* [PATCH] powerpc: add Book E support to 64-bit hibernation
From: Wang Dongsheng @ 2013-03-14 3:36 UTC (permalink / raw)
To: linuxppc-dev; +Cc: scottwood, johannes, Wang Dongsheng
Update the 64-bit hibernation code to support Book E CPUs.
Some registers and instructions are not defined for Book3e
(SDR reg, tlbia instruction).
SDR: Storage Description Register. Book3S and Book3E have different
address translation mode, we do not need HTABORG & HTABSIZE to
translate virtual address to real address.
More registers are saved in BookE-64bit.(TCR, SPRGx)
Signed-off-by: Wang Dongsheng <dongsheng.wang@freescale.com>
---
arch/powerpc/kernel/swsusp_asm64.S | 64 ++++++++++++++++++++++++++++++++++-
1 files changed, 62 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/swsusp_asm64.S b/arch/powerpc/kernel/swsusp_asm64.S
index 86ac1d9..608e4ceb 100644
--- a/arch/powerpc/kernel/swsusp_asm64.S
+++ b/arch/powerpc/kernel/swsusp_asm64.S
@@ -46,10 +46,29 @@
#define SL_r29 0xe8
#define SL_r30 0xf0
#define SL_r31 0xf8
-#define SL_SIZE SL_r31+8
+#define SL_SPRG0 0x100
+#define SL_SPRG1 0x108
+#define SL_SPRG2 0x110
+#define SL_SPRG3 0x118
+#define SL_SPRG4 0x120
+#define SL_SPRG5 0x128
+#define SL_SPRG6 0x130
+#define SL_SPRG7 0x138
+#define SL_TCR 0x140
+#define SL_SIZE SL_TCR+8
/* these macros rely on the save area being
* pointed to by r11 */
+
+#define SAVE_SPR(register) \
+ mfspr r0,SPRN_##register ;\
+ std r0,SL_##register(r11)
+#define RESTORE_SPR(register) \
+ ld r0,SL_##register(r11) ;\
+ mtspr SPRN_##register,r0
+#define RESTORE_SPRG(n) \
+ ld r0,SL_SPRG##n(r11) ;\
+ mtsprg n,r0
#define SAVE_SPECIAL(special) \
mf##special r0 ;\
std r0, SL_##special(r11)
@@ -103,8 +122,21 @@ _GLOBAL(swsusp_arch_suspend)
SAVE_REGISTER(r30)
SAVE_REGISTER(r31)
SAVE_SPECIAL(MSR)
- SAVE_SPECIAL(SDR1)
SAVE_SPECIAL(XER)
+#ifdef CONFIG_PPC_BOOK3S_64
+ SAVE_SPECIAL(SDR1)
+#else
+ SAVE_SPR(TCR)
+ /* Save SPRGs */
+ SAVE_SPR(SPRG0)
+ SAVE_SPR(SPRG1)
+ SAVE_SPR(SPRG2)
+ SAVE_SPR(SPRG3)
+ SAVE_SPR(SPRG4)
+ SAVE_SPR(SPRG5)
+ SAVE_SPR(SPRG6)
+ SAVE_SPR(SPRG7)
+#endif
/* we push the stack up 128 bytes but don't store the
* stack pointer on the stack like a real stackframe */
@@ -151,6 +183,7 @@ copy_page_loop:
bne+ copyloop
nothing_to_copy:
+#ifdef CONFIG_PPC_BOOK3S_64
/* flush caches */
lis r3, 0x10
mtctr r3
@@ -167,6 +200,7 @@ nothing_to_copy:
sync
tlbia
+#endif
ld r11,swsusp_save_area_ptr@toc(r2)
@@ -208,16 +242,42 @@ nothing_to_copy:
RESTORE_REGISTER(r29)
RESTORE_REGISTER(r30)
RESTORE_REGISTER(r31)
+
+#ifdef CONFIG_PPC_BOOK3S_64
/* can't use RESTORE_SPECIAL(MSR) */
ld r0, SL_MSR(r11)
mtmsrd r0, 0
RESTORE_SPECIAL(SDR1)
+#else
+ /* Save SPRGs */
+ RESTORE_SPRG(0)
+ RESTORE_SPRG(1)
+ RESTORE_SPRG(2)
+ RESTORE_SPRG(3)
+ RESTORE_SPRG(4)
+ RESTORE_SPRG(5)
+ RESTORE_SPRG(6)
+ RESTORE_SPRG(7)
+
+ RESTORE_SPECIAL(MSR)
+
+ /* Restore TCR and clear any pending bits in TSR. */
+ RESTORE_SPR(TCR)
+ lis r0, (TSR_ENW | TSR_WIS | TSR_DIS | TSR_FIS)@h
+ mtspr SPRN_TSR,r0
+
+ /* Kick decrementer */
+ li r0,1
+ mtdec r0
+#endif
RESTORE_SPECIAL(XER)
sync
addi r1,r1,-128
+#ifdef CONFIG_PPC_BOOK3S_64
bl slb_flush_and_rebolt
+#endif
bl do_after_copyback
addi r1,r1,128
--
1.7.5.1
^ permalink raw reply related
* Re: [PATCH -V4 2/3] powerpc: Update kernel VSID range
From: Geoff Levand @ 2013-03-13 22:06 UTC (permalink / raw)
To: benh; +Cc: linuxppc-dev, paulus, Aneesh Kumar K.V, David Gibson
In-Reply-To: <1363181695-5315-2-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
Hi,
On Wed, 2013-03-13 at 19:04 +0530, Aneesh Kumar K.V wrote:
> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>
> This patch change the kernel VSID range so that we limit VSID_BITS to 37.
> This enables us to support 64TB with 65 bit VA (37+28). Without this patch
> we have boot hangs on platforms that only support 65 bit VA.
I tested the V4 series on PS3 applied to linux-3.8 and it fixed the boot
hang problem.
Ben, please apply.
Tested-by: Geoff Levand <geoff@infradead.org>
^ permalink raw reply
* Re: [PATCH] powerpc/85xx: workaround for chips with MSI hardware errata
From: Scott Wood @ 2013-03-13 20:24 UTC (permalink / raw)
To: Michael Ellerman; +Cc: B07421, linuxppc-dev, Jia Hongtao
In-Reply-To: <20130313050403.GB9317@concordia>
On 03/13/2013 12:04:03 AM, Michael Ellerman wrote:
> On Tue, Mar 12, 2013 at 03:48:02PM +0800, Jia Hongtao wrote:
> > The MPIC chip with version 2.0 has a MSI errata (errata PIC1 of =20
> mpc8544),
> > It causes that neither MSI nor MSI-X can work fine. This is a =20
> workaround
> > to allow MSI-X to function properly.
>=20
> You say "neither MSI nor MSI-X can work fine", which I take to mean
> "both MSI and MSI-X do not work".
>=20
> But then you say this is a workaround to allow MSI-X to work.
>=20
> So what I think you mean is, the erratum prevents both MSI and MSI-X
> from working. This is a workaround that allows MSI-X to work, and in =20
> addition
> the patch prevents MSI from being used on chips with the erratum -
> because there is no workaround for MSI.
There actually is a workaround for MSI, but it's more complicated and =20
not implemented by this patch.
We could also possibly get away with allowing exactly one MSI =20
(byteswapping doesn't matter if the value is zero) -- not sure how hard =20
that would be.
> > + list_for_each_entry(msi, &msi_head, list)
> > + if (msi->feature & MSI_HW_ERRATA_ENDIAN)
> > + return -EINVAL;
>=20
> I take it you're happy preventing MSI for all devices even if only a
> single chip in the machine has the erratum? In practice you probably
> have all or none with the erratum right?
Yes, it's all integrated onto one chip (the SoC itself). In fact there =20
should only be one MSI block on these chips.
-Scott=
^ permalink raw reply
* Re: [PATCH][v4] PPC: add paravirt idle loop for 64-bit book E
From: Kumar Gala @ 2013-03-13 19:22 UTC (permalink / raw)
To: Stuart Yoder; +Cc: kvm-ppc, agraf, kvm, linuxppc-dev
In-Reply-To: <1360351376-5323-1-git-send-email-stuart.yoder@freescale.com>
On Feb 8, 2013, at 1:22 PM, Stuart Yoder wrote:
> From: Stuart Yoder <stuart.yoder@freescale.com>
>
> Signed-off-by: Stuart Yoder <stuart.yoder@freescale.com>
> ---
>
> -removed KVM prefix to patch subject, patch is not KVM specific
>
> arch/powerpc/kernel/epapr_hcalls.S | 2 ++
> arch/powerpc/kernel/idle_book3e.S | 32 ++++++++++++++++++++++++++++++--
> 2 files changed, 32 insertions(+), 2 deletions(-)
applied to next
- k
^ permalink raw reply
* [RFC][PATCH 2/2] powerpc/fsl-pci Make PCIe hotplug work with Freescale
From: Kumar Gala @ 2013-03-13 19:07 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Rojhalat Ibrahim
In-Reply-To: <1363201636-7318-1-git-send-email-galak@kernel.crashing.org>
From: Rojhalat Ibrahim <imr@rtschenk.de>
Signed-off-by: Rojhalat Ibrahim <imr@rtschenk.de>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
arch/powerpc/include/asm/pci-bridge.h | 6 ++++++
arch/powerpc/sysdev/fsl_pci.c | 37 +++++++++++++++++++++++++++++----
arch/powerpc/sysdev/indirect_pci.c | 10 ++++-----
3 files changed, 43 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/include/asm/pci-bridge.h b/arch/powerpc/include/asm/pci-bridge.h
index c0278f0..ffbc5fd 100644
--- a/arch/powerpc/include/asm/pci-bridge.h
+++ b/arch/powerpc/include/asm/pci-bridge.h
@@ -120,6 +120,12 @@ extern void setup_indirect_pci(struct pci_controller* hose,
resource_size_t cfg_addr,
resource_size_t cfg_data, u32 flags);
+extern int indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val);
+
+extern int indirect_write_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 val);
+
static inline struct pci_controller *pci_bus_to_host(const struct pci_bus *bus)
{
return bus->sysdata;
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 41bbcc4..b18c377 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -74,6 +74,35 @@ static int __init fsl_pcie_check_link(struct pci_controller *hose)
return 0;
}
+static int fsl_indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val)
+{
+ struct pci_controller *hose = pci_bus_to_host(bus);
+
+ /* check the link status */
+ if ((bus->number == hose->first_busno) && (devfn == 0)) {
+ if (fsl_pcie_check_link(hose))
+ hose->indirect_type |= PPC_INDIRECT_TYPE_NO_PCIE_LINK;
+ else
+ hose->indirect_type &= ~PPC_INDIRECT_TYPE_NO_PCIE_LINK;
+ }
+ return indirect_read_config(bus, devfn, offset, len, val);
+}
+
+static struct pci_ops fsl_indirect_pci_ops =
+{
+ .read = fsl_indirect_read_config,
+ .write = indirect_write_config,
+};
+
+static void __init fsl_setup_indirect_pci(struct pci_controller* hose,
+ resource_size_t cfg_addr,
+ resource_size_t cfg_data, u32 flags)
+{
+ setup_indirect_pci(hose, cfg_addr, cfg_data, flags);
+ hose->ops = &fsl_indirect_pci_ops;
+}
+
#if defined(CONFIG_FSL_SOC_BOOKE) || defined(CONFIG_PPC_86xx)
#define MAX_PHYS_ADDR_BITS 40
@@ -469,8 +498,8 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
if (!hose->private_data)
goto no_bridge;
- setup_indirect_pci(hose, rsrc.start, rsrc.start + 0x4,
- PPC_INDIRECT_TYPE_BIG_ENDIAN);
+ fsl_setup_indirect_pci(hose, rsrc.start, rsrc.start + 0x4,
+ PPC_INDIRECT_TYPE_BIG_ENDIAN);
if (in_be32(&pci->block_rev1) < PCIE_IP_REV_3_0)
hose->indirect_type |= PPC_INDIRECT_TYPE_FSL_CFG_REG_LINK;
@@ -779,8 +808,8 @@ int __init mpc83xx_add_bridge(struct device_node *dev)
if (ret)
goto err0;
} else {
- setup_indirect_pci(hose, rsrc_cfg.start,
- rsrc_cfg.start + 4, 0);
+ fsl_setup_indirect_pci(hose, rsrc_cfg.start,
+ rsrc_cfg.start + 4, 0);
}
printk(KERN_INFO "Found FSL PCI host bridge at 0x%016llx. "
diff --git a/arch/powerpc/sysdev/indirect_pci.c b/arch/powerpc/sysdev/indirect_pci.c
index 82fdad8..c6c8b52 100644
--- a/arch/powerpc/sysdev/indirect_pci.c
+++ b/arch/powerpc/sysdev/indirect_pci.c
@@ -20,9 +20,8 @@
#include <asm/pci-bridge.h>
#include <asm/machdep.h>
-static int
-indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
- int len, u32 *val)
+int indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val)
{
struct pci_controller *hose = pci_bus_to_host(bus);
volatile void __iomem *cfg_data;
@@ -78,9 +77,8 @@ indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
return PCIBIOS_SUCCESSFUL;
}
-static int
-indirect_write_config(struct pci_bus *bus, unsigned int devfn, int offset,
- int len, u32 val)
+int indirect_write_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 val)
{
struct pci_controller *hose = pci_bus_to_host(bus);
volatile void __iomem *cfg_data;
--
1.7.9.7
^ permalink raw reply related
* [RFC][PATCH 1/2] powerpc/fsl-pci: Keep PCI SoC controller registers in
From: Kumar Gala @ 2013-03-13 19:07 UTC (permalink / raw)
To: linuxppc-dev
Move to keeping the SoC registers that control and config the PCI
controllers on FSL SoCs in the pci_controller struct. This allows us to
not need to ioremap() the registers in multiple different places that
use them.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
arch/powerpc/include/asm/pci-bridge.h | 5 ++-
arch/powerpc/sysdev/fsl_pci.c | 69 ++++++++++++++-------------------
2 files changed, 34 insertions(+), 40 deletions(-)
diff --git a/arch/powerpc/include/asm/pci-bridge.h b/arch/powerpc/include/asm/pci-bridge.h
index 025a130..c0278f0 100644
--- a/arch/powerpc/include/asm/pci-bridge.h
+++ b/arch/powerpc/include/asm/pci-bridge.h
@@ -70,6 +70,8 @@ struct pci_controller {
* BIG_ENDIAN - cfg_addr is a big endian register
* BROKEN_MRM - the 440EPx/GRx chips have an errata that causes hangs on
* the PLB4. Effectively disable MRM commands by setting this.
+ * FSL_CFG_REG_LINK - Freescale controller version in which the PCIe
+ * link status is in a RC PCIe cfg register (vs being a SoC register)
*/
#define PPC_INDIRECT_TYPE_SET_CFG_TYPE 0x00000001
#define PPC_INDIRECT_TYPE_EXT_REG 0x00000002
@@ -77,6 +79,7 @@ struct pci_controller {
#define PPC_INDIRECT_TYPE_NO_PCIE_LINK 0x00000008
#define PPC_INDIRECT_TYPE_BIG_ENDIAN 0x00000010
#define PPC_INDIRECT_TYPE_BROKEN_MRM 0x00000020
+#define PPC_INDIRECT_TYPE_FSL_CFG_REG_LINK 0x00000040
u32 indirect_type;
/* Currently, we limit ourselves to 1 IO range and 3 mem
* ranges since the common pci_bus structure can't handle more
@@ -90,9 +93,9 @@ struct pci_controller {
#ifdef CONFIG_PPC64
unsigned long buid;
+#endif /* CONFIG_PPC64 */
void *private_data;
-#endif /* CONFIG_PPC64 */
};
/* These are used for config access before all the PCI probing
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 3271177..41bbcc4 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -54,34 +54,22 @@ static void quirk_fsl_pcie_header(struct pci_dev *dev)
return;
}
-static int __init fsl_pcie_check_link(struct pci_controller *hose,
- struct resource *rsrc)
+static int __init fsl_pcie_check_link(struct pci_controller *hose)
{
- struct ccsr_pci __iomem *pci = NULL;
u32 val;
- /* for PCIe IP rev 3.0 or greater use CSR0 for link state */
- if (rsrc) {
- pr_debug("PCI memory map start 0x%016llx, size 0x%016llx\n",
- (u64)rsrc->start, (u64)rsrc->end - (u64)rsrc->start + 1);
- pci = ioremap(rsrc->start, rsrc->end - rsrc->start + 1);
- if (!pci) {
- dev_err(hose->parent, "Unable to map PCIe registers\n");
- return -ENOMEM;
- }
- if (in_be32(&pci->block_rev1) >= PCIE_IP_REV_3_0) {
- val = (in_be32(&pci->pex_csr0) & PEX_CSR0_LTSSM_MASK)
- >> PEX_CSR0_LTSSM_SHIFT;
- if (val != PEX_CSR0_LTSSM_L0)
- return 1;
- iounmap(pci);
- return 0;
- }
- iounmap(pci);
+ if (hose->indirect_type & PPC_INDIRECT_TYPE_FSL_CFG_REG_LINK) {
+ early_read_config_dword(hose, 0, 0, PCIE_LTSSM, &val);
+ if (val < PCIE_LTSSM_L0)
+ return 1;
+ } else {
+ struct ccsr_pci __iomem *pci = hose->private_data;
+ /* for PCIe IP rev 3.0 or greater use CSR0 for link state */
+ val = (in_be32(&pci->pex_csr0) & PEX_CSR0_LTSSM_MASK)
+ >> PEX_CSR0_LTSSM_SHIFT;
+ if (val != PEX_CSR0_LTSSM_L0)
+ return 1;
}
- early_read_config_dword(hose, 0, 0, PCIE_LTSSM, &val);
- if (val < PCIE_LTSSM_L0)
- return 1;
return 0;
}
@@ -148,10 +136,9 @@ static int setup_one_atmu(struct ccsr_pci __iomem *pci,
}
/* atmu setup for fsl pci/pcie controller */
-static void setup_pci_atmu(struct pci_controller *hose,
- struct resource *rsrc)
+static void setup_pci_atmu(struct pci_controller *hose)
{
- struct ccsr_pci __iomem *pci;
+ struct ccsr_pci __iomem *pci = hose->private_data;
int i, j, n, mem_log, win_idx = 3, start_idx = 1, end_idx = 4;
u64 mem, sz, paddr_hi = 0;
u64 paddr_lo = ULLONG_MAX;
@@ -162,15 +149,6 @@ static void setup_pci_atmu(struct pci_controller *hose,
const u64 *reg;
int len;
- pr_debug("PCI memory map start 0x%016llx, size 0x%016llx\n",
- (u64)rsrc->start, (u64)resource_size(rsrc));
-
- pci = ioremap(rsrc->start, resource_size(rsrc));
- if (!pci) {
- dev_err(hose->parent, "Unable to map ATMU registers\n");
- return;
- }
-
if (early_find_capability(hose, 0, 0, PCI_CAP_ID_EXP)) {
if (in_be32(&pci->block_rev1) >= PCIE_IP_REV_2_2) {
win_idx = 2;
@@ -451,6 +429,7 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
const int *bus_range;
u8 hdr_type, progif;
struct device_node *dev;
+ struct ccsr_pci __iomem *pci;
dev = pdev->dev.of_node;
@@ -483,9 +462,19 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
hose->first_busno = bus_range ? bus_range[0] : 0x0;
hose->last_busno = bus_range ? bus_range[1] : 0xff;
+ pr_debug("PCI memory map start 0x%016llx, size 0x%016llx\n",
+ (u64)rsrc.start, (u64)resource_size(&rsrc));
+
+ pci = hose->private_data = ioremap(rsrc.start, resource_size(&rsrc));
+ if (!hose->private_data)
+ goto no_bridge;
+
setup_indirect_pci(hose, rsrc.start, rsrc.start + 0x4,
PPC_INDIRECT_TYPE_BIG_ENDIAN);
+ if (in_be32(&pci->block_rev1) < PCIE_IP_REV_3_0)
+ hose->indirect_type |= PPC_INDIRECT_TYPE_FSL_CFG_REG_LINK;
+
if (early_find_capability(hose, 0, 0, PCI_CAP_ID_EXP)) {
/* For PCIE read HEADER_TYPE to identify controler mode */
early_read_config_byte(hose, 0, 0, PCI_HEADER_TYPE, &hdr_type);
@@ -505,7 +494,7 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
if (early_find_capability(hose, 0, 0, PCI_CAP_ID_EXP)) {
hose->indirect_type |= PPC_INDIRECT_TYPE_EXT_REG |
PPC_INDIRECT_TYPE_SURPRESS_PRIMARY_BUS;
- if (fsl_pcie_check_link(hose, &rsrc))
+ if (fsl_pcie_check_link(hose))
hose->indirect_type |= PPC_INDIRECT_TYPE_NO_PCIE_LINK;
}
@@ -522,11 +511,12 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
pci_process_bridge_OF_ranges(hose, dev, is_primary);
/* Setup PEX window registers */
- setup_pci_atmu(hose, &rsrc);
+ setup_pci_atmu(hose);
return 0;
no_bridge:
+ iounmap(hose->private_data);
/* unmap cfg_data & cfg_addr separately if not on same page */
if (((unsigned long)hose->cfg_data & PAGE_MASK) !=
((unsigned long)hose->cfg_addr & PAGE_MASK))
@@ -703,11 +693,12 @@ static int __init mpc83xx_pcie_setup(struct pci_controller *hose,
WARN_ON(hose->dn->data);
hose->dn->data = pcie;
hose->ops = &mpc83xx_pcie_ops;
+ hose->indirect_type |= PPC_INDIRECT_TYPE_FSL_CFG_REG_LINK;
out_le32(pcie->cfg_type0 + PEX_OUTWIN0_TAH, 0);
out_le32(pcie->cfg_type0 + PEX_OUTWIN0_TAL, 0);
- if (fsl_pcie_check_link(hose, NULL))
+ if (fsl_pcie_check_link(hose))
hose->indirect_type |= PPC_INDIRECT_TYPE_NO_PCIE_LINK;
return 0;
--
1.7.9.7
^ permalink raw reply related
* Re: [PATCH] Make PCIe hotplug work with Freescale PCIe controllers
From: Kumar Gala @ 2013-03-13 19:04 UTC (permalink / raw)
To: Rojhalat Ibrahim; +Cc: linuxppc-dev
In-Reply-To: <1510128.fXMLXu3tfm@pcimr>
On Mar 13, 2013, at 5:06 AM, Rojhalat Ibrahim wrote:
> On Tuesday 12 March 2013 15:48:01 Kumar Gala wrote:
>>> I'd rather we just export indirect_read_config() & =
indirect_write_config()
>>> from indirect_pci.c and call the functions directly. Adding a =
global and
>>> call them via a function pointer seems wrong to me.
>>>=20
>>> - k
>>=20
>> Also, can you base this patch on my powerpc.git next branch as =
fsl_pci.{c,h}
>> have some changes in them.
>>=20
>> - k
>>=20
>=20
> Hi Kumar,
>=20
> here's the patch with direct calls to indirect_{r,w}_config based on =
your powerpc.git next branch.
>=20
> This does not work for PCIe IP rev 3.0 as I'm not sure how to handle =
"struct resource" aside from making it global or changing "struct =
pci_controller".
>=20
> Rojhalat
I'll send a reworked version that hopefully deals with that issue that =
you can test and make sure things work ok. As well as some other minor =
style code cleanups.
- k=
^ permalink raw reply
* [PATCH 5/5 v9] iommu/fsl: Freescale PAMU driver and iommu implementation.
From: Varun Sethi @ 2013-03-13 18:49 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
Following is a brief description of the PAMU hardware:
PAMU determines what action to take and whether to authorize the action on
the basis of the memory address, a Logical IO Device Number (LIODN), and
PAACT table (logically) indexed by LIODN and address. Hardware devices which
need to access memory must provide an LIODN in addition to the memory address.
Peripheral Access Authorization and Control Tables (PAACTs) are the primary
data structures used by PAMU. A PAACT is a table of peripheral access
authorization and control entries (PAACE).Each PAACE defines the range of
I/O bus address space that is accessible by the LIOD and the associated access
capabilities.
There are two types of PAACTs: primary PAACT (PPAACT) and secondary PAACT
(SPAACT).A given physical I/O device may be able to act as one or more
independent logical I/O devices (LIODs). Each such logical I/O device is
assigned an identifier called logical I/O device number (LIODN). A LIODN is
allocated a contiguous portion of the I/O bus address space called the DSA window
for performing DSA operations. The DSA window may optionally be divided into
multiple sub-windows, each of which may be used to map to a region in system
storage space. The first sub-window is referred to as the primary sub-window
and the remaining are called secondary sub-windows.
This patch provides the PAMU driver (fsl_pamu.c) and the corresponding IOMMU
API implementation (fsl_pamu_domain.c). The PAMU hardware driver (fsl_pamu.c)
has been derived from the work done by Ashish Kalra and Timur Tabi.
Signed-off-by: Timur Tabi
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
---
changes in v9:
- Merged and createad a single function to delete
a device from domain list.
- Refactored the add_device API code.
- Renamed the paace and spaace init fucntions.
- Renamed functions for mapping windows and subwindows.
- Changed the MAX LIODN value to MAX value u-boot can
program.
- Hard coded maximum number of subwindows.
changes in v8:
- implemented the new API for window based IOMMUs.
changes in v7:
- Set max_subwidows in the geometry attribute.
- Add checking for maximum supported LIODN value.
- Use upper_32_bits and lower_32_bits macros while
intializing PAMU data structures.
changes in v6:
- Simplified complex conditional statements.
- Fixed indentation issues.
- Added comments for IOMMU API implementation.
changes in v5:
- Addressed comments from Timur.
changes in v4:
- Addressed comments from Timur and Scott.
changes in v3:
- Addressed comments by Kumar Gala
- dynamic fspi allocation
- fixed alignment check in map and unmap
arch/powerpc/sysdev/fsl_pci.h | 5 +
drivers/iommu/Kconfig | 8 +
drivers/iommu/Makefile | 1 +
drivers/iommu/fsl_pamu.c | 1250 +++++++++++++++++++++++++++++++++++++++
drivers/iommu/fsl_pamu.h | 405 +++++++++++++
drivers/iommu/fsl_pamu_domain.c | 1134 +++++++++++++++++++++++++++++++++++
drivers/iommu/fsl_pamu_domain.h | 85 +++
7 files changed, 2888 insertions(+), 0 deletions(-)
create mode 100644 drivers/iommu/fsl_pamu.c
create mode 100644 drivers/iommu/fsl_pamu.h
create mode 100644 drivers/iommu/fsl_pamu_domain.c
create mode 100644 drivers/iommu/fsl_pamu_domain.h
diff --git a/arch/powerpc/sysdev/fsl_pci.h b/arch/powerpc/sysdev/fsl_pci.h
index c495c00..feb34f6 100644
--- a/arch/powerpc/sysdev/fsl_pci.h
+++ b/arch/powerpc/sysdev/fsl_pci.h
@@ -14,6 +14,11 @@
#ifndef __POWERPC_FSL_PCI_H
#define __POWERPC_FSL_PCI_H
+
+/* FSL PCI controller BRR1 register */
+#define PCI_FSL_BRR1 0xbf8
+#define PCI_FSL_BRR1_VER 0xffff
+
#define PCIE_LTSSM 0x0404 /* PCIE Link Training and Status */
#define PCIE_LTSSM_L0 0x16 /* L0 state */
#define PCIE_IP_REV_2_2 0x02080202 /* PCIE IP block version Rev2.2 */
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index 5c514d07..b4b6a99 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -17,6 +17,14 @@ config OF_IOMMU
def_bool y
depends on OF
+config FSL_PAMU
+ bool "Freescale IOMMU support"
+ depends on PPC_E500MC
+ select IOMMU_API
+ select GENERIC_ALLOCATOR
+ help
+ Freescale PAMU support.
+
# MSM IOMMU support
config MSM_IOMMU
bool "MSM IOMMU Support"
diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile
index ef0e520..027d1af 100644
--- a/drivers/iommu/Makefile
+++ b/drivers/iommu/Makefile
@@ -15,3 +15,4 @@ obj-$(CONFIG_TEGRA_IOMMU_SMMU) += tegra-smmu.o
obj-$(CONFIG_EXYNOS_IOMMU) += exynos-iommu.o
obj-$(CONFIG_SHMOBILE_IOMMU) += shmobile-iommu.o
obj-$(CONFIG_SHMOBILE_IPMMU) += shmobile-ipmmu.o
+obj-$(CONFIG_FSL_PAMU) += fsl_pamu.o fsl_pamu_domain.o
diff --git a/drivers/iommu/fsl_pamu.c b/drivers/iommu/fsl_pamu.c
new file mode 100644
index 0000000..93df844
--- /dev/null
+++ b/drivers/iommu/fsl_pamu.c
@@ -0,0 +1,1250 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ *
+ */
+
+#define pr_fmt(fmt) "fsl-pamu: %s: " fmt, __func__
+
+#include <linux/init.h>
+#include <linux/iommu.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/device.h>
+#include <linux/of_platform.h>
+#include <linux/bootmem.h>
+#include <linux/genalloc.h>
+#include <asm/io.h>
+#include <asm/bitops.h>
+#include <asm/fsl_guts.h>
+
+#include "fsl_pamu.h"
+
+/* define indexes for each operation mapping scenario */
+#define OMI_QMAN 0x00
+#define OMI_FMAN 0x01
+#define OMI_QMAN_PRIV 0x02
+#define OMI_CAAM 0x03
+
+/* Handling access violations */
+#define make64(high, low) (((u64)(high) << 32) | (low))
+
+struct pamu_isr_data {
+ void __iomem *pamu_reg_base; /* Base address of PAMU regs*/
+ unsigned int count; /* The number of PAMUs */
+};
+
+static struct paace *ppaact;
+static struct paace *spaact;
+static struct ome *omt;
+
+/* maximum subwindows permitted per liodn */
+static u32 max_subwindow_count;
+
+/* Pool for fspi allocation */
+struct gen_pool *spaace_pool;
+
+/**
+ * pamu_get_max_subwin_cnt() - Return the maximum supported
+ * subwindow count per liodn.
+ *
+ */
+u32 pamu_get_max_subwin_cnt()
+{
+ return max_subwindow_count;
+}
+
+/**
+ * pamu_get_ppaace() - Return the primary PACCE
+ * @liodn: liodn PAACT index for desired PAACE
+ *
+ * Returns the ppace pointer upon success else return
+ * null.
+ */
+static struct paace *pamu_get_ppaace(int liodn)
+{
+ if (!ppaact || liodn >= PAACE_NUMBER_ENTRIES) {
+ pr_err("PPAACT doesn't exist\n");
+ return NULL;
+ }
+
+ return &ppaact[liodn];
+}
+
+/**
+ * pamu_enable_liodn() - Set valid bit of PACCE
+ * @liodn: liodn PAACT index for desired PAACE
+ *
+ * Returns 0 upon success else error code < 0 returned
+ */
+int pamu_enable_liodn(int liodn)
+{
+ struct paace *ppaace;
+
+ ppaace = pamu_get_ppaace(liodn);
+ if (!ppaace) {
+ pr_err("Invalid primary paace entry\n");
+ return -ENOENT;
+ }
+
+ if (!get_bf(ppaace->addr_bitfields, PPAACE_AF_WSE)) {
+ pr_err("liodn %d not configured\n", liodn);
+ return -EINVAL;
+ }
+
+ /* Ensure that all other stores to the ppaace complete first */
+ mb();
+
+ ppaace->addr_bitfields |= PAACE_V_VALID;
+ mb();
+
+ return 0;
+}
+
+/**
+ * pamu_disable_liodn() - Clears valid bit of PACCE
+ * @liodn: liodn PAACT index for desired PAACE
+ *
+ * Returns 0 upon success else error code < 0 returned
+ */
+int pamu_disable_liodn(int liodn)
+{
+ struct paace *ppaace;
+
+ ppaace = pamu_get_ppaace(liodn);
+ if (!ppaace) {
+ pr_err("Invalid primary paace entry\n");
+ return -ENOENT;
+ }
+
+ set_bf(ppaace->addr_bitfields, PAACE_AF_V, PAACE_V_INVALID);
+ mb();
+
+ return 0;
+}
+
+/* Derive the window size encoding for a particular PAACE entry */
+static unsigned int map_addrspace_size_to_wse(phys_addr_t addrspace_size)
+{
+ /* Bug if not a power of 2 */
+ BUG_ON((addrspace_size & (addrspace_size - 1)));
+
+ /* window size is 2^(WSE+1) bytes */
+ return __ffs(addrspace_size >> PAMU_PAGE_SHIFT) + PAMU_PAGE_SHIFT - 1;
+}
+
+/* Derive the PAACE window count encoding for the subwindow count */
+static unsigned int map_subwindow_cnt_to_wce(u32 subwindow_cnt)
+{
+ /* window count is 2^(WCE+1) bytes */
+ return __ffs(subwindow_cnt) - 1;
+}
+
+/*
+ * Set the PAACE type as primary and set the coherency required domain
+ * attribute
+ */
+static void pamu_init_ppaace(struct paace *ppaace)
+{
+ set_bf(ppaace->addr_bitfields, PAACE_AF_PT, PAACE_PT_PRIMARY);
+
+ set_bf(ppaace->domain_attr.to_host.coherency_required, PAACE_DA_HOST_CR,
+ PAACE_M_COHERENCE_REQ);
+}
+
+/*
+ * Set the PAACE type as secondary and set the coherency required domain
+ * attribute.
+ */
+static void pamu_init_spaace(struct paace *spaace)
+{
+ set_bf(spaace->addr_bitfields, PAACE_AF_PT, PAACE_PT_SECONDARY);
+ set_bf(spaace->domain_attr.to_host.coherency_required, PAACE_DA_HOST_CR,
+ PAACE_M_COHERENCE_REQ);
+}
+
+/*
+ * Return the spaace (corresponding to the secondary window index)
+ * for a particular ppaace.
+ */
+static struct paace *pamu_get_spaace(struct paace *paace, u32 wnum)
+{
+ u32 subwin_cnt;
+ struct paace *spaace = NULL;
+
+ subwin_cnt = 1UL << (get_bf(paace->impl_attr, PAACE_IA_WCE) + 1);
+
+ if (wnum < subwin_cnt)
+ spaace = &spaact[paace->fspi + wnum];
+ else
+ pr_err("secondary paace out of bounds\n");
+
+ return spaace;
+}
+
+/**
+ * pamu_get_fspi_and_allocate() - Allocates fspi index and reserves subwindows
+ * required for primary PAACE in the secondary
+ * PAACE table.
+ * @subwin_cnt: Number of subwindows to be reserved.
+ *
+ * A PPAACE entry may have a number of associated subwindows. A subwindow
+ * corresponds to a SPAACE entry in the SPAACT table. Each PAACE entry stores
+ * the index (fspi) of the first SPAACE entry in the SPAACT table. This
+ * function returns the index of the first SPAACE entry. The remaining
+ * SPAACE entries are reserved contiguously from that index.
+ *
+ * Returns a valid fspi index in the range of 0 - SPAACE_NUMBER_ENTRIES on success.
+ * If no SPAACE entry is available or the allocator can not reserve the required
+ * number of contiguous entries function returns ULONG_MAX indicating a failure.
+ *
+*/
+static unsigned long pamu_get_fspi_and_allocate(u32 subwin_cnt)
+{
+ unsigned long spaace_addr;
+
+ spaace_addr = gen_pool_alloc(spaace_pool, subwin_cnt * sizeof(struct paace));
+ if (!spaace_addr)
+ return ULONG_MAX;
+
+ return (spaace_addr - (unsigned long)spaact) / (sizeof(struct paace));
+}
+
+/* Release the subwindows reserved for a particular LIODN */
+void pamu_free_subwins(int liodn)
+{
+ struct paace *ppaace;
+ u32 subwin_cnt, size;
+
+ ppaace = pamu_get_ppaace(liodn);
+ if (!ppaace) {
+ pr_err("Invalid liodn entry\n");
+ return;
+ }
+
+ if (get_bf(ppaace->addr_bitfields, PPAACE_AF_MW)) {
+ subwin_cnt = 1UL << (get_bf(ppaace->impl_attr, PAACE_IA_WCE) + 1);
+ size = (subwin_cnt - 1) * sizeof(struct paace);
+ gen_pool_free(spaace_pool, (unsigned long)&spaact[ppaace->fspi], size);
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_MW, 0);
+ }
+}
+
+/*
+ * Function used for updating stash destination for the coressponding
+ * LIODN.
+ */
+int pamu_update_paace_stash(int liodn, u32 subwin, u32 value)
+{
+ struct paace *paace;
+
+ paace = pamu_get_ppaace(liodn);
+ if (!paace) {
+ pr_err("Invalid liodn entry\n");
+ return -ENOENT;
+ }
+ if (subwin) {
+ paace = pamu_get_spaace(paace, subwin - 1);
+ if (!paace) {
+ return -ENOENT;
+ }
+ }
+ set_bf(paace->impl_attr, PAACE_IA_CID, value);
+
+ mb();
+
+ return 0;
+}
+
+/* Disable a subwindow corresponding to the LIODN */
+int pamu_disable_spaace(int liodn, u32 subwin)
+{
+ struct paace *paace;
+
+ paace = pamu_get_ppaace(liodn);
+ if (!paace) {
+ pr_err("Invalid liodn entry\n");
+ return -ENOENT;
+ }
+ if (subwin) {
+ paace = pamu_get_spaace(paace, subwin - 1);
+ if (!paace) {
+ return -ENOENT;
+ }
+ set_bf(paace->addr_bitfields, PAACE_AF_V,
+ PAACE_V_INVALID);
+ } else {
+ set_bf(paace->addr_bitfields, PAACE_AF_AP,
+ PAACE_AP_PERMS_DENIED);
+ }
+
+ mb();
+
+ return 0;
+}
+
+
+/**
+ * pamu_config_paace() - Sets up PPAACE entry for specified liodn
+ *
+ * @liodn: Logical IO device number
+ * @win_addr: starting address of DSA window
+ * @win-size: size of DSA window
+ * @omi: Operation mapping index -- if ~omi == 0 then omi not defined
+ * @rpn: real (true physical) page number
+ * @stashid: cache stash id for associated cpu -- if ~stashid == 0 then
+ * stashid not defined
+ * @snoopid: snoop id for hardware coherency -- if ~snoopid == 0 then
+ * snoopid not defined
+ * @subwin_cnt: number of sub-windows
+ * @prot: window permissions
+ *
+ * Returns 0 upon success else error code < 0 returned
+ */
+int pamu_config_ppaace(int liodn, phys_addr_t win_addr, phys_addr_t win_size,
+ u32 omi, unsigned long rpn, u32 snoopid, u32 stashid,
+ u32 subwin_cnt, int prot)
+{
+ struct paace *ppaace;
+ unsigned long fspi;
+
+ if ((win_size & (win_size - 1)) || win_size < PAMU_PAGE_SIZE) {
+ pr_err("window size too small or not a power of two %llx\n", win_size);
+ return -EINVAL;
+ }
+
+ if (win_addr & (win_size - 1)) {
+ pr_err("window address is not aligned with window size\n");
+ return -EINVAL;
+ }
+
+ ppaace = pamu_get_ppaace(liodn);
+ if (!ppaace) {
+ return -ENOENT;
+ }
+
+ /* window size is 2^(WSE+1) bytes */
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_WSE,
+ map_addrspace_size_to_wse(win_size));
+
+ pamu_init_ppaace(ppaace);
+
+ ppaace->wbah = win_addr >> (PAMU_PAGE_SHIFT + 20);
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_WBAL,
+ (win_addr >> PAMU_PAGE_SHIFT));
+
+ /* set up operation mapping if it's configured */
+ if (omi < OME_NUMBER_ENTRIES) {
+ set_bf(ppaace->impl_attr, PAACE_IA_OTM, PAACE_OTM_INDEXED);
+ ppaace->op_encode.index_ot.omi = omi;
+ } else if (~omi != 0) {
+ pr_err("bad operation mapping index: %d\n", omi);
+ return -EINVAL;
+ }
+
+ /* configure stash id */
+ if (~stashid != 0)
+ set_bf(ppaace->impl_attr, PAACE_IA_CID, stashid);
+
+ /* configure snoop id */
+ if (~snoopid != 0)
+ ppaace->domain_attr.to_host.snpid = snoopid;
+
+ if (subwin_cnt) {
+ /* The first entry is in the primary PAACE instead */
+ fspi = pamu_get_fspi_and_allocate(subwin_cnt - 1);
+ if (fspi == ULONG_MAX) {
+ pr_err("spaace indexes exhausted\n");
+ return -EINVAL;
+ }
+
+ /* window count is 2^(WCE+1) bytes */
+ set_bf(ppaace->impl_attr, PAACE_IA_WCE,
+ map_subwindow_cnt_to_wce(subwin_cnt));
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_MW, 0x1);
+ ppaace->fspi = fspi;
+ } else {
+ set_bf(ppaace->impl_attr, PAACE_IA_ATM, PAACE_ATM_WINDOW_XLATE);
+ ppaace->twbah = rpn >> 20;
+ set_bf(ppaace->win_bitfields, PAACE_WIN_TWBAL, rpn);
+ set_bf(ppaace->addr_bitfields, PAACE_AF_AP, prot);
+ set_bf(ppaace->impl_attr, PAACE_IA_WCE, 0);
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_MW, 0);
+ }
+ mb();
+
+ return 0;
+}
+
+/**
+ * pamu_config_spaace() - Sets up SPAACE entry for specified subwindow
+ *
+ * @liodn: Logical IO device number
+ * @subwin_cnt: number of sub-windows associated with dma-window
+ * @subwin: subwindow index
+ * @subwin_size: size of subwindow
+ * @omi: Operation mapping index
+ * @rpn: real (true physical) page number
+ * @snoopid: snoop id for hardware coherency -- if ~snoopid == 0 then
+ * snoopid not defined
+ * @stashid: cache stash id for associated cpu
+ * @enable: enable/disable subwindow after reconfiguration
+ * @prot: sub window permissions
+ *
+ * Returns 0 upon success else error code < 0 returned
+ */
+int pamu_config_spaace(int liodn, u32 subwin_cnt, u32 subwin,
+ phys_addr_t subwin_size, u32 omi, unsigned long rpn,
+ u32 snoopid, u32 stashid, int enable, int prot)
+{
+ struct paace *paace;
+
+ /* setup sub-windows */
+ if (!subwin_cnt) {
+ pr_err("Invalid subwindow count\n");
+ return -EINVAL;
+ }
+
+ paace = pamu_get_ppaace(liodn);
+ if (subwin > 0 && subwin < subwin_cnt && paace) {
+ paace = pamu_get_spaace(paace, subwin - 1);
+
+ if (paace && !(paace->addr_bitfields & PAACE_V_VALID)) {
+ pamu_init_spaace(paace);
+ set_bf(paace->addr_bitfields, SPAACE_AF_LIODN, liodn);
+ }
+ }
+
+ if (!paace) {
+ pr_err("Invalid liodn entry\n");
+ return -ENOENT;
+ }
+
+ if (subwin_size & (subwin_size - 1) || subwin_size < PAMU_PAGE_SIZE) {
+ pr_err("subwindow size out of range, or not a power of 2\n");
+ return -EINVAL;
+ }
+
+ if (rpn == ULONG_MAX) {
+ pr_err("real page number out of range\n");
+ return -EINVAL;
+ }
+
+ /* window size is 2^(WSE+1) bytes */
+ set_bf(paace->win_bitfields, PAACE_WIN_SWSE,
+ map_addrspace_size_to_wse(subwin_size));
+
+ set_bf(paace->impl_attr, PAACE_IA_ATM, PAACE_ATM_WINDOW_XLATE);
+ paace->twbah = rpn >> 20;
+ set_bf(paace->win_bitfields, PAACE_WIN_TWBAL, rpn);
+ set_bf(paace->addr_bitfields, PAACE_AF_AP, prot);
+
+ /* configure snoop id */
+ if (~snoopid != 0)
+ paace->domain_attr.to_host.snpid = snoopid;
+
+ /* set up operation mapping if it's configured */
+ if (omi < OME_NUMBER_ENTRIES) {
+ set_bf(paace->impl_attr, PAACE_IA_OTM, PAACE_OTM_INDEXED);
+ paace->op_encode.index_ot.omi = omi;
+ } else if (~omi != 0) {
+ pr_err("bad operation mapping index: %d\n", omi);
+ return -EINVAL;
+ }
+
+ if (~stashid != 0)
+ set_bf(paace->impl_attr, PAACE_IA_CID, stashid);
+
+ smp_wmb();
+
+ if (enable)
+ paace->addr_bitfields |= PAACE_V_VALID;
+
+ mb();
+
+ return 0;
+}
+
+/**
+* get_ome_index() - Returns the index in the operation mapping table
+* for device.
+* @*omi_index: pointer for storing the index value
+*
+*/
+void get_ome_index(u32 *omi_index, struct device *dev)
+{
+ if (of_device_is_compatible(dev->of_node, "fsl,qman-portal"))
+ *omi_index = OMI_QMAN;
+ if (of_device_is_compatible(dev->of_node, "fsl,qman"))
+ *omi_index = OMI_QMAN_PRIV;
+}
+
+/**
+ * get_stash_id - Returns stash destination id corresponding to a
+ * cache type and vcpu.
+ * @stash_dest_hint: L1, L2 or L3
+ * @vcpu: vpcu target for a particular cache type.
+ *
+ * Returs stash on success or ~(u32)0 on failure.
+ *
+ */
+u32 get_stash_id(u32 stash_dest_hint, u32 vcpu)
+{
+ const u32 *prop;
+ struct device_node *node;
+ u32 cache_level;
+ int len;
+
+ /* Fastpath, exit early if L3/CPC cache is target for stashing */
+ if (stash_dest_hint == IOMMU_ATTR_CACHE_L3) {
+ node = of_find_compatible_node(NULL, NULL,
+ "fsl,p4080-l3-cache-controller");
+ if (node) {
+ prop = of_get_property(node, "cache-stash-id", 0);
+ if (!prop) {
+ pr_err("missing cache-stash-id at %s\n", node->full_name);
+ of_node_put(node);
+ return ~(u32)0;
+ }
+ of_node_put(node);
+ return be32_to_cpup(prop);
+ }
+ return ~(u32)0;
+ }
+
+ for_each_node_by_type(node, "cpu") {
+ prop = of_get_property(node, "reg", &len);
+ if (be32_to_cpup(prop) == vcpu)
+ break;
+ }
+
+ /* find the hwnode that represents the cache */
+ for (cache_level = IOMMU_ATTR_CACHE_L1; cache_level < IOMMU_ATTR_CACHE_L3; cache_level++) {
+ if (stash_dest_hint == cache_level) {
+ prop = of_get_property(node, "cache-stash-id", 0);
+ if (!prop) {
+ pr_err("missing cache-stash-id at %s\n", node->full_name);
+ of_node_put(node);
+ return ~(u32)0;
+ }
+ of_node_put(node);
+ return be32_to_cpup(prop);
+ }
+
+ prop = of_get_property(node, "next-level-cache", 0);
+ if (!prop) {
+ pr_err("can't find next-level-cache at %s\n",
+ node->full_name);
+ of_node_put(node);
+ return ~(u32)0; /* can't traverse any further */
+ }
+ of_node_put(node);
+
+ /* advance to next node in cache hierarchy */
+ node = of_find_node_by_phandle(*prop);
+ if (!node) {
+ pr_err("Invalid node for cache hierarchy %s\n",
+ node->full_name);
+ return ~(u32)0;
+ }
+ }
+
+ pr_err("stash dest not found for %d on vcpu %d\n",
+ stash_dest_hint, vcpu);
+ return ~(u32)0;
+}
+
+/* Identify if the PAACT table entry belongs to QMAN, BMAN or QMAN Portal */
+#define QMAN_PAACE 1
+#define QMAN_PORTAL_PAACE 2
+#define BMAN_PAACE 3
+
+/**
+ * Setup operation mapping and stash destinations for QMAN and QMAN portal.
+ * Memory accesses to QMAN and BMAN private memory need not be coherent, so
+ * clear the PAACE entry coherency attribute for them.
+ */
+static void setup_qbman_paace(struct paace *ppaace, int paace_type)
+{
+ switch (paace_type) {
+ case QMAN_PAACE:
+ set_bf(ppaace->impl_attr, PAACE_IA_OTM, PAACE_OTM_INDEXED);
+ ppaace->op_encode.index_ot.omi = OMI_QMAN_PRIV;
+ /* setup QMAN Private data stashing for the L3 cache */
+ set_bf(ppaace->impl_attr, PAACE_IA_CID, get_stash_id(IOMMU_ATTR_CACHE_L3, 0));
+ set_bf(ppaace->domain_attr.to_host.coherency_required, PAACE_DA_HOST_CR,
+ 0);
+ break;
+ case QMAN_PORTAL_PAACE:
+ set_bf(ppaace->impl_attr, PAACE_IA_OTM, PAACE_OTM_INDEXED);
+ ppaace->op_encode.index_ot.omi = OMI_QMAN;
+ /*Set DQRR and Frame stashing for the L3 cache */
+ set_bf(ppaace->impl_attr, PAACE_IA_CID, get_stash_id(IOMMU_ATTR_CACHE_L3, 0));
+ break;
+ case BMAN_PAACE:
+ set_bf(ppaace->domain_attr.to_host.coherency_required, PAACE_DA_HOST_CR,
+ 0);
+ break;
+ }
+}
+
+/**
+ * Setup the operation mapping table for various devices. This is a static
+ * table where each table index corresponds to a particular device. PAMU uses
+ * this table to translate device transaction to appropriate corenet
+ * transaction.
+ */
+static void __init setup_omt(struct ome *omt)
+{
+ struct ome *ome;
+
+ /* Configure OMI_QMAN */
+ ome = &omt[OMI_QMAN];
+
+ ome->moe[IOE_READ_IDX] = EOE_VALID | EOE_READ;
+ ome->moe[IOE_EREAD0_IDX] = EOE_VALID | EOE_RSA;
+ ome->moe[IOE_WRITE_IDX] = EOE_VALID | EOE_WRITE;
+ ome->moe[IOE_EWRITE0_IDX] = EOE_VALID | EOE_WWSAO;
+
+ ome->moe[IOE_DIRECT0_IDX] = EOE_VALID | EOE_LDEC;
+ ome->moe[IOE_DIRECT1_IDX] = EOE_VALID | EOE_LDECPE;
+
+ /* Configure OMI_FMAN */
+ ome = &omt[OMI_FMAN];
+ ome->moe[IOE_READ_IDX] = EOE_VALID | EOE_READI;
+ ome->moe[IOE_WRITE_IDX] = EOE_VALID | EOE_WRITE;
+
+ /* Configure OMI_QMAN private */
+ ome = &omt[OMI_QMAN_PRIV];
+ ome->moe[IOE_READ_IDX] = EOE_VALID | EOE_READ;
+ ome->moe[IOE_WRITE_IDX] = EOE_VALID | EOE_WRITE;
+ ome->moe[IOE_EREAD0_IDX] = EOE_VALID | EOE_RSA;
+ ome->moe[IOE_EWRITE0_IDX] = EOE_VALID | EOE_WWSA;
+
+ /* Configure OMI_CAAM */
+ ome = &omt[OMI_CAAM];
+ ome->moe[IOE_READ_IDX] = EOE_VALID | EOE_READI;
+ ome->moe[IOE_WRITE_IDX] = EOE_VALID | EOE_WRITE;
+}
+
+/*
+ * Get the maximum number of PAACT table entries
+ * and subwindows supported by PAMU
+ */
+static void get_pamu_cap_values(unsigned long pamu_reg_base)
+{
+ u32 pc_val;
+
+ pc_val = in_be32((u32 *)(pamu_reg_base + PAMU_PC3));
+ /* Maximum number of subwindows per liodn */
+ max_subwindow_count = 1 << (1 + PAMU_PC3_MWCE(pc_val));
+}
+
+/* Setup PAMU registers pointing to PAACT, SPAACT and OMT */
+int setup_one_pamu(unsigned long pamu_reg_base, unsigned long pamu_reg_size,
+ phys_addr_t ppaact_phys, phys_addr_t spaact_phys,
+ phys_addr_t omt_phys)
+{
+ u32 *pc;
+ struct pamu_mmap_regs *pamu_regs;
+
+ pc = (u32 *) (pamu_reg_base + PAMU_PC);
+ pamu_regs = (struct pamu_mmap_regs *)
+ (pamu_reg_base + PAMU_MMAP_REGS_BASE);
+
+ /* set up pointers to corenet control blocks */
+
+ out_be32(&pamu_regs->ppbah, upper_32_bits(ppaact_phys));
+ out_be32(&pamu_regs->ppbal, lower_32_bits(ppaact_phys));
+ ppaact_phys = ppaact_phys + PAACT_SIZE;
+ out_be32(&pamu_regs->pplah, upper_32_bits(ppaact_phys));
+ out_be32(&pamu_regs->pplal, lower_32_bits(ppaact_phys));
+
+ out_be32(&pamu_regs->spbah, upper_32_bits(spaact_phys));
+ out_be32(&pamu_regs->spbal, lower_32_bits(spaact_phys));
+ spaact_phys = spaact_phys + SPAACT_SIZE;
+ out_be32(&pamu_regs->splah, upper_32_bits(spaact_phys));
+ out_be32(&pamu_regs->splal, lower_32_bits(spaact_phys));
+
+ out_be32(&pamu_regs->obah, upper_32_bits(omt_phys));
+ out_be32(&pamu_regs->obal, lower_32_bits(omt_phys));
+ omt_phys = omt_phys + OMT_SIZE;
+ out_be32(&pamu_regs->olah, upper_32_bits(omt_phys));
+ out_be32(&pamu_regs->olal, lower_32_bits(omt_phys));
+
+ /*
+ * set PAMU enable bit,
+ * allow ppaact & omt to be cached
+ * & enable PAMU access violation interrupts.
+ */
+
+ out_be32((u32 *)(pamu_reg_base + PAMU_PICS),
+ PAMU_ACCESS_VIOLATION_ENABLE);
+ out_be32(pc, PAMU_PC_PE | PAMU_PC_OCE | PAMU_PC_SPCC | PAMU_PC_PPCC);
+ return 0;
+}
+
+/* Enable all device LIODNS */
+static void __init setup_liodns(void)
+{
+ int i, len;
+ struct paace *ppaace;
+ struct device_node *node = NULL;
+ const u32 *prop;
+
+ for_each_node_with_property(node, "fsl,liodn") {
+ prop = of_get_property(node, "fsl,liodn", &len);
+ for (i = 0; i < len / sizeof(u32); i++) {
+ int liodn;
+
+ liodn = be32_to_cpup(&prop[i]);
+ if (liodn >= PAACE_NUMBER_ENTRIES) {
+ pr_err("Invalid LIODN value %d\n", liodn);
+ continue;
+ }
+ ppaace = pamu_get_ppaace(liodn);
+ pamu_init_ppaace(ppaace);
+ /* window size is 2^(WSE+1) bytes */
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_WSE, 35);
+ ppaace->wbah = 0;
+ set_bf(ppaace->addr_bitfields, PPAACE_AF_WBAL, 0);
+ set_bf(ppaace->impl_attr, PAACE_IA_ATM,
+ PAACE_ATM_NO_XLATE);
+ set_bf(ppaace->addr_bitfields, PAACE_AF_AP,
+ PAACE_AP_PERMS_ALL);
+ if (of_device_is_compatible(node, "fsl,qman-portal"))
+ setup_qbman_paace(ppaace, QMAN_PORTAL_PAACE);
+ if (of_device_is_compatible(node, "fsl,qman"))
+ setup_qbman_paace(ppaace, QMAN_PAACE);
+ if (of_device_is_compatible(node, "fsl,bman"))
+ setup_qbman_paace(ppaace, BMAN_PAACE);
+ mb();
+ pamu_enable_liodn(liodn);
+ }
+ }
+}
+
+/* TBD: PAMU access violation interrupt handler */
+irqreturn_t pamu_av_isr(int irq, void *arg)
+{
+ struct pamu_isr_data *data = arg;
+ phys_addr_t phys;
+ unsigned int i, j;
+
+ pr_emerg("fsl-pamu: access violation interrupt\n");
+
+ for (i = 0; i < data->count; i++) {
+ void __iomem *p = data->pamu_reg_base + i * PAMU_OFFSET;
+ u32 pics = in_be32(p + PAMU_PICS);
+
+ if (pics & PAMU_ACCESS_VIOLATION_STAT) {
+ pr_emerg("POES1=%08x\n", in_be32(p + PAMU_POES1));
+ pr_emerg("POES2=%08x\n", in_be32(p + PAMU_POES2));
+ pr_emerg("AVS1=%08x\n", in_be32(p + PAMU_AVS1));
+ pr_emerg("AVS2=%08x\n", in_be32(p + PAMU_AVS2));
+ pr_emerg("AVA=%016llx\n", make64(in_be32(p + PAMU_AVAH),
+ in_be32(p + PAMU_AVAL)));
+ pr_emerg("UDAD=%08x\n", in_be32(p + PAMU_UDAD));
+ pr_emerg("POEA=%016llx\n", make64(in_be32(p + PAMU_POEAH),
+ in_be32(p + PAMU_POEAL)));
+
+ phys = make64(in_be32(p + PAMU_POEAH),
+ in_be32(p + PAMU_POEAL));
+
+ /* Assume that POEA points to a PAACE */
+ if (phys) {
+ u32 *paace = phys_to_virt(phys);
+
+ /* Only the first four words are relevant */
+ for (j = 0; j < 4; j++)
+ pr_emerg("PAACE[%u]=%08x\n", j, in_be32(paace + j));
+ }
+ }
+ }
+
+ panic("\n");
+ /* NOTREACHED */
+
+ return IRQ_HANDLED;
+}
+
+#define LAWAR_EN 0x80000000
+#define LAWAR_TARGET_MASK 0x0FF00000
+#define LAWAR_TARGET_SHIFT 20
+#define LAWAR_SIZE_MASK 0x0000003F
+#define LAWAR_CSDID_MASK 0x000FF000
+#define LAWAR_CSDID_SHIFT 12
+
+#define LAW_SIZE_4K 0xb
+
+struct ccsr_law {
+ u32 lawbarh; /* LAWn base address high */
+ u32 lawbarl; /* LAWn base address low */
+ u32 lawar; /* LAWn attributes */
+ u32 reserved;
+};
+
+#define make64(high, low) (((u64)(high) << 32) | (low))
+
+/*
+ * Create a coherence subdomain for a given memory block.
+ */
+static int __init create_csd(phys_addr_t phys, size_t size, u32 csd_port_id)
+{
+ struct device_node *np;
+ const __be32 *iprop;
+ void __iomem *lac = NULL; /* Local Access Control registers */
+ struct ccsr_law __iomem *law;
+ void __iomem *ccm = NULL;
+ u32 __iomem *csdids;
+ unsigned int i, num_laws, num_csds;
+ u32 law_target = 0;
+ u32 csd_id = 0;
+ int ret = 0;
+
+ np = of_find_compatible_node(NULL, NULL, "fsl,corenet-law");
+ if (!np)
+ return -ENODEV;
+
+ iprop = of_get_property(np, "fsl,num-laws", NULL);
+ if (!iprop) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ num_laws = be32_to_cpup(iprop);
+ if (!num_laws) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ lac = of_iomap(np, 0);
+ if (!lac) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ /* LAW registers are at offset 0xC00 */
+ law = lac + 0xC00;
+
+ of_node_put(np);
+
+ np = of_find_compatible_node(NULL, NULL, "fsl,corenet-cf");
+ if (!np) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ iprop = of_get_property(np, "fsl,ccf-num-csdids", NULL);
+ if (!iprop) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ num_csds = be32_to_cpup(iprop);
+ if (!num_csds) {
+ ret = -ENODEV;
+ goto error;
+ }
+
+ ccm = of_iomap(np, 0);
+ if (!ccm) {
+ ret = -ENOMEM;
+ goto error;
+ }
+
+ /* The undocumented CSDID registers are at offset 0x600 */
+ csdids = ccm + 0x600;
+
+ of_node_put(np);
+ np = NULL;
+
+ /* Find an unused coherence subdomain ID */
+ for (csd_id = 0; csd_id < num_csds; csd_id++) {
+ if (!csdids[csd_id])
+ break;
+ }
+
+ /* Store the Port ID in the (undocumented) proper CIDMRxx register */
+ csdids[csd_id] = csd_port_id;
+
+ /* Find the DDR LAW that maps to our buffer. */
+ for (i = 0; i < num_laws; i++) {
+ if (law[i].lawar & LAWAR_EN) {
+ phys_addr_t law_start, law_end;
+
+ law_start = make64(law[i].lawbarh, law[i].lawbarl);
+ law_end = law_start +
+ (2ULL << (law[i].lawar & LAWAR_SIZE_MASK));
+
+ if (law_start <= phys && phys < law_end) {
+ law_target = law[i].lawar & LAWAR_TARGET_MASK;
+ break;
+ }
+ }
+ }
+
+ if (i == 0 || i == num_laws) {
+ /* This should never happen*/
+ ret = -ENOENT;
+ goto error;
+ }
+
+ /* Find a free LAW entry */
+ while (law[--i].lawar & LAWAR_EN) {
+ if (i == 0) {
+ /* No higher priority LAW slots available */
+ ret = -ENOENT;
+ goto error;
+ }
+ }
+
+ law[i].lawbarh = upper_32_bits(phys);
+ law[i].lawbarl = lower_32_bits(phys);
+ wmb();
+ law[i].lawar = LAWAR_EN | law_target | (csd_id << LAWAR_CSDID_SHIFT) |
+ (LAW_SIZE_4K + get_order(size));
+ wmb();
+
+error:
+ if (ccm)
+ iounmap(ccm);
+
+ if (lac)
+ iounmap(lac);
+
+ if (np)
+ of_node_put(np);
+
+ return ret;
+}
+
+/*
+ * Table of SVRs and the corresponding PORT_ID values.
+ *
+ * All future CoreNet-enabled SOCs will have this erratum fixed, so this table
+ * should never need to be updated. SVRs are guaranteed to be unique, so
+ * there is no worry that a future SOC will inadvertently have one of these
+ * values.
+ */
+static const struct {
+ u32 svr;
+ u32 port_id;
+} port_id_map[] = {
+ {0x82100010, 0xFF000000}, /* P2040 1.0 */
+ {0x82100011, 0xFF000000}, /* P2040 1.1 */
+ {0x82100110, 0xFF000000}, /* P2041 1.0 */
+ {0x82100111, 0xFF000000}, /* P2041 1.1 */
+ {0x82110310, 0xFF000000}, /* P3041 1.0 */
+ {0x82110311, 0xFF000000}, /* P3041 1.1 */
+ {0x82010020, 0xFFF80000}, /* P4040 2.0 */
+ {0x82000020, 0xFFF80000}, /* P4080 2.0 */
+ {0x82210010, 0xFC000000}, /* P5010 1.0 */
+ {0x82210020, 0xFC000000}, /* P5010 2.0 */
+ {0x82200010, 0xFC000000}, /* P5020 1.0 */
+ {0x82050010, 0xFF800000}, /* P5021 1.0 */
+ {0x82040010, 0xFF800000}, /* P5040 1.0 */
+};
+
+#define SVR_SECURITY 0x80000 /* The Security (E) bit */
+
+static int __init fsl_pamu_probe(struct platform_device *pdev)
+{
+ void __iomem *pamu_regs = NULL;
+ struct ccsr_guts __iomem *guts_regs = NULL;
+ u32 pamubypenr, pamu_counter;
+ unsigned long pamu_reg_off;
+ unsigned long pamu_reg_base;
+ struct pamu_isr_data *data;
+ struct device_node *guts_node;
+ u64 size;
+ struct page *p;
+ int ret = 0;
+ int irq;
+ phys_addr_t ppaact_phys;
+ phys_addr_t spaact_phys;
+ phys_addr_t omt_phys;
+ size_t mem_size = 0;
+ unsigned int order = 0;
+ u32 csd_port_id = 0;
+ unsigned i;
+ /*
+ * enumerate all PAMUs and allocate and setup PAMU tables
+ * for each of them,
+ * NOTE : All PAMUs share the same LIODN tables.
+ */
+
+ pamu_regs = of_iomap(pdev->dev.of_node, 0);
+ if (!pamu_regs) {
+ dev_err(&pdev->dev, "ioremap of PAMU node failed\n");
+ return -ENOMEM;
+ }
+ of_get_address(pdev->dev.of_node, 0, &size, NULL);
+
+ irq = irq_of_parse_and_map(pdev->dev.of_node, 0);
+ if (irq == NO_IRQ) {
+ dev_warn(&pdev->dev, "no interrupts listed in PAMU node\n");
+ goto error;
+ }
+
+ data = kzalloc(sizeof(struct pamu_isr_data), GFP_KERNEL);
+ if (!data) {
+ iounmap(pamu_regs);
+ return -ENOMEM;
+ }
+ data->pamu_reg_base = pamu_regs;
+ data->count = size / PAMU_OFFSET;
+
+ /* The ISR needs access to the regs, so we won't iounmap them */
+ ret = request_irq(irq, pamu_av_isr, 0, "pamu", data);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "error %i installing ISR for irq %i\n",
+ ret, irq);
+ goto error;
+ }
+
+ guts_node = of_find_compatible_node(NULL, NULL,
+ "fsl,qoriq-device-config-1.0");
+ if (!guts_node) {
+ dev_err(&pdev->dev, "could not find GUTS node %s\n",
+ pdev->dev.of_node->full_name);
+ ret = -ENODEV;
+ goto error;
+ }
+
+ guts_regs = of_iomap(guts_node, 0);
+ of_node_put(guts_node);
+ if (!guts_regs) {
+ dev_err(&pdev->dev, "ioremap of GUTS node failed\n");
+ ret = -ENODEV;
+ goto error;
+ }
+
+ /* read in the PAMU capability registers */
+ get_pamu_cap_values((unsigned long)pamu_regs);
+ /*
+ * To simplify the allocation of a coherency domain, we allocate the
+ * PAACT and the OMT in the same memory buffer. Unfortunately, this
+ * wastes more memory compared to allocating the buffers separately.
+ */
+ /* Determine how much memory we need */
+ mem_size = (PAGE_SIZE << get_order(PAACT_SIZE)) +
+ (PAGE_SIZE << get_order(SPAACT_SIZE)) +
+ (PAGE_SIZE << get_order(OMT_SIZE));
+ order = get_order(mem_size);
+
+ p = alloc_pages(GFP_KERNEL | __GFP_ZERO, order);
+ if (!p) {
+ dev_err(&pdev->dev, "unable to allocate PAACT/SPAACT/OMT block\n");
+ ret = -ENOMEM;
+ goto error;
+ }
+
+ ppaact = page_address(p);
+ ppaact_phys = page_to_phys(p);
+
+ /* Make sure the memory is naturally aligned */
+ if (ppaact_phys & ((PAGE_SIZE << order) - 1)) {
+ dev_err(&pdev->dev, "PAACT/OMT block is unaligned\n");
+ ret = -ENOMEM;
+ goto error;
+ }
+
+ spaact = (void *)ppaact + (PAGE_SIZE << get_order(PAACT_SIZE));
+ omt = (void *)spaact + (PAGE_SIZE << get_order(SPAACT_SIZE));
+
+ dev_dbg(&pdev->dev, "ppaact virt=%p phys=0x%llx\n", ppaact,
+ (unsigned long long) ppaact_phys);
+
+ /* Check to see if we need to implement the work-around on this SOC */
+
+ /* Determine the Port ID for our coherence subdomain */
+ for (i = 0; i < ARRAY_SIZE(port_id_map); i++) {
+ if (port_id_map[i].svr == (mfspr(SPRN_SVR) & ~SVR_SECURITY)) {
+ csd_port_id = port_id_map[i].port_id;
+ dev_dbg(&pdev->dev, "found matching SVR %08x\n",
+ port_id_map[i].svr);
+ break;
+ }
+ }
+
+ if (csd_port_id) {
+ dev_dbg(&pdev->dev, "creating coherency subdomain at address "
+ "0x%llx, size %zu, port id 0x%08x", ppaact_phys,
+ mem_size, csd_port_id);
+
+ ret = create_csd(ppaact_phys, mem_size, csd_port_id);
+ if (ret) {
+ dev_err(&pdev->dev, "could not create coherence "
+ "subdomain\n");
+ return ret;
+ }
+ }
+
+ spaact_phys = virt_to_phys(spaact);
+ omt_phys = virt_to_phys(omt);
+
+ spaace_pool = gen_pool_create(ilog2(sizeof(struct paace)), -1);
+ if (!spaace_pool) {
+ ret = -ENOMEM;
+ dev_err(&pdev->dev, "PAMU : failed to allocate spaace gen pool\n");
+ goto error;
+ }
+
+ ret = gen_pool_add(spaace_pool, (unsigned long)spaact, SPAACT_SIZE, -1);
+ if (ret)
+ goto error_genpool;
+
+ pamubypenr = in_be32(&guts_regs->pamubypenr);
+
+ for (pamu_reg_off = 0, pamu_counter = 0x80000000; pamu_reg_off < size;
+ pamu_reg_off += PAMU_OFFSET, pamu_counter >>= 1) {
+
+ pamu_reg_base = (unsigned long) pamu_regs + pamu_reg_off;
+ setup_one_pamu(pamu_reg_base, pamu_reg_off, ppaact_phys,
+ spaact_phys, omt_phys);
+ /* Disable PAMU bypass for this PAMU */
+ pamubypenr &= ~pamu_counter;
+ }
+
+ setup_omt(omt);
+
+ /* Enable all relevant PAMU(s) */
+ out_be32(&guts_regs->pamubypenr, pamubypenr);
+
+ iounmap(guts_regs);
+
+ /* Enable DMA for the LIODNs in the device tree*/
+
+ setup_liodns();
+
+ return 0;
+
+error_genpool:
+ gen_pool_destroy(spaace_pool);
+
+error:
+ if (irq != NO_IRQ)
+ free_irq(irq, 0);
+
+ if (pamu_regs)
+ iounmap(pamu_regs);
+
+ if (guts_regs)
+ iounmap(guts_regs);
+
+ if (ppaact)
+ free_pages((unsigned long)ppaact, order);
+
+ ppaact = NULL;
+
+ return ret;
+}
+
+static const struct of_device_id fsl_of_pamu_ids[] = {
+ {
+ .compatible = "fsl,p4080-pamu",
+ },
+ {
+ .compatible = "fsl,pamu",
+ },
+ {},
+};
+
+static struct platform_driver fsl_of_pamu_driver = {
+ .driver = {
+ .name = "fsl-of-pamu",
+ .owner = THIS_MODULE,
+ },
+ .probe = fsl_pamu_probe,
+};
+
+static __init int fsl_pamu_init(void)
+{
+ struct platform_device *pdev = NULL;
+ struct device_node *np;
+ int ret;
+
+ /*
+ * The normal OF process calls the probe function at some
+ * indeterminate later time, after most drivers have loaded. This is
+ * too late for us, because PAMU clients (like the Qman driver)
+ * depend on PAMU being initialized early.
+ *
+ * So instead, we "manually" call our probe function by creating the
+ * platform devices ourselves.
+ */
+
+ /*
+ * We assume that there is only one PAMU node in the device tree. A
+ * single PAMU node represents all of the PAMU devices in the SOC
+ * already. Everything else already makes that assumption, and the
+ * binding for the PAMU nodes doesn't allow for any parent-child
+ * relationships anyway. In other words, support for more than one
+ * PAMU node would require significant changes to a lot of code.
+ */
+
+ np = of_find_compatible_node(NULL, NULL, "fsl,pamu");
+ if (!np) {
+ pr_err("fsl-pamu: could not find a PAMU node\n");
+ return -ENODEV;
+ }
+
+ ret = platform_driver_register(&fsl_of_pamu_driver);
+ if (ret) {
+ pr_err("fsl-pamu: could not register driver (err=%i)\n", ret);
+ goto error_driver_register;
+ }
+
+ pdev = platform_device_alloc("fsl-of-pamu", 0);
+ if (!pdev) {
+ pr_err("fsl-pamu: could not allocate device %s\n",
+ np->full_name);
+ ret = -ENOMEM;
+ goto error_device_alloc;
+ }
+ pdev->dev.of_node = of_node_get(np);
+
+ ret = pamu_domain_init();
+ if (ret)
+ goto error_device_add;
+
+ ret = platform_device_add(pdev);
+ if (ret) {
+ pr_err("fsl-pamu: could not add device %s (err=%i)\n",
+ np->full_name, ret);
+ goto error_device_add;
+ }
+
+ return 0;
+
+error_device_add:
+ of_node_put(pdev->dev.of_node);
+ pdev->dev.of_node = NULL;
+
+ platform_device_put(pdev);
+
+error_device_alloc:
+ platform_driver_unregister(&fsl_of_pamu_driver);
+
+error_driver_register:
+ of_node_put(np);
+
+ return ret;
+}
+arch_initcall(fsl_pamu_init);
diff --git a/drivers/iommu/fsl_pamu.h b/drivers/iommu/fsl_pamu.h
new file mode 100644
index 0000000..83cbd26
--- /dev/null
+++ b/drivers/iommu/fsl_pamu.h
@@ -0,0 +1,405 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ *
+ */
+
+#ifndef __FSL_PAMU_H
+#define __FSL_PAMU_H
+
+/* Bit Field macros
+ * v = bit field variable; m = mask, m##_SHIFT = shift, x = value to load
+ */
+#define set_bf(v, m, x) (v = ((v) & ~(m)) | (((x) << (m##_SHIFT)) & (m)))
+#define get_bf(v, m) (((v) & (m)) >> (m##_SHIFT))
+
+/* PAMU CCSR space */
+#define PAMU_PGC 0x00000000 /* Allows all peripheral accesses */
+#define PAMU_PE 0x40000000 /* enable PAMU */
+
+/* PAMU_OFFSET to the next pamu space in ccsr */
+#define PAMU_OFFSET 0x1000
+
+#define PAMU_MMAP_REGS_BASE 0
+
+struct pamu_mmap_regs {
+ u32 ppbah;
+ u32 ppbal;
+ u32 pplah;
+ u32 pplal;
+ u32 spbah;
+ u32 spbal;
+ u32 splah;
+ u32 splal;
+ u32 obah;
+ u32 obal;
+ u32 olah;
+ u32 olal;
+};
+
+/* PAMU Error Registers */
+#define PAMU_POES1 0x0040
+#define PAMU_POES2 0x0044
+#define PAMU_POEAH 0x0048
+#define PAMU_POEAL 0x004C
+#define PAMU_AVS1 0x0050
+#define PAMU_AVS1_AV 0x1
+#define PAMU_AVS1_OTV 0x6
+#define PAMU_AVS1_APV 0x78
+#define PAMU_AVS1_WAV 0x380
+#define PAMU_AVS1_LAV 0x1c00
+#define PAMU_AVS1_GCV 0x2000
+#define PAMU_AVS1_PDV 0x4000
+#define PAMU_AV_MASK (PAMU_AVS1_AV | PAMU_AVS1_OTV | PAMU_AVS1_APV | PAMU_AVS1_WAV \
+ | PAMU_AVS1_LAV | PAMU_AVS1_GCV | PAMU_AVS1_PDV)
+#define PAMU_AVS1_LIODN_SHIFT 16
+#define PAMU_LAV_LIODN_NOT_IN_PPAACT 0x400
+
+#define PAMU_AVS2 0x0054
+#define PAMU_AVAH 0x0058
+#define PAMU_AVAL 0x005C
+#define PAMU_EECTL 0x0060
+#define PAMU_EEDIS 0x0064
+#define PAMU_EEINTEN 0x0068
+#define PAMU_EEDET 0x006C
+#define PAMU_EEATTR 0x0070
+#define PAMU_EEAHI 0x0074
+#define PAMU_EEALO 0x0078
+#define PAMU_EEDHI 0X007C
+#define PAMU_EEDLO 0x0080
+#define PAMU_EECC 0x0084
+#define PAMU_UDAD 0x0090
+
+/* PAMU Revision Registers */
+#define PAMU_PR1 0x0BF8
+#define PAMU_PR2 0x0BFC
+
+/* PAMU Capabilities Registers */
+#define PAMU_PC1 0x0C00
+#define PAMU_PC2 0x0C04
+#define PAMU_PC3 0x0C08
+#define PAMU_PC4 0x0C0C
+
+/* PAMU Control Register */
+#define PAMU_PC 0x0C10
+
+/* PAMU control defs */
+#define PAMU_CONTROL 0x0C10
+#define PAMU_PC_PGC 0x80000000 /* PAMU gate closed bit */
+#define PAMU_PC_PE 0x40000000 /* PAMU enable bit */
+#define PAMU_PC_SPCC 0x00000010 /* sPAACE cache enable */
+#define PAMU_PC_PPCC 0x00000001 /* pPAACE cache enable */
+#define PAMU_PC_OCE 0x00001000 /* OMT cache enable */
+
+#define PAMU_PFA1 0x0C14
+#define PAMU_PFA2 0x0C18
+
+#define PAMU_PC2_MLIODN(X) ((X) >> 16)
+#define PAMU_PC3_MWCE(X) (((X) >> 21) & 0xf)
+
+/* PAMU Interrupt control and Status Register */
+#define PAMU_PICS 0x0C1C
+#define PAMU_ACCESS_VIOLATION_STAT 0x8
+#define PAMU_ACCESS_VIOLATION_ENABLE 0x4
+
+/* PAMU Debug Registers */
+#define PAMU_PD1 0x0F00
+#define PAMU_PD2 0x0F04
+#define PAMU_PD3 0x0F08
+#define PAMU_PD4 0x0F0C
+
+#define PAACE_AP_PERMS_DENIED 0x0
+#define PAACE_AP_PERMS_QUERY 0x1
+#define PAACE_AP_PERMS_UPDATE 0x2
+#define PAACE_AP_PERMS_ALL 0x3
+
+#define PAACE_DD_TO_HOST 0x0
+#define PAACE_DD_TO_IO 0x1
+#define PAACE_PT_PRIMARY 0x0
+#define PAACE_PT_SECONDARY 0x1
+#define PAACE_V_INVALID 0x0
+#define PAACE_V_VALID 0x1
+#define PAACE_MW_SUBWINDOWS 0x1
+
+#define PAACE_WSE_4K 0xB
+#define PAACE_WSE_8K 0xC
+#define PAACE_WSE_16K 0xD
+#define PAACE_WSE_32K 0xE
+#define PAACE_WSE_64K 0xF
+#define PAACE_WSE_128K 0x10
+#define PAACE_WSE_256K 0x11
+#define PAACE_WSE_512K 0x12
+#define PAACE_WSE_1M 0x13
+#define PAACE_WSE_2M 0x14
+#define PAACE_WSE_4M 0x15
+#define PAACE_WSE_8M 0x16
+#define PAACE_WSE_16M 0x17
+#define PAACE_WSE_32M 0x18
+#define PAACE_WSE_64M 0x19
+#define PAACE_WSE_128M 0x1A
+#define PAACE_WSE_256M 0x1B
+#define PAACE_WSE_512M 0x1C
+#define PAACE_WSE_1G 0x1D
+#define PAACE_WSE_2G 0x1E
+#define PAACE_WSE_4G 0x1F
+
+#define PAACE_DID_PCI_EXPRESS_1 0x00
+#define PAACE_DID_PCI_EXPRESS_2 0x01
+#define PAACE_DID_PCI_EXPRESS_3 0x02
+#define PAACE_DID_PCI_EXPRESS_4 0x03
+#define PAACE_DID_LOCAL_BUS 0x04
+#define PAACE_DID_SRIO 0x0C
+#define PAACE_DID_MEM_1 0x10
+#define PAACE_DID_MEM_2 0x11
+#define PAACE_DID_MEM_3 0x12
+#define PAACE_DID_MEM_4 0x13
+#define PAACE_DID_MEM_1_2 0x14
+#define PAACE_DID_MEM_3_4 0x15
+#define PAACE_DID_MEM_1_4 0x16
+#define PAACE_DID_BM_SW_PORTAL 0x18
+#define PAACE_DID_PAMU 0x1C
+#define PAACE_DID_CAAM 0x21
+#define PAACE_DID_QM_SW_PORTAL 0x3C
+#define PAACE_DID_CORE0_INST 0x80
+#define PAACE_DID_CORE0_DATA 0x81
+#define PAACE_DID_CORE1_INST 0x82
+#define PAACE_DID_CORE1_DATA 0x83
+#define PAACE_DID_CORE2_INST 0x84
+#define PAACE_DID_CORE2_DATA 0x85
+#define PAACE_DID_CORE3_INST 0x86
+#define PAACE_DID_CORE3_DATA 0x87
+#define PAACE_DID_CORE4_INST 0x88
+#define PAACE_DID_CORE4_DATA 0x89
+#define PAACE_DID_CORE5_INST 0x8A
+#define PAACE_DID_CORE5_DATA 0x8B
+#define PAACE_DID_CORE6_INST 0x8C
+#define PAACE_DID_CORE6_DATA 0x8D
+#define PAACE_DID_CORE7_INST 0x8E
+#define PAACE_DID_CORE7_DATA 0x8F
+#define PAACE_DID_BROADCAST 0xFF
+
+#define PAACE_ATM_NO_XLATE 0x00
+#define PAACE_ATM_WINDOW_XLATE 0x01
+#define PAACE_ATM_PAGE_XLATE 0x02
+#define PAACE_ATM_WIN_PG_XLATE \
+ (PAACE_ATM_WINDOW_XLATE | PAACE_ATM_PAGE_XLATE)
+#define PAACE_OTM_NO_XLATE 0x00
+#define PAACE_OTM_IMMEDIATE 0x01
+#define PAACE_OTM_INDEXED 0x02
+#define PAACE_OTM_RESERVED 0x03
+
+#define PAACE_M_COHERENCE_REQ 0x01
+
+#define PAACE_PID_0 0x0
+#define PAACE_PID_1 0x1
+#define PAACE_PID_2 0x2
+#define PAACE_PID_3 0x3
+#define PAACE_PID_4 0x4
+#define PAACE_PID_5 0x5
+#define PAACE_PID_6 0x6
+#define PAACE_PID_7 0x7
+
+#define PAACE_TCEF_FORMAT0_8B 0x00
+#define PAACE_TCEF_FORMAT1_RSVD 0x01
+/*
+ * Hard coded value for the PAACT size to accomodate
+ * maximum LIODN value generated by u-boot.
+ */
+#define PAACE_NUMBER_ENTRIES 0x500
+/* Hard coded value for the SPAACT size */
+#define SPAACE_NUMBER_ENTRIES 0x800
+
+#define OME_NUMBER_ENTRIES 16
+
+/* PAACE Bit Field Defines */
+#define PPAACE_AF_WBAL 0xfffff000
+#define PPAACE_AF_WBAL_SHIFT 12
+#define PPAACE_AF_WSE 0x00000fc0
+#define PPAACE_AF_WSE_SHIFT 6
+#define PPAACE_AF_MW 0x00000020
+#define PPAACE_AF_MW_SHIFT 5
+
+#define SPAACE_AF_LIODN 0xffff0000
+#define SPAACE_AF_LIODN_SHIFT 16
+
+#define PAACE_AF_AP 0x00000018
+#define PAACE_AF_AP_SHIFT 3
+#define PAACE_AF_DD 0x00000004
+#define PAACE_AF_DD_SHIFT 2
+#define PAACE_AF_PT 0x00000002
+#define PAACE_AF_PT_SHIFT 1
+#define PAACE_AF_V 0x00000001
+#define PAACE_AF_V_SHIFT 0
+
+#define PAACE_DA_HOST_CR 0x80
+#define PAACE_DA_HOST_CR_SHIFT 7
+
+#define PAACE_IA_CID 0x00FF0000
+#define PAACE_IA_CID_SHIFT 16
+#define PAACE_IA_WCE 0x000000F0
+#define PAACE_IA_WCE_SHIFT 4
+#define PAACE_IA_ATM 0x0000000C
+#define PAACE_IA_ATM_SHIFT 2
+#define PAACE_IA_OTM 0x00000003
+#define PAACE_IA_OTM_SHIFT 0
+
+#define PAACE_WIN_TWBAL 0xfffff000
+#define PAACE_WIN_TWBAL_SHIFT 12
+#define PAACE_WIN_SWSE 0x00000fc0
+#define PAACE_WIN_SWSE_SHIFT 6
+
+/* PAMU Data Structures */
+/* primary / secondary paact structure */
+struct paace {
+ /* PAACE Offset 0x00 */
+ u32 wbah; /* only valid for Primary PAACE */
+ u32 addr_bitfields; /* See P/S PAACE_AF_* */
+
+ /* PAACE Offset 0x08 */
+ /* Interpretation of first 32 bits dependent on DD above */
+ union {
+ struct {
+ /* Destination ID, see PAACE_DID_* defines */
+ u8 did;
+ /* Partition ID */
+ u8 pid;
+ /* Snoop ID */
+ u8 snpid;
+ /* coherency_required : 1 reserved : 7 */
+ u8 coherency_required; /* See PAACE_DA_* */
+ } to_host;
+ struct {
+ /* Destination ID, see PAACE_DID_* defines */
+ u8 did;
+ u8 reserved1;
+ u16 reserved2;
+ } to_io;
+ } domain_attr;
+
+ /* Implementation attributes + window count + address & operation translation modes */
+ u32 impl_attr; /* See PAACE_IA_* */
+
+ /* PAACE Offset 0x10 */
+ /* Translated window base address */
+ u32 twbah;
+ u32 win_bitfields; /* See PAACE_WIN_* */
+
+ /* PAACE Offset 0x18 */
+ /* first secondary paace entry */
+ u32 fspi; /* only valid for Primary PAACE */
+ union {
+ struct {
+ u8 ioea;
+ u8 moea;
+ u8 ioeb;
+ u8 moeb;
+ } immed_ot;
+ struct {
+ u16 reserved;
+ u16 omi;
+ } index_ot;
+ } op_encode;
+
+ /* PAACE Offsets 0x20-0x38 */
+ u32 reserved[8]; /* not currently implemented */
+};
+
+/* OME : Operation mapping entry
+ * MOE : Mapped Operation Encodings
+ * The operation mapping table is table containing operation mapping entries (OME).
+ * The index of a particular OME is programmed in the PAACE entry for translation
+ * in bound I/O operations corresponding to an LIODN. The OMT is used for translation
+ * specifically in case of the indexed translation mode. Each OME contains a 128
+ * byte mapped operation encoding (MOE), where each byte represents an MOE.
+ */
+#define NUM_MOE 128
+struct ome {
+ u8 moe[NUM_MOE];
+} __attribute__((packed));
+
+#define PAACT_SIZE (sizeof(struct paace) * PAACE_NUMBER_ENTRIES)
+#define SPAACT_SIZE (sizeof(struct paace) * SPAACE_NUMBER_ENTRIES)
+#define OMT_SIZE (sizeof(struct ome) * OME_NUMBER_ENTRIES)
+
+#define PAMU_PAGE_SHIFT 12
+#define PAMU_PAGE_SIZE 4096ULL
+
+#define IOE_READ 0x00
+#define IOE_READ_IDX 0x00
+#define IOE_WRITE 0x81
+#define IOE_WRITE_IDX 0x01
+#define IOE_EREAD0 0x82 /* Enhanced read type 0 */
+#define IOE_EREAD0_IDX 0x02 /* Enhanced read type 0 */
+#define IOE_EWRITE0 0x83 /* Enhanced write type 0 */
+#define IOE_EWRITE0_IDX 0x03 /* Enhanced write type 0 */
+#define IOE_DIRECT0 0x84 /* Directive type 0 */
+#define IOE_DIRECT0_IDX 0x04 /* Directive type 0 */
+#define IOE_EREAD1 0x85 /* Enhanced read type 1 */
+#define IOE_EREAD1_IDX 0x05 /* Enhanced read type 1 */
+#define IOE_EWRITE1 0x86 /* Enhanced write type 1 */
+#define IOE_EWRITE1_IDX 0x06 /* Enhanced write type 1 */
+#define IOE_DIRECT1 0x87 /* Directive type 1 */
+#define IOE_DIRECT1_IDX 0x07 /* Directive type 1 */
+#define IOE_RAC 0x8c /* Read with Atomic clear */
+#define IOE_RAC_IDX 0x0c /* Read with Atomic clear */
+#define IOE_RAS 0x8d /* Read with Atomic set */
+#define IOE_RAS_IDX 0x0d /* Read with Atomic set */
+#define IOE_RAD 0x8e /* Read with Atomic decrement */
+#define IOE_RAD_IDX 0x0e /* Read with Atomic decrement */
+#define IOE_RAI 0x8f /* Read with Atomic increment */
+#define IOE_RAI_IDX 0x0f /* Read with Atomic increment */
+
+#define EOE_READ 0x00
+#define EOE_WRITE 0x01
+#define EOE_RAC 0x0c /* Read with Atomic clear */
+#define EOE_RAS 0x0d /* Read with Atomic set */
+#define EOE_RAD 0x0e /* Read with Atomic decrement */
+#define EOE_RAI 0x0f /* Read with Atomic increment */
+#define EOE_LDEC 0x10 /* Load external cache */
+#define EOE_LDECL 0x11 /* Load external cache with stash lock */
+#define EOE_LDECPE 0x12 /* Load external cache with preferred exclusive */
+#define EOE_LDECPEL 0x13 /* Load external cache with preferred exclusive and lock */
+#define EOE_LDECFE 0x14 /* Load external cache with forced exclusive */
+#define EOE_LDECFEL 0x15 /* Load external cache with forced exclusive and lock */
+#define EOE_RSA 0x16 /* Read with stash allocate */
+#define EOE_RSAU 0x17 /* Read with stash allocate and unlock */
+#define EOE_READI 0x18 /* Read with invalidate */
+#define EOE_RWNITC 0x19 /* Read with no intention to cache */
+#define EOE_WCI 0x1a /* Write cache inhibited */
+#define EOE_WWSA 0x1b /* Write with stash allocate */
+#define EOE_WWSAL 0x1c /* Write with stash allocate and lock */
+#define EOE_WWSAO 0x1d /* Write with stash allocate only */
+#define EOE_WWSAOL 0x1e /* Write with stash allocate only and lock */
+#define EOE_VALID 0x80
+
+/* Function prototypes */
+int pamu_domain_init(void);
+int pamu_enable_liodn(int liodn);
+int pamu_disable_liodn(int liodn);
+void pamu_free_subwins(int liodn);
+int pamu_config_ppaace(int liodn, phys_addr_t win_addr, phys_addr_t win_size,
+ u32 omi, unsigned long rpn, u32 snoopid, uint32_t stashid,
+ u32 subwin_cnt, int prot);
+int pamu_config_spaace(int liodn, u32 subwin_cnt, u32 subwin_addr,
+ phys_addr_t subwin_size, u32 omi, unsigned long rpn,
+ uint32_t snoopid, u32 stashid, int enable, int prot);
+
+u32 get_stash_id(u32 stash_dest_hint, u32 vcpu);
+void get_ome_index(u32 *omi_index, struct device *dev);
+int pamu_update_paace_stash(int liodn, u32 subwin, u32 value);
+int pamu_disable_spaace(int liodn, u32 subwin);
+u32 pamu_get_max_subwin_cnt(void);
+
+#endif /* __FSL_PAMU_H */
diff --git a/drivers/iommu/fsl_pamu_domain.c b/drivers/iommu/fsl_pamu_domain.c
new file mode 100644
index 0000000..60df7c3
--- /dev/null
+++ b/drivers/iommu/fsl_pamu_domain.c
@@ -0,0 +1,1134 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ * Author: Varun Sethi <varun.sethi@freescale.com>
+ *
+ */
+
+#define pr_fmt(fmt) "fsl-pamu-domain: %s: " fmt, __func__
+
+#include <linux/init.h>
+#include <linux/iommu.h>
+#include <linux/notifier.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/device.h>
+#include <linux/of_platform.h>
+#include <linux/bootmem.h>
+#include <linux/err.h>
+#include <asm/io.h>
+#include <asm/bitops.h>
+
+#include <asm/pci-bridge.h>
+#include <sysdev/fsl_pci.h>
+
+#include "fsl_pamu_domain.h"
+
+/*
+ * Global spinlock that needs to be held while
+ * configuring PAMU.
+ */
+static DEFINE_SPINLOCK(iommu_lock);
+
+static struct kmem_cache *fsl_pamu_domain_cache;
+static struct kmem_cache *iommu_devinfo_cache;
+static DEFINE_SPINLOCK(device_domain_lock);
+
+int __init iommu_init_mempool(void)
+{
+
+ fsl_pamu_domain_cache = kmem_cache_create("fsl_pamu_domain",
+ sizeof(struct fsl_dma_domain),
+ 0,
+ SLAB_HWCACHE_ALIGN,
+
+ NULL);
+ if (!fsl_pamu_domain_cache) {
+ pr_err("Couldn't create fsl iommu_domain cache\n");
+ return -ENOMEM;
+ }
+
+ iommu_devinfo_cache = kmem_cache_create("iommu_devinfo",
+ sizeof(struct device_domain_info),
+ 0,
+ SLAB_HWCACHE_ALIGN,
+ NULL);
+ if (!iommu_devinfo_cache) {
+ pr_err("Couldn't create devinfo cache\n");
+ kmem_cache_destroy(fsl_pamu_domain_cache);
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static phys_addr_t get_phys_addr(struct fsl_dma_domain *dma_domain, u64 iova)
+{
+ u32 win_cnt = dma_domain->win_cnt;
+ struct dma_window *win_ptr =
+ &dma_domain->win_arr[0];
+ struct iommu_domain_geometry *geom;
+
+ geom = &dma_domain->iommu_domain->geometry;
+
+ if (!win_cnt || !dma_domain->geom_size) {
+ pr_err("Number of windows/geometry not configured for the domain\n");
+ return 0;
+ }
+
+ if (win_cnt > 1) {
+ u64 subwin_size;
+ u64 subwin_iova;
+ u32 wnd;
+
+ subwin_size = dma_domain->geom_size >> ilog2(win_cnt);
+ subwin_iova = iova & ~(subwin_size - 1);
+ wnd = (subwin_iova - geom->aperture_start) >> ilog2(subwin_size);
+ win_ptr = &dma_domain->win_arr[wnd];
+ }
+
+ if (win_ptr->valid)
+ return (win_ptr->paddr + (iova & (win_ptr->size - 1)));
+
+ return 0;
+}
+
+static int map_subwins(int liodn, struct fsl_dma_domain *dma_domain)
+{
+ struct dma_window *sub_win_ptr =
+ &dma_domain->win_arr[0];
+ int i, ret;
+ unsigned long rpn;
+
+ for (i = 0; i < dma_domain->win_cnt; i++) {
+ if (sub_win_ptr[i].valid) {
+ rpn = sub_win_ptr[i].paddr >>
+ PAMU_PAGE_SHIFT;
+ spin_lock(&iommu_lock);
+ ret = pamu_config_spaace(liodn, dma_domain->win_cnt, i,
+ sub_win_ptr[i].size,
+ ~(u32)0,
+ rpn,
+ dma_domain->snoop_id,
+ dma_domain->stash_id,
+ (i > 0) ? 1 : 0,
+ sub_win_ptr[i].prot);
+ spin_unlock(&iommu_lock);
+ if (ret) {
+ pr_err("PAMU SPAACE configuration failed for liodn %d\n",
+ liodn);
+ return ret;
+ }
+ }
+ }
+
+ return ret;
+}
+
+static int map_win(int liodn, struct fsl_dma_domain *dma_domain)
+{
+ int ret;
+ struct dma_window *wnd = &dma_domain->win_arr[0];
+ phys_addr_t wnd_addr = dma_domain->iommu_domain->geometry.aperture_start;
+
+ spin_lock(&iommu_lock);
+ ret = pamu_config_ppaace(liodn, wnd_addr,
+ wnd->size,
+ ~(u32)0,
+ wnd->paddr >> PAMU_PAGE_SHIFT,
+ dma_domain->snoop_id, dma_domain->stash_id,
+ 0, wnd->prot);
+ spin_unlock(&iommu_lock);
+ if (ret)
+ pr_err("PAMU PAACE configuration failed for liodn %d\n",
+ liodn);
+
+ return ret;
+}
+
+/* Map the DMA window corresponding to the LIODN */
+static int map_liodn(int liodn, struct fsl_dma_domain *dma_domain)
+{
+ if (dma_domain->win_cnt > 1)
+ return map_subwins(liodn, dma_domain);
+ else
+ return map_win(liodn, dma_domain);
+
+}
+
+/* Update window/subwindow mapping for the LIODN */
+static int update_liodn(int liodn, struct fsl_dma_domain *dma_domain, u32 wnd_nr)
+{
+ int ret;
+ struct dma_window *wnd = &dma_domain->win_arr[wnd_nr];
+
+ spin_lock(&iommu_lock);
+ if (dma_domain->win_cnt > 1) {
+ ret = pamu_config_spaace(liodn, dma_domain->win_cnt, wnd_nr,
+ wnd->size,
+ ~(u32)0,
+ wnd->paddr >> PAMU_PAGE_SHIFT,
+ dma_domain->snoop_id,
+ dma_domain->stash_id,
+ (wnd_nr > 0) ? 1 : 0,
+ wnd->prot);
+ if (ret)
+ pr_err("Subwindow reconfiguration failed for liodn %d\n", liodn);
+ } else {
+ phys_addr_t wnd_addr;
+
+ wnd_addr = dma_domain->iommu_domain->geometry.aperture_start;
+
+ ret = pamu_config_ppaace(liodn, wnd_addr,
+ wnd->size,
+ ~(u32)0,
+ wnd->paddr >> PAMU_PAGE_SHIFT,
+ dma_domain->snoop_id, dma_domain->stash_id,
+ 0, wnd->prot);
+ if (ret)
+ pr_err("Window reconfiguration failed for liodn %d\n", liodn);
+ }
+
+ spin_unlock(&iommu_lock);
+
+ return ret;
+}
+
+static int update_liodn_stash(int liodn, struct fsl_dma_domain *dma_domain,
+ u32 val)
+{
+ int ret = 0, i;
+
+ spin_lock(&iommu_lock);
+ if (!dma_domain->win_cnt) {
+ ret = pamu_update_paace_stash(liodn, 0, val);
+ if (ret) {
+ pr_err("Failed to update PAACE field for liodn %d\n ", liodn);
+ spin_unlock(&iommu_lock);
+ return ret;
+ }
+ } else {
+ for (i = 0; i < dma_domain->win_cnt; i++) {
+ ret = pamu_update_paace_stash(liodn, i, val);
+ if (ret) {
+ pr_err("Failed to update SPAACE %d field for liodn %d\n ", i, liodn);
+ spin_unlock(&iommu_lock);
+ return ret;
+ }
+ }
+ }
+ spin_unlock(&iommu_lock);
+
+ return ret;
+}
+
+/* Set the geometry parameters for a LIODN */
+static int pamu_set_liodn(int liodn, struct device *dev,
+ struct fsl_dma_domain *dma_domain,
+ struct iommu_domain_geometry *geom_attr,
+ u32 win_cnt)
+{
+ phys_addr_t window_addr, window_size;
+ phys_addr_t subwin_size;
+ int ret = 0, i;
+ u32 omi_index = ~(u32)0;
+
+ /*
+ * Configure the omi_index at the geometry setup time.
+ * This is a static value which depends on the type of
+ * device and would not change thereafter.
+ */
+ get_ome_index(&omi_index, dev);
+
+ window_addr = geom_attr->aperture_start;
+ window_size = dma_domain->geom_size;
+
+ spin_lock(&iommu_lock);
+ ret = pamu_disable_liodn(liodn);
+ if (!ret)
+ ret = pamu_config_ppaace(liodn, window_addr, window_size, omi_index,
+ 0, dma_domain->snoop_id,
+ dma_domain->stash_id, win_cnt, 0);
+ spin_unlock(&iommu_lock);
+ if (ret) {
+ pr_err("PAMU PAACE configuration failed for liodn %d, win_cnt =%d\n", liodn, win_cnt);
+ return ret;
+ }
+
+ if (win_cnt > 1) {
+ subwin_size = window_size >> ilog2(win_cnt);
+ for (i = 0; i < win_cnt; i++) {
+ spin_lock(&iommu_lock);
+ ret = pamu_config_spaace(liodn, win_cnt, i, subwin_size,
+ omi_index, 0,
+ dma_domain->snoop_id,
+ dma_domain->stash_id, 0, 0);
+ spin_unlock(&iommu_lock);
+ if (ret) {
+ pr_err("PAMU SPAACE configuration failed for liodn %d\n", liodn);
+ return ret;
+ }
+ }
+ }
+
+ return ret;
+}
+
+static int check_size(u64 size, unsigned long iova)
+{
+ /*
+ * Size must be a power of two and at least be equal
+ * to PAMU page size.
+ */
+ if ((size & (size - 1)) || size < PAMU_PAGE_SIZE) {
+ pr_err("%s: size too small or not a power of two\n", __func__);
+ return -EINVAL;
+ }
+
+ /* iova must be page size aligned*/
+ if (iova & (size - 1)) {
+ pr_err("%s: address is not aligned with window size\n", __func__);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static struct fsl_dma_domain *iommu_alloc_dma_domain(void)
+{
+ struct fsl_dma_domain *domain;
+
+ domain = kmem_cache_zalloc(fsl_pamu_domain_cache, GFP_KERNEL);
+ if (!domain)
+ return NULL;
+
+ domain->stash_id = ~(u32)0;
+ domain->snoop_id = ~(u32)0;
+ domain->win_cnt = pamu_get_max_subwin_cnt();
+ domain->geom_size = 0;
+
+ INIT_LIST_HEAD(&domain->devices);
+
+ spin_lock_init(&domain->domain_lock);
+
+ return domain;
+}
+
+static inline struct device_domain_info *find_domain(struct device *dev)
+{
+ return dev->archdata.iommu_domain;
+}
+
+static void remove_device_ref(struct device_domain_info *info, u32 win_cnt)
+{
+ list_del(&info->link);
+ spin_lock(&iommu_lock);
+ if (win_cnt > 1)
+ pamu_free_subwins(info->liodn);
+ pamu_disable_liodn(info->liodn);
+ spin_unlock(&iommu_lock);
+ spin_lock(&device_domain_lock);
+ info->dev->archdata.iommu_domain = NULL;
+ kmem_cache_free(iommu_devinfo_cache, info);
+ spin_unlock(&device_domain_lock);
+}
+
+static void detach_device(struct device *dev, struct fsl_dma_domain *dma_domain)
+{
+ struct device_domain_info *info;
+ struct list_head *entry, *tmp;
+ unsigned long flags;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ /* Remove the device from the domain device list */
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_safe(entry, tmp, &dma_domain->devices) {
+ info = list_entry(entry, struct device_domain_info, link);
+ if (!dev || (info->dev == dev))
+ remove_device_ref(info, dma_domain->win_cnt);
+ }
+ }
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+}
+
+static void attach_device(struct fsl_dma_domain *dma_domain, int liodn, struct device *dev)
+{
+ struct device_domain_info *info, *old_domain_info;
+
+ spin_lock(&device_domain_lock);
+ /*
+ * Check here if the device is already attached to domain or not.
+ * If the device is already attached to a domain detach it.
+ */
+ old_domain_info = find_domain(dev);
+ if (old_domain_info && old_domain_info->domain != dma_domain) {
+ spin_unlock(&device_domain_lock);
+ detach_device(dev, old_domain_info->domain);
+ spin_lock(&device_domain_lock);
+ }
+
+ info = kmem_cache_zalloc(iommu_devinfo_cache, GFP_KERNEL);
+
+ info->dev = dev;
+ info->liodn = liodn;
+ info->domain = dma_domain;
+
+ list_add(&info->link, &dma_domain->devices);
+ /*
+ * In case of devices with multiple LIODNs just store
+ * the info for the first LIODN as all
+ * LIODNs share the same domain
+ */
+ if (!old_domain_info)
+ dev->archdata.iommu_domain = info;
+ spin_unlock(&device_domain_lock);
+
+}
+
+static phys_addr_t fsl_pamu_iova_to_phys(struct iommu_domain *domain,
+ u64 iova)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+
+ if ((iova < domain->geometry.aperture_start) ||
+ iova > (domain->geometry.aperture_end))
+ return 0;
+
+ return get_phys_addr(dma_domain, iova);
+}
+
+static int fsl_pamu_domain_has_cap(struct iommu_domain *domain,
+ unsigned long cap)
+{
+ return cap == IOMMU_CAP_CACHE_COHERENCY;
+}
+
+static void fsl_pamu_domain_destroy(struct iommu_domain *domain)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+
+ domain->priv = NULL;
+
+ /* remove all the devices from the device list */
+ detach_device(NULL, dma_domain);
+
+ dma_domain->enabled = 0;
+ dma_domain->mapped = 0;
+
+ kmem_cache_free(fsl_pamu_domain_cache, dma_domain);
+}
+
+static int fsl_pamu_domain_init(struct iommu_domain *domain)
+{
+ struct fsl_dma_domain *dma_domain;
+
+ dma_domain = iommu_alloc_dma_domain();
+ if (!dma_domain) {
+ pr_err("dma_domain allocation failed\n");
+ return -ENOMEM;
+ }
+ domain->priv = dma_domain;
+ dma_domain->iommu_domain = domain;
+ /* defaul geometry 64 GB i.e. maximum system address */
+ domain->geometry.aperture_start = 0;
+ domain->geometry.aperture_end = 1ULL << 36;
+ domain->geometry.force_aperture = true;
+
+ return 0;
+}
+
+/* Configure geometry settings for all LIODNs associated with domain */
+static int pamu_set_domain_geometry(struct fsl_dma_domain *dma_domain,
+ struct iommu_domain_geometry *geom_attr,
+ u32 win_cnt)
+{
+ struct device_domain_info *info;
+ int ret = 0;
+
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_entry(info, &dma_domain->devices, link) {
+ ret = pamu_set_liodn(info->liodn, info->dev, dma_domain,
+ geom_attr, win_cnt);
+ if (ret)
+ break;
+ }
+ }
+
+ return ret;
+}
+
+/* Update stash destination for all LIODNs associated with the domain */
+static int update_domain_stash(struct fsl_dma_domain *dma_domain, u32 val)
+{
+ struct device_domain_info *info;
+ int ret = 0;
+
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_entry(info, &dma_domain->devices, link) {
+ ret = update_liodn_stash(info->liodn, dma_domain, val);
+ if (ret)
+ break;
+ }
+ }
+
+ return ret;
+}
+
+/* Update domain mappings for all LIODNs associated with the domain */
+static int update_domain_mapping(struct fsl_dma_domain *dma_domain, u32 wnd_nr)
+{
+ struct device_domain_info *info;
+ int ret = 0;
+
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_entry(info, &dma_domain->devices, link) {
+ ret = update_liodn(info->liodn, dma_domain, wnd_nr);
+ if (ret)
+ break;
+ }
+ }
+ return ret;
+}
+
+static int disable_domain_win(struct fsl_dma_domain *dma_domain, u32 wnd_nr)
+{
+ struct device_domain_info *info;
+ int ret = 0;
+
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_entry(info, &dma_domain->devices, link) {
+ if (dma_domain->win_cnt == 1 && dma_domain->enabled) {
+ ret = pamu_disable_liodn(info->liodn);
+ if (!ret)
+ dma_domain->enabled = 0;
+ } else {
+ ret = pamu_disable_spaace(info->liodn, wnd_nr);
+ }
+ }
+ }
+
+ return ret;
+}
+
+static void fsl_pamu_window_disable(struct iommu_domain *domain, u32 wnd_nr)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ if (!dma_domain->win_arr) {
+ pr_err("Number of windows not configured\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return;
+ }
+
+ if (wnd_nr >= dma_domain->win_cnt) {
+ pr_err("Invalid window index\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return;
+ }
+
+ if (dma_domain->win_arr[wnd_nr].valid) {
+ ret = disable_domain_win(dma_domain, wnd_nr);
+ if (!ret) {
+ dma_domain->win_arr[wnd_nr].valid = 0;
+ dma_domain->mapped--;
+ }
+ }
+
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+}
+
+static int fsl_pamu_window_enable(struct iommu_domain *domain, u32 wnd_nr,
+ phys_addr_t paddr, u64 size, int prot)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ struct dma_window *wnd;
+ int pamu_prot = 0;
+ int ret;
+ unsigned long flags;
+ u64 win_size;
+
+ if (prot & IOMMU_READ)
+ pamu_prot |= PAACE_AP_PERMS_QUERY;
+ if (prot & IOMMU_WRITE)
+ pamu_prot |= PAACE_AP_PERMS_UPDATE;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ if (!dma_domain->win_arr) {
+ pr_err("Number of windows not configured\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -ENODEV;
+ }
+
+ if (wnd_nr >= dma_domain->win_cnt) {
+ pr_err("Invalid window index\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+
+ win_size = dma_domain->geom_size >> ilog2(dma_domain->win_cnt);
+ if (size > win_size) {
+ pr_err("Invalid window size \n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+
+ if (dma_domain->win_cnt == 1) {
+ if (dma_domain->enabled) {
+ pr_err("Disable the window before updating the mapping\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EBUSY;
+ }
+
+ ret = check_size(size, domain->geometry.aperture_start);
+ if (ret) {
+ pr_err("Aperture start not aligned to the size\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+ }
+
+ wnd = &dma_domain->win_arr[wnd_nr];
+ if (!wnd->valid) {
+ wnd->paddr = paddr;
+ wnd->size = size;
+ wnd->prot = pamu_prot;
+
+ ret = update_domain_mapping(dma_domain, wnd_nr);
+ if (!ret) {
+ wnd->valid = 1;
+ dma_domain->mapped++;
+ }
+ } else {
+ pr_err("Disable the window before updating the mapping\n");
+ ret = -EBUSY;
+ }
+
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return ret;
+}
+
+/*
+ * Attach the LIODN to the DMA domain and configure the geometry
+ * and window mappings.
+ */
+static int handle_attach_device(struct fsl_dma_domain *dma_domain,
+ struct device *dev, const u32 *liodn,
+ int num)
+{
+ unsigned long flags;
+ struct iommu_domain *domain = dma_domain->iommu_domain;
+ int ret = 0;
+ int i;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ for (i = 0; i < num; i++) {
+
+ /* Ensure that LIODN value is valid */
+ if (liodn[i] >= PAACE_NUMBER_ENTRIES) {
+ pr_err("Invalid liodn %d, attach device failed for %s\n",
+ liodn[i], dev->of_node->full_name);
+ ret = -EINVAL;
+ break;
+ }
+
+ attach_device(dma_domain, liodn[i], dev);
+ /*
+ * Check if geometry has already been configured
+ * for the domain. If yes, set the geometry for
+ * the LIODN.
+ */
+ if (dma_domain->win_cnt) {
+ u32 win_cnt = dma_domain->win_cnt > 1 ? dma_domain->win_cnt : 0;
+ ret = pamu_set_liodn(liodn[i], dev, dma_domain,
+ &domain->geometry,
+ win_cnt);
+ if (ret)
+ break;
+ if (dma_domain->mapped) {
+ /*
+ * Create window/subwindow mapping for
+ * the LIODN.
+ */
+ ret = map_liodn(liodn[i], dma_domain);
+ if (ret)
+ break;
+ }
+ }
+ }
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return ret;
+}
+
+static int fsl_pamu_attach_device(struct iommu_domain *domain,
+ struct device *dev)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ const u32 *liodn;
+ u32 liodn_cnt;
+ int len, ret = 0;
+ struct pci_dev *pdev = NULL;
+ struct pci_controller *pci_ctl;
+
+ /*
+ * Use LIODN of the PCI controller while attaching a
+ * PCI device.
+ */
+ if (dev->bus == &pci_bus_type) {
+ pdev = to_pci_dev(dev);
+ pci_ctl = pci_bus_to_host(pdev->bus);
+ /*
+ * make dev point to pci controller device
+ * so we can get the LIODN programmed by
+ * u-boot.
+ */
+ dev = pci_ctl->parent;
+ }
+
+ liodn = of_get_property(dev->of_node, "fsl,liodn", &len);
+ if (liodn) {
+ liodn_cnt = len / sizeof(u32);
+ ret = handle_attach_device(dma_domain, dev,
+ liodn, liodn_cnt);
+ } else {
+ pr_err("missing fsl,liodn property at %s\n",
+ dev->of_node->full_name);
+ ret = -EINVAL;
+ }
+
+ return ret;
+}
+
+static void fsl_pamu_detach_device(struct iommu_domain *domain,
+ struct device *dev)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ const u32 *prop;
+ int len;
+ struct pci_dev *pdev = NULL;
+ struct pci_controller *pci_ctl;
+
+ /*
+ * Use LIODN of the PCI controller while detaching a
+ * PCI device.
+ */
+ if (dev->bus == &pci_bus_type) {
+ pdev = to_pci_dev(dev);
+ pci_ctl = pci_bus_to_host(pdev->bus);
+ /*
+ * make dev point to pci controller device
+ * so we can get the LIODN programmed by
+ * u-boot.
+ */
+ dev = pci_ctl->parent;
+ }
+
+ prop = of_get_property(dev->of_node, "fsl,liodn", &len);
+ if (prop)
+ detach_device(dev, dma_domain);
+ else
+ pr_err("missing fsl,liodn property at %s\n",
+ dev->of_node->full_name);
+}
+
+static int configure_domain_geometry(struct iommu_domain *domain, void *data)
+{
+ struct iommu_domain_geometry *geom_attr = data;
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ dma_addr_t geom_size;
+ unsigned long flags;
+
+ geom_size = geom_attr->aperture_end - geom_attr->aperture_start;
+ /*
+ * Sanity check the geometry size. Also, we do not support
+ * DMA outside of the geometry.
+ */
+ if (check_size(geom_size, geom_attr->aperture_start) ||
+ !geom_attr->force_aperture) {
+ pr_err("Invalid PAMU geometry attributes\n");
+ return -EINVAL;
+ }
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ if (dma_domain->enabled) {
+ pr_err("Can't set geometry attributes as domain is active\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EBUSY;
+ }
+
+ /* Copy the domain geometry information */
+ memcpy(&domain->geometry, geom_attr,
+ sizeof(struct iommu_domain_geometry));
+ dma_domain->geom_size = geom_size;
+
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return 0;
+}
+
+/* Set the domain stash attribute */
+static int configure_domain_stash(struct fsl_dma_domain *dma_domain, void *data)
+{
+ struct iommu_stash_attribute *stash_attr = data;
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+
+ memcpy(&dma_domain->dma_stash, stash_attr,
+ sizeof(struct iommu_stash_attribute));
+
+ dma_domain->stash_id = get_stash_id(stash_attr->cache,
+ stash_attr->cpu);
+ if (dma_domain->stash_id == ~(u32)0) {
+ pr_err("Invalid stash attributes\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+
+ ret = update_domain_stash(dma_domain, dma_domain->stash_id);
+
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return ret;
+}
+
+/* Configure domain dma state i.e. enable/disable DMA*/
+static int configure_domain_dma_state(struct fsl_dma_domain *dma_domain, bool enable)
+{
+ struct device_domain_info *info;
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+
+ if (enable && !dma_domain->mapped) {
+ pr_err("Can't enable DMA domain without valid mapping\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -ENODEV;
+ }
+
+ dma_domain->enabled = enable;
+ if (!list_empty(&dma_domain->devices)) {
+ list_for_each_entry(info, &dma_domain->devices,
+ link) {
+ ret = (enable) ? pamu_enable_liodn(info->liodn) :
+ pamu_disable_liodn(info->liodn);
+ if (ret)
+ pr_err("Unable to set dma state for liodn %d",
+ info->liodn);
+ }
+ }
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return 0;
+}
+
+int fsl_pamu_set_domain_attr(struct iommu_domain *domain,
+ enum iommu_attr attr_type, void *data)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ int ret = 0;
+
+
+ switch (attr_type) {
+ case DOMAIN_ATTR_GEOMETRY:
+ ret = configure_domain_geometry(domain, data);
+ break;
+ case DOMAIN_ATTR_PAMU_STASH:
+ ret = configure_domain_stash(dma_domain, data);
+ break;
+ case DOMAIN_ATTR_PAMU_ENABLE:
+ ret = configure_domain_dma_state(dma_domain, *(int *)data);
+ break;
+ default:
+ pr_err("Unsupported attribute type\n");
+ ret = -EINVAL;
+ break;
+ };
+
+ return ret;
+}
+
+int fsl_pamu_get_domain_attr(struct iommu_domain *domain,
+ enum iommu_attr attr_type, void *data)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ int ret = 0;
+
+
+ switch (attr_type) {
+ case DOMAIN_ATTR_PAMU_STASH:
+ memcpy((struct iommu_stash_attribute *) data, &dma_domain->dma_stash,
+ sizeof(struct iommu_stash_attribute));
+ break;
+ case DOMAIN_ATTR_PAMU_ENABLE:
+ *(int *)data = dma_domain->enabled;
+ break;
+ case DOMAIN_ATTR_FSL_PAMUV1:
+ *(int *)data = DOMAIN_ATTR_FSL_PAMUV1;
+ break;
+ default:
+ pr_err("Unsupported attribute type\n");
+ ret = -EINVAL;
+ break;
+ };
+
+ return ret;
+}
+
+static void swap_pci_ref(struct pci_dev **from, struct pci_dev *to)
+{
+ pci_dev_put(*from);
+ *from = to;
+}
+
+static struct iommu_group *get_device_iommu_group(struct device *dev)
+{
+ struct iommu_group *group;
+
+ group = iommu_group_get(dev);
+ if (!group)
+ group = iommu_group_alloc();
+
+ return group;
+}
+
+static bool check_pci_ctl_endpt_part(struct pci_controller *pci_ctl)
+{
+ u32 version;
+
+ /* Check the PCI controller version number by readding BRR1 register */
+ version = in_be32(pci_ctl->cfg_addr + (PCI_FSL_BRR1 >> 2));
+ version &= PCI_FSL_BRR1_VER;
+ /* If PCI controller version is >= 0x204 we can partition endpoints*/
+ if (version >= 0x204)
+ return 1;
+
+ return 0;
+}
+
+static struct iommu_group *get_peer_pci_device_group(struct pci_dev *pdev)
+{
+ struct iommu_group *group = NULL;
+
+ /* check if this is the first device on the bus*/
+ if (pdev->bus_list.next == pdev->bus_list.prev) {
+ struct pci_bus *bus = pdev->bus->parent;
+ /* Traverese the parent bus list to get
+ * pdev & dev for the sibling device.
+ */
+ while (bus) {
+ if (!list_empty(&bus->devices)) {
+ pdev = container_of(bus->devices.next,
+ struct pci_dev, bus_list);
+ group = iommu_group_get(&pdev->dev);
+ break;
+ } else
+ bus = bus->parent;
+ }
+ } else {
+ /*
+ * Get the pdev & dev for the sibling device
+ */
+ pdev = container_of(pdev->bus_list.prev,
+ struct pci_dev, bus_list);
+ group = iommu_group_get(&pdev->dev);
+ }
+
+ return group;
+}
+
+static struct iommu_group *get_pci_device_group(struct pci_dev *pdev)
+{
+ struct iommu_group *group = NULL;
+ struct pci_dev *bridge, *dma_pdev = NULL;
+ struct pci_controller *pci_ctl;
+ bool pci_endpt_partioning;
+
+ pci_ctl = pci_bus_to_host(pdev->bus);
+ pci_endpt_partioning = check_pci_ctl_endpt_part(pci_ctl);
+ /* We can partition PCIe devices so assign device group to the device */
+ if (pci_endpt_partioning) {
+ bridge = pci_find_upstream_pcie_bridge(pdev);
+ if (bridge) {
+ if (pci_is_pcie(bridge))
+ dma_pdev = pci_get_domain_bus_and_slot(
+ pci_domain_nr(pdev->bus),
+ bridge->subordinate->number, 0);
+ if (!dma_pdev)
+ dma_pdev = pci_dev_get(bridge);
+ } else
+ dma_pdev = pci_dev_get(pdev);
+ /* Account for quirked devices */
+ swap_pci_ref(&dma_pdev, pci_get_dma_source(dma_pdev));
+ group = get_device_iommu_group(&pdev->dev);
+ pci_dev_put(pdev);
+ /*
+ * PCIe controller is not a paritionable entity
+ * free the controller device iommu_group.
+ */
+ if (pci_ctl->parent->iommu_group)
+ iommu_group_remove_device(pci_ctl->parent);
+ } else {
+ /*
+ * All devices connected to the controller will share the
+ * PCI controllers device group. If this is the first
+ * device to be probed for the pci controller, copy the
+ * device group information from the PCI controller device
+ * node and remove the PCI controller iommu group.
+ * For subsequent devices, the iommu group information can
+ * be obtained from sibling devices (i.e. from the bus_devices
+ * link list).
+ */
+ if (pci_ctl->parent->iommu_group) {
+ group = get_device_iommu_group(pci_ctl->parent);
+ iommu_group_remove_device(pci_ctl->parent);
+ } else
+ group = get_peer_pci_device_group(pdev);
+ }
+
+ return group;
+}
+
+static int fsl_pamu_add_device(struct device *dev)
+{
+ struct iommu_group *group = NULL;
+ struct pci_dev *pdev;
+ int ret;
+
+ /*
+ * For platform devices we allocate a separate group for
+ * each of the devices.
+ */
+ if (dev->bus == &pci_bus_type) {
+ pdev = to_pci_dev(dev);
+ /* Don't create device groups for virtual PCI bridges */
+ if (pdev->subordinate)
+ return 0;
+
+ group = get_pci_device_group(pdev);
+
+ } else
+ group = get_device_iommu_group(dev);
+
+ if (!group || IS_ERR(group))
+ return PTR_ERR(group);
+
+ ret = iommu_group_add_device(group, dev);
+
+ iommu_group_put(group);
+ return ret;
+}
+
+static void fsl_pamu_remove_device(struct device *dev)
+{
+ iommu_group_remove_device(dev);
+}
+
+static int fsl_pamu_set_windows(struct iommu_domain *domain, u32 w_count)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&dma_domain->domain_lock, flags);
+ /* Ensure domain is inactive i.e. DMA should be disabled for the domain */
+ if (dma_domain->enabled) {
+ pr_err("Can't set geometry attributes as domain is active\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EBUSY;
+ }
+
+ /* Ensure that the geometry has been set for the domain */
+ if (!dma_domain->geom_size) {
+ pr_err("Please configure geometry before setting the number of windows\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+
+ /*
+ * Ensure we have valid window count i.e. it should be less than
+ * maximum permissible limit and should be a power of two.
+ */
+ if (w_count > pamu_get_max_subwin_cnt() || (w_count & (w_count - 1))) {
+ pr_err("Invalid window count\n");
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -EINVAL;
+ }
+
+ ret = pamu_set_domain_geometry(dma_domain, &domain->geometry,
+ ((w_count > 1) ? w_count : 0));
+ if (!ret) {
+ if (dma_domain->win_arr)
+ kfree(dma_domain->win_arr);
+ dma_domain->win_arr = kzalloc(sizeof(struct dma_window) *
+ w_count, GFP_KERNEL);
+ if (!dma_domain->win_arr) {
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+ return -ENOMEM;
+ }
+ dma_domain->win_cnt = w_count;
+ }
+ spin_unlock_irqrestore(&dma_domain->domain_lock, flags);
+
+ return ret;
+}
+
+static u32 fsl_pamu_get_windows(struct iommu_domain *domain)
+{
+ struct fsl_dma_domain *dma_domain = domain->priv;
+
+ return dma_domain->win_cnt;
+}
+
+static struct iommu_ops fsl_pamu_ops = {
+ .domain_init = fsl_pamu_domain_init,
+ .domain_destroy = fsl_pamu_domain_destroy,
+ .attach_dev = fsl_pamu_attach_device,
+ .detach_dev = fsl_pamu_detach_device,
+ .domain_window_enable = fsl_pamu_window_enable,
+ .domain_window_disable = fsl_pamu_window_disable,
+ .domain_get_windows = fsl_pamu_get_windows,
+ .domain_set_windows = fsl_pamu_set_windows,
+ .iova_to_phys = fsl_pamu_iova_to_phys,
+ .domain_has_cap = fsl_pamu_domain_has_cap,
+ .domain_set_attr = fsl_pamu_set_domain_attr,
+ .domain_get_attr = fsl_pamu_get_domain_attr,
+ .add_device = fsl_pamu_add_device,
+ .remove_device = fsl_pamu_remove_device,
+};
+
+int pamu_domain_init()
+{
+ int ret = 0;
+
+ ret = iommu_init_mempool();
+ if (ret)
+ return ret;
+
+ bus_set_iommu(&platform_bus_type, &fsl_pamu_ops);
+ bus_set_iommu(&pci_bus_type, &fsl_pamu_ops);
+
+ return ret;
+}
diff --git a/drivers/iommu/fsl_pamu_domain.h b/drivers/iommu/fsl_pamu_domain.h
new file mode 100644
index 0000000..52dede7
--- /dev/null
+++ b/drivers/iommu/fsl_pamu_domain.h
@@ -0,0 +1,85 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, version 2, as
+ * published by the Free Software Foundation.
+ *
+ * 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, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ *
+ * Copyright (C) 2013 Freescale Semiconductor, Inc.
+ *
+ */
+
+#ifndef __FSL_PAMU_DOMAIN_H
+#define __FSL_PAMU_DOMAIN_H
+
+#include "fsl_pamu.h"
+
+struct dma_window {
+ phys_addr_t paddr;
+ u64 size;
+ int valid;
+ int prot;
+};
+
+struct fsl_dma_domain {
+ /*
+ * Indicates the geometry size for the domain.
+ * This would be set when the geometry is
+ * configured for the domain.
+ */
+ dma_addr_t geom_size;
+ /*
+ * Number of windows assocaited with this domain.
+ * During domain initialization, it is set to the
+ * the maximum number of subwindows allowed for a LIODN.
+ * Minimum value for this is 1 indicating a single PAMU
+ * window, without any sub windows. Value can be set/
+ * queried by set_attr/get_attr API for DOMAIN_ATTR_WINDOWS.
+ * Value can only be set once the geometry has been configured.
+ */
+ u32 win_cnt;
+ /*
+ * win_arr contains information of the configured
+ * windows for a domain. This is allocated only
+ * when the number of windows for the domain are
+ * set.
+ */
+ struct dma_window *win_arr;
+ /* list of devices associated with the domain */
+ struct list_head devices;
+ /* dma_domain states:
+ * mapped - A particular mapping has been created
+ * within the configured geometry.
+ * enabled - DMA has been enabled for the given
+ * domain. This translates to setting of the
+ * valid bit for the primary PAACE in the PAMU
+ * PAACT table. Domain geometry should be set and
+ * it must have a valid mapping before DMA can be
+ * enabled for it.
+ *
+ */
+ int mapped;
+ int enabled;
+ /* stash_id obtained from the stash attribute details */
+ u32 stash_id;
+ struct iommu_stash_attribute dma_stash;
+ u32 snoop_id;
+ struct iommu_domain *iommu_domain;
+ spinlock_t domain_lock;
+};
+
+/* domain-device relationship */
+struct device_domain_info {
+ struct list_head link; /* link to domain siblings */
+ struct device *dev;
+ u32 liodn;
+ struct fsl_dma_domain *domain; /* pointer to domain */
+};
+#endif /* __FSL_PAMU_DOMAIN_H */
--
1.7.4.1
^ permalink raw reply related
* [PATCH 4/5 v9] iommu/fsl: Add additional iommu attributes required by the PAMU driver.
From: Varun Sethi @ 2013-03-13 18:45 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
In-Reply-To: <1363200310-11513-1-git-send-email-Varun.Sethi@freescale.com>
Added the following domain attributes for the FSL PAMU driver:
1. Added new iommu stash attribute, which allows setting of the
LIODN specific stash id parameter through IOMMU API.
2. Added an attribute for enabling/disabling DMA to a particular
memory window.
3. Added domain attribute to check for PAMUV1 specific constraints.
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
---
include/linux/iommu.h | 35 +++++++++++++++++++++++++++++++++++
1 files changed, 35 insertions(+), 0 deletions(-)
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 02ac111..1db9fd3 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -40,6 +40,25 @@ struct notifier_block;
typedef int (*iommu_fault_handler_t)(struct iommu_domain *,
struct device *, unsigned long, int, void *);
+/* cache stash targets */
+enum stash_target {
+ IOMMU_ATTR_CACHE_L1 = 1,
+ IOMMU_ATTR_CACHE_L2,
+ IOMMU_ATTR_CACHE_L3,
+};
+
+/* This attribute corresponds to IOMMUs capable of generating
+ * a stash transaction. A stash transaction is typically a
+ * hardware initiated prefetch of data from memory to cache.
+ * This attribute allows configuring stashig specific parameters
+ * in the IOMMU hardware.
+ */
+
+struct iommu_stash_attribute {
+ u32 cpu; /* cpu number */
+ u32 cache; /* cache to stash to: L1,L2,L3 */
+};
+
struct iommu_domain_geometry {
dma_addr_t aperture_start; /* First address that can be mapped */
dma_addr_t aperture_end; /* Last address that can be mapped */
@@ -57,10 +76,26 @@ struct iommu_domain {
#define IOMMU_CAP_CACHE_COHERENCY 0x1
#define IOMMU_CAP_INTR_REMAP 0x2 /* isolates device intrs */
+/*
+ * Following constraints are specifc to PAMUV1:
+ * -aperture must be power of 2, and naturally aligned
+ * -number of windows must be power of 2, and address space size
+ * of each window is determined by aperture size / # of windows
+ * -the actual size of the mapped region of a window must be power
+ * of 2 starting with 4KB and physical address must be naturally
+ * aligned.
+ * DOMAIN_ATTR_FSL_PAMUV1 corresponds to the above mentioned contraints.
+ * The caller can invoke iommu_domain_get_attr to check if the underlying
+ * iommu implementation supports these constraints.
+ */
+
enum iommu_attr {
DOMAIN_ATTR_GEOMETRY,
DOMAIN_ATTR_PAGING,
DOMAIN_ATTR_WINDOWS,
+ DOMAIN_ATTR_PAMU_STASH,
+ DOMAIN_ATTR_PAMU_ENABLE,
+ DOMAIN_ATTR_FSL_PAMUV1,
DOMAIN_ATTR_MAX,
};
--
1.7.4.1
^ permalink raw reply related
* [PATCH 3/5 v9] iommu/fsl: Add the window permission flag as a parameter to iommu_window_enable API.
From: Varun Sethi @ 2013-03-13 18:45 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
In-Reply-To: <1363200310-11513-1-git-send-email-Varun.Sethi@freescale.com>
Each iommu window can have access permissions associated with it. Extended the
window_enable API to incorporate window access permissions.
In case of PAMU each window can have its specific set of permissions.
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
---
drivers/iommu/iommu.c | 5 +++--
include/linux/iommu.h | 7 ++++---
2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 39106ec..c4d5265 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -853,12 +853,13 @@ EXPORT_SYMBOL_GPL(iommu_unmap);
int iommu_domain_window_enable(struct iommu_domain *domain, u32 wnd_nr,
- phys_addr_t paddr, u64 size)
+ phys_addr_t paddr, u64 size, int prot)
{
if (unlikely(domain->ops->domain_window_enable == NULL))
return -ENODEV;
- return domain->ops->domain_window_enable(domain, wnd_nr, paddr, size);
+ return domain->ops->domain_window_enable(domain, wnd_nr, paddr, size,
+ prot);
}
EXPORT_SYMBOL_GPL(iommu_domain_window_enable);
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index 576cc19..02ac111 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -104,7 +104,7 @@ struct iommu_ops {
/* Window handling functions */
int (*domain_window_enable)(struct iommu_domain *domain, u32 wnd_nr,
- phys_addr_t paddr, u64 size);
+ phys_addr_t paddr, u64 size, int prot);
void (*domain_window_disable)(struct iommu_domain *domain, u32 wnd_nr);
/* Set the numer of window per domain */
int (*domain_set_windows)(struct iommu_domain *domain, u32 w_count);
@@ -169,7 +169,8 @@ extern int iommu_domain_set_attr(struct iommu_domain *domain, enum iommu_attr,
/* Window handling function prototypes */
extern int iommu_domain_window_enable(struct iommu_domain *domain, u32 wnd_nr,
- phys_addr_t offset, u64 size);
+ phys_addr_t offset, u64 size,
+ int prot);
extern void iommu_domain_window_disable(struct iommu_domain *domain, u32 wnd_nr);
/**
* report_iommu_fault() - report about an IOMMU fault to the IOMMU framework
@@ -255,7 +256,7 @@ static inline int iommu_unmap(struct iommu_domain *domain, unsigned long iova,
static inline int iommu_domain_window_enable(struct iommu_domain *domain,
u32 wnd_nr, phys_addr_t paddr,
- u64 size)
+ u64 size, int prot)
{
return -ENODEV;
}
--
1.7.4.1
^ permalink raw reply related
* [PATCH 2/5 v9] powerpc: Add iommu domain pointer to device archdata
From: Varun Sethi @ 2013-03-13 18:45 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
In-Reply-To: <1363200310-11513-1-git-send-email-Varun.Sethi@freescale.com>
Add an iommu domain pointer to device (powerpc) archdata. Devices
are attached to iommu domains and this pointer provides a mechanism
to correlate between a device and the associated iommu domain. This
field is set when a device is attached to a domain.
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
---
arch/powerpc/include/asm/device.h | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/arch/powerpc/include/asm/device.h b/arch/powerpc/include/asm/device.h
index 77e97dd..2d5c1c5 100644
--- a/arch/powerpc/include/asm/device.h
+++ b/arch/powerpc/include/asm/device.h
@@ -28,6 +28,12 @@ struct dev_archdata {
void *iommu_table_base;
} dma_data;
+ /* IOMMU domain information pointer. This would be set
+ * when this device is attached to an iommu_domain.
+ */
+#ifdef CONFIG_IOMMU_API
+ void *iommu_domain;
+#endif
#ifdef CONFIG_SWIOTLB
dma_addr_t max_direct_dma_addr;
#endif
--
1.7.4.1
^ permalink raw reply related
* [PATCH 1/5 v9] iommu/fsl: Make iova u64 in the iommu_iova_to_phys API.
From: Varun Sethi @ 2013-03-13 18:45 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
In-Reply-To: <1363200310-11513-1-git-send-email-Varun.Sethi@freescale.com>
This is required in case of PAMU, as it can support a window size of up
to 64G (even on 32bit).
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
---
drivers/iommu/amd_iommu.c | 2 +-
drivers/iommu/exynos-iommu.c | 2 +-
drivers/iommu/intel-iommu.c | 2 +-
drivers/iommu/iommu.c | 3 +--
drivers/iommu/msm_iommu.c | 2 +-
drivers/iommu/omap-iommu.c | 2 +-
drivers/iommu/shmobile-iommu.c | 2 +-
drivers/iommu/tegra-gart.c | 2 +-
drivers/iommu/tegra-smmu.c | 2 +-
include/linux/iommu.h | 9 +++------
10 files changed, 12 insertions(+), 16 deletions(-)
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index 98f555d..42f6a71 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -3412,7 +3412,7 @@ static size_t amd_iommu_unmap(struct iommu_domain *dom, unsigned long iova,
}
static phys_addr_t amd_iommu_iova_to_phys(struct iommu_domain *dom,
- unsigned long iova)
+ u64 iova)
{
struct protection_domain *domain = dom->priv;
unsigned long offset_mask;
diff --git a/drivers/iommu/exynos-iommu.c b/drivers/iommu/exynos-iommu.c
index 238a3ca..541e81b 100644
--- a/drivers/iommu/exynos-iommu.c
+++ b/drivers/iommu/exynos-iommu.c
@@ -1027,7 +1027,7 @@ done:
}
static phys_addr_t exynos_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+ u64 iova)
{
struct exynos_iommu_domain *priv = domain->priv;
unsigned long *entry;
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 0099667..c9663ac 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4111,7 +4111,7 @@ static size_t intel_iommu_unmap(struct iommu_domain *domain,
}
static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+ u64 iova)
{
struct dmar_domain *dmar_domain = domain->priv;
struct dma_pte *pte;
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index b972d43..39106ec 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -706,8 +706,7 @@ void iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group)
}
EXPORT_SYMBOL_GPL(iommu_detach_group);
-phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, u64 iova)
{
if (unlikely(domain->ops->iova_to_phys == NULL))
return 0;
diff --git a/drivers/iommu/msm_iommu.c b/drivers/iommu/msm_iommu.c
index 6a8870a..fcd14a3 100644
--- a/drivers/iommu/msm_iommu.c
+++ b/drivers/iommu/msm_iommu.c
@@ -554,7 +554,7 @@ fail:
}
static phys_addr_t msm_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long va)
+ u64 va)
{
struct msm_priv *priv;
struct msm_iommu_drvdata *iommu_drvdata;
diff --git a/drivers/iommu/omap-iommu.c b/drivers/iommu/omap-iommu.c
index 6ac02fa..102ae56 100644
--- a/drivers/iommu/omap-iommu.c
+++ b/drivers/iommu/omap-iommu.c
@@ -1219,7 +1219,7 @@ static void omap_iommu_domain_destroy(struct iommu_domain *domain)
}
static phys_addr_t omap_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long da)
+ u64 da)
{
struct omap_iommu_domain *omap_domain = domain->priv;
struct omap_iommu *oiommu = omap_domain->iommu_dev;
diff --git a/drivers/iommu/shmobile-iommu.c b/drivers/iommu/shmobile-iommu.c
index b6e8b57..9216802 100644
--- a/drivers/iommu/shmobile-iommu.c
+++ b/drivers/iommu/shmobile-iommu.c
@@ -296,7 +296,7 @@ done:
}
static phys_addr_t shmobile_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+ u64 iova)
{
struct shmobile_iommu_domain *sh_domain = domain->priv;
uint32_t l1entry = 0, l2entry = 0;
diff --git a/drivers/iommu/tegra-gart.c b/drivers/iommu/tegra-gart.c
index 8643757..17179c0 100644
--- a/drivers/iommu/tegra-gart.c
+++ b/drivers/iommu/tegra-gart.c
@@ -279,7 +279,7 @@ static size_t gart_iommu_unmap(struct iommu_domain *domain, unsigned long iova,
}
static phys_addr_t gart_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+ u64 iova)
{
struct gart_device *gart = domain->priv;
unsigned long pte;
diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
index b34e5fd..eef9460 100644
--- a/drivers/iommu/tegra-smmu.c
+++ b/drivers/iommu/tegra-smmu.c
@@ -757,7 +757,7 @@ static size_t smmu_iommu_unmap(struct iommu_domain *domain, unsigned long iova,
}
static phys_addr_t smmu_iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+ u64 iova)
{
struct smmu_as *as = domain->priv;
unsigned long *pte;
diff --git a/include/linux/iommu.h b/include/linux/iommu.h
index ba3b8a9..576cc19 100644
--- a/include/linux/iommu.h
+++ b/include/linux/iommu.h
@@ -91,8 +91,7 @@ struct iommu_ops {
phys_addr_t paddr, size_t size, int prot);
size_t (*unmap)(struct iommu_domain *domain, unsigned long iova,
size_t size);
- phys_addr_t (*iova_to_phys)(struct iommu_domain *domain,
- unsigned long iova);
+ phys_addr_t (*iova_to_phys)(struct iommu_domain *domain, u64 iova);
int (*domain_has_cap)(struct iommu_domain *domain,
unsigned long cap);
int (*add_device)(struct device *dev);
@@ -134,8 +133,7 @@ extern int iommu_map(struct iommu_domain *domain, unsigned long iova,
phys_addr_t paddr, size_t size, int prot);
extern size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova,
size_t size);
-extern phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova);
+extern phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, u64 iova);
extern int iommu_domain_has_cap(struct iommu_domain *domain,
unsigned long cap);
extern void iommu_set_fault_handler(struct iommu_domain *domain,
@@ -267,8 +265,7 @@ static inline void iommu_domain_window_disable(struct iommu_domain *domain,
{
}
-static inline phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain,
- unsigned long iova)
+static inline phys_addr_t iommu_iova_to_phys(struct iommu_domain *domain, u64 iova)
{
return 0;
}
--
1.7.4.1
^ permalink raw reply related
* [PATCH 0/5 v9] iommu/fsl: Freescale PAMU driver and IOMMU API implementation.
From: Varun Sethi @ 2013-03-13 18:45 UTC (permalink / raw)
To: joro, iommu, linuxppc-dev, linux-kernel, galak, benh, scottwood,
stuart.yoder
Cc: Varun Sethi
This patchset provides the Freescale PAMU (Peripheral Access Management Unit) driver
and the corresponding IOMMU API implementation. PAMU is the IOMMU present on Freescale
QorIQ platforms. PAMU can authorize memory access, remap the memory address, and remap
the I/O transaction type.
This set consists of the following patches:
1. Make iova u64 in the iommu_iova_to_phys API.
2. Addition of new field in the device (powerpc) archdata structure for storing iommu domain information
pointer.
3. Add window permission flags in the iommu_domain_window_enable API.
4. Add domain attributes for FSL PAMU driver.
5. PAMU driver and IOMMU API implementation.
This patch set is based on the master branch (3.9-rc2) of Linus Torvald's git tree.
Varun Sethi (5):
Make iova u64 in the iommu_iova_to_phys API.
Add iommu domain pointer to device archdata
Add the window permission flag as a parameter to iommu_window_enable
API.
Add addition iommu attributes required by the PAMU driver.
FSL PAMU driver.
arch/powerpc/include/asm/device.h | 6 +
arch/powerpc/sysdev/fsl_pci.h | 5 +
drivers/iommu/Kconfig | 8 +
drivers/iommu/Makefile | 1 +
drivers/iommu/amd_iommu.c | 2 +-
drivers/iommu/exynos-iommu.c | 2 +-
drivers/iommu/fsl_pamu.c | 1250 +++++++++++++++++++++++++++++++++++++
drivers/iommu/fsl_pamu.h | 405 ++++++++++++
drivers/iommu/fsl_pamu_domain.c | 1134 +++++++++++++++++++++++++++++++++
drivers/iommu/fsl_pamu_domain.h | 85 +++
drivers/iommu/intel-iommu.c | 2 +-
drivers/iommu/iommu.c | 8 +-
drivers/iommu/msm_iommu.c | 2 +-
drivers/iommu/omap-iommu.c | 2 +-
drivers/iommu/shmobile-iommu.c | 2 +-
drivers/iommu/tegra-gart.c | 2 +-
drivers/iommu/tegra-smmu.c | 2 +-
include/linux/iommu.h | 51 ++-
18 files changed, 2948 insertions(+), 21 deletions(-)
create mode 100644 drivers/iommu/fsl_pamu.c
create mode 100644 drivers/iommu/fsl_pamu.h
create mode 100644 drivers/iommu/fsl_pamu_domain.c
create mode 100644 drivers/iommu/fsl_pamu_domain.h
--
1.7.4.1
^ permalink raw reply
* Re: [PATCH V4] powerpc/85xx: Add machine check handler to fix PCIe erratum on mpc85xx
From: Scott Wood @ 2013-03-13 16:37 UTC (permalink / raw)
To: David Laight
Cc: Wood Scott-B07421, Stuart Yoder, linuxppc-dev, Jia Hongtao-B38951
In-Reply-To: <AE90C24D6B3A694183C094C60CF0A2F6026B7186@saturn3.aculab.com>
On 03/13/2013 04:40:40 AM, David Laight wrote:
> > Hmm, seems there's no probe_user_address() -- for userspace we
> > basically want the same thing minus the KERNEL_DS. See
> > arch/powerpc/perf/callchain.c for an example.
>=20
> Isn't that just copy_from_user() ?
Plus pagefault_disable/enable().
-Scott=
^ permalink raw reply
* [PATCH] powerpc/mpc512x: create SoC devices for more nodes
From: Anatolij Gustschin @ 2013-03-13 14:43 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Anatolij Gustschin
Create devices for mbx, sram, pci and gpio-leds nodes and
also move nfc compatible to of_bus_id list for automatic
nfc device creation.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
---
arch/powerpc/platforms/512x/mpc512x_shared.c | 13 +++++--------
1 files changed, 5 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/platforms/512x/mpc512x_shared.c b/arch/powerpc/platforms/512x/mpc512x_shared.c
index d30235b..2765cd1 100644
--- a/arch/powerpc/platforms/512x/mpc512x_shared.c
+++ b/arch/powerpc/platforms/512x/mpc512x_shared.c
@@ -330,22 +330,19 @@ void __init mpc512x_init_IRQ(void)
static struct of_device_id __initdata of_bus_ids[] = {
{ .compatible = "fsl,mpc5121-immr", },
{ .compatible = "fsl,mpc5121-localbus", },
+ { .compatible = "fsl,mpc5121-mbx", },
+ { .compatible = "fsl,mpc5121-nfc", },
+ { .compatible = "fsl,mpc5121-sram", },
+ { .compatible = "fsl,mpc5121-pci", },
+ { .compatible = "gpio-leds", },
{},
};
void __init mpc512x_declare_of_platform_devices(void)
{
- struct device_node *np;
-
if (of_platform_bus_probe(NULL, of_bus_ids, NULL))
printk(KERN_ERR __FILE__ ": "
"Error while probing of_platform bus\n");
-
- np = of_find_compatible_node(NULL, NULL, "fsl,mpc5121-nfc");
- if (np) {
- of_platform_device_create(np, NULL, NULL);
- of_node_put(np);
- }
}
#define DEFAULT_FIFO_SIZE 16
--
1.7.5.4
^ permalink raw reply related
* Re: [PATCH -V3 2/3] powerpc: Update kernel VSID range
From: Aneesh Kumar K.V @ 2013-03-13 13:44 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, David Gibson
In-Reply-To: <20130313034254.GC21125@iris.ozlabs.ibm.com>
Paul Mackerras <paulus@samba.org> writes:
>> BEGIN_FTR_SECTION
>> b slb_finish_load
>> END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
>> @@ -91,24 +97,19 @@ _GLOBAL(slb_miss_kernel_load_vmemmap)
>> _GLOBAL(slb_miss_kernel_load_io)
>> li r11,0
>> 6:
>> - li r9,0x1
>> /*
>> - * for 1T we shift 12 bits more. slb_finish_load_1T will do
>> - * the necessary adjustment
>> + * context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1
>> */
>> - rldimi r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0
>> + rldicl r9,r3,4,62
>> + addis r9,r9,8
>> + subi r9,r9,4
>
> If you did the context calculation earlier, before the "bne cr7,1f",
> you could save 3 more instructions.
We use r9 later as region id in
#ifdef CONFIG_SPARSEMEM_VMEMMAP
/* Check virtual memmap region. To be patches at kernel boot */
cmpldi cr0,r9,0xf
bne 1f
hence i didn't do the above suggested change.
-aneesh
^ permalink raw reply
* [PATCH -V4 2/3] powerpc: Update kernel VSID range
From: Aneesh Kumar K.V @ 2013-03-13 13:34 UTC (permalink / raw)
To: benh, paulus, David Gibson; +Cc: linuxppc-dev, Aneesh Kumar K.V
In-Reply-To: <1363181695-5315-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
This patch change the kernel VSID range so that we limit VSID_BITS to 37.
This enables us to support 64TB with 65 bit VA (37+28). Without this patch
we have boot hangs on platforms that only support 65 bit VA.
With this patch we now have proto vsid generated as below:
We first generate a 37-bit "proto-VSID". Proto-VSIDs are generated
from mmu context id and effective segment id of the address.
For user processes max context id is limited to ((1ul << 19) - 5)
for kernel space, we use the top 4 context ids to map address as below
0x7fffc - [ 0xc000000000000000 - 0xc0003fffffffffff ]
0x7fffd - [ 0xd000000000000000 - 0xd0003fffffffffff ]
0x7fffe - [ 0xe000000000000000 - 0xe0003fffffffffff ]
0x7ffff - [ 0xf000000000000000 - 0xf0003fffffffffff ]
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/mmu-hash64.h | 115 +++++++++++++++++----------------
arch/powerpc/kernel/exceptions-64s.S | 34 +++++++---
arch/powerpc/mm/hash_utils_64.c | 20 ++++--
arch/powerpc/mm/mmu_context_hash64.c | 11 +---
arch/powerpc/mm/slb_low.S | 50 +++++++-------
arch/powerpc/mm/tlb_hash64.c | 2 +-
6 files changed, 126 insertions(+), 106 deletions(-)
diff --git a/arch/powerpc/include/asm/mmu-hash64.h b/arch/powerpc/include/asm/mmu-hash64.h
index 5f8c2bd..a32461f 100644
--- a/arch/powerpc/include/asm/mmu-hash64.h
+++ b/arch/powerpc/include/asm/mmu-hash64.h
@@ -343,17 +343,16 @@ extern void slb_set_size(u16 size);
/*
* VSID allocation (256MB segment)
*
- * We first generate a 38-bit "proto-VSID". For kernel addresses this
- * is equal to the ESID | 1 << 37, for user addresses it is:
- * (context << USER_ESID_BITS) | (esid & ((1U << USER_ESID_BITS) - 1)
+ * We first generate a 37-bit "proto-VSID". Proto-VSIDs are generated
+ * from mmu context id and effective segment id of the address.
*
- * This splits the proto-VSID into the below range
- * 0 - (2^(CONTEXT_BITS + USER_ESID_BITS) - 1) : User proto-VSID range
- * 2^(CONTEXT_BITS + USER_ESID_BITS) - 2^(VSID_BITS) : Kernel proto-VSID range
- *
- * We also have CONTEXT_BITS + USER_ESID_BITS = VSID_BITS - 1
- * That is, we assign half of the space to user processes and half
- * to the kernel.
+ * For user processes max context id is limited to ((1ul << 19) - 5)
+ * for kernel space, we use the top 4 context ids to map address as below
+ * NOTE: each context only support 64TB now.
+ * 0x7fffc - [ 0xc000000000000000 - 0xc0003fffffffffff ]
+ * 0x7fffd - [ 0xd000000000000000 - 0xd0003fffffffffff ]
+ * 0x7fffe - [ 0xe000000000000000 - 0xe0003fffffffffff ]
+ * 0x7ffff - [ 0xf000000000000000 - 0xf0003fffffffffff ]
*
* The proto-VSIDs are then scrambled into real VSIDs with the
* multiplicative hash:
@@ -363,22 +362,19 @@ extern void slb_set_size(u16 size);
* VSID_MULTIPLIER is prime, so in particular it is
* co-prime to VSID_MODULUS, making this a 1:1 scrambling function.
* Because the modulus is 2^n-1 we can compute it efficiently without
- * a divide or extra multiply (see below).
- *
- * This scheme has several advantages over older methods:
+ * a divide or extra multiply (see below). The scramble function gives
+ * robust scattering in the hash table (at least based on some initial
+ * results).
*
- * - We have VSIDs allocated for every kernel address
- * (i.e. everything above 0xC000000000000000), except the very top
- * segment, which simplifies several things.
+ * We also consider VSID 0 special. We use VSID 0 for slb entries mapping
+ * bad address. This enables us to consolidate bad address handling in
+ * hash_page.
*
- * - We allow for USER_ESID_BITS significant bits of ESID and
- * CONTEXT_BITS bits of context for user addresses.
- * i.e. 64T (46 bits) of address space for up to half a million contexts.
- *
- * - The scramble function gives robust scattering in the hash
- * table (at least based on some initial results). The previous
- * method was more susceptible to pathological cases giving excessive
- * hash collisions.
+ * We also need to avoid the last segment of the last context, because that
+ * would give a protovsid of 0x1fffffffff. That will result in a VSID 0
+ * because of the modulo operation in vsid scramble. But the vmemmap
+ * (which is what uses region 0xf) will never be close to 64TB in size
+ * (it's 56 bytes per page of system memory).
*/
#define CONTEXT_BITS 19
@@ -386,15 +382,25 @@ extern void slb_set_size(u16 size);
#define USER_ESID_BITS_1T 6
/*
+ * 256MB segment
+ * The proto-VSID space has 2^(CONTEX_BITS + USER_ESID_BITS) - 1 segments
+ * available for user + kernel mapping. The top 4 contexts are used for
+ * kernel mapping. Each segment contains 2^28 bytes. Each
+ * context maps 2^46 bytes (64TB) so we can support 2^19-1 contexts
+ * (19 == 37 + 28 - 46).
+ */
+#define MAX_USER_CONTEXT ((ASM_CONST(1) << CONTEXT_BITS) - 5)
+
+/*
* This should be computed such that protovosid * vsid_mulitplier
* doesn't overflow 64 bits. It should also be co-prime to vsid_modulus
*/
#define VSID_MULTIPLIER_256M ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_256M (CONTEXT_BITS + USER_ESID_BITS + 1)
+#define VSID_BITS_256M (CONTEXT_BITS + USER_ESID_BITS)
#define VSID_MODULUS_256M ((1UL<<VSID_BITS_256M)-1)
#define VSID_MULTIPLIER_1T ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_1T (CONTEXT_BITS + USER_ESID_BITS_1T + 1)
+#define VSID_BITS_1T (CONTEXT_BITS + USER_ESID_BITS_1T)
#define VSID_MODULUS_1T ((1UL<<VSID_BITS_1T)-1)
@@ -422,7 +428,8 @@ extern void slb_set_size(u16 size);
srdi rx,rt,VSID_BITS_##size; \
clrldi rt,rt,(64-VSID_BITS_##size); \
add rt,rt,rx; /* add high and low bits */ \
- /* Now, r3 == VSID (mod 2^36-1), and lies between 0 and \
+ /* NOTE: explanation based on VSID_BITS_##size = 36 \
+ * Now, r3 == VSID (mod 2^36-1), and lies between 0 and \
* 2^36-1+2^28-1. That in particular means that if r3 >= \
* 2^36-1, then r3+1 has the 2^36 bit set. So, if r3+1 has \
* the bit clear, r3 already has the answer we want, if it \
@@ -514,34 +521,6 @@ typedef struct {
})
#endif /* 1 */
-/*
- * This is only valid for addresses >= PAGE_OFFSET
- * The proto-VSID space is divided into two class
- * User: 0 to 2^(CONTEXT_BITS + USER_ESID_BITS) -1
- * kernel: 2^(CONTEXT_BITS + USER_ESID_BITS) to 2^(VSID_BITS) - 1
- *
- * With KERNEL_START at 0xc000000000000000, the proto vsid for
- * the kernel ends up with 0xc00000000 (36 bits). With 64TB
- * support we need to have kernel proto-VSID in the
- * [2^37 to 2^38 - 1] range due to the increased USER_ESID_BITS.
- */
-static inline unsigned long get_kernel_vsid(unsigned long ea, int ssize)
-{
- unsigned long proto_vsid;
- /*
- * We need to make sure proto_vsid for the kernel is
- * >= 2^(CONTEXT_BITS + USER_ESID_BITS[_1T])
- */
- if (ssize == MMU_SEGSIZE_256M) {
- proto_vsid = ea >> SID_SHIFT;
- proto_vsid |= (1UL << (CONTEXT_BITS + USER_ESID_BITS));
- return vsid_scramble(proto_vsid, 256M);
- }
- proto_vsid = ea >> SID_SHIFT_1T;
- proto_vsid |= (1UL << (CONTEXT_BITS + USER_ESID_BITS_1T));
- return vsid_scramble(proto_vsid, 1T);
-}
-
/* Returns the segment size indicator for a user address */
static inline int user_segment_size(unsigned long addr)
{
@@ -551,10 +530,15 @@ static inline int user_segment_size(unsigned long addr)
return MMU_SEGSIZE_256M;
}
-/* This is only valid for user addresses (which are below 2^44) */
static inline unsigned long get_vsid(unsigned long context, unsigned long ea,
int ssize)
{
+ /*
+ * Bad address. We return VSID 0 for that
+ */
+ if ((ea & ~REGION_MASK) >= PGTABLE_RANGE)
+ return 0;
+
if (ssize == MMU_SEGSIZE_256M)
return vsid_scramble((context << USER_ESID_BITS)
| (ea >> SID_SHIFT), 256M);
@@ -562,6 +546,25 @@ static inline unsigned long get_vsid(unsigned long context, unsigned long ea,
| (ea >> SID_SHIFT_1T), 1T);
}
+/*
+ * This is only valid for addresses >= PAGE_OFFSET
+ *
+ * For kernel space, we use the top 4 context ids to map address as below
+ * 0x7fffc - [ 0xc000000000000000 - 0xc0003fffffffffff ]
+ * 0x7fffd - [ 0xd000000000000000 - 0xd0003fffffffffff ]
+ * 0x7fffe - [ 0xe000000000000000 - 0xe0003fffffffffff ]
+ * 0x7ffff - [ 0xf000000000000000 - 0xf0003fffffffffff ]
+ */
+static inline unsigned long get_kernel_vsid(unsigned long ea, int ssize)
+{
+ unsigned long context;
+
+ /*
+ * kernel take the top 4 context from the available range
+ */
+ context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1;
+ return get_vsid(context, ea, ssize);
+}
#endif /* __ASSEMBLY__ */
#endif /* _ASM_POWERPC_MMU_HASH64_H_ */
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 4665e82..aec4be9 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1268,20 +1268,36 @@ do_ste_alloc:
_GLOBAL(do_stab_bolted)
stw r9,PACA_EXSLB+EX_CCR(r13) /* save CR in exc. frame */
std r11,PACA_EXSLB+EX_SRR0(r13) /* save SRR0 in exc. frame */
+ mfspr r11,SPRN_DAR /* ea */
+ /*
+ * check for bad kernel/user address
+ * (ea & ~REGION_MASK) >= PGTABLE_RANGE
+ */
+ rldicr. r9,r11,4,(63 - 46 - 4)
+ li r9,0 /* VSID = 0 for bad address */
+ bne- 0f
+
+ /*
+ * Calculate VSID:
+ * This is the kernel vsid, we take the top for context from
+ * the range. context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1
+ * Here we know that (ea >> 60) == 0xc
+ */
+ lis r9,(MAX_USER_CONTEXT + 1)@ha
+ addi r9,r9,(MAX_USER_CONTEXT + 1)@l
+
+ srdi r10,r11,SID_SHIFT
+ rldimi r10,r9,USER_ESID_BITS,0 /* proto vsid */
+ ASM_VSID_SCRAMBLE(r10, r9, 256M)
+ rldic r9,r10,12,16 /* r9 = vsid << 12 */
+
+0:
/* Hash to the primary group */
ld r10,PACASTABVIRT(r13)
- mfspr r11,SPRN_DAR
- srdi r11,r11,28
+ srdi r11,r11,SID_SHIFT
rldimi r10,r11,7,52 /* r10 = first ste of the group */
- /* Calculate VSID */
- /* This is a kernel address, so protovsid = ESID | 1 << 37 */
- li r9,0x1
- rldimi r11,r9,(CONTEXT_BITS + USER_ESID_BITS),0
- ASM_VSID_SCRAMBLE(r11, r9, 256M)
- rldic r9,r11,12,16 /* r9 = vsid << 12 */
-
/* Search the primary group for a free entry */
1: ld r11,0(r10) /* Test valid bit of the current ste */
andi. r11,r11,0x80
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 3a292be..bfeab83 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -194,6 +194,11 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
unsigned long vpn = hpt_vpn(vaddr, vsid, ssize);
unsigned long tprot = prot;
+ /*
+ * If we hit a bad address return error.
+ */
+ if (!vsid)
+ return -1;
/* Make kernel text executable */
if (overlaps_kernel_text(vaddr, vaddr + step))
tprot &= ~HPTE_R_N;
@@ -921,11 +926,6 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap)
DBG_LOW("hash_page(ea=%016lx, access=%lx, trap=%lx\n",
ea, access, trap);
- if ((ea & ~REGION_MASK) >= PGTABLE_RANGE) {
- DBG_LOW(" out of pgtable range !\n");
- return 1;
- }
-
/* Get region & vsid */
switch (REGION_ID(ea)) {
case USER_REGION_ID:
@@ -956,6 +956,11 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap)
}
DBG_LOW(" mm=%p, mm->pgdir=%p, vsid=%016lx\n", mm, mm->pgd, vsid);
+ /* Bad address. */
+ if (!vsid) {
+ DBG_LOW("Bad address!\n");
+ return 1;
+ }
/* Get pgdir */
pgdir = mm->pgd;
if (pgdir == NULL)
@@ -1125,6 +1130,8 @@ void hash_preload(struct mm_struct *mm, unsigned long ea,
/* Get VSID */
ssize = user_segment_size(ea);
vsid = get_vsid(mm->context.id, ea, ssize);
+ if (!vsid)
+ return;
/* Hash doesn't like irqs */
local_irq_save(flags);
@@ -1217,6 +1224,9 @@ static void kernel_map_linear_page(unsigned long vaddr, unsigned long lmi)
hash = hpt_hash(vpn, PAGE_SHIFT, mmu_kernel_ssize);
hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP);
+ /* Don't create HPTE entries for bad address */
+ if (!vsid)
+ return;
ret = ppc_md.hpte_insert(hpteg, vpn, __pa(vaddr),
mode, HPTE_V_BOLTED,
mmu_linear_psize, mmu_kernel_ssize);
diff --git a/arch/powerpc/mm/mmu_context_hash64.c b/arch/powerpc/mm/mmu_context_hash64.c
index 40bc5b0..d1d1b92 100644
--- a/arch/powerpc/mm/mmu_context_hash64.c
+++ b/arch/powerpc/mm/mmu_context_hash64.c
@@ -29,15 +29,6 @@
static DEFINE_SPINLOCK(mmu_context_lock);
static DEFINE_IDA(mmu_context_ida);
-/*
- * 256MB segment
- * The proto-VSID space has 2^(CONTEX_BITS + USER_ESID_BITS) - 1 segments
- * available for user mappings. Each segment contains 2^28 bytes. Each
- * context maps 2^46 bytes (64TB) so we can support 2^19-1 contexts
- * (19 == 37 + 28 - 46).
- */
-#define MAX_CONTEXT ((1UL << CONTEXT_BITS) - 1)
-
int __init_new_context(void)
{
int index;
@@ -56,7 +47,7 @@ again:
else if (err)
return err;
- if (index > MAX_CONTEXT) {
+ if (index > MAX_USER_CONTEXT) {
spin_lock(&mmu_context_lock);
ida_remove(&mmu_context_ida, index);
spin_unlock(&mmu_context_lock);
diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S
index 1a16ca2..77aafaa 100644
--- a/arch/powerpc/mm/slb_low.S
+++ b/arch/powerpc/mm/slb_low.S
@@ -31,10 +31,15 @@
* No other registers are examined or changed.
*/
_GLOBAL(slb_allocate_realmode)
- /* r3 = faulting address */
+ /*
+ * check for bad kernel/user address
+ * (ea & ~REGION_MASK) >= PGTABLE_RANGE
+ */
+ rldicr. r9,r3,4,(63 - 46 - 4)
+ bne- 8f
srdi r9,r3,60 /* get region */
- srdi r10,r3,28 /* get esid */
+ srdi r10,r3,SID_SHIFT /* get esid */
cmpldi cr7,r9,0xc /* cmp PAGE_OFFSET for later use */
/* r3 = address, r10 = esid, cr7 = <> PAGE_OFFSET */
@@ -56,12 +61,14 @@ _GLOBAL(slb_allocate_realmode)
*/
_GLOBAL(slb_miss_kernel_load_linear)
li r11,0
- li r9,0x1
/*
- * for 1T we shift 12 bits more. slb_finish_load_1T will do
- * the necessary adjustment
+ * context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1
+ * r9 = region id.
*/
- rldimi r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0
+ addis r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@ha
+ addi r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@l
+
+
BEGIN_FTR_SECTION
b slb_finish_load
END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
@@ -91,24 +98,19 @@ _GLOBAL(slb_miss_kernel_load_vmemmap)
_GLOBAL(slb_miss_kernel_load_io)
li r11,0
6:
- li r9,0x1
/*
- * for 1T we shift 12 bits more. slb_finish_load_1T will do
- * the necessary adjustment
+ * context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1
+ * r9 = region id.
*/
- rldimi r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0
+ addis r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@ha
+ addi r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@l
+
BEGIN_FTR_SECTION
b slb_finish_load
END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
b slb_finish_load_1T
-0: /* user address: proto-VSID = context << 15 | ESID. First check
- * if the address is within the boundaries of the user region
- */
- srdi. r9,r10,USER_ESID_BITS
- bne- 8f /* invalid ea bits set */
-
-
+0:
/* when using slices, we extract the psize off the slice bitmaps
* and then we need to get the sllp encoding off the mmu_psize_defs
* array.
@@ -164,15 +166,13 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
ld r9,PACACONTEXTID(r13)
BEGIN_FTR_SECTION
cmpldi r10,0x1000
-END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT)
- rldimi r10,r9,USER_ESID_BITS,0
-BEGIN_FTR_SECTION
bge slb_finish_load_1T
END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT)
b slb_finish_load
8: /* invalid EA */
li r10,0 /* BAD_VSID */
+ li r9,0 /* BAD_VSID */
li r11,SLB_VSID_USER /* flags don't much matter */
b slb_finish_load
@@ -221,8 +221,6 @@ _GLOBAL(slb_allocate_user)
/* get context to calculate proto-VSID */
ld r9,PACACONTEXTID(r13)
- rldimi r10,r9,USER_ESID_BITS,0
-
/* fall through slb_finish_load */
#endif /* __DISABLED__ */
@@ -231,9 +229,10 @@ _GLOBAL(slb_allocate_user)
/*
* Finish loading of an SLB entry and return
*
- * r3 = EA, r10 = proto-VSID, r11 = flags, clobbers r9, cr7 = <> PAGE_OFFSET
+ * r3 = EA, r9 = context, r10 = ESID, r11 = flags, clobbers r9, cr7 = <> PAGE_OFFSET
*/
slb_finish_load:
+ rldimi r10,r9,USER_ESID_BITS,0
ASM_VSID_SCRAMBLE(r10,r9,256M)
/*
* bits above VSID_BITS_256M need to be ignored from r10
@@ -298,10 +297,11 @@ _GLOBAL(slb_compare_rr_to_size)
/*
* Finish loading of a 1T SLB entry (for the kernel linear mapping) and return.
*
- * r3 = EA, r10 = proto-VSID, r11 = flags, clobbers r9
+ * r3 = EA, r9 = context, r10 = ESID(256MB), r11 = flags, clobbers r9
*/
slb_finish_load_1T:
- srdi r10,r10,40-28 /* get 1T ESID */
+ srdi r10,r10,(SID_SHIFT_1T - SID_SHIFT) /* get 1T ESID */
+ rldimi r10,r9,USER_ESID_BITS_1T,0
ASM_VSID_SCRAMBLE(r10,r9,1T)
/*
* bits above VSID_BITS_1T need to be ignored from r10
diff --git a/arch/powerpc/mm/tlb_hash64.c b/arch/powerpc/mm/tlb_hash64.c
index 0d82ef5..023ec8a 100644
--- a/arch/powerpc/mm/tlb_hash64.c
+++ b/arch/powerpc/mm/tlb_hash64.c
@@ -82,11 +82,11 @@ void hpte_need_flush(struct mm_struct *mm, unsigned long addr,
if (!is_kernel_addr(addr)) {
ssize = user_segment_size(addr);
vsid = get_vsid(mm->context.id, addr, ssize);
- WARN_ON(vsid == 0);
} else {
vsid = get_kernel_vsid(addr, mmu_kernel_ssize);
ssize = mmu_kernel_ssize;
}
+ WARN_ON(vsid == 0);
vpn = hpt_vpn(addr, vsid, ssize);
rpte = __real_pte(__pte(pte), ptep);
--
1.7.10
^ permalink raw reply related
* [PATCH -V4 3/3] powerpc: rename USER_ESID_BITS* to ESID_BITS*
From: Aneesh Kumar K.V @ 2013-03-13 13:34 UTC (permalink / raw)
To: benh, paulus, David Gibson; +Cc: linuxppc-dev, Aneesh Kumar K.V
In-Reply-To: <1363181695-5315-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Now we use ESID_BITS of kernel address to build proto vsid. So rename
USER_ESIT_BITS to ESID_BITS
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/mmu-hash64.h | 16 ++++++++--------
arch/powerpc/kernel/exceptions-64s.S | 2 +-
arch/powerpc/kvm/book3s_64_mmu_host.c | 4 ++--
arch/powerpc/mm/pgtable_64.c | 2 +-
arch/powerpc/mm/slb_low.S | 4 ++--
5 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/include/asm/mmu-hash64.h b/arch/powerpc/include/asm/mmu-hash64.h
index a32461f..b59e06f 100644
--- a/arch/powerpc/include/asm/mmu-hash64.h
+++ b/arch/powerpc/include/asm/mmu-hash64.h
@@ -378,12 +378,12 @@ extern void slb_set_size(u16 size);
*/
#define CONTEXT_BITS 19
-#define USER_ESID_BITS 18
-#define USER_ESID_BITS_1T 6
+#define ESID_BITS 18
+#define ESID_BITS_1T 6
/*
* 256MB segment
- * The proto-VSID space has 2^(CONTEX_BITS + USER_ESID_BITS) - 1 segments
+ * The proto-VSID space has 2^(CONTEX_BITS + ESID_BITS) - 1 segments
* available for user + kernel mapping. The top 4 contexts are used for
* kernel mapping. Each segment contains 2^28 bytes. Each
* context maps 2^46 bytes (64TB) so we can support 2^19-1 contexts
@@ -396,15 +396,15 @@ extern void slb_set_size(u16 size);
* doesn't overflow 64 bits. It should also be co-prime to vsid_modulus
*/
#define VSID_MULTIPLIER_256M ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_256M (CONTEXT_BITS + USER_ESID_BITS)
+#define VSID_BITS_256M (CONTEXT_BITS + ESID_BITS)
#define VSID_MODULUS_256M ((1UL<<VSID_BITS_256M)-1)
#define VSID_MULTIPLIER_1T ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_1T (CONTEXT_BITS + USER_ESID_BITS_1T)
+#define VSID_BITS_1T (CONTEXT_BITS + ESID_BITS_1T)
#define VSID_MODULUS_1T ((1UL<<VSID_BITS_1T)-1)
-#define USER_VSID_RANGE (1UL << (USER_ESID_BITS + SID_SHIFT))
+#define USER_VSID_RANGE (1UL << (ESID_BITS + SID_SHIFT))
/*
* This macro generates asm code to compute the VSID scramble
@@ -540,9 +540,9 @@ static inline unsigned long get_vsid(unsigned long context, unsigned long ea,
return 0;
if (ssize == MMU_SEGSIZE_256M)
- return vsid_scramble((context << USER_ESID_BITS)
+ return vsid_scramble((context << ESID_BITS)
| (ea >> SID_SHIFT), 256M);
- return vsid_scramble((context << USER_ESID_BITS_1T)
+ return vsid_scramble((context << ESID_BITS_1T)
| (ea >> SID_SHIFT_1T), 1T);
}
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index aec4be9..3684cbd 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -1288,7 +1288,7 @@ _GLOBAL(do_stab_bolted)
addi r9,r9,(MAX_USER_CONTEXT + 1)@l
srdi r10,r11,SID_SHIFT
- rldimi r10,r9,USER_ESID_BITS,0 /* proto vsid */
+ rldimi r10,r9,ESID_BITS,0 /* proto vsid */
ASM_VSID_SCRAMBLE(r10, r9, 256M)
rldic r9,r10,12,16 /* r9 = vsid << 12 */
diff --git a/arch/powerpc/kvm/book3s_64_mmu_host.c b/arch/powerpc/kvm/book3s_64_mmu_host.c
index ead58e3..5d7d29a 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_host.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_host.c
@@ -326,8 +326,8 @@ int kvmppc_mmu_init(struct kvm_vcpu *vcpu)
vcpu3s->context_id[0] = err;
vcpu3s->proto_vsid_max = ((vcpu3s->context_id[0] + 1)
- << USER_ESID_BITS) - 1;
- vcpu3s->proto_vsid_first = vcpu3s->context_id[0] << USER_ESID_BITS;
+ << ESID_BITS) - 1;
+ vcpu3s->proto_vsid_first = vcpu3s->context_id[0] << ESID_BITS;
vcpu3s->proto_vsid_next = vcpu3s->proto_vsid_first;
kvmppc_mmu_hpte_init(vcpu);
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index e212a27..654258f 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -61,7 +61,7 @@
#endif
#ifdef CONFIG_PPC_STD_MMU_64
-#if TASK_SIZE_USER64 > (1UL << (USER_ESID_BITS + SID_SHIFT))
+#if TASK_SIZE_USER64 > (1UL << (ESID_BITS + SID_SHIFT))
#error TASK_SIZE_USER64 exceeds user VSID range
#endif
#endif
diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S
index 77aafaa..17aa6df 100644
--- a/arch/powerpc/mm/slb_low.S
+++ b/arch/powerpc/mm/slb_low.S
@@ -232,7 +232,7 @@ _GLOBAL(slb_allocate_user)
* r3 = EA, r9 = context, r10 = ESID, r11 = flags, clobbers r9, cr7 = <> PAGE_OFFSET
*/
slb_finish_load:
- rldimi r10,r9,USER_ESID_BITS,0
+ rldimi r10,r9,ESID_BITS,0
ASM_VSID_SCRAMBLE(r10,r9,256M)
/*
* bits above VSID_BITS_256M need to be ignored from r10
@@ -301,7 +301,7 @@ _GLOBAL(slb_compare_rr_to_size)
*/
slb_finish_load_1T:
srdi r10,r10,(SID_SHIFT_1T - SID_SHIFT) /* get 1T ESID */
- rldimi r10,r9,USER_ESID_BITS_1T,0
+ rldimi r10,r9,ESID_BITS_1T,0
ASM_VSID_SCRAMBLE(r10,r9,1T)
/*
* bits above VSID_BITS_1T need to be ignored from r10
--
1.7.10
^ permalink raw reply related
* [PATCH -V4 1/3] powerpc: Make VSID_BITS* dependency explicit
From: Aneesh Kumar K.V @ 2013-03-13 13:34 UTC (permalink / raw)
To: benh, paulus, David Gibson; +Cc: linuxppc-dev, Aneesh Kumar K.V
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
VSID_BITS and VSID_BITS_1T depends on the context bits and user esid
bits. Make the dependency explicit
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/mmu-hash64.h | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/mmu-hash64.h b/arch/powerpc/include/asm/mmu-hash64.h
index 2fdb47a..5f8c2bd 100644
--- a/arch/powerpc/include/asm/mmu-hash64.h
+++ b/arch/powerpc/include/asm/mmu-hash64.h
@@ -381,21 +381,22 @@ extern void slb_set_size(u16 size);
* hash collisions.
*/
+#define CONTEXT_BITS 19
+#define USER_ESID_BITS 18
+#define USER_ESID_BITS_1T 6
+
/*
* This should be computed such that protovosid * vsid_mulitplier
* doesn't overflow 64 bits. It should also be co-prime to vsid_modulus
*/
#define VSID_MULTIPLIER_256M ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_256M 38
+#define VSID_BITS_256M (CONTEXT_BITS + USER_ESID_BITS + 1)
#define VSID_MODULUS_256M ((1UL<<VSID_BITS_256M)-1)
#define VSID_MULTIPLIER_1T ASM_CONST(12538073) /* 24-bit prime */
-#define VSID_BITS_1T 26
+#define VSID_BITS_1T (CONTEXT_BITS + USER_ESID_BITS_1T + 1)
#define VSID_MODULUS_1T ((1UL<<VSID_BITS_1T)-1)
-#define CONTEXT_BITS 19
-#define USER_ESID_BITS 18
-#define USER_ESID_BITS_1T 6
#define USER_VSID_RANGE (1UL << (USER_ESID_BITS + SID_SHIFT))
--
1.7.10
^ permalink raw reply related
* Re: [PATCH -V2 07/26] powerpc: Add size argument to pgtable_cache_add
From: Paul Mackerras @ 2013-03-13 11:34 UTC (permalink / raw)
To: Aneesh Kumar K.V; +Cc: linuxppc-dev
In-Reply-To: <877glbd536.fsf@linux.vnet.ibm.com>
On Wed, Mar 13, 2013 at 02:55:17PM +0530, Aneesh Kumar K.V wrote:
> Paul Mackerras <paulus@samba.org> writes:
> > NAK to this (and in fact the rest of this patch).
> >
> > If you need twice the size, use the next slot along in pgtable_cache,
> > which will give you the size you need, rather than breaking the
> > relationship between slot index and allocation size.
>
> That would mean we will have to do
>
> static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
> {
> - return kmem_cache_alloc(PGT_CACHE(PMD_INDEX_SIZE),
> + return kmem_cache_alloc(PGT_CACHE(PMD_INDEX_SIZE + 1),
> GFP_KERNEL|__GFP_REPEAT);
> }
>
> is that ok ?
I would define a symbol such as PMD_CACHE_INDEX and arrange for that
to be either PMD_INDEX_SIZE or PMD_INDEX_SIZE + 1, as needed, and then
use PGT_CACHE(PMD_CACHE_INDEX) everywhere instead of
PGT_CACHE(PMD_INDEX_SIZE). If you want, you could also do
PUD_CACHE_INDEX and PGD_CACHE_INDEX for consistency, but it's not
totally necessary.
Paul.
^ permalink raw reply
* Re: [PATCH] Make PCIe hotplug work with Freescale PCIe controllers
From: Rojhalat Ibrahim @ 2013-03-13 10:06 UTC (permalink / raw)
To: Kumar Gala; +Cc: linuxppc-dev
In-Reply-To: <B392DAB5-F9A7-4B1A-8CAE-015CA5AC2C33@kernel.crashing.org>
On Tuesday 12 March 2013 15:48:01 Kumar Gala wrote:
> > I'd rather we just export indirect_read_config() & indirect_write_config()
> > from indirect_pci.c and call the functions directly. Adding a global and
> > call them via a function pointer seems wrong to me.
> >
> > - k
>
> Also, can you base this patch on my powerpc.git next branch as fsl_pci.{c,h}
> have some changes in them.
>
> - k
>
Hi Kumar,
here's the patch with direct calls to indirect_{r,w}_config based on your powerpc.git next branch.
This does not work for PCIe IP rev 3.0 as I'm not sure how to handle "struct resource" aside from making it global or changing "struct pci_controller".
Rojhalat
Signed-off-by: Rojhalat Ibrahim <imr@rtschenk.de>
---
arch/powerpc/sysdev/fsl_pci.c | 51 ++++++++++++++++++++++++++++++++++---
arch/powerpc/sysdev/indirect_pci.c | 10 ++-----
2 files changed, 51 insertions(+), 10 deletions(-)
diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c
index 3271177..1d7da9d 100644
--- a/arch/powerpc/sysdev/fsl_pci.c
+++ b/arch/powerpc/sysdev/fsl_pci.c
@@ -36,6 +36,11 @@
static int fsl_pcie_bus_fixup, is_mpc83xx_pci;
+extern int indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val);
+extern int indirect_write_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 val);
+
static void quirk_fsl_pcie_header(struct pci_dev *dev)
{
u8 hdr_type;
@@ -86,6 +91,44 @@ static int __init fsl_pcie_check_link(struct pci_controller *hose,
return 0;
}
+static int fsl_indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val)
+{
+ struct pci_controller *hose = pci_bus_to_host(bus);
+
+ // check the link status
+ if ((bus->number == hose->first_busno) && (devfn == 0)) {
+ u32 ltssm = 0;
+ indirect_read_config(bus, 0, PCIE_LTSSM, 4, <ssm);
+ if (ltssm < PCIE_LTSSM_L0) {
+ hose->indirect_type |= PPC_INDIRECT_TYPE_NO_PCIE_LINK;
+ } else {
+ hose->indirect_type &= ~PPC_INDIRECT_TYPE_NO_PCIE_LINK;
+ }
+ }
+ return indirect_read_config(bus, devfn, offset, len, val);
+}
+
+static int fsl_indirect_write_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 val)
+{
+ return indirect_write_config(bus, devfn, offset, len, val);
+}
+
+static struct pci_ops fsl_indirect_pci_ops =
+{
+ .read = fsl_indirect_read_config,
+ .write = fsl_indirect_write_config,
+};
+
+static void __init fsl_setup_indirect_pci(struct pci_controller* hose,
+ resource_size_t cfg_addr,
+ resource_size_t cfg_data, u32 flags)
+{
+ setup_indirect_pci(hose, cfg_addr, cfg_data, flags);
+ hose->ops = &fsl_indirect_pci_ops;
+}
+
#if defined(CONFIG_FSL_SOC_BOOKE) || defined(CONFIG_PPC_86xx)
#define MAX_PHYS_ADDR_BITS 40
@@ -483,8 +526,8 @@ int __init fsl_add_bridge(struct platform_device *pdev, int is_primary)
hose->first_busno = bus_range ? bus_range[0] : 0x0;
hose->last_busno = bus_range ? bus_range[1] : 0xff;
- setup_indirect_pci(hose, rsrc.start, rsrc.start + 0x4,
- PPC_INDIRECT_TYPE_BIG_ENDIAN);
+ fsl_setup_indirect_pci(hose, rsrc.start, rsrc.start + 0x4,
+ PPC_INDIRECT_TYPE_BIG_ENDIAN);
if (early_find_capability(hose, 0, 0, PCI_CAP_ID_EXP)) {
/* For PCIE read HEADER_TYPE to identify controler mode */
@@ -788,8 +831,8 @@ int __init mpc83xx_add_bridge(struct device_node *dev)
if (ret)
goto err0;
} else {
- setup_indirect_pci(hose, rsrc_cfg.start,
- rsrc_cfg.start + 4, 0);
+ fsl_setup_indirect_pci(hose, rsrc_cfg.start,
+ rsrc_cfg.start + 4, 0);
}
printk(KERN_INFO "Found FSL PCI host bridge at 0x%016llx. "
diff --git a/arch/powerpc/sysdev/indirect_pci.c b/arch/powerpc/sysdev/indirect_pci.c
index 82fdad8..c6c8b52 100644
--- a/arch/powerpc/sysdev/indirect_pci.c
+++ b/arch/powerpc/sysdev/indirect_pci.c
@@ -20,9 +20,8 @@
#include <asm/pci-bridge.h>
#include <asm/machdep.h>
-static int
-indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
- int len, u32 *val)
+int indirect_read_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 *val)
{
struct pci_controller *hose = pci_bus_to_host(bus);
volatile void __iomem *cfg_data;
@@ -78,9 +77,8 @@ indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
return PCIBIOS_SUCCESSFUL;
}
-static int
-indirect_write_config(struct pci_bus *bus, unsigned int devfn, int offset,
- int len, u32 val)
+int indirect_write_config(struct pci_bus *bus, unsigned int devfn,
+ int offset, int len, u32 val)
{
struct pci_controller *hose = pci_bus_to_host(bus);
volatile void __iomem *cfg_data;
^ permalink raw reply related
* Re: [PATCH -V3 2/3] powerpc: Update kernel VSID range
From: Benjamin Herrenschmidt @ 2013-03-13 9:49 UTC (permalink / raw)
To: Aneesh Kumar K.V; +Cc: linuxppc-dev, Paul Mackerras, David Gibson
In-Reply-To: <87a9q7d5gt.fsf@linux.vnet.ibm.com>
On Wed, 2013-03-13 at 14:47 +0530, Aneesh Kumar K.V wrote:
> I update the commit message but missed updating this.
>
> Ben,
>
> Let me know if you want me to send V4 to take care of these review
> feedbacks.
Yes please, I'll fix up my local merge branch with your new patch
tomorrow (I had applied but not pushed out yet) and will push out the
result to Linus.
Cheers,
Ben.
^ permalink raw reply
* RE: [PATCH V4] powerpc/85xx: Add machine check handler to fix PCIe erratum on mpc85xx
From: David Laight @ 2013-03-13 9:40 UTC (permalink / raw)
To: Scott Wood, Jia Hongtao-B38951
Cc: Wood Scott-B07421, linuxppc-dev, Stuart Yoder
In-Reply-To: <1363123452.17135.13@snotra>
> Hmm, seems there's no probe_user_address() -- for userspace we
> basically want the same thing minus the KERNEL_DS. See
> arch/powerpc/perf/callchain.c for an example.
Isn't that just copy_from_user() ?
David
^ permalink raw reply
* Re: [PATCH -V2 07/26] powerpc: Add size argument to pgtable_cache_add
From: Aneesh Kumar K.V @ 2013-03-13 9:25 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev
In-Reply-To: <20130313024725.GA21125@iris.ozlabs.ibm.com>
Paul Mackerras <paulus@samba.org> writes:
> On Wed, Mar 06, 2013 at 11:40:08AM +0530, Aneesh Kumar K.V wrote:
>> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>>
>> We will use this later with THP changes to request for pmd table of double the size.
>> THP code does PTE page allocation along with large page request and deposit them
>> for later use. This is to ensure that we won't have any failures when we split
>> huge pages to regular pages.
>>
>> On powerpc we want to use the deposited PTE page for storing hash pte slot and
>> secondary bit information for the HPTEs. Hence we save them in the second half
>> of the pmd table.
>
> That much is fine, but...
>
>> -void pgtable_cache_add(unsigned shift, void (*ctor)(void *));
>> +extern void __pgtable_cache_add(unsigned shift, unsigned long table_size,
>> + void (*ctor)(void *));
>> void pgtable_cache_init(void);
>> +static inline void pgtable_cache_add(unsigned shift, void (*ctor)(void *))
>> +{
>> + return __pgtable_cache_add(shift, sizeof(void *) << shift, ctor);
>> +}
>
> NAK to this (and in fact the rest of this patch).
>
> If you need twice the size, use the next slot along in pgtable_cache,
> which will give you the size you need, rather than breaking the
> relationship between slot index and allocation size.
That would mean we will have to do
static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
{
- return kmem_cache_alloc(PGT_CACHE(PMD_INDEX_SIZE),
+ return kmem_cache_alloc(PGT_CACHE(PMD_INDEX_SIZE + 1),
GFP_KERNEL|__GFP_REPEAT);
}
is that ok ?
-aneesh
^ permalink raw reply
* Re: [PATCH -V3 2/3] powerpc: Update kernel VSID range
From: Aneesh Kumar K.V @ 2013-03-13 9:17 UTC (permalink / raw)
To: Paul Mackerras; +Cc: linuxppc-dev, David Gibson
In-Reply-To: <20130313034254.GC21125@iris.ozlabs.ibm.com>
Paul Mackerras <paulus@samba.org> writes:
> On Tue, Mar 12, 2013 at 05:38:50PM +0530, Aneesh Kumar K.V wrote:
>> From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
>>
>> This patch change the kernel VSID range so that we limit VSID_BITS to 37.
>> This enables us to support 64TB with 65 bit VA (37+28). Without this patch
>> we have boot hangs on platforms that only support 65 bit VA.
>>
>> With this patch we now have proto vsid generated as below:
>>
>> We first generate a 37-bit "proto-VSID". Proto-VSIDs are generated
>> from mmu context id and effective segment id of the address.
>>
>> For user processes max context id is limited to ((1ul << 19) - 5)
>> for kernel space, we use the top 4 context ids to map address as below
>> 0x7fffc - [ 0xc000000000000000 - 0xc0003fffffffffff ]
>> 0x7fffd - [ 0xd000000000000000 - 0xd0003fffffffffff ]
>> 0x7fffe - [ 0xe000000000000000 - 0xe0003fffffffffff ]
>> 0x7ffff - [ 0xf000000000000000 - 0xf0003fffffffffff ]
>>
>> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
>
> Mostly looks OK, and it could go in as is, so
>
> Acked-by: Paul Mackerras <paulus@samba.org>
>
> Some minor comments below...
>
>> + * For user processes max context id is limited to ((1ul << 19) - 6)
>
> should be ((1ul << 19) - 5)
I update the commit message but missed updating this.
Ben,
Let me know if you want me to send V4 to take care of these review
feedbacks.
-aneesh
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox