Xen 4.17.4-pre-patchset-14.17.4-pre-patchset-14.17

Signed-off-by: Tomáš Mózes <hydrapolic@gmail.com>

1 files changed, 86 insertions, 0 deletions

diff --git a/0013-x86-hvm-Fix-fast-singlestep-state-persistence.patch b/0013-x86-hvm-Fix-fast-singlestep-state-persistence.patch
new file mode 100644
index 0000000..2a14354
--- /dev/null
+++ b/0013-x86-hvm-Fix-fast-singlestep-state-persistence.patch
@@ -0,0 +1,86 @@
+From 415f770d23f9fcbc02436560fa6583dcd8e1343f Mon Sep 17 00:00:00 2001
+From: =?UTF-8?q?Petr=20Bene=C5=A1?= <w1benny@gmail.com>
+Date: Tue, 27 Feb 2024 14:07:45 +0100
+Subject: [PATCH 13/67] x86/hvm: Fix fast singlestep state persistence
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+This patch addresses an issue where the fast singlestep setting would persist
+despite xc_domain_debug_control being called with XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_OFF.
+Specifically, if fast singlestep was enabled in a VMI session and that session
+stopped before the MTF trap occurred, the fast singlestep setting remained
+active even though MTF itself was disabled.  This led to a situation where, upon
+starting a new VMI session, the first event to trigger an EPT violation would
+cause the corresponding EPT event callback to be skipped due to the lingering
+fast singlestep setting.
+
+The fix ensures that the fast singlestep setting is properly reset when
+disabling single step debugging operations.
+
+Signed-off-by: Petr Beneš <w1benny@gmail.com>
+Reviewed-by: Tamas K Lengyel <tamas@tklengyel.com>
+master commit: 897def94b56175ce569673a05909d2f223e1e749
+master date: 2024-02-12 09:37:58 +0100
+---
+ xen/arch/x86/hvm/hvm.c | 34 ++++++++++++++++++++++++----------
+ 1 file changed, 24 insertions(+), 10 deletions(-)
+
+diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c
+index d6c6ab8897..558dc3eddc 100644
+--- a/xen/arch/x86/hvm/hvm.c
++++ b/xen/arch/x86/hvm/hvm.c
+@@ -5153,26 +5153,40 @@ long do_hvm_op(unsigned long op, XEN_GUEST_HANDLE_PARAM(void) arg)
+ 
+ int hvm_debug_op(struct vcpu *v, int32_t op)
+ {
+-    int rc;
++    int rc = 0;
+ 
+     switch ( op )
+     {
+         case XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_ON:
+         case XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_OFF:
+-            rc = -EOPNOTSUPP;
+             if ( !cpu_has_monitor_trap_flag )
+-                break;
+-            rc = 0;
+-            vcpu_pause(v);
+-            v->arch.hvm.single_step =
+-                (op == XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_ON);
+-            vcpu_unpause(v); /* guest will latch new state */
++                return -EOPNOTSUPP;
+             break;
+         default:
+-            rc = -ENOSYS;
+-            break;
++            return -ENOSYS;
++    }
++
++    vcpu_pause(v);
++
++    switch ( op )
++    {
++    case XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_ON:
++        v->arch.hvm.single_step = true;
++        break;
++
++    case XEN_DOMCTL_DEBUG_OP_SINGLE_STEP_OFF:
++        v->arch.hvm.single_step = false;
++        v->arch.hvm.fast_single_step.enabled = false;
++        v->arch.hvm.fast_single_step.p2midx = 0;
++        break;
++
++    default: /* Excluded above */
++        ASSERT_UNREACHABLE();
++        return -ENOSYS;
+     }
+ 
++    vcpu_unpause(v); /* guest will latch new state */
++
+     return rc;
+ }
+ 
+-- 
+2.44.0
+