From: Sairaj Kodilkar <sarunkod@amd.com>
To: <seanjc@google.com>
Cc: <baolu.lu@linux.intel.com>, <dmatlack@google.com>,
<dwmw2@infradead.org>, <francescolavra.fl@gmail.com>,
<iommu@lists.linux.dev>, <joao.m.martins@oracle.com>,
<joro@8bytes.org>, <kvm@vger.kernel.org>,
<linux-kernel@vger.kernel.org>, <mlevitsk@redhat.com>,
<pbonzini@redhat.com>, <sarunkod@amd.com>, <vasant.hegde@amd.com>
Subject: Re:[PATCH v2 00/59] KVM: iommu: Overhaul device posted IRQs support
Date: Mon, 9 Jun 2025 17:50:50 +0530 [thread overview]
Message-ID: <20250609122050.28499-1-sarunkod@amd.com> (raw)
In-Reply-To: <20250523010004.3240643-1-seanjc@google.com>
Hi Sean,
Sorry for the delay in testing. All sanity tests are OK. I reran the performance
test on the V2 and noticed that V2 has significantly more GALOG entries than V1
for all three cases. I also noticed that the Guest Nvme interrupt rate has
dropped for the 192 VCPUS.
I haven't figured out what is causing this. I will continue my investigation
further.
VCPUS = 32, Jobs per NVME = 8
==============================================================================================
V2 V1 Percent change
----------------------------------------------------------------------------------------------
Guest Nvme interrupts 124,260,796 124,559,110 -0.20%
IOPS (in kilo) 4,790 4,796 -0.01%
GALOG entries 8117 169 4702%
----------------------------------------------------------------------------------------------
VCPUS = 64, Jobs per NVME = 16
==============================================================================================
V2 V1 Percent change
----------------------------------------------------------------------------------------------
Guest Nvme interrupts 102,394,358 99,800,056 2.00%
IOPS (in kilo) 4,796 4,798 -0.04%
GALOG entries 19,057 11,923 59.83%
----------------------------------------------------------------------------------------------
VCPUS = 192, Jobs per NVME = 48
==============================================================================================
V2 V1 Percent change
----------------------------------------------------------------------------------------------
Guest Nvme interrupts 68,363,232 78,066,512 -12.42%
IOPS (in kilo) 4,751 4,749 -0.04%
GALOG entries 62,768 56,215 11.66%
----------------------------------------------------------------------------------------------
Thanks
Sairaj
next prev parent reply other threads:[~2025-06-09 12:21 UTC|newest]
Thread overview: 68+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-23 0:59 [PATCH v2 00/59] KVM: iommu: Overhaul device posted IRQs support Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 01/59] KVM: x86: Pass new routing entries and irqfd when updating IRTEs Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 02/59] KVM: SVM: Track per-vCPU IRTEs using kvm_kernel_irqfd structure Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 03/59] KVM: SVM: Delete IRTE link from previous vCPU before setting new IRTE Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 04/59] iommu/amd: KVM: SVM: Delete now-unused cached/previous GA tag fields Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 05/59] KVM: SVM: Delete IRTE link from previous vCPU irrespective of new routing Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 06/59] KVM: SVM: Drop pointless masking of default APIC base when setting V_APIC_BAR Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 07/59] KVM: SVM: Drop pointless masking of kernel page pa's with AVIC HPA masks Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 08/59] KVM: SVM: Add helper to deduplicate code for getting AVIC backing page Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 09/59] KVM: SVM: Drop vcpu_svm's pointless avic_backing_page field Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 10/59] KVM: SVM: Inhibit AVIC if ID is too big instead of rejecting vCPU creation Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 11/59] KVM: SVM: Drop redundant check in AVIC code on ID during " Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 12/59] KVM: SVM: Track AVIC tables as natively sized pointers, not "struct pages" Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 13/59] KVM: SVM: Drop superfluous "cache" of AVIC Physical ID entry pointer Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 14/59] KVM: VMX: Move enable_ipiv knob to common x86 Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 15/59] KVM: SVM: Add enable_ipiv param, never set IsRunning if disabled Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 16/59] KVM: SVM: Disable (x2)AVIC IPI virtualization if CPU has erratum #1235 Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 17/59] KVM: VMX: Suppress PI notifications whenever the vCPU is put Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 18/59] KVM: SVM: Add a comment to explain why avic_vcpu_blocking() ignores IRQ blocking Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 19/59] iommu/amd: KVM: SVM: Use pi_desc_addr to derive ga_root_ptr Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 20/59] iommu/amd: KVM: SVM: Pass NULL @vcpu_info to indicate "not guest mode" Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 21/59] KVM: SVM: Stop walking list of routing table entries when updating IRTE Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 22/59] KVM: VMX: " Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 23/59] KVM: SVM: Extract SVM specific code out of get_pi_vcpu_info() Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 24/59] KVM: x86: Move IRQ routing/delivery APIs from x86.c => irq.c Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 25/59] KVM: x86: Nullify irqfd->producer after updating IRTEs Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 26/59] KVM: x86: Dedup AVIC vs. PI code for identifying target vCPU Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 27/59] KVM: x86: Move posted interrupt tracepoint to common code Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 28/59] KVM: SVM: Clean up return handling in avic_pi_update_irte() Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 29/59] iommu: KVM: Split "struct vcpu_data" into separate AMD vs. Intel structs Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 30/59] KVM: Don't WARN if updating IRQ bypass route fails Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 31/59] KVM: Fold kvm_arch_irqfd_route_changed() into kvm_arch_update_irqfd_routing() Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 32/59] KVM: x86: Track irq_bypass_vcpu in common x86 code Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 33/59] KVM: x86: Skip IOMMU IRTE updates if there's no old or new vCPU being targeted Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 34/59] KVM: x86: Don't update IRTE entries when old and new routes were !MSI Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 35/59] KVM: SVM: Revert IRTE to legacy mode if IOMMU doesn't provide IR metadata Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 36/59] KVM: SVM: Take and hold ir_list_lock across IRTE updates in IOMMU Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 37/59] iommu/amd: Document which IRTE fields amd_iommu_update_ga() can modify Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 38/59] iommu/amd: KVM: SVM: Infer IsRun from validity of pCPU destination Sean Christopherson
2025-05-30 10:13 ` Sairaj Kodilkar
2025-06-02 21:59 ` Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 39/59] iommu/amd: Factor out helper for manipulating IRTE GA/CPU info Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 40/59] iommu/amd: KVM: SVM: Set pCPU info in IRTE when setting vCPU affinity Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 41/59] iommu/amd: KVM: SVM: Add IRTE metadata to affined vCPU's list if AVIC is inhibited Sean Christopherson
2025-05-30 10:06 ` Sairaj Kodilkar
2025-06-02 22:26 ` Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 42/59] KVM: SVM: Don't check for assigned device(s) when updating affinity Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 43/59] KVM: SVM: Don't check for assigned device(s) when activating AVIC Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 44/59] KVM: SVM: WARN if (de)activating guest mode in IOMMU fails Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 45/59] KVM: SVM: Process all IRTEs on affinity change even if one update fails Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 46/59] KVM: SVM: WARN if updating IRTE GA fields in IOMMU fails Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 47/59] KVM: x86: Drop superfluous "has assigned device" check in kvm_pi_update_irte() Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 48/59] KVM: x86: WARN if IRQ bypass isn't supported " Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 49/59] KVM: x86: WARN if IRQ bypass routing is updated without in-kernel local APIC Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 50/59] KVM: SVM: WARN if ir_list is non-empty at vCPU free Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 51/59] KVM: x86: Decouple device assignment from IRQ bypass Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 52/59] KVM: VMX: WARN if VT-d Posted IRQs aren't possible when starting " Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 53/59] KVM: SVM: Use vcpu_idx, not vcpu_id, for GA log tag/metadata Sean Christopherson
2025-05-23 0:59 ` [PATCH v2 54/59] iommu/amd: WARN if KVM calls GA IRTE helpers without virtual APIC support Sean Christopherson
2025-05-23 1:00 ` [PATCH v2 55/59] KVM: SVM: Fold avic_set_pi_irte_mode() into its sole caller Sean Christopherson
2025-05-23 1:00 ` [PATCH v2 56/59] KVM: SVM: Don't check vCPU's blocking status when toggling AVIC on/off Sean Christopherson
2025-05-23 1:00 ` [PATCH v2 57/59] KVM: SVM: Consolidate IRTE update " Sean Christopherson
2025-05-23 1:00 ` [PATCH v2 58/59] iommu/amd: KVM: SVM: Allow KVM to control need for GA log interrupts Sean Christopherson
2025-05-23 1:00 ` [PATCH v2 59/59] KVM: SVM: Generate GA log IRQs only if the associated vCPUs is blocking Sean Christopherson
2025-06-04 17:11 ` [PATCH v2 00/59] KVM: iommu: Overhaul device posted IRQs support Paolo Bonzini
2025-06-09 12:20 ` Sairaj Kodilkar [this message]
2025-06-09 14:34 ` Sean Christopherson
2025-06-12 11:03 ` Sairaj Kodilkar
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=20250609122050.28499-1-sarunkod@amd.com \
--to=sarunkod@amd.com \
--cc=baolu.lu@linux.intel.com \
--cc=dmatlack@google.com \
--cc=dwmw2@infradead.org \
--cc=francescolavra.fl@gmail.com \
--cc=iommu@lists.linux.dev \
--cc=joao.m.martins@oracle.com \
--cc=joro@8bytes.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mlevitsk@redhat.com \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=vasant.hegde@amd.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