For dev_ops not supported by the secondary process, either return -EPERM
or return without doing anything. In both cases log a warning.
It's still application's responsibility to avoid calls like that and
those changes are for debugging/informational purposes.
Signed-off-by: Stanislaw Kardach <kda@semihalf.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
rte_pci_device and rte_eth_dev are process-local structures. Therefore
ena_adapter::pdev and ena_adapter::rte_dev cannot be used universally.
Both ena_timer_wd_callback and ena_interrupt_handler_rte needs access to
the rte_eth_dev, but as they are being setup and executed in the primary
process, it is safe to pass there the same pointer, which is used for
the device configuration.
In all other cases, except the eth_ena_dev_init(), the rte_eth_dev_data
is used instead.
Signed-off-by: Stanislaw Kardach <kda@semihalf.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
To make it possible to the app to determine if the hash was calculated
for the packet or not, the PKT_RX_RSS_HASH should be set in the mbuf's
ol_flags.
As the PMD wasn't setting that, the application couldn't check if there
is a hash in a proper way.
The hash is valid only if it's UDP or TCP and the IP packet wasn't
fragmented.
Fixes: e5df9f33db ("net/ena: fix passing RSS hash to mbuf")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Amit Bernstein <amitbern@amazon.com>
This version update contains:
* memcpy mapping to the dpdk-optimized version.
* ena_com (HAL) update to the latest version (from 18.09.2020).
* Bug fixes for the large LLQ headers and devargs parsing.
* Bug fix for the default ring size.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Remove invalid ring size alignment logic and add default Rx and Tx port
ring sizes to the device info spec.
The logic in lines 1297 and 1371 is invalid. The
RTE_ETH_DEV_FALLBACK_RX_RINGSIZE (and the TX counterpart) is a value
that rte_eth_rx_queue_setup() will set if
dev_info.default_rxportconf.ring_size is 0 and user provided 0 in
nb_rx_desc argument. However the current code treats it as a hint for
the PMD to change the ring size to internal defaults.
Additionally since the ENA_DEFAULT_RING_SIZE is defined, report it in
the device capabilities so that both rte_ethdev code and the user can
utilize it for device configuration.
Fixes: ea93d37eb4 ("net/ena: add HW queues depth setup")
Cc: stable@dpdk.org
Signed-off-by: Stanislaw Kardach <kda@semihalf.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
pthread_cond_timedwait() may spuriously wakeup according to POSIX.
Therefore it is required to check whether predicate is actually true
before finishing the waiting loop.
Signed-off-by: Stanislaw Kardach <kda@semihalf.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
swap*_*_le() functions are not used anywhere and besides there are rte
alternatives already present.
Fixes: 1173fca25a ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: Stanislaw Kardach <kda@semihalf.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
As the documentation of rte_kvargs_parse() states, the valid_keys
argument must be NULL terminated. Lack of this feature may cause
segmentation fault if the passed devarg will be different then the
supported value.
Fixes: 8a7a73f26c ("net/ena: support large LLQ headers")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
The code incorrectly checked the return value of comparison when parsing
the argument key name. The return value of strcmp should be compared
to 0 to identify a match.
Fixes: 8a7a73f26c ("net/ena: support large LLQ headers")
Cc: stable@dpdk.org
Signed-off-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
1. As memzone uses unique names, changed alloc coherent macro to use
64 bit size atomic variable to increase the memzone name space
2. "handle" param name change to be consistent with other macros
3. Variable definition displacement
4. Backslash alignment to column 80
Signed-off-by: Amit Bernstein <amitbern@amazon.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
As in the v20.11 rte_prefetch0_write API was added, it should be used
in the platform file for the definition of the macro prefetchw, instead
of using simply prefetch0.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Amit Bernstein <amitbern@amazon.com>
This feature wasn't used by any of the drivers. Because of that, it was
removed.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Shay Agroskin <shayagr@amazon.com>
The RSS indirection table shouldn't be included as a part of the
device features context.
If the driver has to acquire it, it can be done using the separate API
whenever the RSS configuration is needed.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Amit Bernstein <amitbern@amazon.com>
Although the ENA DPDK PMD doesn't have to perform any actions for
destroying the wait event, some other platforms have to.
The macro "ENA_WAIT_EVENT_DESTROY" was renamed to
"ENA_WAIT_EVENTS_DESTROY" and also whole implementation responsible for
that was moved to a separate function for better readability.
Fixes: 3adcba9a89 ("net/ena: update HAL to the newer version")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
To silence error messages from the static code analysis, make the type
conversions explicit where they're intended.
Also fix the type for the DMA width value.
Fixes: 99ecfbf845 ("ena: import communication layer")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
List of changes:
* Comment style was adjusted for the functions
* The keys_num at "struct ena_admin_feature_rss_flow_hash_control" was
renamed to the key_parts to better describe it's meaning
* The RSS indirection table was called "REDIRECTION" -> changed to
INDIRECTION
* Change AENQ field "syndrom" -> "syndrome"
* Calculate number of the RSS key parts or whole key by using the
common way: sizeof of the first element of the RSS key
* Add description of the "enum ena_admin_aq_feature_id"
* Rename "map_rx_buf_bidirectional" field as "rx_buf_mirroring"
* Other minor style fixes (remove extra spaces, add missing line break,
improve indentation)
* Remove unused macros ENA_ADMIN_EXTRA_PROPERTIES_*
* Restructure the "if {} else if {} else" conditional statement for
setting up the meta descriptor
Fixes: 99ecfbf845 ("ena: import communication layer")
Fixes: b68309be44 ("net/ena/base: update communication layer for the ENAv2")
Fixes: b2b02edeb0 ("net/ena/base: upgrade HAL for new HW features")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Some platforms may make use of the device argument to make the logs more
verbose and specific for the appropriate device.
As it's not used by the ENA DPDK PMD for the logging, the type is just
defined, but never used.
It may be reconsidered to change this in the future by adding port ID
to the message logs, but as for now the logging behavior won't change.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Instead of using 'queue' for struct ena_com_admin_queue and 'dev' for
struct ena_com_dev variables, use more descriptive 'admin_queue' and
'ena_dev'.
This also unifies the names of variables of the type struct ena_com_dev
in the driver.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
memcpy is now mapped to rte_memcpy macro on x86 architectures.
Fixes: 9ba7981ec9 ("ena: add communication layer for DPDK")
Cc: stable@dpdk.org
Signed-off-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Artur Rojek <ar@semihalf.com>
Let's try to enforce the convention where most drivers use a pmd. logtype
with their class reflected in it, and libraries use a lib. logtype.
Introduce two new macros:
- RTE_LOG_REGISTER_DEFAULT can be used when a single logtype is
used in a component. It is associated to the default name provided
by the build system,
- RTE_LOG_REGISTER_SUFFIX can be used when multiple logtypes are used,
and then the passed name is appended to the default name,
RTE_LOG_REGISTER is left untouched for existing external users
and for components that do not comply with the convention.
There is a new Meson variable log_prefix to adapt the default name
for baseband (pmd.bb.), bus (no pmd.) and mempool (no pmd.) classes.
Note: achieved with below commands + reverted change on net/bonding +
edits on crypto/virtio, compress/mlx5, regex/mlx5
$ git grep -l RTE_LOG_REGISTER drivers/ |
while read file; do
pattern=${file##drivers/};
class=${pattern%%/*};
pattern=${pattern#$class/};
drv=${pattern%%/*};
case "$class" in
baseband) pattern=pmd.bb.$drv;;
bus) pattern=bus.$drv;;
mempool) pattern=mempool.$drv;;
*) pattern=pmd.$class.$drv;;
esac
sed -i -e 's/RTE_LOG_REGISTER(\(.*\), '$pattern',/RTE_LOG_REGISTER_DEFAULT(\1,/' $file;
sed -i -e 's/RTE_LOG_REGISTER(\(.*\), '$pattern'\.\(.*\),/RTE_LOG_REGISTER_SUFFIX(\1, \2,/' $file;
done
$ git grep -l RTE_LOG_REGISTER lib/ |
while read file; do
pattern=${file##lib/};
pattern=lib.${pattern%%/*};
sed -i -e 's/RTE_LOG_REGISTER(\(.*\), '$pattern',/RTE_LOG_REGISTER_DEFAULT(\1,/' $file;
sed -i -e 's/RTE_LOG_REGISTER(\(.*\), '$pattern'\.\(.*\),/RTE_LOG_REGISTER_SUFFIX(\1, \2,/' $file;
done
Signed-off-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
When ena_tx_queue_release_bufs() frees the mbufs it does not clear
the mbuf pointers. So, when the device starts and stops multiple
times it can cause the application to receive duplicate mbufs for
two different packets. Fix the issue by clearing the mbuf pointer.
Also, while tracking down the "double free" issue the ena calls to
allocate and free mbufs in bulk were migrated to the mbuf based APIs
so the common mbuf alloc/free routines are exercised.
Fixes: 79405ee175 ("net/ena: fix out of order completion")
Fixes: 1173fca25a ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: David Harton <dharton@cisco.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>
Replace copies of size(arr)/size(arr[0]) with RTE_DIM().
Eventually all of these macro definitions should be removed.
Signed-off-by: Andrew Boyer <aboyer@pensando.io>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Add per-tx-ring flag for packets that were pushed to HW but await
doorbell. That is to prevent a situation when a doorbell is sent due to
reaching Tx burst threshold and next send fails (e.g., due to queue
full). In such case we shouldn't send another doorbell because there are
no actual packets waiting for transmission.
Fixes: c7519ea5eb ("net/ena: call additional doorbells if needed")
Cc: stable@dpdk.org
Signed-off-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Before starting transmission of Tx burst, the driver checked the
available space in the sq and limited the number of packets for
transmission accordingly.
The calculation was incorrect for fragmented packets and potentially had
significantly limited the length of Tx bursts.
This patch removes the assessment and pushes packets to the sq as long
as the burst is not exhausted and space is available in the sq.
Correct evaluation of the required space isn't possible before the burst
because it depends on the number of segments of each packet.
This patch adds per-packet space evaluation for each packet before
attempting to process it. In case there is not enough queue space, the
burst will just stop without error.
Fixes: 2061fe41f2 ("net/ena: linearize Tx mbuf")
Cc: stable@dpdk.org
Signed-off-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Instead of verifying the Rx descriptor each time it's being used in the
driver code, now the verification happens on the HAL side.
This simplifies code a lot as instead of doing 2 validations, only
single one is needed. The driver have to check the rc value returned
by the ena_com upon reading the Rx descriptor and trigger the reset
if needed. It was previously the responsibility of the
validate_rx_req_id() function.
As part of the change, the version of the driver was bumped to v2.2.1.
Fixes: 2061fe41f2 ("net/ena: linearize Tx mbuf")
Cc: stable@dpdk.org
Signed-off-by: Ido Segev <idose@amazon.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Increment Tx doorbell statistics on tx_pkt_burst
after writing to doorbell and in case max burst size achieved
Fixes: c7519ea5eb ("net/ena: call additional doorbells if needed")
Cc: stable@dpdk.org
Signed-off-by: Amit Bernstein <amitbern@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
As the refill called as part of ena_start(), we end up the refill
progress with stuck buffers at the caller core cache.
Calling to flush the cache results with invalidate this cache and free
those stuck buffers.
Fixes: 1173fca25a ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: Ido Segev <idose@amazon.com>
Reviewed-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
The rte_ethdev_driver.h, rte_ethdev_vdev.h and rte_ethdev_pci.h files are
for drivers only and should be a private to DPDK and not installed.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Steven Webster <steven.webster@windriver.com>
Allows i40e and mlx5 PMDs to compile on Windows and disable other drivers.
Disable few i40e warnings with Clang such as comparison of integers of
different signs and macro redefinitions.
Signed-off-by: Pallavi Kadam <pallavi.kadam@intel.com>
Reviewed-by: Ranjit Menon <ranjit.menon@intel.com>
Acked-by: Tal Shnaiderman <talshn@nvidia.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
The v2.2.0 adds support for network interface metrics, includes some bug
fixes and updates HAL to the latest version.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Latest generation HW requires IO completion queue descriptors to be
aligned to a 4K in order to achieve the best performance.
Because of that, the new allocation macros were added, which allows
driver to allocate the memory with specified alignment.
The previous allocation macros are now wrappers around the macros
doing the alignment, with the alignment value equal to cacheline size.
Fixes: b68309be44 ("net/ena/base: update communication layer for the ENAv2")
Cc: stable@dpdk.org
Signed-off-by: Ido Segev <idose@amazon.com>
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Amit Bernstein <amitbern@amazon.com>
The ID 0xEC21 is not associated with LLQ feature of the device, so it
would be misleading for the user. Because of that, the current
identifier is more precise.
Together with code update, the documentation was changed to reflect
current changes
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
The driver was never setting PKT_RX_*_CKSUM_GOOD flags, so the only way
of checking if the checksum was checked was by testing for the
PKT_RX_*_CKSUM_BAD. In that situation, the application couldn't detect
if the checksum was valid or unknown, as unknown flag is equal to 0.
Moreover, the l3_csum_err value is only valid if the l3_proto is
indicating IPv4, so it shouldn't be checked for other protocols.
Fixes: 1173fca25a ("ena: add polling-mode driver")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
There was a bug in a code, which was reading stat_offset value from the
ena_stats_rx_strings array instead of ena_stats_global_strings.
It wasn't causing real problems just because ena_stats_rx_strings was
not smaller than ena_stats_global_strings and both arrays hold the same
offsets.
Fixes: 7830e905b7 ("net/ena: expose extended stats")
Cc: stable@dpdk.org
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
This assert macro is not called anymore.
This also fixes an invalid reference to RTE_LOGTYPE_ERR that does not
exist.
Fixes: 3adcba9a89 ("net/ena: update HAL to the newer version")
Fixes: 6f1c9df9e9 ("net/ena: use dynamic log type for debug logging")
Cc: stable@dpdk.org
Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Michal Krawczyk <mk@semihalf.com>
Since each version map file is contained in the subdirectory of the library
it refers to, there is no need to include the library name in the filename.
This makes things simpler in case of library renaming.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Luca Boccassi <bluca@debian.org>
Acked-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Acked-by: Rosen Xu <rosen.xu@intel.com>
Queue stats are stored in 'struct rte_eth_stats' as array and array size
is defined by 'RTE_ETHDEV_QUEUE_STAT_CNTRS' compile time flag.
As a result of technical board discussion, decided to remove the queue
statistics from 'struct rte_eth_stats' in the long term.
Instead PMDs should represent the queue statistics via xstats, this
gives more flexibility on the number of the queues supported.
Currently queue stats in the xstats are filled by ethdev layer, using
some basic stats, when queue stats removed from basic stats the
responsibility to fill the relevant xstats will be pushed to the PMDs.
During the switch period, temporary 'RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS'
device flag is created. Initially all PMDs using xstats set this flag.
The PMDs implemented queue stats in the xstats should clear the flag.
When all PMDs switch to the xstats for the queue stats, queue stats
related fields from 'struct rte_eth_stats' will be removed, as well as
'RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS' flag.
Later 'RTE_ETHDEV_QUEUE_STAT_CNTRS' compile time flag also can be
removed.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Xiao Wang <xiao.w.wang@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
Change eth_dev_stop_t return value from void to int.
Make eth_dev_stop_t implementations across all drivers to return
negative errno values if case of error conditions.
Signed-off-by: Ivan Ilchenko <ivan.ilchenko@oktetlabs.ru>
Signed-off-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
The function rte_eth_dev_release_port() is partially resetting
the struct rte_eth_dev. The drivers were completing this reset
with more pointers set to NULL in the close or remove operations.
More pointers are reset at ethdev level,
and some redundant assignments are removed from PMDs.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
When closing a port, it is supposed to be already stopped,
and marked as such with "dev_started" state zeroed by the stop API.
Resetting "dev_started" before calling the driver close operation
was hiding the case of not properly stopped port being closed.
The flag "dev_started" is not changed anymore in "rte_eth_dev_close()".
In case the "dev_stop" function is called from "dev_close",
bypassing "rte_eth_dev_stop()" API,
the "dev_started" state must be explicitly reset in the PMD
in order to keep the same behaviour.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
The secondary processes are not allowed to release shared resources.
Only process-private resources should be freed in a secondary process.
Most of the time, there is no process-private resource,
so the close operation is just forbidden in a secondary process.
After adding proper check in the port close functions,
some redundant checks in the device remove functions are dropped.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Reviewed-by: Sachin Saxena <sachin.saxena@oss.nxp.com>
Reviewed-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Liron Himi <lironh@marvell.com>
Reviewed-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
The temporary flag RTE_ETH_DEV_CLOSE_REMOVE is removed.
It was introduced in DPDK 18.11 in order to give time for PMDs to migrate.
The old behaviour was to free only queues when closing a port.
The new behaviour is calling rte_eth_dev_release_port() which does
three more tasks:
- trigger event callback
- reset state and few pointers
- free all generic port resources
The private port resources must be released in the .dev_close callback.
The .remove callback should:
- call .dev_close callback
- call rte_eth_dev_release_port()
- free multi-port device shared resources
Despite waiting two years, some drivers have not migrated,
so they may hit issues with the incompatible new behaviour.
After sending emails, adding logs, and announcing the deprecation,
the only last solution is to declare these drivers as unmaintained:
ionic, liquidio, nfp
Below is a summary of what to implement in those drivers.
* The freeing of private port resources must be moved
from the ".remove(device)" function to the ".dev_close(port)" function.
* If a generic resource (.mac_addrs or .hash_mac_addrs) cannot be freed,
it must be set to NULL in ".dev_close" function to protect from
subsequent rte_eth_dev_release_port() freeing.
* Note 1:
The generic resources are freed in rte_eth_dev_release_port(),
after ".dev_close" is called in rte_eth_dev_close(), but not when
calling ".dev_close" directly from the ".remove" PMD function.
That's why rte_eth_dev_release_port() must still be called explicitly
from ".remove(device)" after calling the ".dev_close" PMD function.
* Note 2:
If a device can have multiple ports, the common resources must be freed
only in the ".remove(device)" function.
* Note 3:
The port is supposed to be in a stopped state when it is closed.
If it is not the case, it is free to the PMD implementation
how to react when trying to close a non-stopped port:
either try to stop it automatically or just return an error.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Liron Himi <lironh@marvell.com>
Reviewed-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Acked-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
The device operation .dev_close was returning void.
This driver interface is changed to return an int.
Note that the API rte_eth_dev_close() is still returning void,
although a deprecation notice is pending to change it as well.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Rosen Xu <rosen.xu@intel.com>
Reviewed-by: Sachin Saxena <sachin.saxena@oss.nxp.com>
Reviewed-by: Liron Himi <lironh@marvell.com>
Reviewed-by: Haiyue Wang <haiyue.wang@intel.com>
Acked-by: Jeff Guo <jia.guo@intel.com>
Reviewed-by: Andrew Rybchenko <arybchenko@solarflare.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
New HAL allows driver to read extra ENI stats. Exact meaning of each of
them can be found in base/ena_defs/ena_admin_defs.h file and structure
ena_admin_eni_stats.
The ena_eni_stats structure is exactly the same as ena_admin_eni_stats,
but it was required to be added for compatibility with xstats macros.
Reading ENI stats requires communication with the admin queue.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
There are some cases, where the admin queue commands after the
configuration phase finished - for example, the application could ask
for the driver statistics from multiple cores at once.
As by the design, the admin queue is not multithread safe, the spinlock
was added to protect all usages of the admin queue after the
configuration is done.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
The current ena_com version was generated on 26.04.2020.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Checking for the cdesc not being NULL doesn't have any sense if the idx
argument is not 0, so it can be skipped, as the error won't be detected
anyway.
To simplify that, only the 'i' value is being verified and the code is
breaking from the infinite loop in case when all descriptors were copied
into the buffer.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
When filling out meta descriptor, all values should be converted to the
desired type (u32 in case of the meta descriptor) to prevent losing the
data.
For example, io_sq->phase is of type u8. If
ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT would be greater then 8, all data
would be lost.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Static code analysis showed up, that it's possible for meta_desc being
NULL. To avoid dereference of the NULL pointer, extra check was added if
the pointer is in fact valid.
Signed-off-by: Michal Krawczyk <mk@semihalf.com>
Reviewed-by: Igor Chauskin <igorch@amazon.com>
Reviewed-by: Guy Tzalik <gtzalik@amazon.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
