[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-devel] [PATCH v4 5/8] xen/arm: assign devices to boot domains
On Wed, 11 Sep 2019, Julien Grall wrote: > Hi, > > On 8/21/19 4:53 AM, Stefano Stabellini wrote: > > Scan the user provided dtb fragment at boot. For each device node, map > > memory to guests, and route interrupts and setup the iommu. > > > > The memory region to remap is specified by the "xen,reg" property. > > > > The iommu is setup by passing the node of the device to assign on the > > host device tree. The path is specified in the device tree fragment as > > the "xen,path" string property. > > > > The interrupts are remapped based on the information from the > > corresponding node on the host device tree. Call > > handle_device_interrupts to remap interrupts. Interrupts related device > > tree properties are copied from the device tree fragment, same as all > > the other properties. > > > > Signed-off-by: Stefano Stabellini <stefanos@xxxxxxxxxx> > > --- > > Changes in v4: > > - use unsigned > > - improve commit message > > - code style > > - use dt_prop_cmp > > - use device_tree_get_reg > > - don't copy over xen,reg and xen,path > > - don't create special interrupt properties for domU: copy them from the > > fragment > > - in-code comment > > > > Changes in v3: > > - improve commit message > > - remove superfluous cast > > - merge code with the copy code > > - add interrup-parent > > - demove depth > 2 check > > - reuse code from handle_device_interrupts > > - copy interrupts from host dt > > > > Changes in v2: > > - rename "path" to "xen,path" > > - grammar fix > > - use gaddr_to_gfn and maddr_to_mfn > > - remove depth <= 2 limitation in scanning the dtb fragment > > - introduce and parse xen,reg > > - code style > > - support more than one interrupt per device > > - specify only the GIC is supported > > --- > > xen/arch/arm/domain_build.c | 66 ++++++++++++++++++++++++++++++++++--- > > 1 file changed, 62 insertions(+), 4 deletions(-) > > > > diff --git a/xen/arch/arm/domain_build.c b/xen/arch/arm/domain_build.c > > index c71b9f2889..256c83462a 100644 > > --- a/xen/arch/arm/domain_build.c > > +++ b/xen/arch/arm/domain_build.c > > @@ -1721,6 +1721,9 @@ static int __init handle_prop_pfdt(struct kernel_info > > *kinfo, > > void *fdt = kinfo->fdt; > > int propoff, nameoff, res; > > const struct fdt_property *prop; > > + struct dt_device_node *node; > > + const __be32 *cell; > > + unsigned int i, len; > > for ( propoff = fdt_first_property_offset(pfdt, nodeoff); > > propoff >= 0; > > @@ -1730,10 +1733,65 @@ static int __init handle_prop_pfdt(struct > > kernel_info *kinfo, > > return -FDT_ERR_INTERNAL; > > nameoff = fdt32_to_cpu(prop->nameoff); > > - res = fdt_property(fdt, fdt_string(pfdt, nameoff), > > - prop->data, fdt32_to_cpu(prop->len)); > > - if ( res ) > > - return res; > > + > > + /* xen,reg specifies where to map the MMIO region */ > > + if ( dt_prop_cmp("xen,reg", fdt_string(pfdt, nameoff)) == 0 ) > > The construct fdt_string(pfdt, nameoff) is used quite often within this > function. I think it is warrant for a local variable. OK > > + { > > + paddr_t mstart, size, gstart; > > + cell = (const __be32 *)prop->data; > > + len = fdt32_to_cpu(prop->len) / > > + ((address_cells * 2 + size_cells) * sizeof(uint32_t)); > > + > > + for ( i = 0; i < len; i++ ) > > + { > > + device_tree_get_reg(&cell, address_cells, size_cells, > > + &mstart, &size); > > + gstart = dt_next_cell(address_cells, &cell); > > + > > + res = guest_physmap_add_entry(kinfo->d, > > gaddr_to_gfn(gstart), > > + maddr_to_mfn(mstart), > > + get_order_from_bytes(size), > > guest_physmap_add_entry is definitely not the correct function to call. It > takes an order and means that if the user ask to map 12KB, it will effectively > map 16KB. You want to use map_regions_p2mt as this is done everywhere else for > mapping MMIO regions. OK > It also raises the question what should we do if the size passed in not > page-aligned? Shall we just blindly round up/down? Should we warn? > > This was not important for dom0, but is potentially critical for domU as you > may happen to inadvertently to export more than you hope to a guest. A warning or even a panic would be OK because it is a static misconfiguration. > > + p2m_mmio_direct_dev); > > + if ( res < 0 ) > > + { > > + dprintk(XENLOG_ERR, > > + "Failed to map %"PRIpaddr" to the guest > > at%"PRIpaddr"\n", > > + mstart, gstart); > > + return -EFAULT; > > + } > > + } > > + } > > + /* > > + * xen,path specifies the corresponding node in the host DT. > > + * Both interrupt mappings and IOMMU settings are based on it, > > + * as they are done based on the corresponding host DT node. > > + */ > > + else if ( dt_prop_cmp("xen,path", fdt_string(pfdt, nameoff)) == 0 ) > > + { > > + node = dt_find_node_by_path(prop->data); > > + if ( node == NULL ) > > + { > > + dprintk(XENLOG_ERR, "Couldn't find node %s in host_dt!\n", > > + (char *)prop->data); > > + return -EINVAL; > > + } > > + > > + res = iommu_assign_dt_device(kinfo->d, node); > > + if ( res < 0 ) > > + return res; > > + > > + res = handle_device_interrupts(kinfo->d, node, true); > > + if ( res < 0 ) > > + return res; > > + } > > + /* copy all other properties */ > > + else > > + { > > + res = fdt_property(fdt, fdt_string(pfdt, nameoff), prop->data, > > + fdt32_to_cpu(prop->len)); > > + if ( res ) > > + return res; > > + } > > } > > /* FDT_ERR_NOTFOUND => There is no more properties for this node */ > > > > Cheers, > > -- > Julien Grall > _______________________________________________ Xen-devel mailing list Xen-devel@xxxxxxxxxxxxxxxxxxxx https://lists.xenproject.org/mailman/listinfo/xen-devel
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |