mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Chang S. Bae" <chang.seok.bae@intel.com>
To: seanjc@google.com, jmattson@google.com
Cc: pbonzini@redhat.com, kvm@vger.kernel.org, x86@kernel.org,
	linux-kernel@vger.kernel.org, zhao1.liu@intel.com,
	chang.seok.bae@intel.com
Subject: [PATCH v9 13/20] KVM: x86: Support REX2-prefixed opcode decode
Date: Tue, 17 Nov 2026 18:03:03 +0000	[thread overview]
Message-ID: <20261117180310.921701-14-chang.seok.bae@intel.com> (raw)
In-Reply-To: <20261117180310.921701-1-chang.seok.bae@intel.com>

Extend the instruction decoder to recognize and handle the REX2 prefix,
including validation of prefix sequences and correct opcode table
selection. Prepare the UD opcode flag explicitly with Undefined, instead
of zeroing.

REX2 is a terminal prefix: once 0xD5 is encountered, the following byte
is the opcode. When REX2.M=0, most prefix bytes are invalid after REX2,
including REX, VEX, EVEX, and another REX2. Also, REX2-prefixed
instructions are only valid in 64-bit mode.

All of the invalid prefix combinations after REX2 coincide with opcodes
that are architecturally invalid in 64-bit mode. Thus, marking such
opcodes with No64 in opcode_table[] naturally disallows those illegal
prefix sequences.

The 0x40 - 0x4F opcode row was missing the No64 flag. While NoRex2
already invalidates REX2 for these opcodes, adding No64 makes opcode
attributes explicit and complete.

Note opcodes marked NotImpl alone end up in handle_emulation_failure(),
but that will inject #UD eventually.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: Zhao Liu <zhao1.liu@intel.com>
Signed-off-by: Chang S. Bae <chang.seok.bae@intel.com>
Link: https://lore.kernel.org/CABgObfYYGTvkYpeyqLSr9JgKMDA_STSff2hXBNchLZuKFU+MMA@mail.gmail.com
---
V8 -> V9: Clarify UD vs emul-failure in changelog and UD define (Jim Mattson)
---
 arch/x86/kvm/emulate.c | 40 +++++++++++++++++++++++++++++++++++-----
 1 file changed, 35 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index c6e11f551438..3a587e553f9a 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -3990,7 +3990,7 @@ static int check_perm_out(struct x86_emulate_ctxt *ctxt)
 		I2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock, _e),	\
 		I2bv(((_f) & ~Lock) | DstAcc | SrcImm, _e)
 
-static const struct opcode ud = I(SrcNone, emulate_ud);
+static const struct opcode ud = I(SrcNone | Undefined, emulate_ud);
 
 static const struct opcode group7_rm0[] = {
 	N,
@@ -4272,7 +4272,7 @@ static const struct opcode opcode_table[256] = {
 	/* 0x38 - 0x3F */
 	I6ALU(NoWrite, em_cmp), N, N,
 	/* 0x40 - 0x4F */
-	X8(I(DstReg | NoRex2, em_inc)), X8(I(DstReg | NoRex2, em_dec)),
+	X8(I(DstReg | NoRex2 | No64, em_inc)), X8(I(DstReg | NoRex2 | No64, em_dec)),
 	/* 0x50 - 0x57 */
 	X8(I(SrcReg | Stack, em_push)),
 	/* 0x58 - 0x5F */
@@ -4865,6 +4865,17 @@ static int x86_decode_avx(struct x86_emulate_ctxt *ctxt,
 	return rc;
 }
 
+static inline bool rex2_invalid(struct x86_emulate_ctxt *ctxt)
+{
+	const struct x86_emulate_ops *ops = ctxt->ops;
+	u64 xcr = 0;
+
+	return ctxt->rex_prefix == REX_PREFIX ||
+	       !(ops->get_cr(ctxt, 4) & X86_CR4_OSXSAVE) ||
+	       ops->get_xcr(ctxt, 0, &xcr) ||
+	       !(xcr & XFEATURE_MASK_APX);
+}
+
 int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len, int emulation_type)
 {
 	int rc = X86EMUL_CONTINUE;
@@ -4918,7 +4929,7 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len, int
 	ctxt->op_bytes = def_op_bytes;
 	ctxt->ad_bytes = def_ad_bytes;
 
-	/* Legacy prefixes. */
+	/* Legacy and REX/REX2 prefixes. */
 	for (;;) {
 		switch (ctxt->b = insn_fetch(u8, ctxt)) {
 		case 0x66:	/* operand-size override */
@@ -4964,6 +4975,17 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len, int
 			ctxt->rex_prefix = REX_PREFIX;
 			ctxt->rex_bits   = ctxt->b & 0xf;
 			continue;
+		case 0xd5: /* REX2 */
+			if (mode != X86EMUL_MODE_PROT64)
+				goto done_prefixes;
+			if (rex2_invalid(ctxt)) {
+				opcode = ud;
+				goto done_modrm;
+			}
+			ctxt->rex_prefix = REX2_PREFIX;
+			ctxt->rex_bits   = insn_fetch(u8, ctxt);
+			ctxt->b          = insn_fetch(u8, ctxt);
+			goto done_prefixes;
 		case 0xf0:	/* LOCK */
 			ctxt->lock_prefix = 1;
 			break;
@@ -4986,6 +5008,12 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len, int
 	if (ctxt->rex_bits & REX_W)
 		ctxt->op_bytes = 8;
 
+	/* REX2 opcode is one byte unless M-bit selects the two-byte map */
+	if (ctxt->rex_bits & REX_M)
+		goto decode_twobytes;
+	if (ctxt->rex_prefix == REX2_PREFIX)
+		goto decode_onebyte;
+
 	/* Opcode byte(s). */
 	if (ctxt->b == 0xc4 || ctxt->b == 0xc5) {
 		/* VEX or LDS/LES */
@@ -5003,17 +5031,19 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len, int
 			goto done;
 	} else if (ctxt->b == 0x0f) {
 		/* Two- or three-byte opcode */
-		ctxt->opcode_len = 2;
 		ctxt->b = insn_fetch(u8, ctxt);
+decode_twobytes:
+		ctxt->opcode_len = 2;
 		opcode = twobyte_table[ctxt->b];
 
 		/* 0F_38 opcode map */
-		if (ctxt->b == 0x38) {
+		if (ctxt->b == 0x38 && ctxt->rex_prefix != REX2_PREFIX) {
 			ctxt->opcode_len = 3;
 			ctxt->b = insn_fetch(u8, ctxt);
 			opcode = opcode_map_0f_38[ctxt->b];
 		}
 	} else {
+decode_onebyte:
 		/* Opcode byte(s). */
 		opcode = opcode_table[ctxt->b];
 	}
-- 
2.53.0


  parent reply	other threads:[~2026-11-17 18:32 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-17 18:02 [PATCH v9 00/20] KVM: x86: Enable APX for guest Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 01/20] KVM: x86: Extend VCPU registers for EGPRs Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 02/20] KVM: VMX: Save guest EGPRs in VCPU cache Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 03/20] KVM: x86: Support EGPRs on VCPU RESET/INIT events Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 04/20] KVM: x86: Support APX state for XSAVE ABI Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 05/20] KVM: VMX: Refactor VMX instruction information access Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 06/20] KVM: VMX: Refactor instruction information decoding Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 07/20] KVM: VMX: Remove unused control-register access defines Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 08/20] KVM: VMX: Refactor register index retrieval from exit qualification Chang S. Bae
2026-11-17 18:02 ` [PATCH v9 09/20] KVM: VMX: Support instruction information extension Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 10/20] KVM: nVMX: Propagate extended instruction information Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 11/20] KVM: x86: Support EGPR accessing and tracking for emulator Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 12/20] KVM: x86: Handle EGPR index and REX2-incompatible opcodes Chang S. Bae
2026-11-17 18:03 ` Chang S. Bae [this message]
2026-11-17 18:03 ` [PATCH v9 14/20] KVM: x86: Document EVEX-prefix not implemented Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 15/20] KVM: x86: Move KVM_SUPPORTED_{XCR0,XSS} into kvm_x86_vendor_init() Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 16/20] KVM: x86: Guard valid XCR0.APX settings Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 17/20] KVM: x86: Add APX to supported XCR0 Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 18/20] KVM: x86: Expose APX feature to userspace Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 19/20] KVM: selftests: Add APX state and ABI test Chang S. Bae
2026-11-17 18:03 ` [PATCH v9 20/20] KVM: selftests: Add APX state handling and XCR0 sanity checks Chang S. Bae

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=20261117180310.921701-14-chang.seok.bae@intel.com \
    --to=chang.seok.bae@intel.com \
    --cc=jmattson@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=x86@kernel.org \
    --cc=zhao1.liu@intel.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

all inboxes | Powered by JetHome®