arm64: dma_mapping: allow PCI host driver to limit DMA mask

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host
bridge has limitations on inbound transactions addressing. Example of
such setup is NVME SSD device connected to RCAR PCIe controller.

Previously there was attempt to handle this via bus notifier: after
driver is attached to PCI device, bridge driver gets notifier callback,
and resets dma_mask from there. However, this is racy: PCI device driver
could already allocate buffers and/or start i/o in probe routine.
In NVME case, i/o is started in workqueue context, and this race gives
"sometimes works, sometimes not" effect.

Proper solution should make driver's dma_set_mask() call to fail if host
bridge can't support mask being set.

This patch makes __swiotlb_dma_supported() to check mask being set for
PCI device against dma_mask of struct device corresponding to PCI host
bridge (one with name "pciXXXX:YY"), if that dma_mask is set.

This is the least destructive approach: currently dma_mask of that device
object is not used anyhow, thus all existing setups will work as before,
and modification is required only in actually affected components -
driver of particular PCI host bridge, and dma_map_ops of particular
platform.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
---
 arch/arm64/mm/dma-mapping.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..49645277 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -28,6 +28,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/pci.h>
 
 #include <asm/cacheflush.h>
 

@@ -347,6 +348,16 @@ static int __swiotlb_get_sgtable(struct device *dev, struct sg_table *sgt,
 
 static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
+#ifdef CONFIG_PCI
+	if (dev_is_pci(hwdev)) {
+		struct pci_dev *pdev = to_pci_dev(hwdev);
+		struct pci_host_bridge *br = pci_find_host_bridge(pdev->bus);
+
+		if (br->dev.dma_mask && (*br->dev.dma_mask) &&
+				(mask & (*br->dev.dma_mask)) != mask)
+			return 0;
+	}
+#endif
 	if (swiotlb)
 		return swiotlb_dma_supported(hwdev, mask);
 	return 1;

-- 
2.1.4

This gives platform DMA mapping code a chance to disallow setting device
DMA mask to something that host bridge can't support.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
---
 drivers/pci/host/pcie-rcar.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/pci/host/pcie-rcar.c b/drivers/pci/host/pcie-rcar.c
index aca85be..b1edc3c 100644
--- a/drivers/pci/host/pcie-rcar.c
+++ b/drivers/pci/host/pcie-rcar.c

@@ -451,6 +451,7 @@ static int rcar_pcie_enable(struct rcar_pcie *pcie)
 {
 	struct device *dev = pcie->dev;
 	struct pci_bus *bus, *child;
+	struct pci_host_bridge *bridge;
 	LIST_HEAD(res);
 
 	/* Try setting 5 GT/s link speed */

@@ -480,6 +481,10 @@ static int rcar_pcie_enable(struct rcar_pcie *pcie)
 	list_for_each_entry(child, &bus->children, node)
 		pcie_bus_configure_settings(child);
 
+	bridge = pci_find_host_bridge(bus);
+	bridge->dev.coherent_dma_mask = DMA_BIT_MASK(32);
+	bridge->dev.dma_mask = &bridge->dev.coherent_dma_mask;
+
 	pci_bus_add_devices(bus);
 
 	return 0;

-- 
2.1.4

Hello!

On 12/29/2016 11:45 PM, Nikita Yushchenko wrote:

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host

    Its.

bridge has limitations on inbound transactions addressing. Example of
such setup is NVME

    Isn't it called NVMe?

SSD device connected to RCAR PCIe controller.

    R=Car.

Previously there was attempt to handle this via bus notifier: after
driver is attached to PCI device, bridge driver gets notifier callback,
and resets dma_mask from there. However, this is racy: PCI device driver
could already allocate buffers and/or start i/o in probe routine.
In NVME case, i/o is started in workqueue context, and this race gives
"sometimes works, sometimes not" effect.

Proper solution should make driver's dma_set_mask() call to fail if host
bridge can't support mask being set.

This patch makes __swiotlb_dma_supported() to check mask being set for

    "To" not needed here.

[...]

    Hum, inner parens not necessary?

[...]

MBR, Sergei

On 12/30/2016 12:46 PM, Sergei Shtylyov wrote:

quoted

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host

   Its.

quoted

bridge has limitations on inbound transactions addressing. Example of
such setup is NVME

   Isn't it called NVMe?

quoted

SSD device connected to RCAR PCIe controller.

   R=Car.

    Sorry, R-Car. :-)

[...]

MBR, Sergei

On Thu, Dec 29, 2016 at 11:45:03PM +0300, Nikita Yushchenko wrote:

16 more quoted lines

Hmm, but this makes it look like the problem is both arm64 and swiotlb
specific, when in reality it's not. Perhaps another hack you could try
would be to register a PCI bus notifier in the host bridge looking for
BUS_NOTIFY_BIND_DRIVER, then you could proxy the DMA ops for each child
device before the driver has probed, but adding a dma_set_mask callback
to limit the mask to what you need?

I agree that it would be better if dma_set_mask handled all of this
transparently, but it's all based on the underlying ops rather than the
bus type.

Will

03.01.2017 21:44, Will Deacon ?????:

On Thu, Dec 29, 2016 at 11:45:03PM +0300, Nikita Yushchenko wrote:

quoted

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host
bridge has limitations on inbound transactions addressing. Example of
such setup is NVME SSD device connected to RCAR PCIe controller.

Previously there was attempt to handle this via bus notifier: after
driver is attached to PCI device, bridge driver gets notifier callback,
and resets dma_mask from there. However, this is racy: PCI device driver
could already allocate buffers and/or start i/o in probe routine.
In NVME case, i/o is started in workqueue context, and this race gives
"sometimes works, sometimes not" effect.

Proper solution should make driver's dma_set_mask() call to fail if host
bridge can't support mask being set.

This patch makes __swiotlb_dma_supported() to check mask being set for
PCI device against dma_mask of struct device corresponding to PCI host
bridge (one with name "pciXXXX:YY"), if that dma_mask is set.

This is the least destructive approach: currently dma_mask of that device
object is not used anyhow, thus all existing setups will work as before,
and modification is required only in actually affected components -
driver of particular PCI host bridge, and dma_map_ops of particular
platform.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
---
 arch/arm64/mm/dma-mapping.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..49645277 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -28,6 +28,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/pci.h>

24 more quoted lines

This is what Renesas BSP tries to do and it does not work.

BUS_NOTIFY_BIND_DRIVER arrives after driver's probe routine exits, but
i/o can be started before that.

quoted

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host
bridge has limitations on inbound transactions addressing. Example of
such setup is NVME SSD device connected to RCAR PCIe controller.

Previously there was attempt to handle this via bus notifier: after
driver is attached to PCI device, bridge driver gets notifier callback,
and resets dma_mask from there. However, this is racy: PCI device driver
could already allocate buffers and/or start i/o in probe routine.
In NVME case, i/o is started in workqueue context, and this race gives
"sometimes works, sometimes not" effect.

Proper solution should make driver's dma_set_mask() call to fail if host
bridge can't support mask being set.

This patch makes __swiotlb_dma_supported() to check mask being set for
PCI device against dma_mask of struct device corresponding to PCI host
bridge (one with name "pciXXXX:YY"), if that dma_mask is set.

This is the least destructive approach: currently dma_mask of that device
object is not used anyhow, thus all existing setups will work as before,
and modification is required only in actually affected components -
driver of particular PCI host bridge, and dma_map_ops of particular
platform.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
---
 arch/arm64/mm/dma-mapping.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..49645277 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -28,6 +28,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/pci.h>
 

23 more quoted lines

This is what Renesas BSP tries to do and it does not work.

BUS_NOTIFY_BIND_DRIVER arrives after driver's probe routine exits, but
i/o can be started before that.

On 01/03/2017 01:01 PM, Nikita Yushchenko wrote:

quoted

quoted

It is possible that PCI device supports 64-bit DMA addressing, and thus
it's driver sets device's dma_mask to DMA_BIT_MASK(64), however PCI host
bridge has limitations on inbound transactions addressing. Example of
such setup is NVME SSD device connected to RCAR PCIe controller.

Previously there was attempt to handle this via bus notifier: after
driver is attached to PCI device, bridge driver gets notifier callback,
and resets dma_mask from there. However, this is racy: PCI device driver
could already allocate buffers and/or start i/o in probe routine.
In NVME case, i/o is started in workqueue context, and this race gives
"sometimes works, sometimes not" effect.

Proper solution should make driver's dma_set_mask() call to fail if host
bridge can't support mask being set.

This patch makes __swiotlb_dma_supported() to check mask being set for
PCI device against dma_mask of struct device corresponding to PCI host
bridge (one with name "pciXXXX:YY"), if that dma_mask is set.

This is the least destructive approach: currently dma_mask of that device
object is not used anyhow, thus all existing setups will work as before,
and modification is required only in actually affected components -
driver of particular PCI host bridge, and dma_map_ops of particular
platform.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
---
 arch/arm64/mm/dma-mapping.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..49645277 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -28,6 +28,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/pci.h>
 

28 more quoted lines

Hm. This is strange statement:
 really_probe
 |->driver_sysfs_add
    |-> blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 					     BUS_NOTIFY_BIND_DRIVER, dev);
...
 |- ret = drv->probe(dev);
...
 |- driver_bound(dev);
    |- blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
					     BUS_NOTIFY_BOUND_DRIVER, dev);

Am I missing smth?

-- 
regards,
-grygorii

quoted

quoted

quoted

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..49645277 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -28,6 +28,7 @@
 #include <linux/dma-contiguous.h>
 #include <linux/vmalloc.h>
 #include <linux/swiotlb.h>
+#include <linux/pci.h>
 
 #include <asm/cacheflush.h>
 

@@ -347,6 +348,16 @@ static int __swiotlb_get_sgtable(struct device *dev, struct sg_table *sgt,
 
 static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
+#ifdef CONFIG_PCI
+	if (dev_is_pci(hwdev)) {
+		struct pci_dev *pdev = to_pci_dev(hwdev);
+		struct pci_host_bridge *br = pci_find_host_bridge(pdev->bus);
+
+		if (br->dev.dma_mask && (*br->dev.dma_mask) &&
+				(mask & (*br->dev.dma_mask)) != mask)
+			return 0;
+	}
+#endif

Hmm, but this makes it look like the problem is both arm64 and swiotlb
specific, when in reality it's not. Perhaps another hack you could try
would be to register a PCI bus notifier in the host bridge looking for
BUS_NOTIFY_BIND_DRIVER, then you could proxy the DMA ops for each child
device before the driver has probed, but adding a dma_set_mask callback
to limit the mask to what you need?

This is what Renesas BSP tries to do and it does not work.

BUS_NOTIFY_BIND_DRIVER arrives after driver's probe routine exits, but
i/o can be started before that.

Hm. This is strange statement:

12 more quoted lines

I misinterpreted your message, sorry.

BSP attaches to BUS_NOTIFY_BOUND_DRIVER, not to BUS_NOTIFY_BIND_DRIVER,
and simply overwrites device's dma_mask there.  You are suggesting
something completely different.

I'll check if your approach is practical.


Currently powerpc architecture has one more approach implemented, they
use pci_controller structure provided by host bridge driver, and that
has a set_dma_mask() hook. Maybe extending this beyond powerpc could be
a good idea. However, that will require changing quite a few host bridge
drivers, without any gain for most of those...

On Tuesday, January 3, 2017 6:44:44 PM CET Will Deacon wrote:

quoted

@@ -347,6 +348,16 @@ static int __swiotlb_get_sgtable(struct device *dev, struct sg_table *sgt,
 
 static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 {
+#ifdef CONFIG_PCI
+     if (dev_is_pci(hwdev)) {
+             struct pci_dev *pdev = to_pci_dev(hwdev);
+             struct pci_host_bridge *br = pci_find_host_bridge(pdev->bus);
+
+             if (br->dev.dma_mask && (*br->dev.dma_mask) &&
+                             (mask & (*br->dev.dma_mask)) != mask)
+                     return 0;
+     }
+#endif

Hmm, but this makes it look like the problem is both arm64 and swiotlb
specific, when in reality it's not. Perhaps another hack you could try
would be to register a PCI bus notifier in the host bridge looking for
BUS_NOTIFY_BIND_DRIVER, then you could proxy the DMA ops for each child
device before the driver has probed, but adding a dma_set_mask callback
to limit the mask to what you need?

I agree that it would be better if dma_set_mask handled all of this
transparently, but it's all based on the underlying ops rather than the
bus type.

This is what I prototyped a long time ago when this first came up.
I still think this needs to be solved properly for all of arm64, not
with a PCI specific hack, and in particular not using notifiers.

	Arnd

commit 9a57d58d116800a535510053136c6dd7a9c26e25
Author: Arnd Bergmann [off-list ref]
Date:   Tue Nov 17 14:06:55 2015 +0100

    [EXPERIMENTAL] ARM64: check implement dma_set_mask
    
    Needs work for coherent mask
    
    Signed-off-by: Arnd Bergmann [off-list ref]

diff --git a/arch/arm64/include/asm/device.h b/arch/arm64/include/asm/device.h
index 243ef256b8c9..a57e7bb10e71 100644
--- a/arch/arm64/include/asm/device.h
+++ b/arch/arm64/include/asm/device.h

@@ -22,6 +22,7 @@ struct dev_archdata {
 	void *iommu;			/* private IOMMU data */
 #endif
 	bool dma_coherent;
+	u64 parent_dma_mask;
 };
 
 struct pdev_archdata {

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f3351f..aa65875c611b 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -352,6 +352,31 @@ static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return 1;
 }
 
+static int __swiotlb_set_dma_mask(struct device *dev, u64 mask)
+{
+	/* device is not DMA capable */
+	if (!dev->dma_mask)
+		return -EIO;
+
+	/* mask is below swiotlb bounce buffer, so fail */
+	if (!swiotlb_dma_supported(dev, mask))
+		return -EIO;
+
+	/*
+	 * because of the swiotlb, we can return success for
+	 * larger masks, but need to ensure that bounce buffers
+	 * are used above parent_dma_mask, so set that as
+	 * the effective mask.
+	 */
+	if (mask > dev->archdata.parent_dma_mask)
+		mask = dev->archdata.parent_dma_mask;
+
+
+	*dev->dma_mask = mask;
+
+	return 0;
+}
+
 static struct dma_map_ops swiotlb_dma_ops = {
 	.alloc = __dma_alloc,
 	.free = __dma_free,

@@ -367,6 +392,7 @@ static struct dma_map_ops swiotlb_dma_ops = {
 	.sync_sg_for_device = __swiotlb_sync_sg_for_device,
 	.dma_supported = __swiotlb_dma_supported,
 	.mapping_error = swiotlb_dma_mapping_error,
+	.set_dma_mask = __swiotlb_set_dma_mask,
 };
 
 static int __init atomic_pool_init(void)

@@ -957,6 +983,18 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	if (!dev->archdata.dma_ops)
 		dev->archdata.dma_ops = &swiotlb_dma_ops;
 
+	/*
+	 * we don't yet support buses that have a non-zero mapping.
+	 *  Let's hope we won't need it
+	 */
+	WARN_ON(dma_base != 0);
+
+	/*
+	 * Whatever the parent bus can set. A device must not set
+	 * a DMA mask larger than this.
+	 */
+	dev->archdata.parent_dma_mask = size;
+
 	dev->archdata.dma_coherent = coherent;
 	__iommu_setup_dma_ops(dev, dma_base, size, iommu);
 }

commit 9a57d58d116800a535510053136c6dd7a9c26e25
Author: Arnd Bergmann [off-list ref]
Date:   Tue Nov 17 14:06:55 2015 +0100

    [EXPERIMENTAL] ARM64: check implement dma_set_mask
    
    Needs work for coherent mask
    
    Signed-off-by: Arnd Bergmann [off-list ref]

Unfortunately this is far incomplete

... because size/mask passed here for PCI devices are meaningless.

For OF platforms, this is called via of_dma_configure(), that checks
dma-ranges of node that is *parent* for host bridge. Host bridge
currently does not control this at all.

In current device trees no dma-ranges is defined for nodes that are
parents to pci host bridges. This will make of_dma_configure() to fall
back to 32-bit size for all devices on all current platforms.  Thus
applying this patch will immediately break 64-bit dma masks on all
hardware that supports it.


Also related: dma-ranges property used by several pci host bridges is
*not* compatible with "legacy" dma-ranges parsed by of_get_dma_range() -
former uses additional flags word at beginning.

On Wednesday, January 4, 2017 9:24:09 AM CET Nikita Yushchenko wrote:

quoted

commit 9a57d58d116800a535510053136c6dd7a9c26e25
Author: Arnd Bergmann [off-list ref]
Date:   Tue Nov 17 14:06:55 2015 +0100

    [EXPERIMENTAL] ARM64: check implement dma_set_mask
    
    Needs work for coherent mask
    
    Signed-off-by: Arnd Bergmann [off-list ref]

Unfortunately this is far incomplete

quoted

@@ -957,6 +983,18 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	if (!dev->archdata.dma_ops)
 		dev->archdata.dma_ops = &swiotlb_dma_ops;
 
+	/*
+	 * we don't yet support buses that have a non-zero mapping.
+	 *  Let's hope we won't need it
+	 */
+	WARN_ON(dma_base != 0);
+
+	/*
+	 * Whatever the parent bus can set. A device must not set
+	 * a DMA mask larger than this.
+	 */
+	dev->archdata.parent_dma_mask = size;
+

... because size/mask passed here for PCI devices are meaningless.

For OF platforms, this is called via of_dma_configure(), that checks
dma-ranges of node that is *parent* for host bridge. Host bridge
currently does not control this at all.

We need to think about this a bit. Is it actually the PCI host
bridge that limits the ranges here, or the bus that it is connected
to. In the latter case, the caller needs to be adapted to handle
both.

In current device trees no dma-ranges is defined for nodes that are
parents to pci host bridges. This will make of_dma_configure() to fall
back to 32-bit size for all devices on all current platforms.  Thus
applying this patch will immediately break 64-bit dma masks on all
hardware that supports it.

No, it won't break it, it will just fall back to swiotlb for all the
ones that are lacking the dma-ranges property. I think this is correct
behavior.

Also related: dma-ranges property used by several pci host bridges is
*not* compatible with "legacy" dma-ranges parsed by of_get_dma_range() -
former uses additional flags word at beginning.

Can you elaborate? Do we have PCI host bridges that use wrongly formatted
dma-ranges properties?

	Arnd

quoted

For OF platforms, this is called via of_dma_configure(), that checks
dma-ranges of node that is *parent* for host bridge. Host bridge
currently does not control this at all.

We need to think about this a bit. Is it actually the PCI host
bridge that limits the ranges here, or the bus that it is connected
to. In the latter case, the caller needs to be adapted to handle
both.

In r-car case, I'm not sure what is the source of limitation at physical
level.

pcie-rcar driver configures ranges for PCIe inbound transactions based
on dma-ranges property in it's device tree node. In the current device
tree for this platform, that only contains one range and it is in lower
memory.

NVMe driver tries i/o to kmalloc()ed area. That returns 0x5xxxxxxxx
addresses here. As a quick experiment, I tried to add second range to
pcie-rcar's dma-ranges to cover 0x5xxxxxxxx area - but that did not make
DMA to high addresses working.

My current understanding is that host bridge hardware module can't
handle inbound transactions to PCI addresses above 4G - and this
limitations comes from host bridge itself.

I've read somewhere in the lists that pcie-rcar hardware is "32-bit" -
but I don't remember where, and don't know lowlevel details. Maybe
somebody from linux-renesas can elaborate?

quoted

In current device trees no dma-ranges is defined for nodes that are
parents to pci host bridges. This will make of_dma_configure() to fall
back to 32-bit size for all devices on all current platforms.  Thus
applying this patch will immediately break 64-bit dma masks on all
hardware that supports it.

No, it won't break it, it will just fall back to swiotlb for all the
ones that are lacking the dma-ranges property. I think this is correct
behavior.

I'd say - for all ones that have parents without dma-ranges property.

As of 4.10-rc2, I see only two definitions of wide parent dma-ranges
under arch/arm64/boot/dts/ - in amd/amd-seattle-soc.dtsi and
apm/apm-storm.dtsi

Are these the only arm64 platforms that can to DMA to high addresses?
I'm not arm64 expert but I'd be surprised if that's the case.

quoted

Also related: dma-ranges property used by several pci host bridges is
*not* compatible with "legacy" dma-ranges parsed by of_get_dma_range() -
former uses additional flags word at beginning.

Can you elaborate? Do we have PCI host bridges that use wrongly formatted
dma-ranges properties?

of_dma_get_range() expects <dma_addr cpu_addr size> format.

pcie-rcar.c, pci-rcar-gen2.c, pci-xgene.c and pcie-iproc.c from
drivers/pci/host/ all parse dma-ranges using of_pci_range_parser that
uses <flags pci-addr cpu-addr size> format - i.e. something different
from what of_dma_get_range() uses.


Nikita

On Wednesday, January 4, 2017 5:30:19 PM CET Nikita Yushchenko wrote:

quoted

quoted

For OF platforms, this is called via of_dma_configure(), that checks
dma-ranges of node that is *parent* for host bridge. Host bridge
currently does not control this at all.

We need to think about this a bit. Is it actually the PCI host
bridge that limits the ranges here, or the bus that it is connected
to. In the latter case, the caller needs to be adapted to handle
both.

In r-car case, I'm not sure what is the source of limitation at physical
level.

pcie-rcar driver configures ranges for PCIe inbound transactions based
on dma-ranges property in it's device tree node. In the current device
tree for this platform, that only contains one range and it is in lower
memory.

NVMe driver tries i/o to kmalloc()ed area. That returns 0x5xxxxxxxx
addresses here. As a quick experiment, I tried to add second range to
pcie-rcar's dma-ranges to cover 0x5xxxxxxxx area - but that did not make
DMA to high addresses working.

My current understanding is that host bridge hardware module can't
handle inbound transactions to PCI addresses above 4G - and this
limitations comes from host bridge itself.

I've read somewhere in the lists that pcie-rcar hardware is "32-bit" -
but I don't remember where, and don't know lowlevel details. Maybe
somebody from linux-renesas can elaborate?

Just a guess, but if the inbound translation windows in the host
bridge are wider than 32-bit, the reason for setting up a single
32-bit window is probably because that is what the parent bus supports.

quoted

quoted

In current device trees no dma-ranges is defined for nodes that are
parents to pci host bridges. This will make of_dma_configure() to fall
back to 32-bit size for all devices on all current platforms.  Thus
applying this patch will immediately break 64-bit dma masks on all
hardware that supports it.

No, it won't break it, it will just fall back to swiotlb for all the
ones that are lacking the dma-ranges property. I think this is correct
behavior.

I'd say - for all ones that have parents without dma-ranges property.

As of 4.10-rc2, I see only two definitions of wide parent dma-ranges
under arch/arm64/boot/dts/ - in amd/amd-seattle-soc.dtsi and
apm/apm-storm.dtsi

Are these the only arm64 platforms that can to DMA to high addresses?
I'm not arm64 expert but I'd be surprised if that's the case.

It's likely that a few others also do high DMA, but a lot of arm64
chips are actually derived from earlier 32-bit chips and don't
even support any RAM above 4GB, as well as having a lot of 32-bit
DMA masters.

quoted

quoted

Also related: dma-ranges property used by several pci host bridges is
*not* compatible with "legacy" dma-ranges parsed by of_get_dma_range() -
former uses additional flags word at beginning.

Can you elaborate? Do we have PCI host bridges that use wrongly formatted
dma-ranges properties?

of_dma_get_range() expects <dma_addr cpu_addr size> format.

pcie-rcar.c, pci-rcar-gen2.c, pci-xgene.c and pcie-iproc.c from
drivers/pci/host/ all parse dma-ranges using of_pci_range_parser that
uses <flags pci-addr cpu-addr size> format - i.e. something different
from what of_dma_get_range() uses.

The "dma_addr" here is expressed in terms of #address-cells of the
bus it is in, and that is "3" in case of PCI, where the first 32-bit
word is a bit pattern containing various things, and the other two
cells are a 64-bit address. I think this is correct, but we may
need to add some special handling for parsing PCI host bridges
in of_dma_get_range, to ensure we actually look at translations for
the memory space.

	Arnd

quoted

quoted

quoted

For OF platforms, this is called via of_dma_configure(), that checks
dma-ranges of node that is *parent* for host bridge. Host bridge
currently does not control this at all.

We need to think about this a bit. Is it actually the PCI host
bridge that limits the ranges here, or the bus that it is connected
to. In the latter case, the caller needs to be adapted to handle
both.

In r-car case, I'm not sure what is the source of limitation at physical
level.

pcie-rcar driver configures ranges for PCIe inbound transactions based
on dma-ranges property in it's device tree node. In the current device
tree for this platform, that only contains one range and it is in lower
memory.

NVMe driver tries i/o to kmalloc()ed area. That returns 0x5xxxxxxxx
addresses here. As a quick experiment, I tried to add second range to
pcie-rcar's dma-ranges to cover 0x5xxxxxxxx area - but that did not make
DMA to high addresses working.

My current understanding is that host bridge hardware module can't
handle inbound transactions to PCI addresses above 4G - and this
limitations comes from host bridge itself.

I've read somewhere in the lists that pcie-rcar hardware is "32-bit" -
but I don't remember where, and don't know lowlevel details. Maybe
somebody from linux-renesas can elaborate?

Just a guess, but if the inbound translation windows in the host
bridge are wider than 32-bit, the reason for setting up a single
32-bit window is probably because that is what the parent bus supports.

Well anyway applying patch similar to your's will fix pcie-rcar + nvme
case - thus I don't object :)   But it can break other cases ...

But why do you hook at set_dma_mask() and overwrite mask inside, instead
of hooking at dma_supported() and rejecting unsupported mask?

I think later is better, because it lets drivers to handle unsupported
high-dma case, like documented in DMA-API_HOWTO.

On Wednesday, January 4, 2017 6:29:39 PM CET Nikita Yushchenko wrote:

quoted

Just a guess, but if the inbound translation windows in the host
bridge are wider than 32-bit, the reason for setting up a single
32-bit window is probably because that is what the parent bus supports.

Well anyway applying patch similar to your's will fix pcie-rcar + nvme
case - thus I don't object :)   But it can break other cases ...

But why do you hook at set_dma_mask() and overwrite mask inside, instead
of hooking at dma_supported() and rejecting unsupported mask?

I think later is better, because it lets drivers to handle unsupported
high-dma case, like documented in DMA-API_HOWTO.

I think the behavior I put in there is required for swiotlb to make
sense, otherwise you would rely on the driver to handle dma_set_mask()
failure gracefully with its own bounce buffers (as network and
scsi drivers do but others don't).

Having swiotlb or iommu enabled should result in dma_set_mask() always
succeeding unless the mask is too small to cover the swiotlb
bounce buffer area or the iommu virtual address space. This behavior
is particularly important in case the bus address space is narrower
than 32-bit, as we have to guarantee that the fallback to 32-bit
DMA always succeeds. There are also a lot of drivers that try to
set a 64-bit mask but don't implement bounce buffers for streaming
mappings if that fails, and swiotlb is what we use to make those
drivers work.

And yes, the API is a horrible mess.

	Arnd

quoted

quoted

Just a guess, but if the inbound translation windows in the host
bridge are wider than 32-bit, the reason for setting up a single
32-bit window is probably because that is what the parent bus supports.

I've re-checked rcar-pcie hardware documentation.

It indeed mentions that AXI bus it sits on is 32-bit.

quoted

Well anyway applying patch similar to your's will fix pcie-rcar + nvme
case - thus I don't object :)   But it can break other cases ...

But why do you hook at set_dma_mask() and overwrite mask inside, instead
of hooking at dma_supported() and rejecting unsupported mask?

I think later is better, because it lets drivers to handle unsupported
high-dma case, like documented in DMA-API_HOWTO.

I think the behavior I put in there is required for swiotlb to make
sense, otherwise you would rely on the driver to handle dma_set_mask()
failure gracefully with its own bounce buffers (as network and
scsi drivers do but others don't).

Having swiotlb or iommu enabled should result in dma_set_mask() always
succeeding unless the mask is too small to cover the swiotlb
bounce buffer area or the iommu virtual address space. This behavior
is particularly important in case the bus address space is narrower
than 32-bit, as we have to guarantee that the fallback to 32-bit
DMA always succeeds. There are also a lot of drivers that try to
set a 64-bit mask but don't implement bounce buffers for streaming
mappings if that fails, and swiotlb is what we use to make those
drivers work.

And yes, the API is a horrible mess.

With my patch applied and thus 32bit dma_mask set for NVMe device, I do
see high addresses passed to dma_map_*() routines and handled by
swiotlb. Thus your statement that behavior "succeed 64bit dma_set_mask()
operation but silently replace mask behind the scene" is required for
swiotlb to be used, does not match reality.

It can be interpreted as a breakage elsewhere, but it's hard to point
particular "root cause". The entire infrastructure to allocate and use
DMA memory is messy.

Still current code does not work, thus fix is needed.

Perhaps need to introduce some generic API to "allocate memory best
suited for DMA to particular device", and fix allocation points (in
drivers, filesystems, etc) to use it. Such an API could try to allocate
area that can be DMAed by hardware, and fallback to other memory that
can be used via swiotlb or other bounce buffer implementation.

But for now, have to stay with dma masks. Will follow-up with a patch
based on your but with coherent mask handling added.

Nikita

It is possible that device is capable of 64-bit DMA addresses, and
device driver tries to set wide DMA mask, but bridge or bus used to
connect device to the system can't handle wide addresses.

With swiotlb, memory above 4G still can be used by drivers for streaming
DMA, but *dev->mask and dev->dma_coherent_mask must still keep values
that hardware handles physically.

This patch enforces that. Based on original version by
Arnd Bergmann [off-list ref], extended with coherent mask hadnling.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
CC: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm64/Kconfig              |  3 +++
 arch/arm64/include/asm/device.h |  1 +
 arch/arm64/mm/dma-mapping.c     | 40 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 44 insertions(+)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 1117421..afb2c08 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig

@@ -216,6 +216,9 @@ config NEED_DMA_MAP_STATE
 config NEED_SG_DMA_LENGTH
 	def_bool y
 
+config ARCH_HAS_DMA_SET_COHERENT_MASK
+	def_bool y
+
 config SMP
 	def_bool y
 

diff --git a/arch/arm64/include/asm/device.h b/arch/arm64/include/asm/device.h
index 243ef25..a57e7bb 100644
--- a/arch/arm64/include/asm/device.h
+++ b/arch/arm64/include/asm/device.h

@@ -22,6 +22,7 @@ struct dev_archdata {
 	void *iommu;			/* private IOMMU data */
 #endif
 	bool dma_coherent;
+	u64 parent_dma_mask;
 };
 
 struct pdev_archdata {

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..be3632e 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -352,6 +352,31 @@ static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return 1;
 }
 
+static int __swiotlb_set_dma_mask(struct device *dev, u64 mask)
+{
+	/* device is not DMA capable */
+	if (!dev->dma_mask)
+		return -EIO;
+
+	/* mask is below swiotlb bounce buffer, so fail */
+	if (!swiotlb_dma_supported(dev, mask))
+		return -EIO;
+
+	/*
+	 * because of the swiotlb, we can return success for
+	 * larger masks, but need to ensure that bounce buffers
+	 * are used above parent_dma_mask, so set that as
+	 * the effective mask.
+	 */
+	if (mask > dev->archdata.parent_dma_mask)
+		mask = dev->archdata.parent_dma_mask;
+
+
+	*dev->dma_mask = mask;
+
+	return 0;
+}
+
 static struct dma_map_ops swiotlb_dma_ops = {
 	.alloc = __dma_alloc,
 	.free = __dma_free,

@@ -367,8 +392,23 @@ static struct dma_map_ops swiotlb_dma_ops = {
 	.sync_sg_for_device = __swiotlb_sync_sg_for_device,
 	.dma_supported = __swiotlb_dma_supported,
 	.mapping_error = swiotlb_dma_mapping_error,
+	.set_dma_mask = __swiotlb_set_dma_mask,
 };
 
+int dma_set_coherent_mask(struct device *dev, u64 mask)
+{
+	if (!dma_supported(dev, mask))
+		return -EIO;
+
+	if (get_dma_ops(dev) == &swiotlb_dma_ops &&
+	    mask > dev->archdata.parent_dma_mask)
+		mask = dev->archdata.parent_dma_mask;
+
+	dev->coherent_dma_mask = mask;
+	return 0;
+}
+EXPORT_SYMBOL(dma_set_coherent_mask);
+
 static int __init atomic_pool_init(void)
 {
 	pgprot_t prot = __pgprot(PROT_NORMAL_NC);

-- 
2.1.4

It is possible that device is capable of 64-bit DMA addresses, and
device driver tries to set wide DMA mask, but bridge or bus used to
connect device to the system can't handle wide addresses.

With swiotlb, memory above 4G still can be used by drivers for streaming
DMA, but *dev->mask and dev->dma_coherent_mask must still keep values
that hardware handles physically.

This patch enforces that. Based on original version by
Arnd Bergmann [off-list ref], extended with coherent mask hadnling.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
CC: Arnd Bergmann <arnd@arndb.de>
---

... now with initially missed change in arch_setup_dma_ops() ...

 arch/arm64/Kconfig              |  3 +++
 arch/arm64/include/asm/device.h |  1 +
 arch/arm64/mm/dma-mapping.c     | 52 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 56 insertions(+)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 1117421..afb2c08 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig

@@ -216,6 +216,9 @@ config NEED_DMA_MAP_STATE
 config NEED_SG_DMA_LENGTH
 	def_bool y
 
+config ARCH_HAS_DMA_SET_COHERENT_MASK
+	def_bool y
+
 config SMP
 	def_bool y
 

diff --git a/arch/arm64/include/asm/device.h b/arch/arm64/include/asm/device.h
index 243ef25..a57e7bb 100644
--- a/arch/arm64/include/asm/device.h
+++ b/arch/arm64/include/asm/device.h

@@ -22,6 +22,7 @@ struct dev_archdata {
 	void *iommu;			/* private IOMMU data */
 #endif
 	bool dma_coherent;
+	u64 parent_dma_mask;
 };
 
 struct pdev_archdata {

diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 290a84f..09c7900 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c

@@ -352,6 +352,31 @@ static int __swiotlb_dma_supported(struct device *hwdev, u64 mask)
 	return 1;
 }
 
+static int __swiotlb_set_dma_mask(struct device *dev, u64 mask)
+{
+	/* device is not DMA capable */
+	if (!dev->dma_mask)
+		return -EIO;
+
+	/* mask is below swiotlb bounce buffer, so fail */
+	if (!swiotlb_dma_supported(dev, mask))
+		return -EIO;
+
+	/*
+	 * because of the swiotlb, we can return success for
+	 * larger masks, but need to ensure that bounce buffers
+	 * are used above parent_dma_mask, so set that as
+	 * the effective mask.
+	 */
+	if (mask > dev->archdata.parent_dma_mask)
+		mask = dev->archdata.parent_dma_mask;
+
+
+	*dev->dma_mask = mask;
+
+	return 0;
+}
+
 static struct dma_map_ops swiotlb_dma_ops = {
 	.alloc = __dma_alloc,
 	.free = __dma_free,

@@ -367,8 +392,23 @@ static struct dma_map_ops swiotlb_dma_ops = {
 	.sync_sg_for_device = __swiotlb_sync_sg_for_device,
 	.dma_supported = __swiotlb_dma_supported,
 	.mapping_error = swiotlb_dma_mapping_error,
+	.set_dma_mask = __swiotlb_set_dma_mask,
 };
 
+int dma_set_coherent_mask(struct device *dev, u64 mask)
+{
+	if (!dma_supported(dev, mask))
+		return -EIO;
+
+	if (get_dma_ops(dev) == &swiotlb_dma_ops &&
+	    mask > dev->archdata.parent_dma_mask)
+		mask = dev->archdata.parent_dma_mask;
+
+	dev->coherent_dma_mask = mask;
+	return 0;
+}
+EXPORT_SYMBOL(dma_set_coherent_mask);
+
 static int __init atomic_pool_init(void)
 {
 	pgprot_t prot = __pgprot(PROT_NORMAL_NC);

@@ -957,6 +997,18 @@ void arch_setup_dma_ops(struct device *dev, u64 dma_base, u64 size,
 	if (!dev->archdata.dma_ops)
 		dev->archdata.dma_ops = &swiotlb_dma_ops;
 
+	/*
+	 * we don't yet support buses that have a non-zero mapping.
+	 *  Let's hope we won't need it
+	 */
+	WARN_ON(dma_base != 0);
+
+	/*
+	 * Whatever the parent bus can set. A device must not set
+	 * a DMA mask larger than this.
+	 */
+	dev->archdata.parent_dma_mask = size;
+
 	dev->archdata.dma_coherent = coherent;
 	__iommu_setup_dma_ops(dev, dma_base, size, iommu);
 }

-- 
2.1.4

Hello!

On 1/6/2017 5:45 PM, Nikita Yushchenko wrote:

It is possible that device is capable of 64-bit DMA addresses, and
device driver tries to set wide DMA mask, but bridge or bus used to
connect device to the system can't handle wide addresses.

With swiotlb, memory above 4G still can be used by drivers for streaming
DMA, but *dev->mask and dev->dma_coherent_mask must still keep values
that hardware handles physically.

This patch enforces that. Based on original version by
Arnd Bergmann [off-list ref], extended with coherent mask hadnling.

Signed-off-by: Nikita Yushchenko <nikita.yoush@cogentembedded.com>
CC: Arnd Bergmann <arnd@arndb.de>

[...]

    One empty line is enough...

[...]

    Not 'size - 1'?

MBR, Sergei

quoted

+    if (mask > dev->archdata.parent_dma_mask)
+        mask = dev->archdata.parent_dma_mask;
+
+

   One empty line is enough...

Ok

quoted

+    /*
+     * Whatever the parent bus can set. A device must not set
+     * a DMA mask larger than this.
+     */
+    dev->archdata.parent_dma_mask = size;

   Not 'size - 1'?

Good question.

Indeed of_dma_configure() calls arch_setup_dma_ops() with size, not
mask. Which implies '-1' is needed here. Although better fix may be to
change caller side - to make DMA_BIT_MASK(64) case cleaner.

Will repost path.

Nikita

On Friday, January 6, 2017 4:47:59 PM CET Nikita Yushchenko wrote:

quoted

quoted

quoted

Just a guess, but if the inbound translation windows in the host
bridge are wider than 32-bit, the reason for setting up a single
32-bit window is probably because that is what the parent bus supports.

I've re-checked rcar-pcie hardware documentation.

It indeed mentions that AXI bus it sits on is 32-bit.

quoted

quoted

Well anyway applying patch similar to your's will fix pcie-rcar + nvme
case - thus I don't object :)   But it can break other cases ...

But why do you hook at set_dma_mask() and overwrite mask inside, instead
of hooking at dma_supported() and rejecting unsupported mask?

I think later is better, because it lets drivers to handle unsupported
high-dma case, like documented in DMA-API_HOWTO.

I think the behavior I put in there is required for swiotlb to make
sense, otherwise you would rely on the driver to handle dma_set_mask()
failure gracefully with its own bounce buffers (as network and
scsi drivers do but others don't).

Having swiotlb or iommu enabled should result in dma_set_mask() always
succeeding unless the mask is too small to cover the swiotlb
bounce buffer area or the iommu virtual address space. This behavior
is particularly important in case the bus address space is narrower
than 32-bit, as we have to guarantee that the fallback to 32-bit
DMA always succeeds. There are also a lot of drivers that try to
set a 64-bit mask but don't implement bounce buffers for streaming
mappings if that fails, and swiotlb is what we use to make those
drivers work.

And yes, the API is a horrible mess.

With my patch applied and thus 32bit dma_mask set for NVMe device, I do
see high addresses passed to dma_map_*() routines and handled by
swiotlb. Thus your statement that behavior "succeed 64bit dma_set_mask()
operation but silently replace mask behind the scene" is required for
swiotlb to be used, does not match reality.

See my point about drivers that don't implement bounce buffering.
Apparently NVMe is one of them, unlike the SATA/SCSI/MMC storage
drivers that do their own thing.

The problem again is the inconsistency of the API.

It can be interpreted as a breakage elsewhere, but it's hard to point
particular "root cause". The entire infrastructure to allocate and use
DMA memory is messy.

Absolutely.

What I think happened here in chronological order is:

- In the old days, 64-bit architectures tended to use an IOMMU 
  all the time to work around 32-bit limitations on DMA masters
- Some architectures had no IOMMU that fully solved this and the
  dma-mapping API required drivers to set the right mask and check
  the return code. If this failed, the driver needed to use its
  own bounce buffers as network and scsi do. See also the
  grossly misnamed "PCI_DMA_BUS_IS_PHYS" macro.
- As we never had support for bounce buffers in all drivers, and
  early 64-bit Intel machines had no IOMMU, the swiotlb code was
  introduced as a workaround, so we can use the IOMMU case without
  driver specific bounce buffers everywhere
- As most of the important 64-bit architectures (x86, arm64, powerpc)
  now always have either IOMMU or swiotlb enabled, drivers like
  NVMe started relying on it, and no longer handle a dma_set_mask
  failure properly.

We may need to audit how drivers typically handle dma_set_mask()
failure. The NVMe driver in its current state will probably cause
silent data corruption when used on a 64-bit architecture that has
a 32-bit bus but neither swiotlb nor iommu enabled at runtime.

I would argue that the driver should be fixed to either refuse
working in that configuration to avoid data corruption, or that
it should implement bounce buffering like SCSI does. If we make it
simply not work, then your suggestion of making dma_set_mask()
fail will break your system in a different way.

Still current code does not work, thus fix is needed.

Perhaps need to introduce some generic API to "allocate memory best
suited for DMA to particular device", and fix allocation points (in
drivers, filesystems, etc) to use it. Such an API could try to allocate
area that can be DMAed by hardware, and fallback to other memory that
can be used via swiotlb or other bounce buffer implementation.

The DMA mapping API is meant to do this, but we can definitely improve
it or clarify some of the rules.
 

But for now, have to stay with dma masks. Will follow-up with a patch
based on your but with coherent mask handling added.

Ok.

	Arnd

[CCing NVMe maintainers since we are discussion issues in that driver]

quoted

With my patch applied and thus 32bit dma_mask set for NVMe device, I do
see high addresses passed to dma_map_*() routines and handled by
swiotlb. Thus your statement that behavior "succeed 64bit dma_set_mask()
operation but silently replace mask behind the scene" is required for
swiotlb to be used, does not match reality.

See my point about drivers that don't implement bounce buffering.
Apparently NVMe is one of them, unlike the SATA/SCSI/MMC storage
drivers that do their own thing.

I believe the bounce buffering code you refer to is not in SATA/SCSI/MMC
but in block layer, in particular it should be controlled by
blk_queue_bounce_limit().  [Yes there is CONFIG_MMC_BLOCK_BOUNCE but it
is something completely different, namely it is for request merging for
hw not supporting scatter-gather].  And NVMe also uses block layer and
thus should get same support.

But blk_queue_bounce_limit() is somewhat broken, it has very strange
code under #if BITS_PER_LONG == 64 that makes setting max_addr to
0xffffffff not working if max_low_pfn is above 4G.

Maybe fixing that, together with making NVMe use this API, could stop it
from issuing dma_map()s of addresses beyond mask.

What I think happened here in chronological order is:

- In the old days, 64-bit architectures tended to use an IOMMU 
  all the time to work around 32-bit limitations on DMA masters
- Some architectures had no IOMMU that fully solved this and the
  dma-mapping API required drivers to set the right mask and check
  the return code. If this failed, the driver needed to use its
  own bounce buffers as network and scsi do. See also the
  grossly misnamed "PCI_DMA_BUS_IS_PHYS" macro.
- As we never had support for bounce buffers in all drivers, and
  early 64-bit Intel machines had no IOMMU, the swiotlb code was
  introduced as a workaround, so we can use the IOMMU case without
  driver specific bounce buffers everywhere
- As most of the important 64-bit architectures (x86, arm64, powerpc)
  now always have either IOMMU or swiotlb enabled, drivers like
  NVMe started relying on it, and no longer handle a dma_set_mask
  failure properly.

... and architectures started to add to this breakage, not handling
dma_set_mask() as documented.


As for PCI_DMA_BUS_IS_PHYS - ironically, what all current usages of this
macro in the kernel do is - *disable* bounce buffers in block layer if
PCI_DMA_BUS_IS_PHYS is zero.  Defining it to zero (as arm64 currently
does) on system with memory above 4G makes all block drivers to depend
on swiotlb (or iommu). Affected drivers are SCSI and IDE.

We may need to audit how drivers typically handle dma_set_mask()
failure. The NVMe driver in its current state will probably cause
silent data corruption when used on a 64-bit architecture that has
a 32-bit bus but neither swiotlb nor iommu enabled at runtime.

With current code NVME causes system memory breakage even if swiotlb is
there - because it's dma_set_mask_and_coherent(DMA_BIT_MASK(64)) call
has effect of silent disable of swiotlb.

I would argue that the driver should be fixed to either refuse
working in that configuration to avoid data corruption, or that
it should implement bounce buffering like SCSI does.

Difference from "SCSI" (actually - from block drivers that work) is in
that dma_set_mask_and_coherent(DMA_BIT_MASK(64)) call: driver that does
not do it works, driver that does it fails.

Per documentation, driver *should* do it if it's hardware supports
64-bit dma, and platform *should* either fail this call, or ensure that
64-bit addresses can be dma_map()ed successfully.

So what we have on arm64 is - drivers that follow documented procedure
fail, drivers that don't follow it work, That's nonsense.

If we make it
simply not work, then your suggestion of making dma_set_mask()
fail will break your system in a different way.

Proper fix should fix *both* architecture and NVMe.

- architecture should stop breaking 64-bit DMA when driver attempts to
set 64-bit dma mask,

- NVMe should issue proper blk_queue_bounce_limit() call based on what
is actually set mask,

- and blk_queue_bounce_limit() should also be fixed to actually set
0xffffffff limit, instead of replacing it with (max_low_pfn <<
PAGE_SHIFT) as it does now.

quoted

Still current code does not work, thus fix is needed.

Perhaps need to introduce some generic API to "allocate memory best
suited for DMA to particular device", and fix allocation points (in
drivers, filesystems, etc) to use it. Such an API could try to allocate
area that can be DMAed by hardware, and fallback to other memory that
can be used via swiotlb or other bounce buffer implementation.

The DMA mapping API is meant to do this, but we can definitely improve
it or clarify some of the rules.

DMA mapping API can't help here, it's about mapping, not about allocation.

What I mean is some API to allocate memory for use with streaming DMA in
such way that bounce buffers won't be needed. There are many cases when
at buffer allocation time, it is already known that buffer will be used
for DMA with particular device. Bounce buffers will still be needed
cases when no such information is available at allocation time, or when
there is no directly-DMAable memory available at allocation time.

quoted

But for now, have to stay with dma masks. Will follow-up with a patch
based on your but with coherent mask handling added.

Ok.

Already posted. Can we have that merged? At least it will make things to
stop breaking memory and start working.

Nikita

On Mon, Jan 09, 2017 at 11:34:55PM +0300, Nikita Yushchenko wrote:

I believe the bounce buffering code you refer to is not in SATA/SCSI/MMC
but in block layer, in particular it should be controlled by
blk_queue_bounce_limit().  [Yes there is CONFIG_MMC_BLOCK_BOUNCE but it
is something completely different, namely it is for request merging for
hw not supporting scatter-gather].  And NVMe also uses block layer and
thus should get same support.

NVMe shouldn't have to call blk_queue_bounce_limit - 
blk_queue_bounce_limit is to set the DMA addressing limit of the device.
NVMe devices must support unlimited 64-bit addressing and thus calling
blk_queue_bounce_limit from NVMe does not make sense.

That being said currently the default for a queue without a call
to blk_queue_make_request which does the wrong thing on highmem
setups, so we should fix it.  In fact BLK_BOUNCE_HIGH as-is doesn't
really make much sense these days as no driver should ever dereference
pages passed to it directly.

Maybe fixing that, together with making NVMe use this API, could stop it
from issuing dma_map()s of addresses beyond mask.

NVMe should never bounce, the fact that it currently possibly does
for highmem pages is a bug.

As for PCI_DMA_BUS_IS_PHYS - ironically, what all current usages of this
macro in the kernel do is - *disable* bounce buffers in block layer if
PCI_DMA_BUS_IS_PHYS is zero.

That's not ironic but the whole point of the macro (horrible name and
the fact that it should be a dma_ops setting aside).

- architecture should stop breaking 64-bit DMA when driver attempts to
set 64-bit dma mask,

- NVMe should issue proper blk_queue_bounce_limit() call based on what
is actually set mask,

Or even better remove the call to dma_set_mask_and_coherent with
DMA_BIT_MASK(32).  NVMe is designed around having proper 64-bit DMA
addressing, there is not point in trying to pretent it works without that

- and blk_queue_bounce_limit() should also be fixed to actually set
0xffffffff limit, instead of replacing it with (max_low_pfn <<
PAGE_SHIFT) as it does now.

We need to kill off BLK_BOUNCE_HIGH, it just doesn't make sense to
mix the highmem aspect with the addressing limits.  In fact the whole
block bouncing scheme doesn't make much sense at all these days, we
should rely on swiotlb instead.

What I mean is some API to allocate memory for use with streaming DMA in
such way that bounce buffers won't be needed. There are many cases when
at buffer allocation time, it is already known that buffer will be used
for DMA with particular device. Bounce buffers will still be needed
cases when no such information is available at allocation time, or when
there is no directly-DMAable memory available at allocation time.

For block I/O that is never the case.

On Monday, January 9, 2017 9:57:46 PM CET Christoph Hellwig wrote:

quoted

- architecture should stop breaking 64-bit DMA when driver attempts to
set 64-bit dma mask,

- NVMe should issue proper blk_queue_bounce_limit() call based on what
is actually set mask,

Or even better remove the call to dma_set_mask_and_coherent with
DMA_BIT_MASK(32).  NVMe is designed around having proper 64-bit DMA
addressing, there is not point in trying to pretent it works without that

Agreed, let's just fail the probe() if DMA_BIT_MASK(64) fails, and
have swiotlb work around machines that for some reason need bounce
buffers.

quoted

- and blk_queue_bounce_limit() should also be fixed to actually set
0xffffffff limit, instead of replacing it with (max_low_pfn <<
PAGE_SHIFT) as it does now.

We need to kill off BLK_BOUNCE_HIGH, it just doesn't make sense to
mix the highmem aspect with the addressing limits.  In fact the whole
block bouncing scheme doesn't make much sense at all these days, we
should rely on swiotlb instead.

If we do this, we should probably have another look at the respective
NETIF_F_HIGHDMA support in the network stack, which does the same thing
and mixes up highmem on 32-bit architectures with the DMA address limit.
(side note: there are actually cases in which you have a 31-bit DMA
mask but 3 GB of lowmem using CONFIG_VMSPLIT_1G, so BLK_BOUNCE_HIGH
and !NETIF_F_HIGHDMA are both missing the limit, causing data corruption
without swiotlb).

Before we rely too much on swiotlb, we may also need to consider which
architectures today rely on bouncing in blk and network.

I see that we have CONFIG_ARCH_PHYS_ADDR_T_64BIT on a couple of
32-bit architectures without swiotlb (arc, arm, some mips32), and
there are several 64-bit architectures that do not have swiotlb
(alpha, parisc, s390, sparc). I believe that alpha, s390 and sparc
always use some form of IOMMU, but the other four apparently don't,
so we would need to add swiotlb support there to remove all the
bounce buffering in network and block layers.

	Arnd

On Tue, Jan 10, 2017 at 11:47:42AM +0100, Arnd Bergmann wrote:

I see that we have CONFIG_ARCH_PHYS_ADDR_T_64BIT on a couple of
32-bit architectures without swiotlb (arc, arm, some mips32), and
there are several 64-bit architectures that do not have swiotlb
(alpha, parisc, s390, sparc). I believe that alpha, s390 and sparc
always use some form of IOMMU, but the other four apparently don't,
so we would need to add swiotlb support there to remove all the
bounce buffering in network and block layers.

mips has lots of weird swiotlb wire-up in it's board code (the swiotlb
arch glue really needs some major cleanup..), as does arm.  Not
sure about the others.

Getting rid of highmem bouncing in the block layer will take some time
as various PIO-only drivers rely on it at the moment.  These should
all be convertable to kmap that data, but it needs a careful audit
first.  For 4.11 I'll plan to switch away from bouncing highmem by
default at least, though and maybe also convert a few PIO drivers.

On Tuesday, January 10, 2017 3:44:53 PM CET Christoph Hellwig wrote:

On Tue, Jan 10, 2017 at 11:47:42AM +0100, Arnd Bergmann wrote:

quoted

I see that we have CONFIG_ARCH_PHYS_ADDR_T_64BIT on a couple of
32-bit architectures without swiotlb (arc, arm, some mips32), and
there are several 64-bit architectures that do not have swiotlb
(alpha, parisc, s390, sparc). I believe that alpha, s390 and sparc
always use some form of IOMMU, but the other four apparently don't,
so we would need to add swiotlb support there to remove all the
bounce buffering in network and block layers.

mips has lots of weird swiotlb wire-up in it's board code (the swiotlb
arch glue really needs some major cleanup..),

My reading of the MIPS code was that only the 64-bit platforms use it,
but there are a number of 32-bit platforms that have 64-bit physical
addresses and don't.

as does arm.  Not sure about the others.

32-bit ARM doesn't actually use SWIOTLB at all, despite selecting it
in Kconfig. I think Xen uses it for its own purposes, but nothing
else does.

Most ARM platforms can't actually have RAM beyond 4GB, and the ones
that do have it tend to also come with an IOMMU, but I remember
at least BCM53xx actually needing swiotlb on some chip revisions
that are widely used and that cannot DMA to the second memory bank
from PCI (!).

	Arnd

On Thursday, December 29, 2016 11:45:03 PM CET Nikita Yushchenko wrote:

I think it's wrong to make this a special case for PCI.

Instead, we should follow the dma-ranges properties during dma_set_mask()
to ensure we don't set a mask that any of the parents up to the root
cannot support.

	Arnd