Max Gurtovoy [Tue, 8 Mar 2022 14:55:44 +0000 (16:55 +0200)]
IB/iser: Use iser_fr_desc as registration context
After removing the FMR support in iSER, there is only one type of
registration context. Replace the void pointer with the explicit structure
for registration (struct iser_fr_desc).
Bob Pearson [Fri, 4 Mar 2022 00:08:04 +0000 (18:08 -0600)]
RDMA/rxe: Replace red-black trees by xarrays
Currently the rxe driver uses red-black trees to add indices to the rxe
object pools. Linux xarrays provide a better way to implement the same
functionality for indices. This patch replaces red-black trees by xarrays
for pool objects. Since xarrays already have a spinlock use that in place
of the pool rwlock. Make sure that all changes in the xarray(index) and
kref(ref counnt) occur atomically.
Bob Pearson [Fri, 4 Mar 2022 00:07:59 +0000 (18:07 -0600)]
RDMA/rxe: Reverse the sense of RXE_POOL_NO_ALLOC
There is only one remaining object type that allocates its own memory,
that is mr. So the sense of RXE_POOL_NO_ALLOC is changed to
RXE_POOL_ALLOC. Add checks to rxe_alloc() and rxe_add_to_pool() to make
sure the correct call is used for the setting of this flag.
Bob Pearson [Fri, 4 Mar 2022 00:07:58 +0000 (18:07 -0600)]
RDMA/rxe: Replace mr by rkey in responder resources
Currently rxe saves a copy of MR in responder resources for RDMA reads.
Since the responder resources are never freed just over written if more
are needed this MR may not have a reference freed until the QP is
destroyed. This patch uses the rkey instead of the MR and on subsequent
packets of a multipacket read reply message it looks up the MR from the
rkey for each packet. This makes it possible for a user to deregister an
MR or unbind a MW on the fly and get correct behaviour.
Bob Pearson [Fri, 4 Mar 2022 00:07:57 +0000 (18:07 -0600)]
RDMA/rxe: Fix ref error in rxe_av.c
The commit referenced below can take a reference to the AH which is never
dropped. This only happens in the UD request path. This patch optionally
passes that AH back to the caller so that it can hold the reference while
the AV is being accessed and then drop it. Code to do this is added to
rxe_req.c. The AV is also passed to rxe_prepare in rxe_net.c as an
optimization.
Yixing Liu [Thu, 10 Mar 2022 04:28:35 +0000 (12:28 +0800)]
RDMA/hns: Use the reserved loopback QPs to free MR before destroying MPT
Before destroying MPT, the reserved loopback QPs send loopback IOs (one
write operation per SL). Completing these loopback IOs represents that
there isn't any outstanding request in MPT, then it's safe to destroy MPT.
Mustafa Ismail [Mon, 28 Feb 2022 18:36:50 +0000 (12:36 -0600)]
RDMA/irdma: Add support for address handle re-use
Address handles (AH) are a limited HW resource and some user applications
may create large numbers of identical AH's. Avoid running out of AH's by
reusing existing identical ones.
Dan Carpenter [Mon, 7 Mar 2022 12:59:28 +0000 (15:59 +0300)]
RDMA/irdma: Prevent some integer underflows
My static checker complains that:
drivers/infiniband/hw/irdma/ctrl.c:3605 irdma_sc_ceq_init()
warn: can subtract underflow 'info->dev->hmc_fpm_misc.max_ceqs'?
It appears that "info->dev->hmc_fpm_misc.max_ceqs" comes from the firmware
in irdma_sc_parse_fpm_query_buf() so, yes, there is a chance that it could
be zero. Even if we trust the firmware, it's easy enough to change the
condition just as a hardenning measure.
Wenpeng Liang [Wed, 2 Mar 2022 06:48:28 +0000 (14:48 +0800)]
RDMA/hns: Clean up the return value check of hns_roce_alloc_cmd_mailbox()
hns_roce_alloc_cmd_mailbox() never returns NULL, so the check should be
IS_ERR(). And the error code should be converted as the function's return
value.
Chengchang Tang [Wed, 2 Mar 2022 06:48:26 +0000 (14:48 +0800)]
RDMA/hns: Refactor mailbox functions
The current mailbox functions have too many parameters, making the code
difficult to maintain. So construct a new structure mbox_msg to pass the
information needed by mailbox.
Link: https://lore.kernel.org/r/20220302064830.61706-6-liangwenpeng@huawei.com Signed-off-by: Chengchang Tang <tangchengchang@huawei.com> Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Chengchang Tang [Wed, 2 Mar 2022 06:48:23 +0000 (14:48 +0800)]
RDMA/hns: Remove fixed parameter “timeout” in the mailbox
The value of the function parameter “timeout” is unique. Therefore,
it is unnecessary to specify the parameter “timeout” value each time.
So remove it.
Chengchang Tang [Wed, 2 Mar 2022 06:48:22 +0000 (14:48 +0800)]
RDMA/hns: Remove the unused parameter "op_modifier" in mailbox
The parameter "op_modifier" is only used for HIP06. It is useless for HIP08
and later versions. After removing HIP06, this parameter is no longer used,
so remove it.
Mustafa Ismail [Fri, 25 Feb 2022 16:32:11 +0000 (10:32 -0600)]
RDMA/irdma: Remove incorrect masking of PD
The PD id is masked with 0x7fff, while PD can be 18 bits for GEN2 HW.
Remove the masking as it should not be needed and can cause incorrect PD
id to be used.
Mustafa Ismail [Fri, 25 Feb 2022 16:32:10 +0000 (10:32 -0600)]
RDMA/irdma: Fix Passthrough mode in VM
Using PCI_FUNC macro in a VM, when the device is in passthrough mode does
not provide the real function instance. This means that currently, devices
will not probe unless the instance in the VM matches the instance in the
host.
Fix this by getting the pf_id from the LAN during the probe.
Fixes: 8498a30e1b94 ("RDMA/irdma: Register auxiliary driver and implement private channel OPs") Link: https://lore.kernel.org/r/20220225163211.127-3-shiraz.saleem@intel.com Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Bob Pearson [Wed, 23 Feb 2022 23:07:04 +0000 (17:07 -0600)]
RDMA/rxe: Collect mca init code in a subroutine
Collect initialization code for struct rxe_mca into a subroutine,
__rxe_init_mca(), to cleanup rxe_attach_mcg() in rxe_mcast.c. Check
limit on total number of attached qp's.
Aharon Landau [Tue, 15 Feb 2022 17:55:33 +0000 (19:55 +0200)]
RDMA/mlx5: Reorder calls to pcie_relaxed_ordering_enabled()
The mkc is the key for the mkey cache, hence, created in each attempt to
get a cache mkey, while pcie_relaxed_ordering_enabled() is called during
the setting of the mkc, but used only for cases where
IB_ACCESS_RELAXED_ORDERING is set.
pcie_relaxed_ordering_enabled() is an expensive call (26 us). Reorder the
code so the driver will call it only when it is needed.
Aharon Landau [Tue, 15 Feb 2022 17:55:32 +0000 (19:55 +0200)]
RDMA/mlx5: Store ndescs instead of the translation table size
Currently, ent->xlt stores the translation table size. This data should
not be stored in the cache entry but be written directly to the mailbox.
Store ndescs instead, and deduce the translation table size from it
according to the access mode.
Aharon Landau [Tue, 15 Feb 2022 17:55:31 +0000 (19:55 +0200)]
RDMA/mlx5: Merge similar flows of allocating MR from the cache
When allocating a MR from the cache, the driver calls to get_cache_mr(),
and in case of failure, retries with create_cache_mr(). This is the flow
of mlx5_mr_cache_alloc(), so use it instead.
Aharon Landau [Tue, 15 Feb 2022 17:55:30 +0000 (19:55 +0200)]
RDMA/mlx5: Fix the flow of a miss in the allocation of a cache ODP MR
When an ODP MR cache entry is empty and trying to allocate it, increment
the ent->miss counter and call to queue_adjust_cache_locked() to verify
the entry is balanced.
Aharon Landau [Tue, 15 Feb 2022 17:55:29 +0000 (19:55 +0200)]
RDMA/mlx5: Remove redundant work in struct mlx5_cache_ent
delayed_cache_work_func() and the cache_work_func() are both wrappers of
__cache_work_func(). Instead of having a special not delayed work, use the
delayed work with delay = 0.
Bob Pearson [Tue, 8 Feb 2022 21:16:40 +0000 (15:16 -0600)]
RDMA/rxe: Remove key'ed object support
Now that rxe_mcast.c has it's own red-black tree support there is no
longer any requirement for key'ed objects in rxe pools. This patch removes
the key APIs and related code.
Bob Pearson [Tue, 8 Feb 2022 21:16:39 +0000 (15:16 -0600)]
RDMA/rxe: Replace pool key by rxe->mcg_tree
Continuing to decouple mcg from rxe pools. Create red-black tree code in
rxe_mcast.c to hold mcg index. Replace pool key calls by calls to local
red-black routines.
Bob Pearson [Tue, 8 Feb 2022 21:16:36 +0000 (15:16 -0600)]
RDMA/rxe: Use kzmalloc/kfree for mca
Remove rxe_mca (was rxe_mc_elem) from rxe pools and use kzmalloc and kfree
to allocate and free in rxe_mcast.c. Call kzalloc outside of spinlocks to
avoid having to use GFP_ATOMIC.
Bob Pearson [Tue, 8 Feb 2022 21:16:35 +0000 (15:16 -0600)]
RDMA/rxe: Move mcg_lock to rxe
Replace mcg->mcg_lock and mc_grp_pool->pool_lock by rxe->mcg_lock. This
is the first step of several intended to decouple the mc_grp and mc_elem
objects from the rxe pool code.
Bob Pearson [Tue, 15 Feb 2022 19:44:49 +0000 (13:44 -0600)]
RDMA/rxe: Revert changes from irqsave to bh locks
A previous patch replaced all irqsave locks in rxe with bh locks. This
ran into problems because rdmacm has a bad habit of calling rdma verbs
APIs while disabling irqs. This is not allowed during spin_unlock_bh()
causing programs that use rdmacm to fail. This patch reverts the changes
to locks that had this problem or got dragged into the same mess. After
this patch blktests/check -q srp now runs correctly.
Link: https://lore.kernel.org/r/20220215194448.44369-1-rpearsonhpe@gmail.com Fixes: 21adfa7a3c4e ("RDMA/rxe: Replace irqsave locks with bh locks") Reported-by: Guoqing Jiang <guoqing.jiang@linux.dev> Reported-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Tested-by: Bart Van Assche <bvanassche@acm.org> Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Yury Norov [Thu, 10 Feb 2022 22:49:03 +0000 (14:49 -0800)]
RDMA/hfi: Replace cpumask_weight with cpumask_empty where appropriate
drivers/infiniband/hw/hfi1/affinity.c code calls cpumask_weight() to check
if any bit of a given cpumask is set. We can do it more efficiently with
cpumask_empty() because cpumask_empty() stops traversing the cpumask as
soon as it finds first set bit, while cpumask_weight() counts all bits
unconditionally.
Mustafa Ismail [Wed, 2 Feb 2022 19:19:21 +0000 (13:19 -0600)]
RDMA/irdma: Add support for DSCP
Add DSCP support for the Intel Ethernet 800 Series devices. Setup VSI
DSCP info when PCI driver indicates DSCP mode during driver probe or as
notification event.
Mustafa Ismail [Wed, 2 Feb 2022 19:19:20 +0000 (13:19 -0600)]
RDMA/irdma: Refactor DCB bits in prep for DSCP support
Rename dcb flag to dcb_vlan_mode in irdma_device struct. Add a new helper
function, irdma_set_qos_info, to set the VSI QoS information passed by the
PCI driver.
Xiao Yang [Wed, 29 Dec 2021 03:44:38 +0000 (11:44 +0800)]
RDMA/rxe: Check the last packet by RXE_END_MASK
It's wrong to check the last packet by RXE_COMP_MASK because the flag is
to indicate if responder needs to generate a completion.
Fixes: 9fcd67d1772c ("IB/rxe: increment msn only when completing a request") Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20211229034438.1854908-1-yangx.jy@fujitsu.com Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Dave Ertman [Tue, 23 Nov 2021 18:25:36 +0000 (10:25 -0800)]
ice: add support for DSCP QoS for IDC
The ice driver provides QoS information to auxiliary drivers
through the exported function ice_get_qos_params. This function
doesn't currently support L3 DSCP QoS.
Add the necessary defines, structure elements and code to support
DSCP QoS through the IIDC functions.
Signed-off-by: Dave Ertman <david.m.ertman@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Bob Pearson [Thu, 27 Jan 2022 21:37:35 +0000 (15:37 -0600)]
RDMA/rxe: Remove qp->grp_lock and qp->grp_list
Since it is no longer required to cleanup attachments to multicast
groups when a QP is destroyed qp->grp_lock and qp->grp_list are
no longer needed and are removed.
Bob Pearson [Thu, 27 Jan 2022 21:37:32 +0000 (15:37 -0600)]
RDMA/rxe: Rename rxe_mc_grp and rxe_mc_elem
Rename rxe_mc_grp to rxe_mcg. Rename rxe_mc_elem to rxe_mca.
These can be read 'multicast group' and 'multicast attachment'.
'elem' collided with the use of elem in rxe pools and was a little
confusing.
Bob Pearson [Thu, 27 Jan 2022 21:37:31 +0000 (15:37 -0600)]
RDMA/rxe: Move rxe_mcast_attach/detach to rxe_mcast.c
Move rxe_mcast_attach and rxe_mcast_detach from rxe_verbs.c to rxe_mcast.c,
Make non-static and add declarations to rxe_loc.h. Make the subroutines
in rxe_mcast.c referenced by these routines static and remove their
declarations from rxe_loc.h.
Jack Wang [Fri, 14 Jan 2022 15:47:53 +0000 (16:47 +0100)]
RDMA/rtrs-clt: Do stop and failover outside reconnect work.
We can't do instant reconnect, not to DDoS server, but we should stop and
failover earlier, so there is less service interruption.
To avoid deadlock, as error_recovery is called from different callback
like rdma event or hb error handler, add a new err recovery_work.
Link: https://lore.kernel.org/r/20220114154753.983568-6-haris.iqbal@ionos.com Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Reviewed-by: Aleksei Marov <aleksei.marov@ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Linus Torvalds [Sun, 23 Jan 2022 06:14:21 +0000 (08:14 +0200)]
Merge tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull more perf tools updates from Arnaldo Carvalho de Melo:
- Fix printing 'phys_addr' in 'perf script'.
- Fix failure to add events with 'perf probe' in ppc64 due to not
removing leading dot (ppc64 ABIv1).
- Fix cpu_map__item() python binding building.
- Support event alias in form foo-bar-baz, add pmu-events and
parse-event tests for it.
- No need to setup affinities when starting a workload or attaching to
a pid.
- Use path__join() to compose a path instead of ad-hoc snprintf()
equivalent.
- Override attr->sample_period for non-libpfm4 events.
- Use libperf cpumap APIs instead of accessing the internal state
directly.
- Sync x86 arch prctl headers and files changed by the new
set_mempolicy_home_node syscall with the kernel sources.
- Remove duplicate include in cpumap.h.
- Remove redundant err variable.
* tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
perf tools: Remove redundant err variable
perf test: Add parse-events test for aliases with hyphens
perf test: Add pmu-events test for aliases with hyphens
perf parse-events: Support event alias in form foo-bar-baz
perf evsel: Override attr->sample_period for non-libpfm4 events
perf cpumap: Remove duplicate include in cpumap.h
perf cpumap: Migrate to libperf cpumap api
perf python: Fix cpu_map__item() building
perf script: Fix printing 'phys_addr' failure issue
tools headers UAPI: Sync files changed by new set_mempolicy_home_node syscall
tools headers UAPI: Sync x86 arch prctl headers with the kernel sources
perf machine: Use path__join() to compose a path instead of snprintf(dir, '/', filename)
perf evlist: No need to setup affinities when disabling events for pid targets
perf evlist: No need to setup affinities when enabling events for pid targets
perf stat: No need to setup affinities when starting a workload
perf affinity: Allow passing a NULL arg to affinity__cleanup()
perf probe: Fix ppc64 'perf probe add events failed' case
Linus Torvalds [Sun, 23 Jan 2022 06:07:02 +0000 (08:07 +0200)]
Merge tag 'trace-v5.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull ftrace fix from Steven Rostedt:
"Fix s390 breakage from sorting mcount tables.
The latest merge of the tracing tree sorts the mcount table at build
time. But s390 appears to do things differently (like always) and
replaces the sorted table back to the original unsorted one. As the
ftrace algorithm depends on it being sorted, bad things happen when it
is not, and s390 experienced those bad things.
Add a new config to tell the boot if the mcount table is sorted or
not, and allow s390 to opt out of it"
* tag 'trace-v5.17-3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Fix assuming build time sort works for s390
Steven Rostedt (Google) [Sat, 22 Jan 2022 14:17:10 +0000 (09:17 -0500)]
ftrace: Fix assuming build time sort works for s390
To speed up the boot process, as mcount_loc needs to be sorted for ftrace
to work properly, sorting it at build time is more efficient than boot up
and can save milliseconds of time. Unfortunately, this change broke s390
as it will modify the mcount_loc location after the sorting takes place
and will put back the unsorted locations. Since the sorting is skipped at
boot up if it is believed that it was sorted at run time, ftrace can crash
as its algorithms are dependent on the list being sorted.
Add a new config BUILDTIME_MCOUNT_SORT that is set when
BUILDTIME_TABLE_SORT but not if S390 is set. Use this config to determine
if sorting should take place at boot up.
Link: https://lore.kernel.org/all/yt9dee51ctfn.fsf@linux.ibm.com/ Fixes: 72b3942a173c ("scripts: ftrace - move the sort-processing in ftrace_init") Reported-by: Sven Schnelle <svens@linux.ibm.com> Tested-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Linus Torvalds [Sun, 23 Jan 2022 04:32:29 +0000 (06:32 +0200)]
Merge tag 'kbuild-fixes-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild
Pull Kbuild fixes from Masahiro Yamada:
- Bring include/uapi/linux/nfc.h into the UAPI compile-test coverage
- Revert the workaround of CONFIG_CC_IMPLICIT_FALLTHROUGH
- Fix build errors in certs/Makefile
* tag 'kbuild-fixes-v5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
certs: Fix build error when CONFIG_MODULE_SIG_KEY is empty
certs: Fix build error when CONFIG_MODULE_SIG_KEY is PKCS#11 URI
Revert "Makefile: Do not quote value for CONFIG_CC_IMPLICIT_FALLTHROUGH"
usr/include/Makefile: add linux/nfc.h to the compile-test coverage
Linus Torvalds [Sun, 23 Jan 2022 04:20:44 +0000 (06:20 +0200)]
Merge tag 'bitmap-5.17-rc1' of git://github.com/norov/linux
Pull bitmap updates from Yury Norov:
- introduce for_each_set_bitrange()
- use find_first_*_bit() instead of find_next_*_bit() where possible
- unify for_each_bit() macros
* tag 'bitmap-5.17-rc1' of git://github.com/norov/linux:
vsprintf: rework bitmap_list_string
lib: bitmap: add performance test for bitmap_print_to_pagebuf
bitmap: unify find_bit operations
mm/percpu: micro-optimize pcpu_is_populated()
Replace for_each_*_bit_from() with for_each_*_bit() where appropriate
find: micro-optimize for_each_{set,clear}_bit()
include/linux: move for_each_bit() macros from bitops.h to find.h
cpumask: replace cpumask_next_* with cpumask_first_* where appropriate
tools: sync tools/bitmap with mother linux
all: replace find_next{,_zero}_bit with find_first{,_zero}_bit where appropriate
cpumask: use find_first_and_bit()
lib: add find_first_and_bit()
arch: remove GENERIC_FIND_FIRST_BIT entirely
include: move find.h from asm_generic to linux
bitops: move find_bit_*_le functions from le.h to find.h
bitops: protect find_first_{,zero}_bit properly
projects / users / jedix / linux-maple.git / log