Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Josh Hilke <jrhilke@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
	kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	 David Matlack <dmatlack@google.com>,
	Alex Williamson <alex.williamson@redhat.com>
Subject: Re: [PATCH v5 07/21] KVM: selftests: Verify IRQ bypass works in IRQ test
Date: Thu, 4 Jun 2026 09:22:07 -0700	[thread overview]
Message-ID: <aiGmL-8RKlVPbBwZ@google.com> (raw)
In-Reply-To: <20260604020143.748245-8-jrhilke@google.com>

On Thu, Jun 04, 2026, Josh Hilke wrote:
> @@ -76,9 +109,10 @@ static void kvm_route_msi(struct kvm_vm *vm, u32 gsi, struct kvm_vcpu *vcpu,
>  
>  static void help(const char *name)
>  {
> -	printf("Usage: %s [-h]\n", name);
> +	printf("Usage: %s [-d <segment:bus:device.function>] [-h]\n", name);
>  	printf("\n");
>  	printf("Tests KVM IRQ injection via irqfd using an emulated eventfd.\n");
> +	printf("-d	Use a VFIO device to send MSI-X interrupts instead of using an emulated eventfd\n");


	printf("-d	Use a VFIO device to send MSI-X interrupts instead of manually signaling the eventfd\n");

>  	printf("\n");
>  	exit(KSFT_FAIL);
>  }
> @@ -100,14 +134,21 @@ int main(int argc, char **argv)
>  	u32 gsi = kvm_random_u64_in_range(&kvm_rng, 24, KVM_MAX_IRQ_ROUTES - 1);
>  	u8 vector = kvm_random_u64_in_range(&kvm_rng, 32, UINT8_MAX);
>  
> -	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
>  	pthread_t vcpu_threads[KVM_MAX_VCPUS];
> +	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
> +	struct vfio_pci_device *device = NULL;
>  	int nr_irqs = 1000, nr_vcpus = 1;
> -	int i, j, c, eventfd;
> +	const char *device_bdf = NULL;
> +	int i, j, c, msix, eventfd;
> +	struct iommu *iommu;
>  	struct kvm_vm *vm;
> +	unsigned int irq;

I think it makes sense to track "irq" as a signed int, so that the !device path
can set it to -1.  Then it can be printed somewhat blindly without being outright
wrong.

> -	while ((c = getopt(argc, argv, "h")) != -1) {
> +	while ((c = getopt(argc, argv, "d:h")) != -1) {
>  		switch (c) {
> +		case 'd':
> +			device_bdf = optarg;
> +			break;
>  		case 'h':
>  		default:
>  			help(argv[0]);
> @@ -119,7 +160,17 @@ int main(int argc, char **argv)
>  	vm = vm_create_with_vcpus(nr_vcpus, guest_code, vcpus);
>  	vm_install_exception_handler(vm, vector, guest_irq_handler);
>  
> -	eventfd = kvm_new_eventfd();
> +	if (device_bdf) {
> +		iommu = iommu_init(default_iommu_mode);
> +		device = vfio_pci_device_init(device_bdf, iommu);
> +		msix = vfio_setup_msi(device);
> +		irq = vfio_msix_to_host_irq(device_bdf, msix);
> +		eventfd = device->msi_eventfds[msix];
> +		printf("Using device %s MSI-X[%d] (IRQ-%u)\n", device_bdf, msix,
> +		       irq);
> +	} else {
> +		eventfd = kvm_new_eventfd();
> +	}
>  
>  	printf("Injecting interrupts for GSI %d (Vector 0x%x) %d times\n",
>  	       gsi, vector, nr_irqs);
> @@ -147,8 +198,7 @@ int main(int argc, char **argv)
>  				    "IRQ flag for vCPU %d not clear prior to test",
>  				    vcpus[j]->id);

Print the IRQ in the assert.  Ditto for the pCPU in the future.  Providing more
debug information is nice, but it should *always* be secondary to providing info
in the assert/error itself.

		TEST_ASSERT(GUEST_RECEIVED_INTERRUPT(vcpu, do_use_nmi),
			    "vCPU %d timed out waiting for %s (vector 0x%x) from GSI %d (via CPU %d)\n",
			    vcpu->id, do_use_nmi ? "NMI" : "IRQ",
			    do_use_nmi ? NMI_VECTOR : vector, gsi, irq_cpu);


>  
> -		/* Trigger interrupt */
> -		eventfd_write(eventfd, 1);
> +		trigger_interrupt(device, eventfd);
>  
>  		clock_gettime(CLOCK_MONOTONIC, &start);
>  		for (;;) {
> -- 
> 2.54.0.1032.g2f8565e1d1-goog
> 

  reply	other threads:[~2026-06-04 16:22 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-04  2:01 [PATCH v5 00/21] KVM: selftests: Link with VFIO selftests lib and test device interrupts Josh Hilke
2026-06-04  2:01 ` [PATCH v5 01/21] KVM: selftests: Build and link selftests/vfio/lib into KVM selftests Josh Hilke
2026-06-04  2:01 ` [PATCH v5 02/21] KVM: selftests: Add guest read/write macros Josh Hilke
2026-06-04  2:01 ` [PATCH v5 03/21] KVM: selftests: Rename guest_rng to kvm_rng Josh Hilke
2026-06-04  2:01 ` [PATCH v5 04/21] KVM: selftests: Add helper to generate random u64 in range [min,max] Josh Hilke
2026-06-04  2:12   ` sashiko-bot
2026-06-04 16:12     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 05/21] KVM: selftests: Add IRQ injection test Josh Hilke
2026-06-04 16:19   ` Sean Christopherson
2026-06-04 16:26     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 06/21] KVM: selftests: Add helper to get host IRQ from device MSIX for IRQ bypass test Josh Hilke
2026-06-04  2:09   ` sashiko-bot
2026-06-04 20:01     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 07/21] KVM: selftests: Verify IRQ bypass works in IRQ test Josh Hilke
2026-06-04 16:22   ` Sean Christopherson [this message]
2026-06-04 17:56   ` Sean Christopherson
2026-06-04 19:25     ` Sean Christopherson
2026-06-04 19:52       ` Sean Christopherson
2026-06-04 23:14       ` Sean Christopherson
2026-06-04 23:35         ` David Matlack
2026-06-04  2:01 ` [PATCH v5 08/21] KVM: selftests: Add helpers to write proc IRQ affinity for " Josh Hilke
2026-06-04  2:09   ` sashiko-bot
2026-06-04 19:35   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 09/21] KVM: selftests: Add helpers to print " Josh Hilke
2026-06-04  2:09   ` sashiko-bot
2026-06-04 17:17     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 10/21] KVM: selftests: Verify interrupts are received when IRQ affinity changes in " Josh Hilke
2026-06-04  2:13   ` sashiko-bot
2026-06-04  2:01 ` [PATCH v5 11/21] KVM: selftests: Verify IRQs wake up halted vCPUs " Josh Hilke
2026-06-04  2:15   ` sashiko-bot
2026-06-04 16:11     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 12/21] KVM: selftests: Verify interrupts are received after modifying IRQ routes " Josh Hilke
2026-06-04 17:22   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 13/21] KVM: selftests: Make number of IRQs configurable " Josh Hilke
2026-06-04 17:35   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 14/21] KVM: selftests: Verify non-postable IRQ remapping " Josh Hilke
2026-06-04 17:22   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 15/21] KVM: selftests: Add kvm_gettid() wrapper and convert users Josh Hilke
2026-06-04  2:01 ` [PATCH v5 16/21] KVM: selftests: Add kvm_sched_getaffinity() " Josh Hilke
2026-06-04 17:23   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 17/21] KVM: selftests: Add pin_task_to_random_cpu() helper function for IRQ test Josh Hilke
2026-06-04  2:01 ` [PATCH v5 18/21] KVM: selftests: Verify vCPU migration during IRQ delivery in " Josh Hilke
2026-06-04 17:27   ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 19/21] KVM: selftests: Print vCPU affinity on timeout during " Josh Hilke
2026-06-04  2:21   ` sashiko-bot
2026-06-04 17:28     ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 20/21] KVM: selftests: Make number of vCPUs configurable in " Josh Hilke
2026-06-04  2:20   ` sashiko-bot
2026-06-04 17:29     ` Sean Christopherson
2026-06-04 17:41       ` Sean Christopherson
2026-06-04  2:01 ` [PATCH v5 21/21] KVM: selftests: Add xAPIC support " Josh Hilke
2026-06-04  2:22   ` sashiko-bot
2026-06-04 17:34     ` Sean Christopherson
2026-06-04 20:22 ` [PATCH v5 00/21] KVM: selftests: Link with VFIO selftests lib and test device interrupts Sean Christopherson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=aiGmL-8RKlVPbBwZ@google.com \
    --to=seanjc@google.com \
    --cc=alex.williamson@redhat.com \
    --cc=dmatlack@google.com \
    --cc=jrhilke@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox