[RFC,3/4] dma-direct: add dma_direct_min_mask
diff mbox series

Message ID 20190717153135.15507-4-nsaenzjulienne@suse.de
State New
Headers show
Series
  • Raspberry Pi 4 DMA addressing support
Related show

Commit Message

Nicolas Saenz Julienne July 17, 2019, 3:31 p.m. UTC
Historically devices with ZONE_DMA32 have been assumed to be able to
address at least the lower 4GB of ram for DMA. This is still the defualt
behavior yet the Raspberry Pi 4 is limited to the first GB of memory.
This has been observed to trigger failures in dma_direct_supported() as
the 'min_mask' isn't properly set.

We create 'dma_direct_min_mask' in order for the arch init code to be
able to fine-tune dma direct's 'min_dma' mask.

Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
---
 kernel/dma/direct.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Comments

hch@lst.de July 18, 2019, 9:15 a.m. UTC | #1
On Wed, Jul 17, 2019 at 05:31:34PM +0200, Nicolas Saenz Julienne wrote:
> Historically devices with ZONE_DMA32 have been assumed to be able to
> address at least the lower 4GB of ram for DMA. This is still the defualt
> behavior yet the Raspberry Pi 4 is limited to the first GB of memory.
> This has been observed to trigger failures in dma_direct_supported() as
> the 'min_mask' isn't properly set.
> 
> We create 'dma_direct_min_mask' in order for the arch init code to be
> able to fine-tune dma direct's 'min_dma' mask.

Normally we use ZONE_DMA for that case.
Nicolas Saenz Julienne July 18, 2019, 11:18 a.m. UTC | #2
On Thu, 2019-07-18 at 11:15 +0200, Christoph Hellwig wrote:
> On Wed, Jul 17, 2019 at 05:31:34PM +0200, Nicolas Saenz Julienne wrote:
> > Historically devices with ZONE_DMA32 have been assumed to be able to
> > address at least the lower 4GB of ram for DMA. This is still the defualt
> > behavior yet the Raspberry Pi 4 is limited to the first GB of memory.
> > This has been observed to trigger failures in dma_direct_supported() as
> > the 'min_mask' isn't properly set.
> > 
> > We create 'dma_direct_min_mask' in order for the arch init code to be
> > able to fine-tune dma direct's 'min_dma' mask.
> 
> Normally we use ZONE_DMA for that case.

Fair enough, I didn't think of that possibility.

So would the arm64 maintainers be happy with something like this:

- ZONE_DMA: Follows standard definition, 16MB in size. ARCH_ZONE_DMA_BITS is
	    left as is.
- ZONE_DMA32: Will honor the most constraining 'dma-ranges'. Which so far for
	      most devices is 4G, except for RPi4.
- ZONE_NORMAL: The rest of the memory.
Nicolas Saenz Julienne July 19, 2019, 1:08 p.m. UTC | #3
On Thu, 2019-07-18 at 13:18 +0200, Nicolas Saenz Julienne wrote:
> On Thu, 2019-07-18 at 11:15 +0200, Christoph Hellwig wrote:
> > On Wed, Jul 17, 2019 at 05:31:34PM +0200, Nicolas Saenz Julienne wrote:
> > > Historically devices with ZONE_DMA32 have been assumed to be able to
> > > address at least the lower 4GB of ram for DMA. This is still the defualt
> > > behavior yet the Raspberry Pi 4 is limited to the first GB of memory.
> > > This has been observed to trigger failures in dma_direct_supported() as
> > > the 'min_mask' isn't properly set.
> > > 
> > > We create 'dma_direct_min_mask' in order for the arch init code to be
> > > able to fine-tune dma direct's 'min_dma' mask.
> > 
> > Normally we use ZONE_DMA for that case.
> 
> Fair enough, I didn't think of that possibility.
> 
> So would the arm64 maintainers be happy with something like this:
> 
> - ZONE_DMA: Follows standard definition, 16MB in size. ARCH_ZONE_DMA_BITS is
> 	    left as is.
> - ZONE_DMA32: Will honor the most constraining 'dma-ranges'. Which so far for
> 	      most devices is 4G, except for RPi4.
> - ZONE_NORMAL: The rest of the memory.

Never mind this suggestion, I don't think it makes any sense. If anything arm64
seems to fit the ZONE_DMA usage pattern of arm and powerpc: where ZONE_DMA's
size is decided based on ram size and/or board configuration. It was actually
set-up like this until Christoph's ad67f5a6545f7 ("arm64: replace ZONE_DMA with
ZONE_DMA32").

So the easy solution would be to simply revert that commit. On one hand I feel
it would be a step backwards as most 64 bit architectures have been moving to
use ZONE_DMA32. On the other, current ZONE_DMA32 usage seems to be heavily
rooted on having a 32 bit DMA mask*, which will no longer be the case on arm64
if we want to support the RPi 4.

So the way I see it and lacking a better solution, the argument is stronger on
moving back arm64 to using ZONE_DMA. Any comments/opinions?

Note that I've been looking at all the DMA/CMA/swiotlb code to see if this
would break anything or change behaviors and couldn't find anything obvious. I
also tested the revert on my RPi4 and nothing seems to fail.

* A good example is dma-direct's implementation.
Catalin Marinas July 24, 2019, 1:51 p.m. UTC | #4
On Fri, Jul 19, 2019 at 03:08:52PM +0200, Nicolas Saenz Julienne wrote:
> On Thu, 2019-07-18 at 13:18 +0200, Nicolas Saenz Julienne wrote:
> > On Thu, 2019-07-18 at 11:15 +0200, Christoph Hellwig wrote:
> > > On Wed, Jul 17, 2019 at 05:31:34PM +0200, Nicolas Saenz Julienne wrote:
> > > > Historically devices with ZONE_DMA32 have been assumed to be able to
> > > > address at least the lower 4GB of ram for DMA. This is still the defualt
> > > > behavior yet the Raspberry Pi 4 is limited to the first GB of memory.
> > > > This has been observed to trigger failures in dma_direct_supported() as
> > > > the 'min_mask' isn't properly set.
> > > > 
> > > > We create 'dma_direct_min_mask' in order for the arch init code to be
> > > > able to fine-tune dma direct's 'min_dma' mask.
> > > 
> > > Normally we use ZONE_DMA for that case.
> > 
> > Fair enough, I didn't think of that possibility.
> > 
> > So would the arm64 maintainers be happy with something like this:
> > 
> > - ZONE_DMA: Follows standard definition, 16MB in size. ARCH_ZONE_DMA_BITS is
> > 	    left as is.
> > - ZONE_DMA32: Will honor the most constraining 'dma-ranges'. Which so far for
> > 	      most devices is 4G, except for RPi4.
> > - ZONE_NORMAL: The rest of the memory.
> 
> Never mind this suggestion, I don't think it makes any sense. If anything arm64
> seems to fit the ZONE_DMA usage pattern of arm and powerpc: where ZONE_DMA's
> size is decided based on ram size and/or board configuration. It was actually
> set-up like this until Christoph's ad67f5a6545f7 ("arm64: replace ZONE_DMA with
> ZONE_DMA32").
> 
> So the easy solution would be to simply revert that commit. On one hand I feel
> it would be a step backwards as most 64 bit architectures have been moving to
> use ZONE_DMA32. On the other, current ZONE_DMA32 usage seems to be heavily
> rooted on having a 32 bit DMA mask*, which will no longer be the case on arm64
> if we want to support the RPi 4.
> 
> So the way I see it and lacking a better solution, the argument is stronger on
> moving back arm64 to using ZONE_DMA. Any comments/opinions?

As it was suggested in this or the previous thread, I'm not keen on
limiting ZONE_DMA32 to the smalles RPi4 can cover, as the naming
implies this zone should cover 32-bit devices that can deal with a full
32-bit mask.

I think it may be better if we have both ZONE_DMA and ZONE_DMA32 on
arm64. ZONE_DMA would be based on the smallest dma-ranges as described
in the DT while DMA32 covers the first naturally aligned 4GB of RAM
(unchanged). When a smaller ZONE_DMA is not needed, it could be expanded
to cover what would normally be ZONE_DMA32 (or could we have ZONE_DMA as
0-bytes? I don't think GFP_DMA can still allocate memory in this case).

We'd probably have to define ARCH_ZONE_DMA_BITS for arm64 to something
smaller than 32-bit but sufficient to cover the known platforms like
RPi4 (the current 24 is too small, so maybe 30). AFAICT,
__dma_direct_optimal_gfp_mask() figures out whether GFP_DMA or GFP_DMA32
should be passed.
hch@lst.de July 24, 2019, 1:56 p.m. UTC | #5
On Wed, Jul 24, 2019 at 02:51:24PM +0100, Catalin Marinas wrote:
> I think it may be better if we have both ZONE_DMA and ZONE_DMA32 on
> arm64. ZONE_DMA would be based on the smallest dma-ranges as described
> in the DT while DMA32 covers the first naturally aligned 4GB of RAM
> (unchanged). When a smaller ZONE_DMA is not needed, it could be expanded
> to cover what would normally be ZONE_DMA32 (or could we have ZONE_DMA as
> 0-bytes? I don't think GFP_DMA can still allocate memory in this case).
> 
> We'd probably have to define ARCH_ZONE_DMA_BITS for arm64 to something
> smaller than 32-bit but sufficient to cover the known platforms like
> RPi4 (the current 24 is too small, so maybe 30). AFAICT,
> __dma_direct_optimal_gfp_mask() figures out whether GFP_DMA or GFP_DMA32
> should be passed.

ARCH_ZONE_DMA_BITS should probably become a variable.  That way we can
just initialize it to the default 24 bits in kernel/dma/direct.c and
allow architectures to override it in their early boot code.
Nicolas Saenz Julienne July 24, 2019, 2:27 p.m. UTC | #6
On Wed, 2019-07-24 at 15:56 +0200, Christoph Hellwig wrote:
> On Wed, Jul 24, 2019 at 02:51:24PM +0100, Catalin Marinas wrote:
> > I think it may be better if we have both ZONE_DMA and ZONE_DMA32 on
> > arm64. ZONE_DMA would be based on the smallest dma-ranges as described
> > in the DT while DMA32 covers the first naturally aligned 4GB of RAM
> > (unchanged). When a smaller ZONE_DMA is not needed, it could be expanded
> > to cover what would normally be ZONE_DMA32 (or could we have ZONE_DMA as
> > 0-bytes? I don't think GFP_DMA can still allocate memory in this case).
> > 
> > We'd probably have to define ARCH_ZONE_DMA_BITS for arm64 to something
> > smaller than 32-bit but sufficient to cover the known platforms like
> > RPi4 (the current 24 is too small, so maybe 30). AFAICT,
> > __dma_direct_optimal_gfp_mask() figures out whether GFP_DMA or GFP_DMA32
> > should be passed.
> 
> ARCH_ZONE_DMA_BITS should probably become a variable.  That way we can
> just initialize it to the default 24 bits in kernel/dma/direct.c and
> allow architectures to override it in their early boot code.

Thanks both for your feedback. I'll start preparing a proper series.

Patch
diff mbox series

diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index b90e1aede743..3c8cd730648b 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -23,6 +23,8 @@ 
 #define ARCH_ZONE_DMA_BITS 24
 #endif
 
+u64 dma_direct_min_mask __ro_after_init = DMA_BIT_MASK(32);
+
 /*
  * For AMD SEV all DMA must be to unencrypted addresses.
  */
@@ -393,7 +395,7 @@  int dma_direct_supported(struct device *dev, u64 mask)
 	if (IS_ENABLED(CONFIG_ZONE_DMA))
 		min_mask = DMA_BIT_MASK(ARCH_ZONE_DMA_BITS);
 	else
-		min_mask = DMA_BIT_MASK(32);
+		min_mask = dma_direct_min_mask;
 
 	min_mask = min_t(u64, min_mask, (max_pfn - 1) << PAGE_SHIFT);