From: Paolo Bonzini <pbonzini@redhat.com>
To: Sean Christopherson <seanjc@google.com>
Cc: kvm@vger.kernel.org
Subject: Re: [kvm-unit-tests PATCH 00/39] x86/access: nVMX: Big overhaul
Date: Fri, 26 Nov 2021 19:43:12 +0100 [thread overview]
Message-ID: <34ff357d-c073-4a68-117d-63ccff1085cb@redhat.com> (raw)
In-Reply-To: <20211125012857.508243-1-seanjc@google.com>
On 11/25/21 02:28, Sean Christopherson wrote:
> This started out as a very simple test (patch 39/39) to expose a KVM bug
> where KVM doesn't sync a shadow MMU on a vmcs12->vpid change. Except the
> test didn't fail. And it turns out, completely removing INVLPG from the
> base access test doesn't fail when using shadow paging either.
>
> The underlying problem in both cases is that the access test is flat out
> stupid when it comes to handling page tables. Instead of allocating page
> tables once and manipulating them on each iteration, it "allocates" a new
> paging structure when necessary on every. single. iteration. In addition
> to being incredibly inefficient (allocation also zeros the entire 4kb page,
> so the test zeros absurd amounts of memory), writing upper level PTEs on
> every iteration triggers write-protection mechanisms in KVM. In effect,
> KVM ends up synchronizing the relevant SPTEs on every iteration, which
> again is ridiculously slow and makes it all but impossible to actually
> test that KVM handles other TLB invalidation scenarios.
>
> Trying to solve that mess by pre-allocating the page tables exposed a
> whole pile of 5-level paging issues. I'd say the test's 5-level support
> is held together by duct tape, but I've fixed many things with duct tape
> that are far less fragile.
>
> The second half of this series is cleanups in the nVMX code to prepare
> for adding the (INV)VPID variants. Not directly related to the access
> tests, but it annoyed me to no end that simply checking if INVVPID is
> supported was non-trivial.
Queued, thanks. The new tests are pretty slow on debug kernels (about 3
minutes each). I'll check next week if there's any low hanging
fruit---or anything broken.
Paolo
> Sean Christopherson (39):
> x86/access: Add proper defines for hardcoded addresses
> x86/access: Cache CR3 to improve performance
> x86/access: Use do-while loop for what is obviously a do-while loop
> x86/access: Stop pretending the test is SMP friendly
> x86/access: Refactor so called "page table pool" logic
> x86/access: Stash root page table level in test environment
> x86/access: Hoist page table allocator helpers above "init" helper
> x86/access: Rename variables in page table walkers
> x86/access: Abort if page table insertion hits an unexpected level
> x86/access: Make SMEP place nice with 5-level paging
> x86/access: Use upper half of virtual address space
> x86/access: Print the index when dumping PTEs
> x86/access: Pre-allocate all page tables at (sub)test init
> x86/access: Don't write page tables if desired PTE is same as current
> PTE
> x86/access: Preserve A/D bits when writing paging structure entries
> x86/access: Make toggling of PRESENT bit a "higher order" action
> x86/access: Manually override PMD in effective permissions sub-test
> x86/access: Remove manual override of PUD/PMD in prefetch sub-test
> x86/access: Remove PMD/PT target overrides
> x86/access: Remove timeout overrides now that performance doesn't suck
> nVMX: Skip EPT tests if INVEPT(SINGLE_CONTEXT) is unsupported
> nVMX: Hoist assert macros to the top of vmx.h
> nVMX: Add a non-reporting assertion macro
> nVMX: Assert success in unchecked INVEPT/INVVPID helpers
> nVMX: Drop less-than-useless ept_sync() wrapper
> nVMX: Move EPT capability check helpers to vmx.h
> nVMX: Drop unused and useless vpid_sync() helper
> nVMX: Remove "v1" version of INVVPID test
> nVMX: Add helper to check if INVVPID type is supported
> nVMX: Add helper to check if INVVPID is supported
> nVMX: Add helper to get first supported INVVPID type
> nVMX: Use helper to check for EPT A/D support
> nVMX: Add helpers to check for 4/5-level EPT support
> nVMX: Fix name of macro defining EPT execute only capability
> nVMX: Add helper to check if a memtype is supported for EPT structures
> nVMX: Get rid of horribly named "ctrl" boolean in test_ept_eptp()
> nVMX: Rename awful "ctrl" booleans to "is_ctrl_valid"
> nVMX: Add helper to check if VPID is supported
> x86/access: nVMX: Add "access" test variants to invalidate via
> (INV)VPID
>
> x86/access.c | 391 ++++++++++++++++++++++++++++------------------
> x86/unittests.cfg | 10 +-
> x86/vmx.c | 71 +--------
> x86/vmx.h | 229 ++++++++++++++++++---------
> x86/vmx_tests.c | 327 +++++++++++++++++---------------------
> 5 files changed, 543 insertions(+), 485 deletions(-)
>
next prev parent reply other threads:[~2021-11-26 19:26 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-25 1:28 [kvm-unit-tests PATCH 00/39] x86/access: nVMX: Big overhaul Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 01/39] x86/access: Add proper defines for hardcoded addresses Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 02/39] x86/access: Cache CR3 to improve performance Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 03/39] x86/access: Use do-while loop for what is obviously a do-while loop Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 04/39] x86/access: Stop pretending the test is SMP friendly Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 05/39] x86/access: Refactor so called "page table pool" logic Sean Christopherson
2021-11-26 18:03 ` Paolo Bonzini
2021-11-25 1:28 ` [kvm-unit-tests PATCH 06/39] x86/access: Stash root page table level in test environment Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 07/39] x86/access: Hoist page table allocator helpers above "init" helper Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 08/39] x86/access: Rename variables in page table walkers Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 09/39] x86/access: Abort if page table insertion hits an unexpected level Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 10/39] x86/access: Make SMEP place nice with 5-level paging Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 11/39] x86/access: Use upper half of virtual address space Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 12/39] x86/access: Print the index when dumping PTEs Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 13/39] x86/access: Pre-allocate all page tables at (sub)test init Sean Christopherson
2021-11-26 18:15 ` Paolo Bonzini
2021-11-25 1:28 ` [kvm-unit-tests PATCH 14/39] x86/access: Don't write page tables if desired PTE is same as current PTE Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 15/39] x86/access: Preserve A/D bits when writing paging structure entries Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 16/39] x86/access: Make toggling of PRESENT bit a "higher order" action Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 17/39] x86/access: Manually override PMD in effective permissions sub-test Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 18/39] x86/access: Remove manual override of PUD/PMD in prefetch sub-test Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 19/39] x86/access: Remove PMD/PT target overrides Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 20/39] x86/access: Remove timeout overrides now that performance doesn't suck Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 21/39] nVMX: Skip EPT tests if INVEPT(SINGLE_CONTEXT) is unsupported Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 22/39] nVMX: Hoist assert macros to the top of vmx.h Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 23/39] nVMX: Add a non-reporting assertion macro Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 24/39] nVMX: Assert success in unchecked INVEPT/INVVPID helpers Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 25/39] nVMX: Drop less-than-useless ept_sync() wrapper Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 26/39] nVMX: Move EPT capability check helpers to vmx.h Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 27/39] nVMX: Drop unused and useless vpid_sync() helper Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 28/39] nVMX: Remove "v1" version of INVVPID test Sean Christopherson
2021-11-26 18:28 ` Paolo Bonzini
2021-11-25 1:28 ` [kvm-unit-tests PATCH 29/39] nVMX: Add helper to check if INVVPID type is supported Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 30/39] nVMX: Add helper to check if INVVPID " Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 31/39] nVMX: Add helper to get first supported INVVPID type Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 32/39] nVMX: Use helper to check for EPT A/D support Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 33/39] nVMX: Add helpers to check for 4/5-level EPT support Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 34/39] nVMX: Fix name of macro defining EPT execute only capability Sean Christopherson
2021-11-26 18:31 ` Paolo Bonzini
2021-11-25 1:28 ` [kvm-unit-tests PATCH 35/39] nVMX: Add helper to check if a memtype is supported for EPT structures Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 36/39] nVMX: Get rid of horribly named "ctrl" boolean in test_ept_eptp() Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 37/39] nVMX: Rename awful "ctrl" booleans to "is_ctrl_valid" Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 38/39] nVMX: Add helper to check if VPID is supported Sean Christopherson
2021-11-25 1:28 ` [kvm-unit-tests PATCH 39/39] x86/access: nVMX: Add "access" test variants to invalidate via (INV)VPID Sean Christopherson
2021-11-26 18:43 ` Paolo Bonzini [this message]
2021-11-29 19:04 ` [kvm-unit-tests PATCH 00/39] x86/access: nVMX: Big overhaul Sean Christopherson
2021-11-29 19:15 ` Paolo Bonzini
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=34ff357d-c073-4a68-117d-63ccff1085cb@redhat.com \
--to=pbonzini@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=seanjc@google.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;
as well as URLs for NNTP newsgroup(s).