Commit Graph

115 Commits

Author SHA1 Message Date
Tal Shnaiderman
b762221ac2 bus/pci: support Windows with bifurcated drivers
Uses SetupAPI.h functions to scan PCI tree.
Uses DEVPKEY_Device_Numa_Node to get the PCI NUMA node.
Uses SPDRP_BUSNUMBER and SPDRP_BUSNUMBER to get the BDF.
scanning currently supports types RTE_KDRV_NONE.

Signed-off-by: Tal Shnaiderman <talshn@mellanox.com>
2020-06-30 00:02:54 +02:00
Tal Shnaiderman
33031608e8 bus/pci: introduce Windows support with stubs
Addition of stub eal and bus/pci functions to compile
bus/pci for Windows.

Signed-off-by: Tal Shnaiderman <talshn@mellanox.com>
2020-06-30 00:02:54 +02:00
Tal Shnaiderman
b137f95366 pci: build on Windows
Added <sys/types.h> in rte_pci header file
to include off_t type since it is missing for Windows.

Define the implementation of the Linux function rte_pci_get_sysfs_path
in pci_common.c for Linux OS only as it is unneeded for other OSs
and to avoid the warning on deprecated call to getenv() on Windows:

"warning: 'getenv' is deprecated: This function or variable may be unsafe.
Consider using _dupenv_s instead."

Signed-off-by: Tal Shnaiderman <talshn@mellanox.com>
2020-06-30 00:02:54 +02:00
Tal Shnaiderman
2fd3567e54 pci: use OS generic memory mapping functions
Changing all of PCIs Unix memory mapping to the
new memory allocation API wrapper.

Change all of PCI mapping function usage in
bus/pci to support the new API.

Signed-off-by: Tal Shnaiderman <talshn@mellanox.com>
2020-06-30 00:02:54 +02:00
Haiyue Wang
54f3fb127d bus/pci: fix VF memory access
To fix CVE-2020-12888, the linux vfio-pci module will invalidate mmaps
and block MMIO access on disabled memory, it will send a SIGBUS to the
application:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=abafbc551fdd

When the application opens the vfio PCI device, the vfio-pci module will
enable the bus memory space through PCI read/write access. According to
the PCIe specification, the 'Memory Space Enable' is always zero for VF:

             Table 9-13 Command Register Changes

Bit Location | PF and VF Register Differences | PF         | VF
             | From Base                      | Attributes | Attributes
-------------+--------------------------------+------------+-----------
             | Memory Space Enable - Does not |            |
             | apply to VFs. Must be hardwired|  Base      |  0b
     1       | to 0b for VFs. VF Memory Space |            |
             | is controlled by the VF MSE bit|            |
             | in the VF Control register.    |            |
-------------+--------------------------------+------------+-----------

Afterwards the vfio-pci will initialize its own virtual PCI config space
data ('vconfig') by reading the VF's physical PCI config space, then the
'Memory Space Enable' bit in vconfig will always be 0b value. This will
make the vfio-pci treat the BAR memory space as disabled, and the SIGBUS
will be triggered if access these BARs.

By investigation, the VF PCI device *passthrough* into the Guest OS by
QEMU has the 'Memory Space Enable' with 1b value. That's because every
PCI driver will start to enable the memory space, and this action will
be hooked by vfio-pci virtual PCI read/write to set the 'Memory Space
Enable' in vconfig space to 1b. So VF runs in guest OS has 'Mem+', but
VF runs in host OS has 'Mem-'.

Align with PCI working mode in Guest/QEMU/Host, in DPDK, enable the PCI
bus memory space explicitly to avoid access on disabled memory.

Fixes: 33604c3135 ("vfio: refactor PCI BAR mapping")
Cc: stable@dpdk.org

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Tested-by: Harman Kalra <hkalra@marvell.com>
Tested-by: David Marchand <david.marchand@redhat.com>
Tested-by: Thierry Martin <thierry.martin.public@gmail.com>
2020-06-25 15:22:51 +02:00
Jerin Jacob
c79a1c6746 bus/pci: optimize bus scan
In order to optimize the PCI management, RTE_KDRV_NONE based
device driver probing removed by not adding them to list in
the scan phase.

The legacy virtio is the only consumer of RTE_KDRV_NONE based device
driver probe scheme. The legacy virtio support will be available
through the existing VFIO/UIO based kernel driver scheme.

This patch also removes the deprecation notice for the same.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
2020-06-24 23:49:15 +02:00
Sunil Kumar Kori
463a5245d5 bus/pci: optimise scanning with whitelist/blacklist
rte_bus_scan API scans all the available PCI devices irrespective of white
or black listing parameters then further devices are probed based on white
or black listing parameters. So unnecessary CPU cycles are wasted during
rte_pci_scan.

For Octeontx2 platform with core frequency 2.4 Ghz, rte_bus_scan consumes
around 26ms to scan around 90 PCI devices but all may not be used by the
application. So for the application which uses 2 NICs, rte_bus_scan
consumes few microseconds and rest time is saved with this patch.

Patch restricts devices to be scanned as per below mentioned conditions:
 - All devices will be scanned if no parameters are passed.
 - Only white listed devices will be scanned if white list is available.
 - All devices, except black listed, will be scanned if black list is
   available.

Signed-off-by: Sunil Kumar Kori <skori@marvell.com>
Acked-by: Gaetan Rivet <grive@u256.net>
2020-05-11 16:59:58 +02:00
David Marchand
87db93e07a remove references to private PCI probe function
rte_pci_probe() is private to the PCI bus.
Clean the remaining references in the documentation and comments.

Fixes: c752998b5e ("pci: introduce library and driver")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Gaetan Rivet <grive@u256.net>
2020-05-11 16:59:58 +02:00
Jerin Jacob
e4f27af0f4 bus/pci: reduce boot-up logs to absolute minimum
Some machines may have a lot of PCI devices and all of them are
not bound to DPDK. In such case the logs from EAL creates a lot of
clutter on boot-up, typically one needs to scroll the screen to
find other issues in boot-up.

This patch changes the following to reduce the clutter in
the default boot-up logs.

- Change the log-level of PCI probes to `debug`
- Introduce new driver probe as `info` log-level for the successful probe.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
2020-05-11 16:36:00 +02:00
Vijaya Mohan Guvva
e6cf7bee1c bus/pci: fix UIO resource access from secondary process
For PCI devices presented through igb_uio, pcidev->mem_resource[] is
not populated when the device is initialized for secondary process.

Initialize pcidev->mem_resource[] with pci-bar mapped addresses.

Fixes: eee16c964c ("pci: support multiple PCI regions per device")
Cc: stable@dpdk.org

Signed-off-by: Vijaya Mohan Guvva <vijay1054@gmail.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
2020-04-25 22:41:58 +02:00
David Christensen
9052157318 bus/pci: support IOVA as VA on PowerNV systems
All recent POWER systems, Power 8 and 9 specifically, support an IOMMU
(it can't be disabled). The functionality of the IOMMU is different
depending on whether it's running on a bare metal PowerNV system or in
a virtual environment (PowerVM LPAR or KVM/QEMU).  DPDK currently
supports the IOMMU found on PowerNV platforms, sPAPRv2, so IOVA=VA
mode can be enabled when the correct platform is detected.

The POWER IOMMU type can't be detected through mechanisms such as
parsing files in the /sys hierarchy like x86_64 systems so the
/proc/cpuinfo file is parsed to determine whether Linux is running
on bare metal (i.e. PowerNV) or in a virtual environment (KVM/QEMU).

Signed-off-by: David Christensen <drc@linux.vnet.ibm.com>
2020-04-25 17:01:01 +02:00
Pavan Nikhilesh
acec04c4b2 build: disable experimental API check internally
Remove setting ALLOW_EXPERIMENTAL_API individually for each Makefile and
meson.build. Instead, enable ALLOW_EXPERIMENTAL_API flag across app, lib
and drivers.
This changes reduces the clutter across the project while still
maintaining the functionality of ALLOW_EXPERIMENTAL_API i.e. warning
external applications about experimental API usage.

Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
2020-04-14 16:22:34 +02:00
Somnath Kotur
fc67ae91eb bus/pci: fix devargs on probing again
As per the comments in this code section, since there is a matching device,
it is now its responsibility to manage the devargs we've just inserted.
But the matching device ptr's devargs is still uninitialized or not pointing
to the newest dev_args that were passed as a parameter to local_dev_probe().
This is needed particularly in the case when *probe is called again* on an
already probed device as part of adding a representor port to OVS-DPDK.

Fixes: 7e8b266501 ("eal: fix hotplug add / remove")
Cc: stable@dpdk.org

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Acked-by: Gaetan Rivet <grive@u256.net>
2020-03-31 02:54:29 +02:00
Matan Azrad
fd03a747b0 bus/pci: clear driver on detach
When probing, a bus is responsible for filling the driver field to
indicate the device is bound.
Let's clear this field when detaching to keep a consistent behavior.

This is not a fix per se, since the device is freed when detaching.
But at least clearing the field has been added to remind that the
driver field has a meaning for the EAL.

Signed-off-by: Matan Azrad <matan@mellanox.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-11-21 00:18:02 +01:00
Pawel Modrak
85ff364f3b build: align symbols with global ABI version
Merge all versions in linker version script files to DPDK_20.0.

This commit was generated by running the following command:

:~/DPDK$ buildtools/update-abi.sh 20.0

Signed-off-by: Pawel Modrak <pawelx.modrak@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-11-20 23:05:39 +01:00
Anatoly Burakov
fbaf943887 build: remove individual library versions
Since the library versioning for both stable and experimental ABI's is
now managed globally, the LIBABIVER and version variables no longer
serve any useful purpose, and can be removed.

The replacement in Makefiles was done using the following regex:

	^(#.*\n)?LIBABIVER\s*:=\s*\d+\n(\s*\n)?

(LIBABIVER := numbers, optionally preceded by a comment and optionally
succeeded by an empty line)

The replacement for meson files was done using the following regex:

	^(#.*\n)?version\s*=\s*\d+\n(\s*\n)?

(version = numbers, optionally preceded by a comment and optionally
succeeded by an empty line)

[David]: those variables are manually removed for the files:
- drivers/common/qat/Makefile
- lib/librte_eal/meson.build
[David]: the LIBABIVER is restored for the external ethtool example
library.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
2019-11-20 23:05:39 +01:00
Wangyu (Eric)
d25ab4b7f1 bus/pci: align next mapping address on page boundary
Currently, the next address picked by PCI mapping infrastructure
may be page-unaligned due to BAR length being smaller than page size.
This leads to a situation where the requested map address is invalid,
resulting in mmap() call returning an arbitrary address,
which will later interfere with device BAR mapping in secondary processes.

Fix it by always aligning the next requested address on page boundary.

Fixes: c752998b5e ("pci: introduce library and driver")
Cc: stable@dpdk.org

Signed-off-by: Xiaofeng Deng <dengxiaofeng@huawei.com>
Signed-off-by: Wangyu (Eric) <seven.wangyu@huawei.com>
Acked-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Acked-by: Min Hu (Connor) <humin29@huawei.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Gavin Hu <gavin.hu@arm.com>
2019-11-14 14:17:48 +01:00
Michal Krawczyk
8108393d98 vfio: fix truncated BAR offset for 32-bit
When 32-bit application is built on 64-bit system it is possible that
the offset of the resource is outside of the 32-bit value.

The problem with the unsigned long is, that it is 32-bit and not 64-bit
when using armhf compiler. Although the system is returning u64 value,
we are losing it's value if it's higher than 32-bit in the conversion
process. It can further cause mmap to fail due to offset being 0 or to
map not intended memory region.

To make it more portable, the uint64_t value is now being used for
storing offset instead of unsigned long. The size of being 32-bit seems
to be fine as the 32-bit application won't be able to access bigger
memory and it is further converted to size_t anyway. But for better
readability and to be consistent, it's type was changed to size_t as
well.

Fixes: 0205f87355 ("vfio: fix overflow of BAR region offset and size")
Cc: stable@dpdk.org

Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
2019-10-26 17:30:17 +02:00
David Marchand
e02b661b51 bus/pci: check IO permissions for UIO only
On x86, calling inb/outb special instructions (used in UIO ioport
read/write parts) is only possible if the right IO permissions has been
granted.

The only user of this API (the net/virtio pmd) checks this
unconditionnaly but this should be hidden by the rte_pci_ioport API
itself and only checked when the device is bound to a UIO driver.

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
2019-10-25 11:54:10 +02:00
Seth Howell
c345c7d1ac bus/pci: remove useless link dependency on ethdev
The makefile in drivers/bus/pci specified rte_ethdev as a dependency for
the library. However there are no actual symbols from librte_ethdev used
in librte_bus_pci.

Including librte_ethdev as a dependency only becomes a problem in some
niche cases like when attempting to build the rte_bus_pci library as a
shared object without building the rte_ethdev library.

I specifically ran into this when trying to build the DPDK included as
an SPDK submodule on a FreeBSD machine. I figure that since there are no
real dependencies between the two, we should enable building
librte_bus_pci without librte_ethdev.

Fixes: c752998b5e ("pci: introduce library and driver")
Cc: stable@dpdk.org

Signed-off-by: Seth Howell <seth.howell@intel.com>
Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>
2019-10-25 10:51:15 +02:00
Stephen Hemminger
2e8d5cf763 bus/pci: fix Intel IOMMU sysfs access check
Just open the sysfs file and handle failure, rather than using access().
This eliminates Coverity warnings about TOCTOU
"time of check versus time of use"; although for this sysfs file that is
not really an issue anyway.

Coverity issue: 347276
Fixes: 54a328f552 ("bus/pci: forbid IOVA mode if IOMMU address width too small")
Cc: stable@dpdk.org

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: David Marchand <david.marchand@redhat.com>
2019-10-09 10:22:45 +02:00
David Marchand
8ac3591694 remove useless include of EAL memory config header
Restrict this header inclusion to its real users.

Fixes: 028669bc9f ("eal: hide shared memory config")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
2019-10-09 10:22:24 +02:00
David Marchand
66d3724b2c bus/pci: always check IOMMU capabilities
IOMMU capabilities won't change and must be checked even if no PCI device
seem to be supported yet when EAL initialised.

This is to accommodate with SPDK that registers its drivers after
rte_eal_init(), especially on PPC platform where the IOMMU does not
support VA.

Fixes: 703458e19c ("bus/pci: consider only usable devices for IOVA mode")

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Tested-by: Jerin Jacob <jerinj@marvell.com>
Tested-by: Takeshi Yoshimura <tyos@jp.ibm.com>
2019-08-05 12:08:15 +02:00
David Marchand
b62f3aff91 bus/pci: remove unused x86 Linux constant
This macro is unused after a previous fix.

Fixes: fe822eb8c5 ("bus/pci: use IOVA DMA mask check when setting IOVA mode")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
2019-08-05 12:02:20 +02:00
Nithin Dabilpuram
33543fb3b6 vfio: revert interrupt eventfd setup at probe
This reverts commit 89aac60e0b.
"vfio: fix interrupts race condition"

The above mentioned commit moves the interrupt's eventfd setup
to probe time but only enables one interrupt for all types of
interrupt handles i.e VFIO_MSI, VFIO_LEGACY, VFIO_MSIX, UIO.
It works fine with default case but breaks below cases specifically
for MSIX based interrupt handles.

* Applications like l3fwd-power that request rxq interrupts
  while ethdev setup.
* Drivers that need > 1 MSIx interrupts to be configured for
  functionality to work.

VFIO PCI for MSIx expects all the possible vectors to be setup up
when using VFIO_IRQ_SET_ACTION_TRIGGER so that they can be
allocated from kernel pci subsystem. Only way to increase the number
of vectors later is first free all by using VFIO_IRQ_SET_DATA_NONE
with action trigger and then enable new vector count.

Above commit changes the behavior of rte_intr_[enable|disable] to
only mask and unmask unlike earlier behavior and thereby
breaking above two scenarios.

Fixes: 89aac60e0b ("vfio: fix interrupts race condition")
Cc: stable@dpdk.org

Signed-off-by: Nithin Dabilpuram <ndabilpuram@marvell.com>
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Tested-by: Stephen Hemminger <stephen@networkplumber.org>
Tested-by: Shahed Shaikh <shshaikh@marvell.com>
Tested-by: Lei Yao <lei.a.yao@intel.com>
Acked-by: David Marchand <david.marchand@redhat.com>
2019-07-23 12:00:14 +02:00
Jerin Jacob
d622cad892 bus/pci: change IOVA as VA flag name
In order to align name with other PCI driver flag such as
RTE_PCI_DRV_NEED_MAPPING and to reflect its purpose, change
RTE_PCI_DRV_IOVA_AS_VA flag name as RTE_PCI_DRV_NEED_IOVA_AS_VA.

Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
2019-07-22 17:46:32 +02:00
David Marchand
b76fafb174 eal: fix IOVA mode selection as VA for PCI drivers
The incriminated commit broke the use of RTE_PCI_DRV_IOVA_AS_VA which
was intended to mean "driver only supports VA" but had been understood
as "driver supports both PA and VA" by most net drivers and used to let
dpdk processes to run as non root (which do not have access to physical
addresses on recent kernels).

The check on physical addresses actually closed the gap for those
drivers. We don't need to mark them with RTE_PCI_DRV_IOVA_AS_VA and this
flag can retain its intended meaning.
Document explicitly its meaning.

We can check that a driver requirement wrt to IOVA mode is fulfilled
before trying to probe a device.

Finally, document the heuristic used to select the IOVA mode and hope
that we won't break it again.

Fixes: 703458e19c ("bus/pci: consider only usable devices for IOVA mode")

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
Tested-by: Jerin Jacob <jerinj@marvell.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
2019-07-22 17:45:52 +02:00
David Marchand
62f8f5ace5 bus/pci: remove Mellanox kernel driver type
This reverts commit 0cb86518db.

The PCI bus now reports DC when faced with a device bound to an unknown
driver and, in such a case, the IOVA mode is selected against physical
address availability.

As a consequence, there is no reason for this special case for Mellanox
drivers.

Fixes: 703458e19c ("bus/pci: consider only usable devices for IOVA mode")

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Jerin Jacob <jerinj@marvell.com>
2019-07-22 17:44:08 +02:00
David Marchand
89aac60e0b vfio: fix interrupts race condition
Populating the eventfd in rte_intr_enable in each request to vfio
triggers a reconfiguration of the interrupt handler on the kernel side.
The problem is that rte_intr_enable is often used to re-enable masked
interrupts from drivers interrupt handlers.

This reconfiguration leaves a window during which a device could send
an interrupt and then the kernel logs this (unsolicited from the kernel
point of view) interrupt:
[158764.159833] do_IRQ: 9.34 No irq handler for vector

VFIO api makes it possible to set the fd at setup time.
Make use of this and then we only need to ask for masking/unmasking
legacy interrupts and we have nothing to do for MSI/MSIX.

"rxtx" interrupts are left untouched but are most likely subject to the
same issue.

Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=1654824
Fixes: 5c782b3928 ("vfio: interrupts")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Tested-by: Shahed Shaikh <shshaikh@marvell.com>
2019-07-10 18:53:47 +02:00
Anatoly Burakov
028669bc9f eal: hide shared memory config
Now that everything that has ever accessed the shared memory
config is doing so through the public API's, we can make it
internal. Since we're removing quite a few headers from
rte_eal_memconfig.h, we need to add them back in places
where this header is used.

This bumps the ABI, so also change all build files and make
update documentation.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: David Marchand <david.marchand@redhat.com>
2019-07-06 10:32:34 +02:00
Ben Walker
703458e19c bus/pci: consider only usable devices for IOVA mode
When selecting the preferred IOVA mode of the pci bus, the current
heuristic ("are devices bound?", "are devices bound to UIO?", "are pmd
drivers supporting IOVA as VA?" etc..) should honor the device
white/blacklist so that an unwanted device does not impact the decision.

There is no reason to consider a device which has no driver available.

This applies to all OS, so implements this in common code then call a
OS specific callback.

On Linux side:
- the VFIO special considerations should be evaluated only if VFIO
  support is built,
- there is no strong requirement on using VA rather than PA if a driver
  supports VA, so defaulting to DC in such a case.

Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Anatoly Burakov <anatoly.burakov@intel.com>
2019-07-05 16:56:00 +02:00
Stephen Hemminger
c530aa78e3 bus/pci: fix TOCTOU for sysfs access
Using access followed by open causes a static analysis warning
about Time of check versus Time of use. Also, access() and
open() have different UID permission checks.

This is not a serious problem; but easy to fix by using errno instead.

Coverity issue: 300870
Fixes: 4a928ef9f6 ("bus/pci: enable write combining during mapping")
Cc: stable@dpdk.org

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: David Marchand <david.marchand@redhat.com>
2019-06-14 16:33:56 +09:00
Yongseok Koh
0cb86518db bus/pci: add Mellanox kernel driver type
When checking RTE_PCI_DRV_IOVA_AS_VA flag to determine IOVA mode,
pci_one_device_has_iova_va() returns true only if kernel driver of the
device is vfio. However, Mellanox mlx4/5 PMD doesn't need to be detached
from kernel driver and attached to VFIO/UIO. Control path still goes
through the existing kernel driver, which is mlx4_core/mlx5_core. In order
to make RTE_PCI_DRV_IOVA_AS_VA effective for mlx4/mlx5 PMD, a new kernel
driver type has to be introduced.

Cc: stable@dpdk.org

Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
2019-06-04 00:33:06 +02:00
Bruce Richardson
adf93ca564 build: increase readability via shortcut variables
Define variables for "is_linux", "is_freebsd" and "is_windows"
to make the code shorter for comparisons and more readable.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Acked-by: Luca Boccassi <bluca@debian.org>
2019-04-17 18:09:52 +02:00
Bruce Richardson
6723c0fc72 replace snprintf with strlcpy
Do a global replace of snprintf(..."%s",...) with strlcpy, adding in the
rte_string_fns.h header if needed.  The function changes in this patch were
auto-generated via command:

  spatch --sp-file devtools/cocci/strlcpy.cocci --dir . --in-place

and then the files edited using awk to add in the missing header:

  gawk -i inplace '/include <rte_/ && ! seen { \
  	print "#include <rte_string_fns.h>"; seen=1} {print}'

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
2019-04-04 22:46:05 +02:00
Bruce Richardson
f9acaf84e9 replace snprintf with strlcpy without adding extra include
For files that already have rte_string_fns.h included in them, we can
do a straight replacement of snprintf(..."%s",...) with strlcpy. The
changes in this patch were auto-generated via command:

spatch --sp-file devtools/cocci/strlcpy-with-header.cocci --dir . --in-place

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
2019-04-04 22:45:54 +02:00
David Marchand
27893e4eee drivers: remove Linux EAL from include path
None of those drivers require EAL linux specific headers.

Signed-off-by: David Marchand <david.marchand@redhat.com>
2019-04-04 22:06:16 +02:00
Shahaf Shuler
c33a675b62 bus: introduce device level DMA memory mapping
The DPDK APIs expose 3 different modes to work with memory used for DMA:

1. Use the DPDK owned memory (backed by the DPDK provided hugepages).
This memory is allocated by the DPDK libraries, included in the DPDK
memory system (memseg lists) and automatically DMA mapped by the DPDK
layers.

2. Use memory allocated by the user and register to the DPDK memory
systems. Upon registration of memory, the DPDK layers will DMA map it
to all needed devices. After registration, allocation of this memory
will be done with rte_*malloc APIs.

3. Use memory allocated by the user and not registered to the DPDK memory
system. This is for users who wants to have tight control on this
memory (e.g. avoid the rte_malloc header).
The user should create a memory, register it through rte_extmem_register
API, and call DMA map function in order to register such memory to
the different devices.

The scope of the patch focus on #3 above.

Currently the only way to map external memory is through VFIO
(rte_vfio_dma_map). While VFIO is common, there are other vendors
which use different ways to map memory (e.g. Mellanox and NXP).

The work in this patch moves the DMA mapping to vendor agnostic APIs.
Device level DMA map and unmap APIs were added. Implementation of those
APIs was done currently only for PCI devices.

For PCI bus devices, the pci driver can expose its own map and unmap
functions to be used for the mapping. In case the driver doesn't provide
any, the memory will be mapped, if possible, to IOMMU through VFIO APIs.

Application usage with those APIs is quite simple:
* allocate memory
* call rte_extmem_register on the memory chunk.
* take a device, and query its rte_device.
* call the device specific mapping function for this device.

Future work will deprecate the rte_vfio_dma_map and rte_vfio_dma_unmap
APIs, leaving the rte device APIs as the preferred option for the user.

Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Gaetan Rivet <gaetan.rivet@6wind.com>
2019-03-30 16:48:56 +01:00
Bruce Richardson
5fbc1d498f build/freebsd: rename macro BSDPAPP to FREEBSD
Rename the macro and all instances in DPDK code, but keep a copy of
the old macro defined for legacy code linking against DPDK

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
2019-03-12 23:01:14 +01:00
Bruce Richardson
742bde12f3 build/linux: rename macro from LINUXAPP to LINUX
Rename the macro to make things shorter and more comprehensible. For
both meson and make builds, keep the old macro around for backward
compatibility.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
2019-03-12 17:31:22 +01:00
Alejandro Lucero
43f2b3d250 vfio: fix error message
The message refers to uio driver.

Fixes: ff0b67d1c8 ("vfio: DMA mapping")
Cc: stable@dpdk.org

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
2019-01-23 22:49:11 +01:00
Tone Zhang
9cea8774cf vfio: support 64KB kernel page size
With a larger PAGE_SIZE it is possible for the MSI table to very
close to the end of the BAR s.t. when we align the start and end
of the MSI table to the PAGE_SIZE, the end offset of the MSI
table is out of the PCI BAR boundary.

This patch addresses the issue by comparing both the start and the
end offset of the MSI table with the BAR size, and skip the mapping
if it is out of Bar scope.

The patch fixes the debug log as below:
EAL: Skipping BAR0

Signed-off-by: Tone Zhang <tone.zhang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Steve Capper <steve.capper@arm.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
2018-12-20 00:12:20 +01:00
Darek Stojaczyk
047e3f9f2a vfio: do not needlessly setup device in secondary process
Setting up a device that wasn't setup in the primary
process will possibly break the primary process. That's
because the IPC message to retrieve the group fd in the
primary will also *open* that group if it wasn't opened
before. Even though the secondary process closes that fd
soon after as a part of its error handling path, the
primary process leaks it.

What's worse, opening that fd on the primary will
increment the process-local counter of opened groups.
If it was 0 before, then the group will never be added
to the vfio container, nor dpdk memory will be ever
mapped.

This patch moves the proper error checks earlier in the
code to fully prevent setting up devices in secondary
processes that weren't setup in the primary process.

Fixes: 2f4adfad0a ("vfio: add multiprocess support")

Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
2018-11-25 13:09:05 +01:00
Ferruh Yigit
d3110b124a bus/pci: fix allocation of device path
The pci_resource_by_index called strlen() on uninitialized
memory which would lead to the wrong size of memory allocated
for the path portion of the resource map. This would either cause
excessively large allocation, or worse memory corruption.

Coverity issue: 300868
Fixes: ea9d56226e ("pci: introduce function to map uio resource by index")
Cc: stable@dpdk.org

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
2018-11-25 11:51:11 +01:00
Thomas Monjalon
b91bc6f357 vfio: fix build with Linux < 4.0
drivers/bus/pci/linux/pci_vfio.c:45:23: error:
‘failure_handle_lock’ defined but not used

Fixes: 8ffe738651 ("vfio: add lock for hot-unplug failure handler")

Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
2018-11-18 22:31:30 +01:00
Jeff Guo
8ffe738651 vfio: add lock for hot-unplug failure handler
When the sigbus handler be enabled for hot-unplug, whatever hot-unplug
sigbus or origin sigbus occur, the sigbus handler will be invoked and
it will access the bus and device. While in the control path, the vfio
req handler also will process the bus and device, so a protection of
the resources in vfio req handler should be need. This patch add a lock
in vfio req handler when process bus and device resource, to avoid the
synchronization issue when device be hot-unplugged.

Fixes: c115fd000c ("vfio: handle hotplug request notifier")

Signed-off-by: Jeff Guo <jia.guo@intel.com>
2018-11-18 17:16:42 +01:00
Fan Zhang
a38eafedda bus/pci: fix config r/w access
The recent change to rte_pci_read/write_config() missed
uio_pci_generic case.

Fixes: 630deed612 ("bus/pci: compare kernel driver instead of interrupt handler")
Cc: stable@dpdk.org

Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
2018-11-06 02:11:25 +01:00
Alejandro Lucero
ec20068713 bus/pci: avoid call to DMA mask check
Calling rte_mem_check_dma_mask when memory has not been initialized
yet is wrong. This patch use rte_mem_set_dma_mask instead.

Once memory initialization is done, the dma mask set will be used
for checking memory mapped is within the specified mask.

Fixes: fe822eb8c5 ("bus/pci: use IOVA DMA mask check when setting IOVA mode")

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-11-05 01:02:08 +01:00
Alejandro Lucero
0de9eb6138 mem: rename DMA mask check with proper prefix
Current name rte_eal_check_dma_mask does not follow the naming
used in the rest of the file.

Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Tested-by: Ferruh Yigit <ferruh.yigit@intel.com>
2018-11-05 01:01:54 +01:00
Qi Zhang
55e411b301 bus/pci: fix resource mapping override
When scanning an already plugged device, the virtual address
of mapped PCI resource in rte_pci_device will be overridden
with 0, that may cause driver does not work correctly.
The fix is not to update any rte_pci_device's field if the being
scanned device's driver is already probed.

Bugzilla ID: 85
Fixes: c752998b5e ("pci: introduce library and driver")
Cc: stable@dpdk.org

Reported-by: Geoffrey Lv <geoffrey.lv@gmail.com>
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
2018-10-31 19:43:34 +01:00