
hold the resource lock already.

-v2: according to Linus, using "bool lock" as parameter
     aka "conditionally take lock" is *wrong*.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>

---
 kernel/resource.c |   41 +++++++++++++++++++++++++++++++++--------
 1 file changed, 33 insertions(+), 8 deletions(-)

Index: linux-2.6/kernel/resource.c
===================================================================
--- linux-2.6.orig/kernel/resource.c
+++ linux-2.6/kernel/resource.c
@@ -470,7 +470,7 @@ static int find_resource(struct resource
  * @newsize: new size of the resource descriptor
  * @constraint: the size and alignment constraints to be met.
  */
-int reallocate_resource(struct resource *root, struct resource *old,
+static int __reallocate_resource(struct resource *root, struct resource *old,
 			resource_size_t newsize,
 			struct resource_constraint  *constraint)
 {
@@ -478,7 +478,6 @@ int reallocate_resource(struct resource
 	struct resource new = *old;
 	struct resource *conflict;
 
-	write_lock(&resource_lock);
 
 	if ((err = __find_resource(root, old, &new, newsize, constraint)))
 		goto out;
@@ -504,11 +503,20 @@ int reallocate_resource(struct resource
 		BUG_ON(conflict);
 	}
 out:
-	write_unlock(&resource_lock);
 	return err;
 }
+int reallocate_resource(struct resource *root, struct resource *old,
+			resource_size_t newsize,
+			struct resource_constraint  *constraint)
+{
+	int ret;
 
+	write_lock(&resource_lock);
+	ret = __reallocate_resource(root, old, newsize, constraint);
+	write_unlock(&resource_lock);
 
+	return ret;
+}
 /**
  * allocate_resource - allocate empty slot in the resource tree given range & alignment.
  * 	The resource will be reallocated with a new size if it was already allocated
@@ -521,7 +529,7 @@ out:
  * @alignf: alignment function, optional, called if not NULL
  * @alignf_data: arbitrary data to pass to the @alignf function
  */
-int allocate_resource(struct resource *root, struct resource *new,
+static int __allocate_resource(struct resource *root, struct resource *new,
 		      resource_size_t size, resource_size_t min,
 		      resource_size_t max, resource_size_t align,
 		      resource_size_t (*alignf)(void *,
@@ -542,19 +550,36 @@ int allocate_resource(struct resource *r
 	constraint.alignf = alignf;
 	constraint.alignf_data = alignf_data;
 
-	if ( new->parent ) {
+	if (new->parent) {
 		/* resource is already allocated, try reallocating with
 		   the new constraints */
-		return reallocate_resource(root, new, size, &constraint);
+		return __reallocate_resource(root, new, size, &constraint);
 	}
 
-	write_lock(&resource_lock);
 	err = find_resource(root, new, size, &constraint);
 	if (err >= 0 && __request_resource(root, new))
 		err = -EBUSY;
-	write_unlock(&resource_lock);
+
 	return err;
 }
+int allocate_resource(struct resource *root, struct resource *new,
+		      resource_size_t size, resource_size_t min,
+		      resource_size_t max, resource_size_t align,
+		      resource_size_t (*alignf)(void *,
+						const struct resource *,
+						resource_size_t,
+						resource_size_t),
+		      void *alignf_data)
+{
+	int ret;
+
+	write_lock(&resource_lock);
+	ret = __allocate_resource(root, new, size, min, max, align,
+				   alignf, alignf_data);
+	write_unlock(&resource_lock);
+
+	return ret;
+}
 
 EXPORT_SYMBOL(allocate_resource);
 

  parent reply	other threads:[~2012-06-26 20:33 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-26 18:53 [PATCH -v12 00/15] PCI: allocate pci bus num range for unassigned bridge busn Yinghai Lu
2012-06-26 18:53 ` [PATCH -v12 01/15] resources: Split out __allocate_resource() Yinghai Lu
2012-06-26 18:53   ` Yinghai Lu
2012-06-26 19:01   ` Linus Torvalds
2012-06-26 19:01     ` Linus Torvalds
2012-06-26 20:33     ` Yinghai Lu [this message]
2012-06-26 20:33       ` Yinghai Lu
2012-06-26 20:43       ` Linus Torvalds
2012-06-26 20:43         ` Linus Torvalds
2012-06-26 18:53 ` [PATCH -v12 02/15] resources: Add probe_resource() Yinghai Lu
2012-06-26 18:53   ` Yinghai Lu
2012-06-26 22:07   ` Yinghai Lu
2012-08-28 16:09     ` Yinghai Lu
2012-08-28 16:09       ` Yinghai Lu
2012-08-28 17:05       ` Linus Torvalds
2012-08-29  0:10         ` Linus Torvalds
2012-08-29  0:10           ` Linus Torvalds
2012-08-29 10:14           ` Ram Pai
2012-08-29 16:02             ` Yinghai Lu
2012-08-29 16:02               ` Yinghai Lu
2012-08-29 15:57           ` Yinghai Lu
2012-08-29 15:57             ` Yinghai Lu
2012-08-29 17:36             ` Yinghai Lu
2012-08-29 17:36               ` Yinghai Lu
2012-08-31  0:40               ` Bjorn Helgaas
2012-08-31  0:40                 ` Bjorn Helgaas
2012-06-26 18:53 ` [PATCH -v12 03/15] resources: Replace registered resource in tree Yinghai Lu
2012-06-26 18:53   ` Yinghai Lu
2012-06-26 18:53 ` [PATCH -v12 04/15] PCI: Add pci_bus_extend/shrink_top() Yinghai Lu
2012-06-26 18:53   ` Yinghai Lu
2012-06-26 18:53 ` [PATCH -v12 05/15] PCI: Probe safe range that we can use for unassigned bridge Yinghai Lu
2012-06-26 18:53   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 06/15] PCI: Add pci_bus_replace_busn_res() Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 07/15] PCI: Allocate bus range instead of use max blindly Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 08/15] PCI: Strict checking of valid range for bridge Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 09/15] PCI: Kill pci_fixup_parent_subordinate_busnr() Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 10/15] PCI: Seperate child bus scanning to two passes overall Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 11/15] pcmcia: Remove workaround for fixing pci parent bus subordinate Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 12/15] PCI: Double checking setting for bus register and bus struct Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 13/15] PCI, pciehp: Remove not needed bus number range checking Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 14/15] PCI: More strict checking of valid range for bridge Yinghai Lu
2012-06-26 18:54 ` [PATCH -v12 15/15] PCI: Don't shrink too much for hotplug bridge Yinghai Lu
2012-06-26 18:54   ` Yinghai Lu

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='CAE9FiQVBPoEinWC__9DF-oi8T-78YJRn95J=xXdL2GzNoKoCaQ@mail.gmail.com' \
    --to=yinghai@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=benh@kernel.crashing.org \
    --cc=bhelgaas@google.com \
    --cc=davem@davemloft.net \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux@dominikbrodowski.net \
    --cc=tony.luck@intel.com \
    --cc=torvalds@linux-foundation.org \
    --cc=x86@kernel.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).