[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v1 16/27] xen/riscv: implement IRQ mapping for device passthrough





On 4/16/26 2:51 PM, Jan Beulich wrote:
On 14.04.2026 13:29, Oleksii Kurochko wrote:
On 4/2/26 2:22 PM, Jan Beulich wrote:
On 10.03.2026 18:08, Oleksii Kurochko wrote:
--- a/xen/arch/riscv/include/asm/setup.h
+++ b/xen/arch/riscv/include/asm/setup.h
@@ -5,6 +5,10 @@
#include <xen/types.h> +struct domain;
+struct dt_device_node;
+struct rangeset;
+
   #define max_init_domid (0)
void setup_mm(void);
@@ -13,6 +17,19 @@ void copy_from_paddr(void *dst, paddr_t paddr, unsigned long 
len);
void init_csr_masks(void); +/* TODO: move somewhere to common header? */

Counter question: Why ...

+/*
+ * Retrieves the interrupts configuration from a device tree node and maps
+ * those interrupts to the target domain.
+ *
+ * Returns:
+ *   < 0 error
+ *   0   success
+ */
+int map_device_irqs_to_domain(struct domain *d, struct dt_device_node *dev,
+                              bool need_mapping,
+                              struct rangeset *irq_ranges);

... is this not an inline function, when ...

--- a/xen/arch/riscv/intc.c
+++ b/xen/arch/riscv/intc.c
@@ -79,3 +79,11 @@ int __init intc_make_domu_dt_node(const struct kernel_info 
*kinfo)
return -ENOSYS;
   }
+
+int map_device_irqs_to_domain(struct domain *d, struct dt_device_node *dev,
+                              bool need_mapping,
+                              struct rangeset *irq_ranges)
+{
+    return d->arch.vintc->ops->map_device_irqs_to_domain(d, dev, need_mapping,
+                                                         irq_ranges);
+}

... it's merely a wrapper around an indirect function call? And then the
function isn't used anywhere anyway.

It is used by dom0less common code and it is a wrapper because Arm has
different implementation and Arm doesn't have
map_device_irqs_to_domain() in its virtual interrupt controller operations.

But the question wasn't why this is a wrapper, but why this wrapper isn't an
inline function.

Got you, it could be really inline.


+int vaplic_map_device_irqs_to_domain(struct domain *d,
+                                     struct dt_device_node *dev,
+                                     bool need_mapping,
+                                     struct rangeset *irq_ranges)
+{
+    unsigned int i, nirq;
+    int res, irq;
+    struct dt_raw_irq rirq;
+    uint32_t *auth_irq_bmp = d->arch.vintc->private;
+    unsigned int reg_num;
+
+    nirq = dt_number_of_irq(dev);
+
+    /* Give permission and map IRQs */
+    for ( i = 0; i < nirq; i++ )
+    {
+        res = dt_device_get_raw_irq(dev, i, &rirq);
+        if ( res )
+        {
+            printk(XENLOG_ERR "Unable to retrieve irq %u for %s\n",
+                   i, dt_node_full_name(dev));
+            return res;
+        }
+
+        /*
+         * Don't map IRQ that have no physical meaning
+         * ie: IRQ whose controller is not APLIC/IMSIC/PLIC.
+         */
+        if ( rirq.controller != dt_interrupt_controller )
+        {
+            dt_dprintk("irq %u not connected to primary controller."
+                       "Connected to %s\n", i,
+                       dt_node_full_name(rirq.controller));
+            continue;
+        }
+
+        irq = platform_get_irq(dev, i);
+        if ( irq < 0 )
+        {
+            printk("Unable to get irq %u for %s\n", i, dt_node_full_name(dev));
+            return irq;
+        }
+
+        res = irq_permit_access(d, irq);
+        if ( res )
+        {
+            printk(XENLOG_ERR "Unable to permit to %pd access to IRQ %u\n", d,
+                   irq);

This time the other way around: %d please with plain int. (Again at least
once further down.)

+            return res;
+        }
+
+        reg_num = irq / APLIC_NUM_REGS;
+
+        if ( is_irq_shared_among_domains(d, irq) )
+        {
+            printk("%s: Shared IRQ isn't supported\n", __func__);
+            return -EINVAL;
+        }
+
+        auth_irq_bmp[reg_num] |= BIT(irq % APLIC_NUM_REGS, U);

... all of this leaves me with the impression that IRQ numbering isn't really
virtualized. IRQs are merely split into groups, one group per domain (and
maybe some unused). How are you going to fit in truly virtual IRQs?

What do you mean by truly virtual IRQs?

Ones where no aspects are represented by any piece of hardware.

I can't totally agree that the current approach isn't use virtual IRQs,
yes, they are 1:1 mapped but on the other side Xen is responsible to
give an IRQ number for guest's device and Xen is responsible that guest
isn't trying to reach IRQ which not belongs to it.

In a non-virtualized environment I expect IRQs are going to be "sparse"
(i.e. with perhaps large blocks of items used elsewhere). If you had
proper translation of IRQ numbers, the same could be true for your
guests.

Partial FDT, which is used to tell which device be passthroughed to guest, is using physical IRQ number (which I am just considering for simplicity to be 1:1 mapped to virtual IRQ number). So if we have the following configuration:
  Physical (bare-metal) IRQ layout is sparse:
    IRQ 5  → UART -> domU0
    IRQ 23 → Ethernet -> domU1
    IRQ 47 → PCIe -> domU0
    IRQ 100 → Storage -> domU1
(gaps everywhere, driven by hardware wiring)

For such configuration we will have for each domain auth_irq_bmp[] which contains:
 IRQ 5 and IRQ47 for domU0
and
 IRQ 23 and IRQ 100 for domU1

And here vIRQ5 = pIRQ5, vIRQ47 = pIRQ47 and so on. auth_irq_bmp just transform xIRQ number to bit position which it will have in real APLIC register. Just as an example, lets take vIRQ5 and vIRQ47.

As reading or writing register setie[k] reads or potentially modifies the enable bits for interrupt sources k × 32 through k × 32 + 31. For an implemented interrupt source i within that range, the enable bit for source i corresponds with register bit (i mod 32).
So for:
 - vIRQ5 == pIRQ5 we have to set bit 5 in setie[0]
 - vIRQ47 == pIRQ47 we have to set bit 15 in setie[1]

Probably it was not the best idea to declare auth_irq_bmp as it will look in h/w and maybe just 'bool auth_irq_bmp[1024]' would be more clearer.

So irqs number are still stay "sparsed" in guest.


+        dt_dprintk("  - IRQ: %u\n", irq);
+
+        if ( irq_ranges )
+        {
+            res = rangeset_add_singleton(irq_ranges, irq);
+            if ( res )
+                return res;
+        }

What is irq_ranges?

IIUC based on Arm code irq_ranges is an optional output accumulator, the
caller allocates and passes it in when it needs to track which IRQs were
mapped (overlay use case), or passes NULL when that tracking is not needed.

I added here as map_device_irqs_to_domain() is called from the common
code and so maybe one day someone will decide to pass irq_ranges to this
functions. At the moment, for RISC-V it is the only one user of
map_device_irqs_to_domain() and it passes NULL.

Simply assert then that it's NULL?

Won't BUG_ON() be better here as it BUG_ON() macros is always defined and doesn't matter if release or debug build are used.

Or maybe you meant:
 if ( irq_ranges )
     assert_failed("irq_ranges arg isn't supported\n");


@@ -34,6 +142,7 @@ static int __init cf_check vcpu_vaplic_init(struct vcpu *v)
static const struct vintc_ops vaplic_ops = {
       .vcpu_init = vcpu_vaplic_init,
+    .map_device_irqs_to_domain = vaplic_map_device_irqs_to_domain,
   };

What about the inverse function, needed for domain cleanup?

I planned to add it when it will be really needed. At the momemnt, I
don't have such use cases.

I.e. if any domain needs re-starting, the entire system needs rebooting?
Recall that "dom0less" is slightly misleading a name, as it only allows
there to not be a Dom0. One can be there, and hence re-starting a crashed
domain ought to be possible. For that, you need to correctly clean up
after the crashed one.

I haven't investigated yet what will be code path to do a reboot or re-start crashed domain but based on an implementation of vaplic_map_device_irqs_to_domain() as nothing is allocated there and filling of auth_irq_bmp[] is happening there which will stay the same after reboot if it is just re-used or it will be allocated new one if crashed domain will just recreate this domain from scratch. So it seems like it is enough to have only map_device_irqs_to_domain().

~ Oleksii




 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.