[dpdk-dev] [PATCH v2 0/6] use IOVAs check based on DMA mask

[dpdk-dev] [PATCH v2 0/6] use IOVAs check based on DMA mask

[Alejandro Lucero]

This patchset adds, mainly, a check for ensuring IOVAs are within a
restricted range due to addressing limitations with some devices. There
are two known cases: NFP and IOMMU VT-d emulation.

With this check IOVAs out of range are detected and PMDs can abort
initialization. For the VT-d case, IOVA VA mode is allowed as long as
IOVAs are within the supported range, avoiding to forbid IOVA VA by
default.

For the addressing limitations known cases, there are just 40(NFP) or
39(VT-d) bits for handling IOVAs. When using IOVA PA, those limitations
imply 1TB(NFP) or 512M(VT-d) as upper limits, which is likely enough for
most systems. With machines using more memory, the added check will
ensure IOVAs within the range.

With IOVA VA, and because the way the Linux kernel serves mmap calls
in 64 bits systems, 39 or 40 bits are not enough. It is possible to
give an address hint with a lower starting address than the default one
used by the kernel, and then ensuring the mmap uses that hint or hint plus
some offset. With 64 bits systems, the process virtual address space is
large enoguh for doing the hugepages mmaping within the supported range
when those addressing limitations exist. This patchset also adds a change
for using such a hint making the use of IOVA VA a more than likely
possibility when there are those addressing limitations.

The check is not done by default but just when it is required. This
patchset adds the check for NFP initialization and for setting the IOVA
mode is an emulated VT-d is detected.

This patchset applies on 17.11.3.

Similar changes will be submitted to main DPDK branch soon.

v2:
 - add get_addr_hint function
 - call munmap when hint given and not used by mmap
 - create dma mask in one step
 - refactor logs

[dpdk-dev] [PATCH v2 1/6] mem: add function for checking memsegs IOVAs addresses

[Alejandro Lucero]

A device can suffer addressing limitations. This functions checks
memsegs have iovas within the supported range based on dma mask.

PMD should use this during initialization if supported devices
suffer addressing limitations, returning an error if this function
returns memsegs out of range.

Another potential usage is for emulated IOMMU hardware with addressing
limitations.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 lib/librte_eal/common/eal_common_memory.c  | 33 ++++++++++++++++++++++++++++++
 lib/librte_eal/common/include/rte_memory.h |  3 +++
 2 files changed, 36 insertions(+)

diff --git a/lib/librte_eal/common/eal_common_memory.c b/lib/librte_eal/common/eal_common_memory.c
index fc6c44d..f5efebe 100644
--- a/lib/librte_eal/common/eal_common_memory.c
+++ b/lib/librte_eal/common/eal_common_memory.c
@@ -109,6 +109,39 @@
 	}
 }
 
+/* check memseg iovas are within the required range based on dma mask */
+int
+rte_eal_check_dma_mask(uint8_t maskbits)
+{
+
+	const struct rte_mem_config *mcfg;
+	uint64_t mask;
+	int i;
+
+	/* create dma mask */
+	mask = ~((1ULL << maskbits) - 1);
+
+	/* get pointer to global configuration */
+	mcfg = rte_eal_get_configuration()->mem_config;
+
+	for (i = 0; i < RTE_MAX_MEMSEG; i++) {
+		if (mcfg->memseg[i].addr == NULL)
+			break;
+
+		if (mcfg->memseg[i].iova & mask) {
+			RTE_LOG(INFO, EAL,
+				"memseg[%d] iova %"PRIx64" out of range:\n",
+				i, mcfg->memseg[i].iova);
+
+			RTE_LOG(INFO, EAL, "\tusing dma mask %"PRIx64"\n",
+				mask);
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
 /* return the number of memory channels */
 unsigned rte_memory_get_nchannel(void)
 {
diff --git a/lib/librte_eal/common/include/rte_memory.h b/lib/librte_eal/common/include/rte_memory.h
index 80a8fc0..b2a0168 100644
--- a/lib/librte_eal/common/include/rte_memory.h
+++ b/lib/librte_eal/common/include/rte_memory.h
@@ -209,6 +209,9 @@ struct rte_memseg {
  */
 unsigned rte_memory_get_nrank(void);
 
+/* check memsegs iovas are within a range based on dma mask */
+int rte_eal_check_dma_mask(uint8_t maskbits);
+
 /**
  * Drivers based on uio will not load unless physical
  * addresses are obtainable. It is only possible to get
-- 
1.9.1

Re: [dpdk-dev] [PATCH v2 1/6] mem: add function for checking memsegs IOVAs addresses

[Burakov, Anatoly]

On 03-Jul-18 1:09 PM, Alejandro Lucero wrote:
> A device can suffer addressing limitations. This functions checks
> memsegs have iovas within the supported range based on dma mask.
> 
> PMD should use this during initialization if supported devices
> suffer addressing limitations, returning an error if this function
> returns memsegs out of range.
> 
> Another potential usage is for emulated IOMMU hardware with addressing
> limitations.
> 
> Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
> ---

Missing .map file addition. Other than that,

Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

-- 
Thanks,
Anatoly

[dpdk-dev] [PATCH v2 2/6] ethdev: add function for checking IOVAs by a device

[Alejandro Lucero]

A PMD should invoke this function for checking memsegs iovas are within
the supported range by the device.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 lib/librte_ether/rte_ethdev.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index eba11ca..e51a432 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -2799,6 +2799,19 @@ int rte_eth_dev_set_vlan_ether_type(uint16_t port_id,
 int rte_eth_dev_set_vlan_pvid(uint16_t port_id, uint16_t pvid, int on);
 
 /**
+ * check device dma mask within expected range based on dma mask.
+ *
+ * @param maskbits
+ *  mask length in bits
+ *
+ */
+static inline int
+rte_eth_dev_check_dma_mask(uint8_t maskbits)
+{
+	return rte_eal_check_dma_mask(maskbits);
+}
+
+/**
  *
  * Retrieve a burst of input packets from a receive queue of an Ethernet
  * device. The retrieved packets are stored in *rte_mbuf* structures whose
-- 
1.9.1

[dpdk-dev] [PATCH v2 3/6] bus/pci: use IOVAs check when setting IOVA mode

[Alejandro Lucero]

Although VT-d emulation currently only supports 39 bits, it could
be iovas being within that supported range. This patch allows
IOVA mode in such a case.

Indeed, memory initialization code can be modified for using lower
virtual addresses than those used by the kernel for 64 bits processes
by default, and therefore memsegs iovas can use 39 bits or less for
most system. And this is likely 100% true for VMs.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/bus/pci/linux/pci.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
index 74deef3..792c819 100644
--- a/drivers/bus/pci/linux/pci.c
+++ b/drivers/bus/pci/linux/pci.c
@@ -43,6 +43,7 @@
 #include <rte_devargs.h>
 #include <rte_memcpy.h>
 #include <rte_vfio.h>
+#include <rte_memory.h>
 
 #include "eal_private.h"
 #include "eal_filesystem.h"
@@ -613,10 +614,12 @@
 	fclose(fp);
 
 	mgaw = ((vtd_cap_reg & VTD_CAP_MGAW_MASK) >> VTD_CAP_MGAW_SHIFT) + 1;
-	if (mgaw < X86_VA_WIDTH)
+
+	if (!rte_eal_check_dma_mask(mgaw))
+		return true;
+	else
 		return false;
 
-	return true;
 }
 #elif defined(RTE_ARCH_PPC_64)
 static bool
@@ -640,13 +643,17 @@
 {
 	struct rte_pci_device *dev = NULL;
 	struct rte_pci_driver *drv = NULL;
+	int iommu_dma_mask_check_done = 0;
 
 	FOREACH_DRIVER_ON_PCIBUS(drv) {
 		FOREACH_DEVICE_ON_PCIBUS(dev) {
 			if (!rte_pci_match(drv, dev))
 				continue;
-			if (!pci_one_device_iommu_support_va(dev))
-				return false;
+			if (!iommu_dma_mask_check_done) {
+				if (pci_one_device_iommu_support_va(dev) < 0)
+					return false;
+				iommu_dma_mask_check_done  = 1;
+			}
 		}
 	}
 	return true;
-- 
1.9.1

[dpdk-dev] [PATCH v2 4/6] mem: use address hint for mapping hugepages

[Alejandro Lucero]

Linux kernel uses a really high address as starting address for
serving mmaps calls. If there exists addressing limitations and
IOVA mode is VA, this starting address is likely too high for
those devices. However, it is possible to use a lower address in
the process virtual address space as with 64 bits there is a lot
of available space.

This patch adds an address hint as starting address for 64 bits
systems.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 58 +++++++++++++++++++++++++++-----
 1 file changed, 49 insertions(+), 9 deletions(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 17c20d4..2ad16b0 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -88,6 +88,23 @@
 
 static uint64_t baseaddr_offset;
 
+#ifdef RTE_ARCH_64
+/*
+ * Linux kernel uses a really high address as starting address for serving
+ * mmaps calls. If there exists addressing limitations and IOVA mode is VA,
+ * this starting address is likely too high for those devices. However, it
+ * is possible to use a lower address in the process virtual address space
+ * as with 64 bits there is a lot of available space.
+ *
+ * Current known limitations are 39 or 40 bits. Setting the starting address
+ * at 4GB implies there are 508GB or 1020GB for mapping the available
+ * hugepages. This is likely enough for most systems, although a device with
+ * addressing limitations should call rte_dev_check_dma_mask for ensuring all
+ * memory is within supported range.
+ */
+static uint64_t baseaddr = 0x100000000;
+#endif
+
 static bool phys_addrs_available = true;
 
 #define RANDOMIZE_VA_SPACE_FILE "/proc/sys/kernel/randomize_va_space"
@@ -250,6 +267,26 @@
 	}
 }
 
+static void *
+get_addr_hint(void)
+{
+	if (internal_config.base_virtaddr != 0) {
+		return (void *) (uintptr_t)
+			    (internal_config.base_virtaddr +
+			     baseaddr_offset);
+	}
+#ifdef RTE_ARCH_64
+	else {
+		return (void *) (uintptr_t) (baseaddr +
+				baseaddr_offset);
+	}
+#else
+	else {
+		return NULL;
+	}
+#endif
+}
+
 /*
  * Try to mmap *size bytes in /dev/zero. If it is successful, return the
  * pointer to the mmap'd area and keep *size unmodified. Else, retry
@@ -260,16 +297,10 @@
 static void *
 get_virtual_area(size_t *size, size_t hugepage_sz)
 {
-	void *addr;
+	void *addr, *addr_hint;
 	int fd;
 	long aligned_addr;
 
-	if (internal_config.base_virtaddr != 0) {
-		addr = (void*) (uintptr_t) (internal_config.base_virtaddr +
-				baseaddr_offset);
-	}
-	else addr = NULL;
-
 	RTE_LOG(DEBUG, EAL, "Ask a virtual area of 0x%zx bytes\n", *size);
 
 	fd = open("/dev/zero", O_RDONLY);
@@ -278,7 +309,9 @@
 		return NULL;
 	}
 	do {
-		addr = mmap(addr,
+		addr_hint = get_addr_hint();
+
+		addr = mmap(addr_hint,
 				(*size) + hugepage_sz, PROT_READ,
 #ifdef RTE_ARCH_PPC_64
 				MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB,
@@ -286,8 +319,15 @@
 				MAP_PRIVATE,
 #endif
 				fd, 0);
-		if (addr == MAP_FAILED)
+		if (addr == MAP_FAILED) {
+			/* map failed. Let's try with less memory */
 			*size -= hugepage_sz;
+		} else if (addr_hint && addr != addr_hint) {
+			/* hint was not used. Try with another offset */
+			munmap(addr, (*size) + hugepage_sz);
+			addr = MAP_FAILED;
+			baseaddr_offset += 0x100000000;
+		}
 	} while (addr == MAP_FAILED && *size > 0);
 
 	if (addr == MAP_FAILED) {
-- 
1.9.1

Re: [dpdk-dev] [PATCH v2 4/6] mem: use address hint for mapping hugepages

[Burakov, Anatoly]

On 03-Jul-18 1:09 PM, Alejandro Lucero wrote:
> Linux kernel uses a really high address as starting address for
> serving mmaps calls. If there exists addressing limitations and
> IOVA mode is VA, this starting address is likely too high for
> those devices. However, it is possible to use a lower address in
> the process virtual address space as with 64 bits there is a lot
> of available space.
> 
> This patch adds an address hint as starting address for 64 bits
> systems.
> 
> Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
> ---

<snip>

> +static void *
> +get_addr_hint(void)
> +{
> +	if (internal_config.base_virtaddr != 0) {
> +		return (void *) (uintptr_t)
> +			    (internal_config.base_virtaddr +
> +			     baseaddr_offset);
> +	}
> +#ifdef RTE_ARCH_64
> +	else {
> +		return (void *) (uintptr_t) (baseaddr +
> +				baseaddr_offset);
> +	}
> +#else
> +	else {
> +		return NULL;
> +	}
> +#endif

Nitpicking, but probably ifdefs should be inside else clause? Two else's 
look weird.

> +}
> +
>   /*
>    * Try to mmap *size bytes in /dev/zero. If it is successful, return the
>    * pointer to the mmap'd area and keep *size unmodified. Else, retry
> @@ -260,16 +297,10 @@
>   static void *
>   get_virtual_area(size_t *size, size_t hugepage_sz)
>   {

<snip>

> -		addr = mmap(addr,
> +		addr_hint = get_addr_hint();
> +
> +		addr = mmap(addr_hint,
>   				(*size) + hugepage_sz, PROT_READ,
>   #ifdef RTE_ARCH_PPC_64
>   				MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB,
> @@ -286,8 +319,15 @@
>   				MAP_PRIVATE,
>   #endif
>   				fd, 0);
> -		if (addr == MAP_FAILED)
> +		if (addr == MAP_FAILED) {
> +			/* map failed. Let's try with less memory */
>   			*size -= hugepage_sz;
> +		} else if (addr_hint && addr != addr_hint) {
> +			/* hint was not used. Try with another offset */
> +			munmap(addr, (*size) + hugepage_sz);

Maybe store mapping size in a variable.

Otherwise,

Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>

-- 
Thanks,
Anatoly

[dpdk-dev] [PATCH v2 5/6] net/nfp: check hugepages IOVAs based on DMA mask

[Alejandro Lucero]

NFP devices can not handle DMA addresses requiring more than
40 bits. This patch uses rte_dev_check_dma_mask with 40 bits
and avoids device initialization if memory out of NFP range.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/net/nfp/nfp_net.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index d9cd047..5976f37 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -2649,6 +2649,14 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 
 	pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
 
+	/* NFP can not handle DMA addresses requiring more than 40 bits */
+	if (rte_eth_dev_check_dma_mask(40) < 0) {
+		RTE_LOG(INFO, PMD, "device %s can not be used:",
+				   pci_dev->device.name);
+		RTE_LOG(INFO, PMD, "\trestricted dma mask to 40 bits!\n");
+		return -ENODEV;
+	};
+
 	if ((pci_dev->id.device_id == PCI_DEVICE_ID_NFP4000_PF_NIC) ||
 	    (pci_dev->id.device_id == PCI_DEVICE_ID_NFP6000_PF_NIC)) {
 		port = get_pf_port_number(eth_dev->data->name);
-- 
1.9.1

[dpdk-dev] [PATCH v2 6/6] net/nfp: support IOVA VA mode

[Alejandro Lucero]

NFP can handle IOVA as VA. It requires to check those IOVAs
being in the supported range what is done during initialization.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
---
 drivers/net/nfp/nfp_net.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 5976f37..354dec3 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -3053,14 +3053,16 @@ static int eth_nfp_pci_remove(struct rte_pci_device *pci_dev)
 
 static struct rte_pci_driver rte_nfp_net_pf_pmd = {
 	.id_table = pci_id_nfp_pf_net_map,
-	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
+	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC |
+		     RTE_PCI_DRV_IOVA_AS_VA,
 	.probe = nfp_pf_pci_probe,
 	.remove = eth_nfp_pci_remove,
 };
 
 static struct rte_pci_driver rte_nfp_net_vf_pmd = {
 	.id_table = pci_id_nfp_vf_net_map,
-	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
+	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC |
+		     RTE_PCI_DRV_IOVA_AS_VA,
 	.probe = eth_nfp_pci_probe,
 	.remove = eth_nfp_pci_remove,
 };
-- 
1.9.1

Re: [dpdk-dev] [PATCH v2 0/6] use IOVAs check based on DMA mask

[Burakov, Anatoly]

On 03-Jul-18 1:09 PM, Alejandro Lucero wrote:
> This patchset adds, mainly, a check for ensuring IOVAs are within a
> restricted range due to addressing limitations with some devices. There
> are two known cases: NFP and IOMMU VT-d emulation.
> 
> With this check IOVAs out of range are detected and PMDs can abort
> initialization. For the VT-d case, IOVA VA mode is allowed as long as
> IOVAs are within the supported range, avoiding to forbid IOVA VA by
> default.
> 
> For the addressing limitations known cases, there are just 40(NFP) or
> 39(VT-d) bits for handling IOVAs. When using IOVA PA, those limitations
> imply 1TB(NFP) or 512M(VT-d) as upper limits, which is likely enough for
> most systems. With machines using more memory, the added check will
> ensure IOVAs within the range.
> 
> With IOVA VA, and because the way the Linux kernel serves mmap calls
> in 64 bits systems, 39 or 40 bits are not enough. It is possible to
> give an address hint with a lower starting address than the default one
> used by the kernel, and then ensuring the mmap uses that hint or hint plus
> some offset. With 64 bits systems, the process virtual address space is
> large enoguh for doing the hugepages mmaping within the supported range
> when those addressing limitations exist. This patchset also adds a change
> for using such a hint making the use of IOVA VA a more than likely
> possibility when there are those addressing limitations.
> 
> The check is not done by default but just when it is required. This
> patchset adds the check for NFP initialization and for setting the IOVA
> mode is an emulated VT-d is detected.
> 
> This patchset applies on 17.11.3.
> 
> Similar changes will be submitted to main DPDK branch soon.
> 
> v2:
>   - add get_addr_hint function
>   - call munmap when hint given and not used by mmap
>   - create dma mask in one step
>   - refactor logs
> 
> 

General comment - new API's should probably be added to the .map files?

-- 
Thanks,
Anatoly

[dpdk-dev] [PATCH v2 0/6] use IOVAs check based on DMA mask

[Alejandro Lucero]

I sent a patchset about this to be applied on 17.11 stable. The memory
code has had main changes since that version, so here it is the patchset
adjusted to current master repo.

This patchset adds, mainly, a check for ensuring IOVAs are within a
restricted range due to addressing limitations with some devices. There
are two known cases: NFP and IOMMU VT-d emulation.

With this check IOVAs out of range are detected and PMDs can abort
initialization. For the VT-d case, IOVA VA mode is allowed as long as
IOVAs are within the supported range, avoiding to forbid IOVA VA by
default.

For the addressing limitations known cases, there are just 40(NFP) or
39(VT-d) bits for handling IOVAs. When using IOVA PA, those limitations
imply 1TB(NFP) or 512M(VT-d) as upper limits, which is likely enough for
most systems. With machines using more memory, the added check will
ensure IOVAs within the range.

With IOVA VA, and because the way the Linux kernel serves mmap calls
in 64 bits systems, 39 or 40 bits are not enough. It is possible to
give an address hint with a lower starting address than the default one
used by the kernel, and then ensuring the mmap uses that hint or hint plus
some offset. With 64 bits systems, the process virtual address space is
large enoguh for doing the hugepages mmaping within the supported range
when those addressing limitations exist. This patchset also adds a change
for using such a hint making the use of IOVA VA a more than likely
possibility when there are those addressing limitations.

The check is not done by default but just when it is required. This
patchset adds the check for NFP initialization and for setting the IOVA
mode is an emulated VT-d is detected. Also, because the recent patchset
adding dynamic memory allocation, the check is also invoked for ensuring
the new memsegs are within the required range.

This patchset could be applied to stable 18.05.

v2:
 - change logs from INFO to DEBUG
 - only keeps dma mask if device capable of addressing allocated memory
 - add ABI changes
 - change hint address increment to page size
 - split pci/bus commit in two
 - fix commits