public inbox for linux-acpi@vger.kernel.org
 help / color / mirror / Atom feed
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
To: Jesse Barnes <jbarnes@virtuousgeek.org>, Len Brown <lenb@kernel.org>
Cc: Jiri Slaby <jirislaby@gmail.com>, Dan Williams <dcbw@redhat.com>,
	linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-acpi@vger.kernel.org, "H. Peter Anvin" <hpa@zytor.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Ingo Molnar <mingo@elte.hu>, Adam Belay <abelay@mit.edu>,
	Matthew Garrett <mjg@redhat.com>
Subject: [PATCH 3/5] x86: avoid PNP resources when allocating address space
Date: Wed, 08 Dec 2010 14:36:16 -0700	[thread overview]
Message-ID: <20101208213616.13026.77390.stgit@bob.kio> (raw)
In-Reply-To: <20101208213606.13026.47657.stgit@bob.kio>


Remove PNP resources from "available space" used by allocate_resource().
This would be better done by putting the PNP resources directly in the
resource maps (iomem_resource, etc), but there are some issues that
keep us from doing that yet.

This patch keeps us from handing out PNP device address space to other
callers of allocate_resource().

Reference: https://bugzilla.kernel.org/show_bug.cgi?id=23332
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=23542
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=23802
Reported-by: Matthew Garrett <mjg@redhat.com>
Reported-by: Dan Williams <dcbw@redhat.com>
Reported-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
---

 arch/x86/kernel/resource.c |   78 ++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 78 insertions(+), 0 deletions(-)


diff --git a/arch/x86/kernel/resource.c b/arch/x86/kernel/resource.c
index 5dd6473..1daee92 100644
--- a/arch/x86/kernel/resource.c
+++ b/arch/x86/kernel/resource.c
@@ -1,6 +1,80 @@
 #include <linux/ioport.h>
+#include <linux/pnp.h>
 #include <asm/e820.h>
 
+#ifdef CONFIG_PNP
+static bool resource_conflict(struct resource *res, resource_size_t start,
+			      resource_size_t end)
+{
+	/*
+	 * Return true if and only if "res" conflicts with the [start-end]
+	 * range.
+	 */
+	return res->start <= end && res->end >= start;
+}
+
+static void resource_split(struct resource *res, resource_size_t start,
+			   resource_size_t end, struct resource *low,
+			   struct resource *high)
+{
+	/*
+	 * If "res" conflicts with [start-end], split "res" into the
+	 * part below "start" (low) and the part above "end" (high),
+	 * either (or both) of which may be empty.
+	 *
+	 * If there's no conflict, return the entire "res" as "low".
+	 */
+	*low = *res;
+	low->start = res->start;
+	low->end = res->start - 1;	/* default to empty (size 0) */
+
+	*high = *res;
+	high->end = res->end;
+	high->start = res->end + 1;	/* default to empty (size 0) */
+
+	if (!resource_conflict(res, start, end)) {
+		low->end = res->end;
+		return;
+	}
+
+	if (res->start < start)
+		low->end = start - 1;
+
+	if (res->end > end)
+		high->start = end + 1;
+}
+
+static void pnp_remove_reservations(struct resource *avail)
+{
+	unsigned long type = resource_type(avail);
+	struct pnp_dev *dev;
+	int i;
+	struct resource *res, low, high;
+
+	/*
+	 * Clip the available region to avoid PNP devices.  The PNP
+	 * resources really should be in the resource map to begin with,
+	 * but there are still some issues preventing that.
+	 */
+	pnp_for_each_dev(dev) {
+		i = 0;
+		res = pnp_get_resource(dev, type, i++);
+		while (res) {
+			if (!(res->flags & IORESOURCE_WINDOW)) {
+				resource_split(avail, res->start, res->end,
+					       &low, &high);
+				if (resource_size(&low) > resource_size(&high))
+					*avail = low;
+				else
+					*avail = high;
+			}
+
+			res = pnp_get_resource(dev, type, i++);
+		}
+	}
+}
+#endif
+
 void arch_remove_reservations(struct resource *avail)
 {
 	/*
@@ -11,4 +85,8 @@ void arch_remove_reservations(struct resource *avail)
 		if (avail->start < BIOS_END)
 			avail->start = BIOS_END;
 	}
+
+#ifdef CONFIG_PNP
+	pnp_remove_reservations(avail);
+#endif
 }


  parent reply	other threads:[~2010-12-08 21:36 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-08 21:36 [PATCH 1/5] resources: add arch hook for preventing allocation in reserved areas Bjorn Helgaas
2010-12-08 21:36 ` [PATCH 2/5] x86: avoid BIOS area when allocating address space Bjorn Helgaas
2010-12-08 21:36 ` Bjorn Helgaas [this message]
2010-12-08 21:36 ` [PATCH 4/5] PNP: add framework for platform PNP quirks Bjorn Helgaas
2010-12-08 21:36 ` [PATCH 5/5] PNP: HP nx6325 fixup: reserve unreported resources Bjorn Helgaas
2010-12-12  3:30   ` Linus Torvalds
2010-12-12  5:23     ` Dave Airlie
2010-12-12  6:17     ` Bjorn Helgaas
2010-12-14 20:34       ` Linus Torvalds
2010-12-14 20:44         ` Linus Torvalds
2010-12-14 23:57           ` Bjorn Helgaas
2010-12-15  6:02           ` Bjorn Helgaas
2010-12-15  6:26         ` Bjorn Helgaas
2010-12-15  7:03           ` Linus Torvalds
2010-12-15 18:18             ` Bjorn Helgaas
2010-12-15 18:27               ` H. Peter Anvin
2010-12-15 19:21               ` Linus Torvalds
2010-12-08 21:37 ` [PATCH 0/5] resources: add arch hook for preventing allocation in reserved areas Bjorn Helgaas
2010-12-10 20:30 ` [PATCH 1/5] " Jesse Barnes
2010-12-10 20:36   ` Jesse Barnes
2010-12-10 21:07     ` Bjorn Helgaas
2010-12-11  1:37       ` Jesse Barnes
2010-12-12  3:34         ` Linus Torvalds
2010-12-12  4:16           ` Jesse Barnes
2010-12-12 13:20             ` Rafael J. Wysocki
2010-12-13  5:43               ` Bjorn Helgaas
2010-12-13 13:47                 ` Ingo Molnar
2010-12-15  0:09                   ` Bjorn Helgaas

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=20101208213616.13026.77390.stgit@bob.kio \
    --to=bjorn.helgaas@hp.com \
    --cc=abelay@mit.edu \
    --cc=dcbw@redhat.com \
    --cc=hpa@zytor.com \
    --cc=jbarnes@virtuousgeek.org \
    --cc=jirislaby@gmail.com \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=mjg@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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