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
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
|
/*
* linux/arch/arm/mach-sa1100/pci-nanoengine.c
*
* PCI functions for BSE nanoEngine PCI
*
* Copyright (C) 2010 Marcelo Roberto Jimenez <mroberto@cpti.cetuc.puc-rio.br>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#include <linux/kernel.h>
#include <linux/irq.h>
#include <linux/pci.h>
#include <linux/spinlock.h>
#include <asm/mach/pci.h>
#include <asm/mach-types.h>
#include <mach/nanoengine.h>
#include <mach/hardware.h>
static DEFINE_SPINLOCK(nano_lock);
static int nanoengine_get_pci_address(struct pci_bus *bus,
unsigned int devfn, int where, unsigned long *address)
{
int ret = PCIBIOS_DEVICE_NOT_FOUND;
unsigned int busnr = bus->number;
*address = NANO_PCI_CONFIG_SPACE_VIRT +
((bus->number << 16) | (devfn << 8) | (where & ~3));
ret = (busnr > 255 || devfn > 255 || where > 255) ?
PCIBIOS_DEVICE_NOT_FOUND : PCIBIOS_SUCCESSFUL;
return ret;
}
static int nanoengine_read_config(struct pci_bus *bus, unsigned int devfn, int where,
int size, u32 *val)
{
int ret;
unsigned long address;
unsigned long flags;
u32 v;
/* nanoEngine PCI bridge does not return -1 for a non-existing
* device. We must fake the answer. We know that the only valid
* device is device zero at bus 0, which is the network chip. */
if (bus->number != 0 || (devfn >> 3) != 0) {
v = -1;
nanoengine_get_pci_address(bus, devfn, where, &address);
goto exit_function;
}
spin_lock_irqsave(&nano_lock, flags);
ret = nanoengine_get_pci_address(bus, devfn, where, &address);
if (ret != PCIBIOS_SUCCESSFUL)
return ret;
v = __raw_readl(address);
spin_unlock_irqrestore(&nano_lock, flags);
v >>= ((where & 3) * 8);
v &= (unsigned long)(-1) >> ((4 - size) * 8);
exit_function:
*val = v;
return PCIBIOS_SUCCESSFUL;
}
static int nanoengine_write_config(struct pci_bus *bus, unsigned int devfn, int where,
int size, u32 val)
{
int ret;
unsigned long address;
unsigned long flags;
unsigned shift;
u32 v;
shift = (where & 3) * 8;
spin_lock_irqsave(&nano_lock, flags);
ret = nanoengine_get_pci_address(bus, devfn, where, &address);
if (ret != PCIBIOS_SUCCESSFUL)
return ret;
v = __raw_readl(address);
switch (size) {
case 1:
v &= ~(0xFF << shift);
v |= val << shift;
break;
case 2:
v &= ~(0xFFFF << shift);
v |= val << shift;
break;
case 4:
v = val;
break;
}
__raw_writel(v, address);
spin_unlock_irqrestore(&nano_lock, flags);
return PCIBIOS_SUCCESSFUL;
}
static struct pci_ops pci_nano_ops = {
.read = nanoengine_read_config,
.write = nanoengine_write_config,
};
static int __init pci_nanoengine_map_irq(const struct pci_dev *dev, u8 slot,
u8 pin)
{
return NANOENGINE_IRQ_GPIO_PCI;
}
struct pci_bus * __init pci_nanoengine_scan_bus(int nr, struct pci_sys_data *sys)
{
return pci_scan_root_bus(NULL, sys->busnr, &pci_nano_ops, sys,
&sys->resources);
}
static struct resource pci_io_ports = {
.name = "PCI IO",
.start = 0x400,
.end = 0x7FF,
.flags = IORESOURCE_IO,
};
static struct resource pci_non_prefetchable_memory = {
.name = "PCI non-prefetchable",
.start = NANO_PCI_MEM_RW_PHYS,
/* nanoEngine documentation says there is a 1 Megabyte window here,
* but PCI reports just 128 + 8 kbytes. */
.end = NANO_PCI_MEM_RW_PHYS + NANO_PCI_MEM_RW_SIZE - 1,
/* .end = NANO_PCI_MEM_RW_PHYS + SZ_128K + SZ_8K - 1,*/
.flags = IORESOURCE_MEM,
};
/*
* nanoEngine PCI reports 1 Megabyte of prefetchable memory, but it
* overlaps with previously defined memory.
*
* Here is what happens:
*
# dmesg
...
pci 0000:00:00.0: [8086:1209] type 0 class 0x000200
pci 0000:00:00.0: reg 10: [mem 0x00021000-0x00021fff]
pci 0000:00:00.0: reg 14: [io 0x0000-0x003f]
pci 0000:00:00.0: reg 18: [mem 0x00000000-0x0001ffff]
pci 0000:00:00.0: reg 30: [mem 0x00000000-0x000fffff pref]
pci 0000:00:00.0: supports D1 D2
pci 0000:00:00.0: PME# supported from D0 D1 D2 D3hot
pci 0000:00:00.0: PME# disabled
PCI: bus0: Fast back to back transfers enabled
pci 0000:00:00.0: BAR 6: can't assign mem pref (size 0x100000)
pci 0000:00:00.0: BAR 2: assigned [mem 0x18600000-0x1861ffff]
pci 0000:00:00.0: BAR 2: set to [mem 0x18600000-0x1861ffff] (PCI address [0x0-0x1ffff])
pci 0000:00:00.0: BAR 0: assigned [mem 0x18620000-0x18620fff]
pci 0000:00:00.0: BAR 0: set to [mem 0x18620000-0x18620fff] (PCI address [0x20000-0x20fff])
pci 0000:00:00.0: BAR 1: assigned [io 0x0400-0x043f]
pci 0000:00:00.0: BAR 1: set to [io 0x0400-0x043f] (PCI address [0x0-0x3f])
*
* On the other hand, if we do not request the prefetchable memory resource,
* linux will alloc it first and the two non-prefetchable memory areas that
* are our real interest will not be mapped. So we choose to map it to an
* unused area. It gets recognized as expansion ROM, but becomes disabled.
*
* Here is what happens then:
*
# dmesg
...
pci 0000:00:00.0: [8086:1209] type 0 class 0x000200
pci 0000:00:00.0: reg 10: [mem 0x00021000-0x00021fff]
pci 0000:00:00.0: reg 14: [io 0x0000-0x003f]
pci 0000:00:00.0: reg 18: [mem 0x00000000-0x0001ffff]
pci 0000:00:00.0: reg 30: [mem 0x00000000-0x000fffff pref]
pci 0000:00:00.0: supports D1 D2
pci 0000:00:00.0: PME# supported from D0 D1 D2 D3hot
pci 0000:00:00.0: PME# disabled
PCI: bus0: Fast back to back transfers enabled
pci 0000:00:00.0: BAR 6: assigned [mem 0x78000000-0x780fffff pref]
pci 0000:00:00.0: BAR 2: assigned [mem 0x18600000-0x1861ffff]
pci 0000:00:00.0: BAR 2: set to [mem 0x18600000-0x1861ffff] (PCI address [0x0-0x1ffff])
pci 0000:00:00.0: BAR 0: assigned [mem 0x18620000-0x18620fff]
pci 0000:00:00.0: BAR 0: set to [mem 0x18620000-0x18620fff] (PCI address [0x20000-0x20fff])
pci 0000:00:00.0: BAR 1: assigned [io 0x0400-0x043f]
pci 0000:00:00.0: BAR 1: set to [io 0x0400-0x043f] (PCI address [0x0-0x3f])
# lspci -vv -s 0000:00:00.0
00:00.0 Class 0200: Device 8086:1209 (rev 09)
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR+ <PERR+ INTx-
Latency: 0 (2000ns min, 14000ns max), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 0
Region 0: Memory at 18620000 (32-bit, non-prefetchable) [size=4K]
Region 1: I/O ports at 0400 [size=64]
Region 2: [virtual] Memory at 18600000 (32-bit, non-prefetchable) [size=128K]
[virtual] Expansion ROM at 78000000 [disabled] [size=1M]
Capabilities: [dc] Power Management version 2
Flags: PMEClk- DSI+ D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=2 PME-
Kernel driver in use: e100
Kernel modules: e100
*
*/
static struct resource pci_prefetchable_memory = {
.name = "PCI prefetchable",
.start = 0x78000000,
.end = 0x78000000 + NANO_PCI_MEM_RW_SIZE - 1,
.flags = IORESOURCE_MEM | IORESOURCE_PREFETCH,
};
static int __init pci_nanoengine_setup_resources(struct pci_sys_data *sys)
{
if (request_resource(&ioport_resource, &pci_io_ports)) {
printk(KERN_ERR "PCI: unable to allocate io port region\n");
return -EBUSY;
}
if (request_resource(&iomem_resource, &pci_non_prefetchable_memory)) {
release_resource(&pci_io_ports);
printk(KERN_ERR "PCI: unable to allocate non prefetchable\n");
return -EBUSY;
}
if (request_resource(&iomem_resource, &pci_prefetchable_memory)) {
release_resource(&pci_io_ports);
release_resource(&pci_non_prefetchable_memory);
printk(KERN_ERR "PCI: unable to allocate prefetchable\n");
return -EBUSY;
}
pci_add_resource(&sys->resources, &pci_io_ports);
pci_add_resource(&sys->resources, &pci_non_prefetchable_memory);
pci_add_resource(&sys->resources, &pci_prefetchable_memory);
return 1;
}
int __init pci_nanoengine_setup(int nr, struct pci_sys_data *sys)
{
int ret = 0;
pcibios_min_io = 0;
pcibios_min_mem = 0;
if (nr == 0) {
sys->mem_offset = NANO_PCI_MEM_RW_PHYS;
sys->io_offset = 0x400;
ret = pci_nanoengine_setup_resources(sys);
/* Enable alternate memory bus master mode, see
* "Intel StrongARM SA1110 Developer's Manual",
* section 10.8, "Alternate Memory Bus Master Mode". */
GPDR = (GPDR & ~GPIO_MBREQ) | GPIO_MBGNT;
GAFR |= GPIO_MBGNT | GPIO_MBREQ;
TUCR |= TUCR_MBGPIO;
}
return ret;
}
static struct hw_pci nanoengine_pci __initdata = {
.map_irq = pci_nanoengine_map_irq,
.nr_controllers = 1,
.scan = pci_nanoengine_scan_bus,
.setup = pci_nanoengine_setup,
};
static int __init nanoengine_pci_init(void)
{
if (machine_is_nanoengine())
pci_common_init(&nanoengine_pci);
return 0;
}
subsys_initcall(nanoengine_pci_init);
|