Re: [Qemu-devel] [PATCH qemu v10 14/14] spapr_pci/spapr_pci_vfio: Support Dynamic DMA Windows (DDW)

[David Gibson <david@gibson.dropbear.id.au>]

[-- Attachment #1: Type: text/plain, Size: 4933 bytes --]

On Mon, Jul 06, 2015 at 12:11:10PM +1000, Alexey Kardashevskiy wrote:
> This adds support for Dynamic DMA Windows (DDW) option defined by
> the SPAPR specification which allows to have additional DMA window(s)
> 
> This implements DDW for emulated and VFIO devices. As all TCE root regions
> are mapped at 0 and 64bit long (and actual tables are child regions),
> this replaces memory_region_add_subregion() with _overlap() to make
> QEMU memory API happy.
> 
> This reserves RTAS token numbers for DDW calls.
> 
> This implements helpers to interact with VFIO kernel interface.
> 
> This changes the TCE table migration descriptor to support dynamic
> tables as from now on, PHB will create as many stub TCE table objects
> as PHB can possibly support but not all of them might be initialized at
> the time of migration because DDW might or might not be requested by
> the guest.
> 
> The "ddw" property is enabled by default on a PHB but for compatibility
> the pseries-2.3 machine and older disable it.
> 
> This implements DDW for VFIO. The host kernel support is required.
> This adds a "levels" property to PHB to control the number of levels
> in the actual TCE table allocated by the host kernel, 0 is the default
> value to tell QEMU to calculate the correct value. Current hardware
> supports up to 5 levels.
> 
> The existing linux guests try creating one additional huge DMA window
> with 64K or 16MB pages and map the entire guest RAM to. If succeeded,
> the guest switches to dma_direct_ops and never calls TCE hypercalls
> (H_PUT_TCE,...) again. This enables VFIO devices to use the entire RAM
> and not waste time on map/unmap later. This adds a "dma64_win_addr"
> property which is a bus address for the 64bit window and by default
> set to 0x800.0000.0000.0000 as this is what the modern POWER8 hardware
> uses and this allows having emulated and VFIO devices on the same bus.
> 
> This adds 4 RTAS handlers:
> * ibm,query-pe-dma-window
> * ibm,create-pe-dma-window
> * ibm,remove-pe-dma-window
> * ibm,reset-pe-dma-window
> These are registered from type_init() callback.
> 
> These RTAS handlers are implemented in a separate file to avoid polluting
> spapr_iommu.c with PCI.


> diff --git a/trace-events b/trace-events
> index b300e94..a1234dd 100644
> --- a/trace-events
> +++ b/trace-events
> @@ -1302,6 +1302,8 @@ spapr_pci_lsi_set(const char *busname, int pin, uint32_t irq) "%s PIN%d IRQ %u"
>  spapr_pci_msi_retry(unsigned config_addr, unsigned req_num, unsigned max_irqs) "Guest device at %x asked %u, have only %u"
>  spapr_pci_dma_update(uint64_t liobn, long ret) "liobn=%"PRIx64" ret=%ld"
>  spapr_pci_dma_realloc_update(uint64_t liobn, long ret) "liobn=%"PRIx64" tcet=%ld"
> +spapr_pci_vfio_init_window(int ps, uint64_t ws, uint64_t off) "pageshift=0x%x winsize=0x%"PRIx64" offset=0x%"PRIx64
> +spapr_pci_vfio_remove_window(uint64_t off) "offset=%"PRIx64
>  
>  # hw/pci/pci.c
>  pci_update_mappings_del(void *d, uint32_t bus, uint32_t func, uint32_t slot, int bar, uint64_t addr, uint64_t size) "d=%p %02x:%02x.%x %d,%#"PRIx64"+%#"PRIx64
> @@ -1365,6 +1367,10 @@ spapr_iommu_pci_indirect(uint64_t liobn, uint64_t ioba, uint64_t tce, uint64_t i
>  spapr_iommu_pci_stuff(uint64_t liobn, uint64_t ioba, uint64_t tce_value, uint64_t npages, uint64_t ret) "liobn=%"PRIx64" ioba=0x%"PRIx64" tcevalue=0x%"PRIx64" npages=%"PRId64" ret=%"PRId64
>  spapr_iommu_xlate(uint64_t liobn, uint64_t ioba, uint64_t tce, unsigned perm, unsigned pgsize) "liobn=%"PRIx64" 0x%"PRIx64" -> 0x%"PRIx64" perm=%u mask=%x"
>  spapr_iommu_alloc_table(uint64_t liobn, void *table, int fd) "liobn=%"PRIx64" table=%p fd=%d"
> +spapr_iommu_ddw_query(uint64_t buid, uint32_t cfgaddr, unsigned wa, uint64_t win_size, uint32_t pgmask) "buid=%"PRIx64" addr=%"PRIx32", %u windows available, max window size=%"PRIx64", mask=%"PRIx32

Turns out the dtrace trace backend barfs on the "long long" here :(

$ ./configure --target-list=ppc64-softmmu --enable-trace-backends=dtrace
[...]
$ make
  GEN   config-host.h
  GEN   trace/generated-tracers.h
  GEN   trace/generated-tracers-dtrace.dtrace
  GEN   trace/generated-tracers-dtrace.h
Warning: /bin/dtrace:trace/generated-tracers-dtrace.dtrace:2212: syntax	error near:
probe spapr_iommu_ddw_query

Warning: Proceeding as if --no-pyparsing was given.

  GEN   trace/generated-tcg-tracers.h
  GEN   trace/generated-helpers-wrappers.h
  GEN   trace/generated-helpers.h
  CC    trace/generated-events.o
  GEN   trace/generated-tracers-dtrace.o
Warning: /bin/dtrace:trace/generated-tracers-dtrace.dtrace:2212: syntax error near:
probe spapr_iommu_ddw_query

Warning: Proceeding as if --no-pyparsing was given.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]