blk_rq_bytes does the wrong thing for special payloads like discards and
might cause the driver to not set up a SGL.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
blk_rq_bytes does the wrong thing for special payloads like discards and
might cause the driver to not set up a SGL.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
THIS_MODULE evaluates to NULL when used from code built into the kernel,
thus breaking built-in transport modules. Remove the bogus check.
Fixes: 0de5cd36 ("nvme-fabrics: protect against module unload during create_ctrl")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Keith Busch <keith.busch@intel.com>
We have expected busses to set up a coherent mask to properly use the
common dma mapping code for a long time, and now that I've added a warning
macio turned out to not set one up yet. This sets it to the same value
as the dma_mask, which seems to be what the drivers expect.
Reported-by: Mathieu Malaterre <malat@debian.org>
Tested-by: Mathieu Malaterre <malat@debian.org>
Reported-by: Meelis Roos <mroos@linux.ee>
Tested-by: Meelis Roos <mroos@linux.ee>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Various PCI_DMA_BUS_IS_PHYS implementations rely on this flag to make proper
decisions for block and networking addressability.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Setting values in struct sock directly is the usual method. Remove
the long dead code using set_fs() and the related comment.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
satp is the name used by the current privileged spec 1.10, use it
instead of the old name. The most recent release binutils release
(2.29) doesn't know about the satp name yet, so stick to the name from
the previous privileged ISA release and comment on the fact.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
init_mm.pgd (aka swapped_pgd) gets relocated like all other kernel
symbols by the elf loader, so there is no need to reload it from satp.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
This patch allows devices that require memory that can be addressed
using 32-bit addresses to work easily on RISC-V systems. The newly
improved dma-direct ops will tap into this pool automatically for
32-bit addressing.
Based on an earlier patch from Wesley W. Terpstra.
CC: Wesley W. Terpstra <terpstra@sifive.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
The SUM bit is enabled at the beginning of the copy_{to,from}_user and
{get,put}_user routines, and cleared before they return. But these user
copy helper can be interrupted by exceptions, in which case the SUM bit
will remain set, which leads to elevated privileges for the code running
in exception context, as that can now access userspace address space
unconditionally. This frequently happens when the user copy routines
access freshly allocated user memory that hasn't been faulted in, and a
pagefault needs to be taken before the user copy routines can continue.
Fix this by unconditionally clearing SUM when the exception handler is
called - the restore code will automatically restore it based on the
saved value.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
But reject reflink + DAX file systems for now until the code to
support reflinks on DAX is actually implemented.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
[darrick: port to 4.16]
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
All other discussions related to the dma mapping interfaces are on the
iommu list, so let's make it the official list for swiotlb and the
second list for xen-swiotlb.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Historically some ISA drivers used the old PCI DMA API with a NULL pdev
argument, but these days this isn't used and not too useful due to the
per-device DMA ops, so remove it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
We need to pass a struct device to the DMA API, even if some architectures
still support that for legacy reasons, and should not mix it with the old
PCI DMA API.
Note that the driver also seems to never actually unmap its DMA mappings,
but to fix that we'll need someone more familar with the driver.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Switch to a plain kzalloc() instead of pci_zalloc_coherent() to allocate
memory for the USB DMA.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Switch to a plain kzalloc() instead of pci_zalloc_coherent() to allocate
memory for the USB DMA.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Some iommu implementations can merge physically and/or virtually
contiguous segments inside sg_map_dma. The NVMe SGL support does not take
this into account and will warn because of falling off a loop. Pass the
number of mapped segments to nvme_pci_setup_sgls so that the SGL setup
can take the number of mapped segments into account.
Reported-by: Fangjian (Turing) <f.fangjian@huawei.com>
Fixes: a7a7cbe3 ("nvme-pci: add SGL support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@rimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Define the bit positions instead of macros using the magic values,
and move the expanded helpers to calculate the size and size unit into
the implementation C file.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Refactor the call to nvme_map_cmb, and change the conditions for probing
for the CMB. First remove the version check as NVMe TPs always apply
to earlier versions of the spec as well. Second check for the whole CMBSZ
register for support of the CMB feature instead of just the size field
inside of it to simplify the code a bit.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
The generic swiotlb_alloc and swiotlb_free routines already take care
of CMA allocations and adding GFP_DMA32 where needed, so use them
instead of the arm specific helpers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
arm64 uses ZONE_DMA for allocations below 32-bits. These days we
name the zone for that ZONE_DMA32, which will allow to use the
dma-direct and generic swiotlb code as-is, so rename it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
These already include the GFP_DMA/GFP_DMA32 usage, and will use CMA
memory if enabled, thus avoiding the GFP_NORETRY hack.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
nlm_swiotlb_dma_ops is unused code, so the whole swiotlb support is dead.
If it gets resurrected at some point it should use the generic
swiotlb_dma_ops instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
These are identical to the tile ops, and would also support CMA
if enabled on tile.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
tile uses ZONE_DMA for allocations below 32-bits. These days we
name the zone for that ZONE_DMA32, which will allow to use the
dma-direct and generic swiotlb code as-is, so rename it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
These are identical to the unicore32 ops, and would also support CMA
if enabled on unicore32.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
The file is only compiled if CONFIG_INTEL_IOMMU is set to start with.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Move the few remaining bits of swiotlb glue towards their callers,
and remove the pointless on ia64 swiotlb variable.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
These are identical to the ia64 ops, and would also support CMA
if enabled on ia64.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
ia64 uses ZONE_DMA for allocations below 32-bits. These days we
name the zone for that ZONE_DMA32, which will allow to use the
dma-direct and generic swiotlb code as-is, so rename it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
All these symbols are only used by arch dma_ops implementations or
xen-swiotlb. None of which can be modular.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Factor out a new swiotlb_alloc_buffer helper that allocates DMA coherent
memory from the swiotlb bounce buffer.
This allows to simplify the swiotlb_alloc implemenation that uses
dma_direct_alloc to try to allocate a reachable buffer first.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Factor out a new swiotlb_free_buffer helper that checks if an address
is allocated from the swiotlb bounce buffer, and if yes frees it.
This allows to simplify the swiotlb_free implemenation that uses
dma_direct_free to free the non-bounce buffer allocations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
To properly reject too small DMA masks based on the addressability of the
bounce buffer.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Currently all architectures that want to use swiotlb have to implement
their own dma_map_ops instances. Provide a generic one based on the
x86 implementation which first calls into dma_direct to try a full blown
direct mapping implementation (including e.g. CMA) before falling back
allocating from the swiotlb buffer.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christian König <christian.koenig@amd.com>
Replace the bare-bones h8300 direct dma mapping implementation with
the fully featured generic dma-direct one.
Signed-off-by: Christoph Hellwig <hch@lst.de>
cris currently has an incomplete direct mapping dma_map_ops implementation
if PCI support is enabled. Replace it with the fully feature generic
dma-direct implementation.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
So that they don't need to indirect through the operation vector.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
If an attempt to allocate memory succeeded, but isn't inside the
supported DMA mask, retry the allocation with GFP_DMA set as a
last resort.
Based on the x86 code, but an off by one error in what is now
dma_coherent_ok has been fixed vs the x86 code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
This allows to dip into zones for lower memory if they are available.
If one of the zones is not available the corresponding GFP_* flag
will evaluate to 0 so they won't change anything. We provide an
arch tunable for those architectures that do not use GFP_DMA for
the lowest 24-bits, given that there are a few.
Roughly based on the x86 code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
This means it uses whatever linear remapping scheme that the architecture
provides is used in the generic dma_direct ops.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
The trivial direct mapping implementation already does a virtual to
physical translation which isn't strictly a noop, and will soon learn
to do non-direct but linear physical to dma translations through the
device offset and a few small tricks. Rename it to a better fitting
name.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
For architectures that just use the generic dma_noop_ops we can provide
a generic version of dma-mapping.h.
Signed-off-by: Christoph Hellwig <hch@lst.de>
To implement the x86 forbid_dac and iommu_sac_force we want an arch hook
so that it can apply the global options across all dma_map_ops
implementations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Lift the code from x86 so that we behave consistently. In the future we
should probably warn if any of these is set.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k]
These days all devices should have a DMA coherent mask, and most dma_ops
implementations rely on that fact. But just to be sure add an assert to
ring the warning bell if that is not the case.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Vladimir Murzin <vladimir.murzin@arm.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
The previous patch assigns interrupt vectors to all possible CPUs, so
now hctx can be mapped to possible CPUs, this patch applies this fact
to simplify queue mapping & schedule so that we don't need to handle
CPU hotplug for dealing with physical CPU plug & unplug. With this
simplication, we can work well on physical CPU plug & unplug, which
is a normal use case for VM at least.
Make sure we allocate blk_mq_ctx structures for all possible CPUs, and
set hctx->numa_node for possible CPUs which are mapped to this hctx. And
only choose the online CPUs for schedule.
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Fixes: 4b855ad371 ("blk-mq: Create hctx for each present CPU")
(merged the three into one because any single one may not work, and fix
selecting online CPUs for scheduler)
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Currently we assign managed interrupt vectors to all present CPUs. This
works fine for systems were we only online/offline CPUs. But in case of
systems that support physical CPU hotplug (or the virtualized version of
it) this means the additional CPUs covered for in the ACPI tables or on
the command line are not catered for. To fix this we'd either need to
introduce new hotplug CPU states just for this case, or we can start
assining vectors to possible but not present CPUs.
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Tested-by: Stefan Haberland <sth@linux.vnet.ibm.com>
Fixes: 4b855ad371 ("blk-mq: Create hctx for each present CPU")
Cc: linux-kernel@vger.kernel.org
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
We want to use the dma_direct_ namespace for a generic implementation,
so rename powerpc to the second best choice: dma_nommu_.
Signed-off-by: Christoph Hellwig <hch@lst.de>
And unlike the other helpers we don't require a <asm/dma-direct.h> as
this helper is a special case for ia64 only, and this keeps it as
simple as possible.
Signed-off-by: Christoph Hellwig <hch@lst.de>
phys_to_dma, dma_to_phys and dma_capable are helpers published by
architecture code for use of swiotlb and xen-swiotlb only. Drivers are
not supposed to use these directly, but use the DMA API instead.
Move these to a new asm/dma-direct.h helper, included by a
linux/dma-direct.h wrapper that provides the default linear mapping
unless the architecture wants to override it.
In the MIPS case the existing dma-coherent.h is reused for now as
untangling it will take a bit of work.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Robin Murphy <robin.murphy@arm.com>
The generic version now takes dma_pfn_offset into account, so there is no
more need for an architecture override.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
This makes sure the generic version can be used with architectures /
devices that have a DMA offset in the direct mapping.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
We always use the stub definitions, so remove the unused other code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
CONFIG_ALPHA_JENSEN has failed to compile since commit 6aca0503
("alpha/dma: use common noop dma ops"), so mark it as broken.
Signed-off-by: Christoph Hellwig <hch@lst.de>
There is only one caller of the trivial function find_dev_pagemap left,
so just merge it into the caller.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This new interface is similar to how struct device (and many others)
work. The caller initializes a 'struct dev_pagemap' as required
and calls 'devm_memremap_pages'. This allows the pagemap structure to
be embedded in another structure and thus container_of can be used. In
this way application specific members can be stored in a containing
struct.
This will be used by the P2P infrastructure and HMM could probably
be cleaned up to use it as well (instead of having it's own, similar
'hmm_devmem_pages_create' function).
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
__radix_tree_insert already checks for duplicates and returns -EEXIST in
that case, so remove the duplicate (and racy) duplicates check.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Change the calling convention so that get_dev_pagemap always consumes the
previous reference instead of doing this using an explicit earlier call to
put_dev_pagemap in the callers.
The callers will still need to put the final reference after finishing the
loop over the pages.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This is a pretty big function, which should be out of line in general,
and a no-op stub if CONFIG_ZONE_DEVICЕ is not set.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
There is no clear separation between the two, so merge them.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
No functional changes, just untangling the call chain and document
why the altmap is passed around the hotplug code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Pass the vmem_altmap two levels down instead of needing a lookup.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
We can just pass this on instead of having to do a radix tree lookup
without proper locking a few levels into the callchain.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
We can just pass this on instead of having to do a radix tree lookup
without proper locking 2 levels into the callchain.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
We can just pass this on instead of having to do a radix tree lookup
without proper locking a few levels into the callchain.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
We can just pass this on instead of having to do a radix tree lookup
without proper locking 2 levels into the callchain.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This function isn't used by any modules, and is only to be called
from core MM code. This includes the calls for the add_pages wrapper
that might be inlined.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Only x86_64 and sh export this symbol, and it is not used by any
modular code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Currently all calls to those functions are eliminated by the compiler when
CONFIG_ZONE_DEVICE is not set, but this soon won't be the case.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The RISC-V port doesn't suport a nommu mode, so there is no reason
to provide some code only under a CONFIG_MMU ifdef.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
Components relying only on the request_queue structure for accessing
block devices (e.g. I/O schedulers) have a limited knowledged of the
device characteristics. In particular, the device capacity cannot be
easily discovered, which for a zoned block device also result in the
inability to easily know the number of zones of the device (the zone
size is indicated by the chunk_sectors field of the queue limits).
Introduce the nr_zones field to the request_queue structure to simplify
access to this information. Also, add the bitmap seq_zone_bitmap which
indicates which zones of the device are sequential zones (write
preferred or write required) and the bitmap seq_zones_wlock which
indicates if a zone is write locked, that is, if a write request
targeting a zone was dispatched to the device. These fields are
initialized by the low level block device driver (sd.c for ZBC/ZAC
disks). They are not initialized by stacking drivers (device mappers)
handling zoned block devices (e.g. dm-linear).
Using this, I/O schedulers can introduce zone write locking to control
request dispatching to a zoned block device and avoid write request
reordering by limiting to at most a single write request per zone
outside of the scheduler at any time.
Based on previous patches from Damien Le Moal.
Signed-off-by: Christoph Hellwig <hch@lst.de>
[Damien]
* Fixed comments and identation in blkdev.h
* Changed helper functions
* Fixed this commit message
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Christoph Hellwig