mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Avi Kivity <avi@qumranet.com>
To: linux-kernel@vger.kernel.org, kvm-devel@lists.sourceforge.net
Subject: [PATCH 16/54] KVM: x86 emulator: Move rep processing before instruction execution
Date: Tue,  1 Jan 2008 17:35:42 +0200	[thread overview]
Message-ID: <1199201780-14001-17-git-send-email-avi@qumranet.com> (raw)
In-Reply-To: <1199201780-14001-1-git-send-email-avi@qumranet.com>

Currently rep processing is handled somewhere in the middle of instruction
processing.  Move it to a sensible place.

Signed-off-by: Avi Kivity <avi@qumranet.com>
---
 drivers/kvm/x86_emulate.c |   70 +++++++++++++++++++++++----------------------
 1 files changed, 36 insertions(+), 34 deletions(-)

diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c
index 84e5366..53377f0 100644
--- a/drivers/kvm/x86_emulate.c
+++ b/drivers/kvm/x86_emulate.c
@@ -64,6 +64,7 @@
 #define Mov         (1<<7)
 #define BitOp       (1<<8)
 #define MemAbs      (1<<9)      /* Memory operand is absolute displacement */
+#define String      (1<<10)     /* String instruction (rep capable) */
 
 static u16 opcode_table[256] = {
 	/* 0x00 - 0x07 */
@@ -133,12 +134,12 @@ static u16 opcode_table[256] = {
 	/* 0xA0 - 0xA7 */
 	ByteOp | DstReg | SrcMem | Mov | MemAbs, DstReg | SrcMem | Mov | MemAbs,
 	ByteOp | DstMem | SrcReg | Mov | MemAbs, DstMem | SrcReg | Mov | MemAbs,
-	ByteOp | ImplicitOps | Mov, ImplicitOps | Mov,
-	ByteOp | ImplicitOps, ImplicitOps,
+	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | String, ImplicitOps | String,
 	/* 0xA8 - 0xAF */
-	0, 0, ByteOp | ImplicitOps | Mov, ImplicitOps | Mov,
-	ByteOp | ImplicitOps | Mov, ImplicitOps | Mov,
-	ByteOp | ImplicitOps, ImplicitOps,
+	0, 0, ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | Mov | String, ImplicitOps | Mov | String,
+	ByteOp | ImplicitOps | String, ImplicitOps | String,
 	/* 0xB0 - 0xBF */
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 	/* 0xC0 - 0xC7 */
@@ -1228,6 +1229,36 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops)
 	if (((c->d & ModRM) && (c->modrm_mod != 3)) || (c->d & MemAbs))
 		memop = c->modrm_ea;
 
+	if (c->rep_prefix && (c->d & String)) {
+		/* All REP prefixes have the same first termination condition */
+		if (c->regs[VCPU_REGS_RCX] == 0) {
+			ctxt->vcpu->rip = c->eip;
+			goto done;
+		}
+		/* The second termination condition only applies for REPE
+		 * and REPNE. Test if the repeat string operation prefix is
+		 * REPE/REPZ or REPNE/REPNZ and if it's the case it tests the
+		 * corresponding termination condition according to:
+		 * 	- if REPE/REPZ and ZF = 0 then done
+		 * 	- if REPNE/REPNZ and ZF = 1 then done
+		 */
+		if ((c->b == 0xa6) || (c->b == 0xa7) ||
+				(c->b == 0xae) || (c->b == 0xaf)) {
+			if ((c->rep_prefix == REPE_PREFIX) &&
+				((ctxt->eflags & EFLG_ZF) == 0)) {
+					ctxt->vcpu->rip = c->eip;
+					goto done;
+			}
+			if ((c->rep_prefix == REPNE_PREFIX) &&
+				((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) {
+				ctxt->vcpu->rip = c->eip;
+				goto done;
+			}
+		}
+		c->regs[VCPU_REGS_RCX]--;
+		c->eip = ctxt->vcpu->rip;
+	}
+
 	if (c->src.type == OP_MEM) {
 		c->src.ptr = (unsigned long *)memop;
 		c->src.val = 0;
@@ -1534,35 +1565,6 @@ special_insn:
 		c->dst.type = OP_NONE;	/* Disable writeback. */
 		break;
 	}
-	if (c->rep_prefix) {
-		/* All REP prefixes have the same first termination condition */
-		if (c->regs[VCPU_REGS_RCX] == 0) {
-			ctxt->vcpu->rip = c->eip;
-			goto done;
-		}
-		/* The second termination condition only applies for REPE
-		 * and REPNE. Test if the repeat string operation prefix is
-		 * REPE/REPZ or REPNE/REPNZ and if it's the case it tests the
-		 * corresponding termination condition according to:
-		 * 	- if REPE/REPZ and ZF = 0 then done
-		 * 	- if REPNE/REPNZ and ZF = 1 then done
-		 */
-		if ((c->b == 0xa6) || (c->b == 0xa7) ||
-				(c->b == 0xae) || (c->b == 0xaf)) {
-			if ((c->rep_prefix == REPE_PREFIX) &&
-				((ctxt->eflags & EFLG_ZF) == 0)) {
-					ctxt->vcpu->rip = c->eip;
-					goto done;
-			}
-			if ((c->rep_prefix == REPNE_PREFIX) &&
-				((ctxt->eflags & EFLG_ZF) == EFLG_ZF)) {
-				ctxt->vcpu->rip = c->eip;
-				goto done;
-			}
-		}
-		c->regs[VCPU_REGS_RCX]--;
-		c->eip = ctxt->vcpu->rip;
-	}
 	switch (c->b) {
 	case 0xa4 ... 0xa5:	/* movs */
 		c->dst.type = OP_MEM;
-- 
1.5.3.7


  parent reply	other threads:[~2008-01-01 15:41 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-01-01 15:35 [PATCH 00/54] KVM patch queue review for 2.6.25 merge window (part IV) Avi Kivity
2008-01-01 15:35 ` [PATCH 01/54] KVM: MMU: Remove gva_to_hpa() Avi Kivity
2008-01-01 15:35 ` [PATCH 02/54] KVM: Remove gpa_to_hpa() Avi Kivity
2008-01-01 15:35 ` [PATCH 03/54] KVM: MMU: Rename variables of type 'struct kvm_mmu_page *' Avi Kivity
2008-01-01 15:35 ` [PATCH 04/54] KVM: MMU: Rename 'release_page' Avi Kivity
2008-01-01 15:35 ` [PATCH 05/54] KVM: Disallow fork() and similar games when using a VM Avi Kivity
2008-01-01 15:35 ` [PATCH 06/54] KVM: Enhance guest cpuid management Avi Kivity
2008-01-01 15:35 ` [PATCH 07/54] KVM: VMX: Remove the secondary execute control dependency on irqchip Avi Kivity
2008-01-01 15:35 ` [PATCH 08/54] KVM: Portability: Move unalias_gfn to arch dependent file Avi Kivity
2008-01-01 15:35 ` [PATCH 09/54] KVM: x86 emulator: Make a distinction between repeat prefixes F3 and F2 Avi Kivity
2008-01-01 15:35 ` [PATCH 10/54] KVM: x86 emulator: address size and operand size overrides are sticky Avi Kivity
2008-01-01 15:35 ` [PATCH 11/54] KVM: Remove misleading check for mmio during event injection Avi Kivity
2008-01-01 15:35 ` [PATCH 12/54] KVM: MMU: mark pages that were inserted to the shadow pages table as accessed Avi Kivity
2008-01-01 15:35 ` [PATCH 13/54] KVM: x86 emulator: Rename 'cr2' to 'memop' Avi Kivity
2008-01-01 15:35 ` [PATCH 14/54] KVM: x86 emulator: cmps instruction Avi Kivity
2008-01-01 15:35 ` [PATCH 15/54] KVM: Add ifdef in irqchip struct for x86 only structures Avi Kivity
2008-01-01 15:35 ` Avi Kivity [this message]
2008-01-01 15:35 ` [PATCH 17/54] KVM: x86 emulator: unify two switches Avi Kivity
2008-01-01 15:35 ` [PATCH 18/54] KVM: x86 emulator: unify four switch statements into two Avi Kivity
2008-01-01 15:35 ` [PATCH 19/54] KVM: Portability: Move KVM_INTERRUPT vcpu ioctl to x86.c Avi Kivity
2008-01-01 15:35 ` [PATCH 20/54] KVM: Correct kvm_init() error paths not freeing bad_pge Avi Kivity
2008-01-01 15:35 ` [PATCH 21/54] KVM: Export include/linux/kvm.h only if $ARCH actually supports KVM Avi Kivity
2008-01-01 16:22   ` Sam Ravnborg
2008-01-01 16:35     ` Avi Kivity
2008-01-01 15:35 ` [PATCH 22/54] KVM: SVM: Remove KVM specific defines for MSR_EFER Avi Kivity
2008-01-01 15:35 ` [PATCH 23/54] KVM: Replace kvm_lapic with kvm_vcpu in ioapic/lapic interface Avi Kivity
2008-01-01 15:35 ` [PATCH 24/54] KVM: Replace dest_Lowest_Prio and dest_Fixed with self-defined macros Avi Kivity
2008-01-01 15:35 ` [PATCH 25/54] KVM: Extend ioapic code to support iosapic Avi Kivity
2008-01-01 15:35 ` [PATCH 26/54] KVM: Portability: Move address types to their own header file Avi Kivity
2008-01-01 15:35 ` [PATCH 27/54] KVM: Portability: Move IO device definitions to its " Avi Kivity
2008-01-01 15:35 ` [PATCH 28/54] KVM: Portability: Stop including x86-specific headers in kvm_main.c Avi Kivity
2008-01-01 15:35 ` [PATCH 29/54] KVM: Portability: Create kvm_arch_vcpu_runnable() function Avi Kivity
2008-01-01 15:35 ` [PATCH 30/54] KVM: Convert KVM from ->nopage() to ->fault() Avi Kivity
2008-01-01 15:35 ` [PATCH 31/54] KVM: MMU: Remove unused prev_shadow_ent variable from fetch() Avi Kivity
2008-01-01 15:35 ` [PATCH 32/54] KVM: Generalize exception injection mechanism Avi Kivity
2008-01-01 15:35 ` [PATCH 33/54] KVM: Replace page fault injection by the generalized exception queue Avi Kivity
2008-01-01 15:36 ` [PATCH 34/54] KVM: Replace #GP " Avi Kivity
2008-01-01 15:36 ` [PATCH 35/54] KVM: Use generalized exception queue for injecting #UD Avi Kivity
2008-01-01 15:36 ` [PATCH 36/54] KVM: x86 emulator: fix eflags preparation for emulation Avi Kivity
2008-01-01 15:36 ` [PATCH 37/54] KVM: VMX: Avoid exit when setting cr8 if the local apic is in the kernel Avi Kivity
2008-01-01 15:36 ` [PATCH 38/54] KVM: SVM: Emulate read/write access to cr8 Avi Kivity
2008-01-01 15:36 ` [PATCH 39/54] KVM: x86 emulator: Fix stack instructions on 64-bit mode Avi Kivity
2008-01-01 15:36 ` [PATCH 40/54] KVM: SVM: Trap access to the cr8 register Avi Kivity
2008-01-01 15:36 ` [PATCH 41/54] KVM: MMU: Use cmpxchg for pte updates on walk_addr() Avi Kivity
2008-01-01 15:36 ` [PATCH 42/54] KVM: MMU: Simplify calculation of pte access Avi Kivity
2008-01-01 15:36 ` [PATCH 43/54] KVM: MMU: Set nx bit correctly on shadow ptes Avi Kivity
2008-01-01 15:36 ` [PATCH 44/54] KVM: MMU: Move pte access calculation into a helper function Avi Kivity
2008-01-01 15:36 ` [PATCH 45/54] KVM: MMU: Fix inherited permissions for emulated guest pte updates Avi Kivity
2008-01-01 15:36 ` [PATCH 46/54] KVM: MMU: No need to pick up nx bit from guest pte Avi Kivity
2008-01-01 15:36 ` [PATCH 47/54] KVM: MMU: Pass pte dirty flag to set_pte() instead of calculating it on-site Avi Kivity
2008-01-01 15:36 ` [PATCH 48/54] KVM: MMU: Remove walker argument to set_pte() Avi Kivity
2008-01-01 15:36 ` [PATCH 49/54] KVM: MMU: Move set_pte() into guest paging mode independent code Avi Kivity
2008-01-01 15:36 ` [PATCH 50/54] KVM: MMU: Adjust mmu_set_spte() debug code for gpte removal Avi Kivity
2008-01-01 15:36 ` [PATCH 51/54] KVM: MMU: Use mmu_set_spte() for real-mode shadows Avi Kivity
2008-01-01 15:36 ` [PATCH 52/54] KVM: SVM: Exit to userspace if write to cr8 and not using in-kernel apic Avi Kivity
2008-01-01 15:36 ` [PATCH 53/54] KVM: MMU: Fix SMP shadow instantiation race Avi Kivity
2008-01-01 15:36 ` [PATCH 54/54] KVM: LAPIC: minor debugging compile fix Avi Kivity

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=1199201780-14001-17-git-send-email-avi@qumranet.com \
    --to=avi@qumranet.com \
    --cc=kvm-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    /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

all inboxes | Powered by JetHome®