Create separate functions for the implicit ODP initialization
which is different from the explicit ODP initialization.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reuse newly added DMA API to cache IOVA and only link/unlink pages
in fast path for UMEM ODP flow.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
As a preparation to remove dma_list, store access mask in PFN pointer
and not in dma_addr_t.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
HMM callers use PFN list to populate range while calling
to hmm_range_fault(), the conversion from PFN to DMA address
is done by the callers with help of another DMA list. However,
it is wasteful on any modern platform and by doing the right
logic, that DMA list can be avoided.
Provide generic logic to manage these lists and gave an interface
to map/unmap PFNs to DMA addresses, without requiring from the callers
to be an experts in DMA core API.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Introduce new sticky flag (HMM_PFN_DMA_MAPPED), which isn't overwritten
by HMM range fault. Such flag allows users to tag specific PFNs with
information if this specific PFN was already DMA mapped.
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Currently the only efficient way to map a complex memory description through
the DMA API is by using the scatter list APIs. The SG APIs are unique in that
they efficiently combine the two fundamental operations of sizing and allocating
a large IOVA window from the IOMMU and processing all the per-address
swiotlb/flushing/p2p/map details.
This uniqueness has been a long standing pain point as the scatter list API
is mandatory, but expensive to use. It prevents any kind of optimization or
feature improvement (such as avoiding struct page for P2P) due to the
impossibility of improving the scatter list.
Several approaches have been explored to expand the DMA API with additional
scatterlist-like structures (BIO, rlist), instead split up the DMA API
to allow callers to bring their own data structure.
The API is split up into parts:
- Allocate IOVA space:
To do any pre-allocation required. This is done based on the caller
supplying some details about how much IOMMU address space it would need
in worst case.
- Map and unmap relevant structures to pre-allocated IOVA space:
Perform the actual mapping into the pre-allocated IOVA. This is very
similar to dma_map_page().
Thanks
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add helper that allows a driver to skip calling dma_unmap_*
if the DMA layer can guarantee that they are no-nops.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Introduce new DMA APIs to perform DMA linkage of buffers
in layers higher than DMA.
In proposed API, the callers will perform the following steps.
In map path:
if (dma_can_use_iova(...))
dma_iova_alloc()
for (page in range)
dma_iova_link_next(...)
dma_iova_sync(...)
else
/* Fallback to legacy map pages */
for (all pages)
dma_map_page(...)
In unmap path:
if (dma_can_use_iova(...))
dma_iova_destroy()
else
for (all pages)
dma_unmap_page(...)
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Split the iommu logic from iommu_dma_map_page into a separate helper.
This not only keeps the code neatly separated, but will also allow for
reuse in another caller.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
The existing .map_pages() callback provides both allocating of IOVA
and linking DMA pages. That combination works great for most of the
callers who use it in control paths, but is less effective in fast
paths where there may be multiple calls to map_page().
These advanced callers already manage their data in some sort of
database and can perform IOVA allocation in advance, leaving range
linkage operation to be in fast path.
Provide an interface to allocate/deallocate IOVA and next patch
link/unlink DMA ranges to that specific IOVA.
In the new API a DMA mapping transaction is identified by a
struct dma_iova_state, which holds some recomputed information
for the transaction which does not change for each page being
mapped, so add a check if IOVA can be used for the specific
transaction.
The API is exported from dma-iommu as it is the only implementation
supported, the namespace is clearly different from iommu_* functions
which are not allowed to be used. This code layout allows us to save
function call per API call used in datapath as well as a lot of boilerplate
code.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
For the upcoming IOVA-based DMA API we want to batch the
ops->iotlb_sync_map() call after mapping multiple IOVAs from
dma-iommu without having a scatterlist. Improve the API.
Add a wrapper for the map_sync as iommu_sync_map() so that callers
don't need to poke into the methods directly.
Formalize __iommu_map() into iommu_map_nosync() which requires the
caller to call iommu_sync_map() after all maps are completed.
Refactor the existing sanity checks from all the different layers
into iommu_map_nosync().
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Will Deacon <will@kernel.org>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
To support the upcoming non-scatterlist mapping helpers, we need to go
back to have them called outside of the DMA API. Thus move them out of
dma-map-ops.h, which is only for DMA API implementations to pci-p2pdma.h,
which is for driver use.
Note that the core helper is still not exported as the mapping is
expected to be done only by very highlevel subsystem code at least for
now.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
The current scheme with a single helper to determine the P2P status
and map a scatterlist segment force users to always use the map_sg
helper to DMA map, which we're trying to get away from because they
are very cache inefficient.
Refactor the code so that there is a single helper that checks the P2P
state for a page, including the result that it is not a P2P page to
simplify the callers, and a second one to perform the address translation
for a bus mapped P2P transfer that does not depend on the scatterlist
structure.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Tested-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
In preparation for making the kmalloc family of allocators type aware,
we need to make sure that the returned type from the allocation matches
the type of the variable being assigned. (Before, the allocator would
always return "void *", which can be implicitly cast to any pointer type.)
The assigned type is "struct tid_rb_node **", but the return type will be
"struct rb_node **". These are the same allocation size (pointer size),
but the types do not match. Adjust the allocation type to match the
assignment.
Signed-off-by: Kees Cook <kees@kernel.org>
Link: https://patch.msgid.link/20250426061247.work.261-kees@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
In preparation for making the kmalloc family of allocators type aware,
we need to make sure that the returned type from the allocation matches
the type of the variable being assigned. (Before, the allocator would
always return "void *", which can be implicitly cast to any pointer type.)
The assigned type is "unsigned long **", but the returned type will be
"long **". These are the same allocation size (pointer size), but the
types do not match. Adjust the allocation type to match the assignment.
Signed-off-by: Kees Cook <kees@kernel.org>
Link: https://patch.msgid.link/20250426061208.work.000-kees@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
hns_roce_hw_v2.h has a direct dependency on hnae3.h due to the
inline function hns_roce_write64(), but it doesn't include this
header currently. This leads to that files including
hns_roce_hw_v2.h must also include hnae3.h to avoid compilation
errors, even if they themselves don't really rely on hnae3.h.
This doesn't make sense, hns_roce_hw_v2.h should include hnae3.h
directly.
Fixes: d3743fa94c ("RDMA/hns: Fix the chip hanging caused by sending doorbell during reset")
Signed-off-by: Junxian Huang <huangjunxian6@hisilicon.com>
Link: https://patch.msgid.link/20250421132750.1363348-6-huangjunxian6@hisilicon.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:94 [inline]
dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120
assign_lock_key kernel/locking/lockdep.c:986 [inline]
register_lock_class+0x4a3/0x4c0 kernel/locking/lockdep.c:1300
__lock_acquire+0x99/0x1ba0 kernel/locking/lockdep.c:5110
lock_acquire kernel/locking/lockdep.c:5866 [inline]
lock_acquire+0x179/0x350 kernel/locking/lockdep.c:5823
__timer_delete_sync+0x152/0x1b0 kernel/time/timer.c:1644
rxe_qp_do_cleanup+0x5c3/0x7e0 drivers/infiniband/sw/rxe/rxe_qp.c:815
execute_in_process_context+0x3a/0x160 kernel/workqueue.c:4596
__rxe_cleanup+0x267/0x3c0 drivers/infiniband/sw/rxe/rxe_pool.c:232
rxe_create_qp+0x3f7/0x5f0 drivers/infiniband/sw/rxe/rxe_verbs.c:604
create_qp+0x62d/0xa80 drivers/infiniband/core/verbs.c:1250
ib_create_qp_kernel+0x9f/0x310 drivers/infiniband/core/verbs.c:1361
ib_create_qp include/rdma/ib_verbs.h:3803 [inline]
rdma_create_qp+0x10c/0x340 drivers/infiniband/core/cma.c:1144
rds_ib_setup_qp+0xc86/0x19a0 net/rds/ib_cm.c:600
rds_ib_cm_initiate_connect+0x1e8/0x3d0 net/rds/ib_cm.c:944
rds_rdma_cm_event_handler_cmn+0x61f/0x8c0 net/rds/rdma_transport.c:109
cma_cm_event_handler+0x94/0x300 drivers/infiniband/core/cma.c:2184
cma_work_handler+0x15b/0x230 drivers/infiniband/core/cma.c:3042
process_one_work+0x9cc/0x1b70 kernel/workqueue.c:3238
process_scheduled_works kernel/workqueue.c:3319 [inline]
worker_thread+0x6c8/0xf10 kernel/workqueue.c:3400
kthread+0x3c2/0x780 kernel/kthread.c:464
ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:153
ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
</TASK>
The root cause is as below:
In the function rxe_create_qp, the function rxe_qp_from_init is called
to create qp, if this function rxe_qp_from_init fails, rxe_cleanup will
be called to handle all the allocated resources, including the timers:
retrans_timer and rnr_nak_timer.
The function rxe_qp_from_init calls the function rxe_qp_init_req to
initialize the timers: retrans_timer and rnr_nak_timer.
But these timers are initialized in the end of rxe_qp_init_req.
If some errors occur before the initialization of these timers, this
problem will occur.
The solution is to check whether these timers are initialized or not.
If these timers are not initialized, ignore these timers.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Reported-by: syzbot+4edb496c3cad6e953a31@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=4edb496c3cad6e953a31
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Link: https://patch.msgid.link/20250419080741.1515231-1-yanjun.zhu@linux.dev
Signed-off-by: Leon Romanovsky <leon@kernel.org>
In the past %pK was preferable to %p as it would not leak raw pointer
values into the kernel log.
Since commit ad67b74d24 ("printk: hash addresses printed with %p")
the regular %p has been improved to avoid this issue.
Furthermore, restricted pointers ("%pK") were never meant to be used
through printk(). They can still unintentionally leak raw pointers or
acquire sleeping looks in atomic contexts.
Switch to the regular pointer formatting which is safer and
easier to reason about.
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Link: https://patch.msgid.link/20250407-restricted-pointers-infiniband-v1-1-22b20504b84d@linutronix.de
Signed-off-by: Leon Romanovsky <leon@kernel.org>
In workloads where there are many processes establishing connections using
RDMA CM in parallel (large scale MPI), there can be heavy contention for
mad_agent_lock in cm_alloc_msg.
This contention can occur while inside of a spin_lock_irq region, leading
to interrupts being disabled for extended durations on many
cores. Furthermore, it leads to the serialization of rdma_create_ah calls,
which has negative performance impacts for NICs which are capable of
processing multiple address handle creations in parallel.
The end result is the machine becoming unresponsive, hung task warnings,
netdev TX timeouts, etc.
Since the lock appears to be only for protection from cm_remove_one, it
can be changed to a rwlock to resolve these issues.
Reproducer:
Server:
for i in $(seq 1 512); do
ucmatose -c 32 -p $((i + 5000)) &
done
Client:
for i in $(seq 1 512); do
ucmatose -c 32 -p $((i + 5000)) -s 10.2.0.52 &
done
Fixes: 76039ac909 ("IB/cm: Protect cm_dev, cm_ports and mad_agent with kref and lock")
Link: https://patch.msgid.link/r/20250220175612.2763122-1-jmoroni@google.com
Signed-off-by: Jacob Moroni <jmoroni@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
-Wflex-array-member-not-at-end was introduced in GCC-14, and we are
getting ready to enable it, globally.
Remove unused flex-array member `class_data` from
`struct opa_mad_notice_attr`.
Fix the following warning:
drivers/infiniband/hw/hfi1/mad.c:23:36: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
Link: https://patch.msgid.link/r/Z-wiYkll8Vo3ME3P@kspp
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
