summaryrefslogtreecommitdiff
path: root/recipes/linux/linux-omap-2.6.28/fix-irq33.diff
blob: 709f042ab7780a0394dcee9d74a070c3cc4a5ee3 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
From: "Nathan Monson" <nmonson@gmail.com>
To: "linux-omap@vger.kernel.org List" <linux-omap@vger.kernel.org>
Subject: Re: omapfb: help from userspace
Cc: "TK, Pratheesh Gangadhar" <pratheesh@ti.com>

On Wed, Oct 8, 2008 at 11:36 AM, Nathan Monson <nmonson@gmail.com> wrote:
> "Felipe Contreras" <felipe.contreras@gmail.com> writes:
>> irq -33, desc: c0335cf8, depth: 0, count: 0, unhandled: 0
>
> On the BeagleBoard list, Pratheesh Gangadhar said that mapping I/O
> regions as Strongly Ordered suppresses this problem:
> http://groups.google.com/group/beagleboard/browse_thread/thread/23e1c95b4bfb09b5/70d12dca569ca503?show_docid=70d12dca569ca503

Pratheesh helped me make a patch against the latest linux-omap git to
try this.

With this patch, my IRQ -33 problems with the DSP have disappeared.
Before, I would end up in IRQ -33 loop after 10 invocations of the DSP
Bridge 'ping.out' utility.  I just finished running it 50,000 times
without error.

As stated before, this patch is just a workaround for testing
purposes, not a fix.  Who knows what performance side effects it
has...

---
diff --git a/arch/arm/include/asm/mach/map.h b/arch/arm/include/asm/mach/map.h
index 9eb936e..5cb4f5f 100644
--- a/arch/arm/include/asm/mach/map.h
+++ b/arch/arm/include/asm/mach/map.h
@@ -25,6 +25,7 @@ struct map_desc {
 #define MT_HIGH_VECTORS		8
 #define MT_MEMORY		9
 #define MT_ROM			10
+#define MT_MEMORY_SO		11

 #define MT_NONSHARED_DEVICE	MT_DEVICE_NONSHARED
 #define MT_IXP2000_DEVICE	MT_DEVICE_IXP2000
diff --git a/arch/arm/mach-omap2/io.c b/arch/arm/mach-omap2/io.c
index adbe21f..c11c0e8 100644
--- a/arch/arm/mach-omap2/io.c
+++ b/arch/arm/mach-omap2/io.c
@@ -119,13 +119,13 @@ static struct map_desc omap34xx_io_desc[] __initdata = {
 		.virtual	= L3_34XX_VIRT,
 		.pfn		= __phys_to_pfn(L3_34XX_PHYS),
 		.length		= L3_34XX_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 	{
 		.virtual	= L4_34XX_VIRT,
 		.pfn		= __phys_to_pfn(L4_34XX_PHYS),
 		.length		= L4_34XX_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 	{
 		.virtual	= L4_WK_34XX_VIRT,
@@ -137,19 +137,19 @@ static struct map_desc omap34xx_io_desc[] __initdata = {
 		.virtual	= OMAP34XX_GPMC_VIRT,
 		.pfn		= __phys_to_pfn(OMAP34XX_GPMC_PHYS),
 		.length		= OMAP34XX_GPMC_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 	{
 		.virtual	= OMAP343X_SMS_VIRT,
 		.pfn		= __phys_to_pfn(OMAP343X_SMS_PHYS),
 		.length		= OMAP343X_SMS_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 	{
 		.virtual	= OMAP343X_SDRC_VIRT,
 		.pfn		= __phys_to_pfn(OMAP343X_SDRC_PHYS),
 		.length		= OMAP343X_SDRC_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 	{
 		.virtual	= L4_PER_34XX_VIRT,
@@ -161,7 +161,7 @@ static struct map_desc omap34xx_io_desc[] __initdata = {
 		.virtual	= L4_EMU_34XX_VIRT,
 		.pfn		= __phys_to_pfn(L4_EMU_34XX_PHYS),
 		.length		= L4_EMU_34XX_SIZE,
-		.type		= MT_DEVICE
+		.type		= MT_MEMORY_SO
 	},
 };
 #endif
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index a713e40..d5f25ad 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -245,6 +245,10 @@ static struct mem_type mem_types[] = {
 		.prot_sect = PMD_TYPE_SECT,
 		.domain    = DOMAIN_KERNEL,
 	},
+	[MT_MEMORY_SO] = {
+		.prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE | PMD_SECT_UNCACHED,
+		.domain    = DOMAIN_KERNEL,
+	},
 };

 const struct mem_type *get_mem_type(unsigned int type)
--
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html