linux-toradex.git/arch/ia64/kernel/dma-mapping.c, branch v5.18

dma-mapping: split

2020-10-06T05:07:03+00:00

Split out all the bits that are purely for dma_map_ops implementations
and related code into a new  header so that they
don't get pulled into all the drivers.  That also means the architecture
specific  is not pulled in by 
any more, which leads to a missing includes that were pulled in by the
x86 or arm versions in a few not overly portable drivers.

Signed-off-by: Christoph Hellwig

dma-mapping: add (back) arch_dma_mark_clean for ia64

2020-09-11T07:10:17+00:00

Add back a hook to optimize dcache flushing after reading executable
code using DMA.  This gets ia64 out of the business of pretending to
be dma incoherent just for this optimization.

Signed-off-by: Christoph Hellwig

dma-direct: provide mmap and get_sgtable method overrides

2019-11-11T09:52:15+00:00

For dma-direct we know that the DMA address is an encoding of the
physical address that we can trivially decode.  Use that fact to
provide implementations that do not need the arch_dma_coherent_to_pfn
architecture hook.  Note that we still can only support mmap of
non-coherent memory only if the architecture provides a way to set an
uncached bit in the page tables.  This must be true for architectures
that use the generic remap helpers, but other architectures can also
manually select it.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Max Filippov

ia64: remove CONFIG_SWIOTLB ifdefs

2019-08-16T21:32:26+00:00

CONFIG_SWIOTLB is now unconditionally selected on ia64, so remove the
ifdefs.

Signed-off-by: Christoph Hellwig 
Link: https://lkml.kernel.org/r/20190813072514.23299-28-hch@lst.de
Signed-off-by: Tony Luck

ia64: rework iommu probing

2019-08-16T18:33:57+00:00

ia64 currently organizes the iommu probing along machves, which isn't
very helpful.  Instead just try to probe for Intel IOMMUs in mem_init
as they are properly described in ACPI and if none was found initialize
the swiotlb buffer.  The HP SBA handling is then only done delayed when
the actual hardware is probed. Only in the case that we actually found
usable IOMMUs we then set up the DMA ops and free the not needed
swiotlb buffer.  This scheme gets rid of the need for the dma_init
machvec operation, and the dig_vtd machvec.

Signed-off-by: Christoph Hellwig 
Link: https://lkml.kernel.org/r/20190813072514.23299-24-hch@lst.de
Signed-off-by: Tony Luck

ia64: remove the zx1 swiotlb machvec

2019-08-16T18:33:57+00:00

The aim of this machvec is to support devices with < 32-bit dma
masks.  But given that ia64 only has a ZONE_DMA32 and not a ZONE_DMA
that isn't supported by swiotlb either.

Signed-off-by: Christoph Hellwig 
Link: https://lkml.kernel.org/r/20190813072514.23299-21-hch@lst.de
Signed-off-by: Tony Luck

dma-mapping: bypass indirect calls for dma-direct

2018-12-13T20:06:18+00:00

Avoid expensive indirect calls in the fast path DMA mapping
operations by directly calling the dma_direct_* ops if we are using
the directly mapped DMA operations.

Signed-off-by: Christoph Hellwig 
Acked-by: Jesper Dangaard Brouer 
Tested-by: Jesper Dangaard Brouer 
Tested-by: Tony Luck

dma-direct: merge swiotlb_dma_ops into the dma_direct code

2018-12-13T20:06:17+00:00

While the dma-direct code is (relatively) clean and simple we actually
have to use the swiotlb ops for the mapping on many architectures due
to devices with addressing limits.  Instead of keeping two
implementations around this commit allows the dma-direct
implementation to call the swiotlb bounce buffering functions and
thus share the guts of the mapping implementation.  This also
simplified the dma-mapping setup on a few architectures where we
don't have to differenciate which implementation to use.

Signed-off-by: Christoph Hellwig 
Acked-by: Jesper Dangaard Brouer 
Tested-by: Jesper Dangaard Brouer 
Tested-by: Tony Luck

swiotlb: remove dma_mark_clean

2018-12-13T20:06:14+00:00

Instead of providing a special dma_mark_clean hook just for ia64, switch
ia64 to use the normal arch_sync_dma_for_cpu hooks instead.

This means that we now also set the PG_arch_1 bit for pages in the
swiotlb buffer, which isn't stricly needed as we will never execute code
out of the swiotlb buffer, but otherwise harmless.

Signed-off-by: Christoph Hellwig 
Acked-by: Jesper Dangaard Brouer 
Tested-by: Jesper Dangaard Brouer 
Tested-by: Tony Luck

dma-debug: move initialization to common code

2018-05-08T11:02:42+00:00

Most mainstream architectures are using 65536 entries, so lets stick to
that.  If someone is really desperate to override it that can still be
done through , but I'd rather see a really good
rationale for that.

dma_debug_init is now called as a core_initcall, which for many
architectures means much earlier, and provides dma-debug functionality
earlier in the boot process.  This should be safe as it only relies
on the memory allocator already being available.

Signed-off-by: Christoph Hellwig 
Acked-by: Marek Szyprowski 
Reviewed-by: Robin Murphy