[PATCH v2 23/23] x86/pv: Adjust eflags handling for FRED mode

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

ERETU, unlike IRET, requires the sticky-1 bit (bit 2) be set, and reserved
bits to be clear.  Notably this means that dom0_construct() must set
X86_EFLAGS_MBS it in order for a PV dom0 to start.

Adjust arch_set_info_guest*() and hypercall_iret() which consume flags to
clamp the reserved bits.

This is a minor ABI change, but by the same argument as commit
9f892f84c279 ("x86/domctl: Stop using XLAT_cpu_user_regs()"), this change will
happen naturally when the vCPU schedules.

Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
---
CC: Jan Beulich <JBeulich@xxxxxxxx>
CC: Roger Pau Monné <roger.pau@xxxxxxxxxx>

v2:
 * New

The handling of VM is complicated.

It turns out that it's simply ignored by IRET in Long Mode (i.e. clearing it
commit 0e47f92b0725 ("x86: force EFLAGS.IF on when exiting to PV guests")
wasn't actually necessary) but ERETU does care.

But, it's unclear how to handle this in in arch_set_info().  We must preserve
it for HVM guests (whih can use vm86 mode).  PV32 has special handling but
only in hypercall_iret(), not in arch_set_info().
---
 xen/arch/x86/domain.c                | 4 ++--
 xen/arch/x86/hvm/domain.c            | 4 ++--
 xen/arch/x86/include/asm/x86-defns.h | 7 +++++++
 xen/arch/x86/pv/dom0_build.c         | 2 +-
 xen/arch/x86/pv/iret.c               | 8 +++++---
 5 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c
index 64922869a625..c1880324f7a9 100644
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
@@ -1273,7 +1273,7 @@ int arch_set_info_guest(
         v->arch.user_regs.rax               = c.nat->user_regs.rax;
         v->arch.user_regs.rip               = c.nat->user_regs.rip;
         v->arch.user_regs.cs                = c.nat->user_regs.cs;
-        v->arch.user_regs.rflags            = c.nat->user_regs.rflags;
+        v->arch.user_regs.rflags            = (c.nat->user_regs.rflags & 
X86_EFLAGS_ALL) | X86_EFLAGS_MBS;
         v->arch.user_regs.rsp               = c.nat->user_regs.rsp;
         v->arch.user_regs.ss                = c.nat->user_regs.ss;
         v->arch.pv.es                       = c.nat->user_regs.es;
@@ -1297,7 +1297,7 @@ int arch_set_info_guest(
         v->arch.user_regs.eax               = c.cmp->user_regs.eax;
         v->arch.user_regs.eip               = c.cmp->user_regs.eip;
         v->arch.user_regs.cs                = c.cmp->user_regs.cs;
-        v->arch.user_regs.eflags            = c.cmp->user_regs.eflags;
+        v->arch.user_regs.eflags            = (c.cmp->user_regs.eflags & 
X86_EFLAGS_ALL) | X86_EFLAGS_MBS;
         v->arch.user_regs.esp               = c.cmp->user_regs.esp;
         v->arch.user_regs.ss                = c.cmp->user_regs.ss;
         v->arch.pv.es                       = c.cmp->user_regs.es;
diff --git a/xen/arch/x86/hvm/domain.c b/xen/arch/x86/hvm/domain.c
index 048f29ae4911..1e874d598952 100644
--- a/xen/arch/x86/hvm/domain.c
+++ b/xen/arch/x86/hvm/domain.c
@@ -194,7 +194,7 @@ int arch_set_info_hvm_guest(struct vcpu *v, const struct 
vcpu_hvm_context *ctx)
         uregs->rsi    = regs->esi;
         uregs->rdi    = regs->edi;
         uregs->rip    = regs->eip;
-        uregs->rflags = regs->eflags;
+        uregs->rflags = (regs->eflags & X86_EFLAGS_ALL) | X86_EFLAGS_MBS;
 
         v->arch.hvm.guest_cr[0] = regs->cr0;
         v->arch.hvm.guest_cr[3] = regs->cr3;
@@ -245,7 +245,7 @@ int arch_set_info_hvm_guest(struct vcpu *v, const struct 
vcpu_hvm_context *ctx)
         uregs->rsi    = regs->rsi;
         uregs->rdi    = regs->rdi;
         uregs->rip    = regs->rip;
-        uregs->rflags = regs->rflags;
+        uregs->rflags = (regs->rflags & X86_EFLAGS_ALL) | X86_EFLAGS_MBS;
 
         v->arch.hvm.guest_cr[0] = regs->cr0;
         v->arch.hvm.guest_cr[3] = regs->cr3;
diff --git a/xen/arch/x86/include/asm/x86-defns.h 
b/xen/arch/x86/include/asm/x86-defns.h
index 0a0ba83de786..edeb0b4ff95a 100644
--- a/xen/arch/x86/include/asm/x86-defns.h
+++ b/xen/arch/x86/include/asm/x86-defns.h
@@ -27,6 +27,13 @@
     (X86_EFLAGS_CF | X86_EFLAGS_PF | X86_EFLAGS_AF |   \
      X86_EFLAGS_ZF | X86_EFLAGS_SF | X86_EFLAGS_OF)
 
+#define X86_EFLAGS_ALL                                          \
+    (X86_EFLAGS_ARITH_MASK | X86_EFLAGS_TF | X86_EFLAGS_IF |    \
+     X86_EFLAGS_DF | X86_EFLAGS_OF | X86_EFLAGS_IOPL |          \
+     X86_EFLAGS_NT | X86_EFLAGS_RF | X86_EFLAGS_VM |            \
+     X86_EFLAGS_AC | X86_EFLAGS_VIF | X86_EFLAGS_VIP |          \
+     X86_EFLAGS_ID)
+
 /*
  * Intel CPU flags in CR0
  */
diff --git a/xen/arch/x86/pv/dom0_build.c b/xen/arch/x86/pv/dom0_build.c
index 21158ce1812e..f9bbbea2ff70 100644
--- a/xen/arch/x86/pv/dom0_build.c
+++ b/xen/arch/x86/pv/dom0_build.c
@@ -1021,7 +1021,7 @@ static int __init dom0_construct(const struct boot_domain 
*bd)
     regs->rip = parms.virt_entry;
     regs->rsp = vstack_end;
     regs->rsi = vstartinfo_start;
-    regs->eflags = X86_EFLAGS_IF;
+    regs->eflags = X86_EFLAGS_IF | X86_EFLAGS_MBS;
 
     /*
      * We don't call arch_set_info_guest(), so some initialisation needs doing
diff --git a/xen/arch/x86/pv/iret.c b/xen/arch/x86/pv/iret.c
index d3a1fb2c685b..39ce316b8d91 100644
--- a/xen/arch/x86/pv/iret.c
+++ b/xen/arch/x86/pv/iret.c
@@ -80,8 +80,9 @@ long do_iret(void)
 
     regs->rip    = iret_saved.rip;
     regs->cs     = iret_saved.cs | 3; /* force guest privilege */
-    regs->rflags = ((iret_saved.rflags & ~(X86_EFLAGS_IOPL|X86_EFLAGS_VM))
-                    | X86_EFLAGS_IF);
+    regs->rflags = ((iret_saved.rflags & X86_EFLAGS_ALL &
+                     ~(X86_EFLAGS_IOPL | X86_EFLAGS_VM)) |
+                    X86_EFLAGS_IF | X86_EFLAGS_MBS);
     regs->rsp    = iret_saved.rsp;
     regs->ss     = iret_saved.ss | 3; /* force guest privilege */
 
@@ -143,7 +144,8 @@ int compat_iret(void)
     if ( VM_ASSIST(v->domain, architectural_iopl) )
         v->arch.pv.iopl = eflags & X86_EFLAGS_IOPL;
 
-    regs->eflags = (eflags & ~X86_EFLAGS_IOPL) | X86_EFLAGS_IF;
+    regs->eflags = ((eflags & X86_EFLAGS_ALL & ~X86_EFLAGS_IOPL) |
+                    X86_EFLAGS_IF | X86_EFLAGS_MBS);
 
     if ( unlikely(eflags & X86_EFLAGS_VM) )
     {
-- 
2.39.5