linux-kernel - [PATCH v1 5/6] resource: Set type of "reserve=" user-specified resources

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20171205004044.12535.85511.stgit@bhelgaas-glaptop.roam.corp.google.com>
Date:   Mon, 04 Dec 2017 18:40:44 -0600
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     unlisted-recipients:; (no To-header on input)
Cc:     Jason Cooper <jason@...edaemon.net>,
        Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>,
        Marc Zyngier <marc.zyngier@....com>,
        Benjamin Herrenschmidt <benh@...nel.crashing.org>,
        linux-kernel@...r.kernel.org, Ralf Baechle <ralf@...ux-mips.org>,
        Paul Mackerras <paulus@...ba.org>,
        Michael Ellerman <mpe@...erman.id.au>,
        Ivan Kokshaysky <ink@...assic.park.msu.ru>,
        Thomas Gleixner <tglx@...utronix.de>,
        Yinghai Lu <yinghai@...nel.org>,
        Matt Turner <mattst88@...il.com>,
        Richard Henderson <rth@...ddle.net>
Subject: [PATCH v1 5/6] resource: Set type of "reserve=" user-specified
 resources

From: Bjorn Helgaas <bhelgaas@...gle.com>

When we reserve regions because the user specified a "reserve=" parameter,
set the resource type to either IORESOURCE_IO (for regions below 0x10000)
or IORESOURCE_MEM.  The test for 0x10000 is just a heuristic; obviously
there can be memory below 0x10000 as well.

Improve documentation of the "reserve=" parameter.

Signed-off-by: Bjorn Helgaas <bhelgaas@...gle.com>
---
 Documentation/admin-guide/kernel-parameters.txt |    6 +++++-
 kernel/resource.c                               |   24 +++++++++++++++++------
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 6571fbfdb2a1..78cdf6a637fc 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -3675,7 +3675,11 @@
 			[KNL, SMP] Set scheduler's default relax_domain_level.
 			See Documentation/cgroup-v1/cpusets.txt.
 
-	reserve=	[KNL,BUGS] Force the kernel to ignore some iomem area
+	reserve=	[KNL,BUGS] Force kernel to ignore I/O ports or memory
+			Format: <base1>,<size1>[,<base2>,<size2>,...]
+			Reserve I/O ports or memory so the kernel won't use
+			them.  If <base> is less than 0x10000, the region
+			is assumed to be I/O ports; otherwise it is memory.
 
 	reservetop=	[X86-32]
 			Format: nn[KMG]
diff --git a/kernel/resource.c b/kernel/resource.c
index 54ba6de3757c..ba3252f7c319 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -1478,7 +1478,7 @@ void __devm_release_region(struct device *dev, struct resource *parent,
 EXPORT_SYMBOL(__devm_release_region);
 
 /*
- * Called from init/main.c to reserve IO ports.
+ * Reserve I/O ports or memory based on "reserve=" kernel parameter.
  */
 #define MAXRESERVE 4
 static int __init reserve_setup(char *str)
@@ -1489,26 +1489,38 @@ static int __init reserve_setup(char *str)
 	for (;;) {
 		unsigned int io_start, io_num;
 		int x = reserved;
+		struct resource *parent;
 
-		if (get_option (&str, &io_start) != 2)
+		if (get_option(&str, &io_start) != 2)
 			break;
-		if (get_option (&str, &io_num)   == 0)
+		if (get_option(&str, &io_num) == 0)
 			break;
 		if (x < MAXRESERVE) {
 			struct resource *res = reserve + x;
+
+			/*
+			 * If the region starts below 0x10000, we assume it's
+			 * I/O port space; otherwise assume it's memory.
+			 */
+			if (io_start < 0x10000) {
+				res->flags = IORESOURCE_IO;
+				parent = &ioport_resource;
+			} else {
+				res->flags = IORESOURCE_MEM;
+				parent = &iomem_resource;
+			}
 			res->name = "reserved";
 			res->start = io_start;
 			res->end = io_start + io_num - 1;
-			res->flags = IORESOURCE_BUSY;
+			res->flags |= IORESOURCE_BUSY;
 			res->desc = IORES_DESC_NONE;
 			res->child = NULL;
-			if (request_resource(res->start >= 0x10000 ? &iomem_resource : &ioport_resource, res) == 0)
+			if (request_resource(parent, res) == 0)
 				reserved = x+1;
 		}
 	}
 	return 1;
 }
-
 __setup("reserve=", reserve_setup);
 
 /*