linux_dsm_epyc7002/arch/sh/drivers/pci/pcie-sh7786.c

580 lines
14 KiB
C
Raw Normal View History

/*
* Low-Level PCI Express Support for the SH7786
*
* Copyright (C) 2009 - 2011 Paul Mundt
*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*/
#define pr_fmt(fmt) "PCI: " fmt
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/io.h>
#include <linux/async.h>
#include <linux/delay.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 15:04:11 +07:00
#include <linux/slab.h>
#include <linux/clk.h>
#include <linux/sh_clk.h>
#include <linux/sh_intc.h>
#include "pcie-sh7786.h"
#include <asm/sizes.h>
struct sh7786_pcie_port {
struct pci_channel *hose;
struct clk *fclk, phy_clk;
unsigned int index;
int endpoint;
int link;
};
static struct sh7786_pcie_port *sh7786_pcie_ports;
static unsigned int nr_ports;
static struct sh7786_pcie_hwops {
int (*core_init)(void);
async_func_ptr *port_init_hw;
} *sh7786_pcie_hwops;
static struct resource sh7786_pci0_resources[] = {
{
.name = "PCIe0 IO",
.start = 0xfd000000,
.end = 0xfd000000 + SZ_8M - 1,
.flags = IORESOURCE_IO,
}, {
.name = "PCIe0 MEM 0",
.start = 0xc0000000,
.end = 0xc0000000 + SZ_512M - 1,
.flags = IORESOURCE_MEM | IORESOURCE_MEM_32BIT,
}, {
.name = "PCIe0 MEM 1",
.start = 0x10000000,
.end = 0x10000000 + SZ_64M - 1,
.flags = IORESOURCE_MEM,
}, {
.name = "PCIe0 MEM 2",
.start = 0xfe100000,
.end = 0xfe100000 + SZ_1M - 1,
.flags = IORESOURCE_MEM,
},
};
static struct resource sh7786_pci1_resources[] = {
{
.name = "PCIe1 IO",
.start = 0xfd800000,
.end = 0xfd800000 + SZ_8M - 1,
.flags = IORESOURCE_IO,
}, {
.name = "PCIe1 MEM 0",
.start = 0xa0000000,
.end = 0xa0000000 + SZ_512M - 1,
.flags = IORESOURCE_MEM | IORESOURCE_MEM_32BIT,
}, {
.name = "PCIe1 MEM 1",
.start = 0x30000000,
.end = 0x30000000 + SZ_256M - 1,
.flags = IORESOURCE_MEM | IORESOURCE_MEM_32BIT,
}, {
.name = "PCIe1 MEM 2",
.start = 0xfe300000,
.end = 0xfe300000 + SZ_1M - 1,
.flags = IORESOURCE_MEM,
},
};
static struct resource sh7786_pci2_resources[] = {
{
.name = "PCIe2 IO",
.start = 0xfc800000,
.end = 0xfc800000 + SZ_4M - 1,
.flags = IORESOURCE_IO,
}, {
.name = "PCIe2 MEM 0",
.start = 0x80000000,
.end = 0x80000000 + SZ_512M - 1,
.flags = IORESOURCE_MEM | IORESOURCE_MEM_32BIT,
}, {
.name = "PCIe2 MEM 1",
.start = 0x20000000,
.end = 0x20000000 + SZ_256M - 1,
.flags = IORESOURCE_MEM | IORESOURCE_MEM_32BIT,
}, {
.name = "PCIe2 MEM 2",
.start = 0xfcd00000,
.end = 0xfcd00000 + SZ_1M - 1,
.flags = IORESOURCE_MEM,
},
};
extern struct pci_ops sh7786_pci_ops;
#define DEFINE_CONTROLLER(start, idx) \
{ \
.pci_ops = &sh7786_pci_ops, \
.resources = sh7786_pci##idx##_resources, \
.nr_resources = ARRAY_SIZE(sh7786_pci##idx##_resources), \
.reg_base = start, \
.mem_offset = 0, \
.io_offset = 0, \
}
static struct pci_channel sh7786_pci_channels[] = {
DEFINE_CONTROLLER(0xfe000000, 0),
DEFINE_CONTROLLER(0xfe200000, 1),
DEFINE_CONTROLLER(0xfcc00000, 2),
};
static struct clk fixed_pciexclkp = {
.rate = 100000000, /* 100 MHz reference clock */
};
static void __devinit sh7786_pci_fixup(struct pci_dev *dev)
{
/*
* Prevent enumeration of root complex resources.
*/
if (pci_is_root_bus(dev->bus) && dev->devfn == 0) {
int i;
for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
dev->resource[i].start = 0;
dev->resource[i].end = 0;
dev->resource[i].flags = 0;
}
}
}
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_RENESAS, PCI_DEVICE_ID_RENESAS_SH7786,
sh7786_pci_fixup);
static int __init phy_wait_for_ack(struct pci_channel *chan)
{
unsigned int timeout = 100;
while (timeout--) {
if (pci_read_reg(chan, SH4A_PCIEPHYADRR) & (1 << BITS_ACK))
return 0;
udelay(100);
}
return -ETIMEDOUT;
}
static int __init pci_wait_for_irq(struct pci_channel *chan, unsigned int mask)
{
unsigned int timeout = 100;
while (timeout--) {
if ((pci_read_reg(chan, SH4A_PCIEINTR) & mask) == mask)
return 0;
udelay(100);
}
return -ETIMEDOUT;
}
static void __init phy_write_reg(struct pci_channel *chan, unsigned int addr,
unsigned int lane, unsigned int data)
{
unsigned long phyaddr;
phyaddr = (1 << BITS_CMD) + ((lane & 0xf) << BITS_LANE) +
((addr & 0xff) << BITS_ADR);
/* Set write data */
pci_write_reg(chan, data, SH4A_PCIEPHYDOUTR);
pci_write_reg(chan, phyaddr, SH4A_PCIEPHYADRR);
phy_wait_for_ack(chan);
/* Clear command */
pci_write_reg(chan, 0, SH4A_PCIEPHYDOUTR);
pci_write_reg(chan, 0, SH4A_PCIEPHYADRR);
phy_wait_for_ack(chan);
}
static int __init pcie_clk_init(struct sh7786_pcie_port *port)
{
struct pci_channel *chan = port->hose;
struct clk *clk;
char fclk_name[16];
int ret;
/*
* First register the fixed clock
*/
ret = clk_register(&fixed_pciexclkp);
if (unlikely(ret != 0))
return ret;
/*
* Grab the port's function clock, which the PHY clock depends
* on. clock lookups don't help us much at this point, since no
* dev_id is available this early. Lame.
*/
snprintf(fclk_name, sizeof(fclk_name), "pcie%d_fck", port->index);
port->fclk = clk_get(NULL, fclk_name);
if (IS_ERR(port->fclk)) {
ret = PTR_ERR(port->fclk);
goto err_fclk;
}
clk_enable(port->fclk);
/*
* And now, set up the PHY clock
*/
clk = &port->phy_clk;
memset(clk, 0, sizeof(struct clk));
clk->parent = &fixed_pciexclkp;
clk->enable_reg = (void __iomem *)(chan->reg_base + SH4A_PCIEPHYCTLR);
clk->enable_bit = BITS_CKE;
ret = sh_clk_mstp32_register(clk, 1);
if (unlikely(ret < 0))
goto err_phy;
return 0;
err_phy:
clk_disable(port->fclk);
clk_put(port->fclk);
err_fclk:
clk_unregister(&fixed_pciexclkp);
return ret;
}
static int __init phy_init(struct sh7786_pcie_port *port)
{
struct pci_channel *chan = port->hose;
unsigned int timeout = 100;
clk_enable(&port->phy_clk);
/* Initialize the phy */
phy_write_reg(chan, 0x60, 0xf, 0x004b008b);
phy_write_reg(chan, 0x61, 0xf, 0x00007b41);
phy_write_reg(chan, 0x64, 0xf, 0x00ff4f00);
phy_write_reg(chan, 0x65, 0xf, 0x09070907);
phy_write_reg(chan, 0x66, 0xf, 0x00000010);
phy_write_reg(chan, 0x74, 0xf, 0x0007001c);
phy_write_reg(chan, 0x79, 0xf, 0x01fc000d);
phy_write_reg(chan, 0xb0, 0xf, 0x00000610);
/* Deassert Standby */
phy_write_reg(chan, 0x67, 0x1, 0x00000400);
/* Disable clock */
clk_disable(&port->phy_clk);
while (timeout--) {
if (pci_read_reg(chan, SH4A_PCIEPHYSR))
return 0;
udelay(100);
}
return -ETIMEDOUT;
}
static void __init pcie_reset(struct sh7786_pcie_port *port)
{
struct pci_channel *chan = port->hose;
pci_write_reg(chan, 1, SH4A_PCIESRSTR);
pci_write_reg(chan, 0, SH4A_PCIETCTLR);
pci_write_reg(chan, 0, SH4A_PCIESRSTR);
pci_write_reg(chan, 0, SH4A_PCIETXVC0SR);
}
static int __init pcie_init(struct sh7786_pcie_port *port)
{
struct pci_channel *chan = port->hose;
unsigned int data;
phys_addr_t memphys;
size_t memsize;
int ret, i, win;
/* Begin initialization */
pcie_reset(port);
/*
* Initial header for port config space is type 1, set the device
* class to match. Hardware takes care of propagating the IDSETR
* settings, so there is no need to bother with a quirk.
*/
pci_write_reg(chan, PCI_CLASS_BRIDGE_PCI << 16, SH4A_PCIEIDSETR1);
/* Initialize default capabilities. */
data = pci_read_reg(chan, SH4A_PCIEEXPCAP0);
data &= ~(PCI_EXP_FLAGS_TYPE << 16);
if (port->endpoint)
data |= PCI_EXP_TYPE_ENDPOINT << 20;
else
data |= PCI_EXP_TYPE_ROOT_PORT << 20;
data |= PCI_CAP_ID_EXP;
pci_write_reg(chan, data, SH4A_PCIEEXPCAP0);
/* Enable data link layer active state reporting */
pci_write_reg(chan, PCI_EXP_LNKCAP_DLLLARC, SH4A_PCIEEXPCAP3);
/* Enable extended sync and ASPM L0s support */
data = pci_read_reg(chan, SH4A_PCIEEXPCAP4);
data &= ~PCI_EXP_LNKCTL_ASPMC;
data |= PCI_EXP_LNKCTL_ES | 1;
pci_write_reg(chan, data, SH4A_PCIEEXPCAP4);
/* Write out the physical slot number */
data = pci_read_reg(chan, SH4A_PCIEEXPCAP5);
data &= ~PCI_EXP_SLTCAP_PSN;
data |= (port->index + 1) << 19;
pci_write_reg(chan, data, SH4A_PCIEEXPCAP5);
/* Set the completion timer timeout to the maximum 32ms. */
data = pci_read_reg(chan, SH4A_PCIETLCTLR);
data &= ~0x3f00;
data |= 0x32 << 8;
pci_write_reg(chan, data, SH4A_PCIETLCTLR);
/*
* Set fast training sequences to the maximum 255,
* and enable MAC data scrambling.
*/
data = pci_read_reg(chan, SH4A_PCIEMACCTLR);
data &= ~PCIEMACCTLR_SCR_DIS;
data |= (0xff << 16);
pci_write_reg(chan, data, SH4A_PCIEMACCTLR);
memphys = __pa(memory_start);
memsize = roundup_pow_of_two(memory_end - memory_start);
/*
* If there's more than 512MB of memory, we need to roll over to
* LAR1/LAMR1.
*/
if (memsize > SZ_512M) {
pci_write_reg(chan, memphys + SZ_512M, SH4A_PCIELAR1);
pci_write_reg(chan, ((memsize - SZ_512M) - SZ_256) | 1,
SH4A_PCIELAMR1);
memsize = SZ_512M;
} else {
/*
* Otherwise just zero it out and disable it.
*/
pci_write_reg(chan, 0, SH4A_PCIELAR1);
pci_write_reg(chan, 0, SH4A_PCIELAMR1);
}
/*
* LAR0/LAMR0 covers up to the first 512MB, which is enough to
* cover all of lowmem on most platforms.
*/
pci_write_reg(chan, memphys, SH4A_PCIELAR0);
pci_write_reg(chan, (memsize - SZ_256) | 1, SH4A_PCIELAMR0);
/* Finish initialization */
data = pci_read_reg(chan, SH4A_PCIETCTLR);
data |= 0x1;
pci_write_reg(chan, data, SH4A_PCIETCTLR);
/* Let things settle down a bit.. */
mdelay(100);
/* Enable DL_Active Interrupt generation */
data = pci_read_reg(chan, SH4A_PCIEDLINTENR);
data |= PCIEDLINTENR_DLL_ACT_ENABLE;
pci_write_reg(chan, data, SH4A_PCIEDLINTENR);
/* Disable MAC data scrambling. */
data = pci_read_reg(chan, SH4A_PCIEMACCTLR);
data |= PCIEMACCTLR_SCR_DIS | (0xff << 16);
pci_write_reg(chan, data, SH4A_PCIEMACCTLR);
/*
* This will timeout if we don't have a link, but we permit the
* port to register anyways in order to support hotplug on future
* hardware.
*/
ret = pci_wait_for_irq(chan, MASK_INT_TX_CTRL);
data = pci_read_reg(chan, SH4A_PCIEPCICONF1);
data &= ~(PCI_STATUS_DEVSEL_MASK << 16);
data |= PCI_COMMAND_IO | PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
(PCI_STATUS_CAP_LIST | PCI_STATUS_DEVSEL_FAST) << 16;
pci_write_reg(chan, data, SH4A_PCIEPCICONF1);
pci_write_reg(chan, 0x80888000, SH4A_PCIETXVC0DCTLR);
pci_write_reg(chan, 0x00222000, SH4A_PCIERXVC0DCTLR);
wmb();
if (ret == 0) {
data = pci_read_reg(chan, SH4A_PCIEMACSR);
printk(KERN_NOTICE "PCI: PCIe#%d x%d link detected\n",
port->index, (data >> 20) & 0x3f);
} else
printk(KERN_NOTICE "PCI: PCIe#%d link down\n",
port->index);
for (i = win = 0; i < chan->nr_resources; i++) {
struct resource *res = chan->resources + i;
resource_size_t size;
u32 mask;
/*
* We can't use the 32-bit mode windows in legacy 29-bit
* mode, so just skip them entirely.
*/
if ((res->flags & IORESOURCE_MEM_32BIT) && __in_29bit_mode())
continue;
pci_write_reg(chan, 0x00000000, SH4A_PCIEPTCTLR(win));
/*
* The PAMR mask is calculated in units of 256kB, which
* keeps things pretty simple.
*/
size = resource_size(res);
mask = (roundup_pow_of_two(size) / SZ_256K) - 1;
pci_write_reg(chan, mask << 18, SH4A_PCIEPAMR(win));
pci_write_reg(chan, upper_32_bits(res->start),
SH4A_PCIEPARH(win));
pci_write_reg(chan, lower_32_bits(res->start),
SH4A_PCIEPARL(win));
mask = MASK_PARE;
if (res->flags & IORESOURCE_IO)
mask |= MASK_SPC;
pci_write_reg(chan, mask, SH4A_PCIEPTCTLR(win));
win++;
}
return 0;
}
PCI: Make the struct pci_dev * argument of pci_fixup_irqs const. Aside of the usual motivation for constification, this function has a history of being abused a hook for interrupt and other fixups so I turned this function const ages ago in the MIPS code but it should be done treewide. Due to function pointer passing in varous places a few other functions had to be constified as well. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> To: Anton Vorontsov <avorontsov@mvista.com> To: Chris Metcalf <cmetcalf@tilera.com> To: Colin Cross <ccross@android.com> Acked-by: "David S. Miller" <davem@davemloft.net> To: Eric Miao <eric.y.miao@gmail.com> To: Erik Gilling <konkers@android.com> Acked-by: Guan Xuetao <gxt@mprc.pku.edu.cn> To: "H. Peter Anvin" <hpa@zytor.com> To: Imre Kaloz <kaloz@openwrt.org> To: Ingo Molnar <mingo@redhat.com> To: Ivan Kokshaysky <ink@jurassic.park.msu.ru> To: Jesse Barnes <jbarnes@virtuousgeek.org> To: Krzysztof Halasa <khc@pm.waw.pl> To: Lennert Buytenhek <kernel@wantstofly.org> To: Matt Turner <mattst88@gmail.com> To: Nicolas Pitre <nico@fluxnic.net> To: Olof Johansson <olof@lixom.net> Acked-by: Paul Mundt <lethal@linux-sh.org> To: Richard Henderson <rth@twiddle.net> To: Russell King <linux@arm.linux.org.uk> To: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: linux-alpha@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org Cc: linux-mips@linux-mips.org Cc: linux-pci@vger.kernel.org Cc: linux-sh@vger.kernel.org Cc: linux-tegra@vger.kernel.org Cc: sparclinux@vger.kernel.org Cc: x86@kernel.org Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2011-06-10 21:30:21 +07:00
int __init pcibios_map_platform_irq(const struct pci_dev *pdev, u8 slot, u8 pin)
{
return evt2irq(0xae0);
}
static int __init sh7786_pcie_core_init(void)
{
/* Return the number of ports */
return test_mode_pin(MODE_PIN12) ? 3 : 2;
}
static void __init sh7786_pcie_init_hw(void *data, async_cookie_t cookie)
{
struct sh7786_pcie_port *port = data;
int ret;
/*
* Check if we are configured in endpoint or root complex mode,
* this is a fixed pin setting that applies to all PCIe ports.
*/
port->endpoint = test_mode_pin(MODE_PIN11);
/*
* Setup clocks, needed both for PHY and PCIe registers.
*/
ret = pcie_clk_init(port);
if (unlikely(ret < 0)) {
pr_err("clock initialization failed for port#%d\n",
port->index);
return;
}
ret = phy_init(port);
if (unlikely(ret < 0)) {
pr_err("phy initialization failed for port#%d\n",
port->index);
return;
}
ret = pcie_init(port);
if (unlikely(ret < 0)) {
pr_err("core initialization failed for port#%d\n",
port->index);
return;
}
/* In the interest of preserving device ordering, synchronize */
async_synchronize_cookie(cookie);
register_pci_controller(port->hose);
}
static struct sh7786_pcie_hwops sh7786_65nm_pcie_hwops __initdata = {
.core_init = sh7786_pcie_core_init,
.port_init_hw = sh7786_pcie_init_hw,
};
static int __init sh7786_pcie_init(void)
{
struct clk *platclk;
int i;
printk(KERN_NOTICE "PCI: Starting initialization.\n");
sh7786_pcie_hwops = &sh7786_65nm_pcie_hwops;
nr_ports = sh7786_pcie_hwops->core_init();
BUG_ON(nr_ports > ARRAY_SIZE(sh7786_pci_channels));
if (unlikely(nr_ports == 0))
return -ENODEV;
sh7786_pcie_ports = kzalloc(nr_ports * sizeof(struct sh7786_pcie_port),
GFP_KERNEL);
if (unlikely(!sh7786_pcie_ports))
return -ENOMEM;
/*
* Fetch any optional platform clock associated with this block.
*
* This is a rather nasty hack for boards with spec-mocking FPGAs
* that have a secondary set of clocks outside of the on-chip
* ones that need to be accounted for before there is any chance
* of touching the existing MSTP bits or CPG clocks.
*/
platclk = clk_get(NULL, "pcie_plat_clk");
if (IS_ERR(platclk)) {
/* Sane hardware should probably get a WARN_ON.. */
platclk = NULL;
}
clk_enable(platclk);
printk(KERN_NOTICE "PCI: probing %d ports.\n", nr_ports);
for (i = 0; i < nr_ports; i++) {
struct sh7786_pcie_port *port = sh7786_pcie_ports + i;
port->index = i;
port->hose = sh7786_pci_channels + i;
port->hose->io_map_base = port->hose->resources[0].start;
async_schedule(sh7786_pcie_hwops->port_init_hw, port);
}
async_synchronize_full();
return 0;
}
arch_initcall(sh7786_pcie_init);