Skip to contentSkip to conversation

KVM: x86: TIF_NEED_FPU_LOAD bug fixes

PATCH Applied Sean Christopherson 2 participants 6 messages Jan 17, 2020 – Jan 18, 2020

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v2 cover KVM: x86: TIF_NEED_FPU_LOAD bug fixes

Apologies for the quick v2, I'm hoping the fixes can squeak into 5.5 and I'm about to go offline for a long weekend.

v2: Add a helper to handle TIF_NEED_FPU_LOAD when saving from the current
FPU to the guest/userspace FPU (patch 01). [Dave]

*** v1 cover letter ***

Three bug fixes related to deferring FPU loading to return to userspace, or in this case, deferring to entering a KVM guest. And a cleanup patch I couldn't resist throwing on top.

The crux of the matter is that TIF_FPU_NEED_LOAD can be set any time control is transferred out of KVM, e.g. via IRQ->softirq, not just when KVM is preempted. The previous attempt to fix the underlying bug(s) by handling TIF_FPU_NEED_LOAD during kvm_sched_in() only made the bug(s) harder to hit, i.e. it resolved the preemption case but only shrunk the window where a softirq could corrupt state.

Paolo, patch 01 will conflict with commit 95145c25a78c ("KVM: x86: Add a WARN on TIF_NEED_FPU_LOAD in kvm_load_guest_fpu()") that is sitting in kvm/queue. The kvm/queue commit should simply be dropped.

Patch 01 fixes the original underlying bug, which is that KVM doesn't handle TIF_FPU_NEED_LOAD when swapping between guest and userspace FPU state.

Patch 02 fixes (unconfirmed) issues similar to the reported bug where KVM doesn't ensure CPU FPU state is fresh when accessing it during emulation. This patch is smoke tested only (via kvm-unit-tests' emulator tests).

Patch 03 reverts the preemption bug fix, which simultaneously restores the handling of TIF_FPU_NEED_LOAD in vcpu_enter_guest() to fix the reported bugs[1][2] and removes the now-unnecessary preemption workaround.

Alternatively, the handling of TIF_NEED_FPU_LOAD in the kvm_sched_in() path could be left in place, i.e it doesn't cause immediate damage, but as stated before, restoring FPU state after KVM is preempted only makes it harder to find the actual bugs. Case in point, I originally split the revert into two separate patches (so that removing the kvm_sched_in() call wouldn't be tagged for stable), but that only hid the underlying kvm_put_guest_fpu() bug until I fully reverted the commit.

Patch 04 removes an unused emulator context param from several of the functions that are fixed in patch 02. The param was left behind during a previous KVM FPU state rework.

Tested via Thomas Lambertz's mprime reproducer[3], which detected issues with buggy KVMs on my system in under a minute. Ran clean for ~30 minutes on each of the first two patches and several hours with all three patches applied.

[1] https://lore.kernel.org/kvm/[email protected]
[2] https://lore.kernel.org/kvm/[email protected]%2F
[3] https://lore.kernel.org/lkml/[email protected]

Sean Christopherson (4):
KVM: x86: Handle TIF_NEED_FPU_LOAD in kvm_{load,put}_guest_fpu()
KVM: x86: Ensure guest's FPU state is loaded when accessing for
emulation
KVM: x86: Revert "KVM: X86: Fix fpu state crash in kvm guest"
KVM: x86: Remove unused ctxt param from emulator's FPU accessors

 arch/x86/kvm/emulate.c | 67 ++++++++++++++++++++++++++++++++----------
 arch/x86/kvm/x86.c     | 28 +++++++++++++-----
 2 files changed, 72 insertions(+), 23 deletions(-)
Signature · 1 line
2.24.1
#1 lore raw

PATCH v2 1/4 KVM: x86: Handle TIF_NEED_FPU_LOAD in kvm_{load,put}_guest_fpu()

Handle TIF_NEED_FPU_LOAD similar to how fpu__copy() handles the flag when duplicating FPU state to a new task struct. TIF_NEED_FPU_LOAD can be set any time control is transferred out of KVM, be it voluntarily, e.g. if I/O is triggered during a KVM call to get_user_pages, or involuntarily, e.g. if softirq runs after an IRQ occurs. Therefore, KVM must account for TIF_NEED_FPU_LOAD whenever it is (potentially) accessing CPU FPU state.

Fixes 5f409e20b7945 ("x86/fpu: Defer FPU state load until return to userspace")
---
 arch/x86/kvm/x86.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)
arch/x86/kvm/x86.c +17 -2
index cf917139de6b..0bbf0110bef7 100644
@@ -8471,12 +8471,26 @@ static int complete_emulated_mmio(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static void kvm_save_current_fpu(struct fpu *fpu)
+{
+	/*
+	 * If the target FPU state is not resident in the CPU registers, just
+	 * memcpy() from current, else save CPU state directly to the target.
+	 */
+	if (test_thread_flag(TIF_NEED_FPU_LOAD))
+		memcpy(&fpu->state, &current->thread.fpu.state,
+		       fpu_kernel_xstate_size);
+	else
+		copy_fpregs_to_fpstate(fpu);
+}
+
 /* Swap (qemu) user FPU context for the guest FPU context. */
 static void kvm_load_guest_fpu(struct kvm_vcpu *vcpu)
 {
 	fpregs_lock();
 
-	copy_fpregs_to_fpstate(vcpu->arch.user_fpu);
+	kvm_save_current_fpu(vcpu->arch.user_fpu);
+
 	/* PKRU is separately restored in kvm_x86_ops->run.  */
 	__copy_kernel_to_fpregs(&vcpu->arch.guest_fpu->state,
 				~XFEATURE_MASK_PKRU);
@@ -8492,7 +8506,8 @@ static void kvm_put_guest_fpu(struct kvm_vcpu *vcpu)
 {
 	fpregs_lock();
 
-	copy_fpregs_to_fpstate(vcpu->arch.guest_fpu);
+	kvm_save_current_fpu(vcpu->arch.guest_fpu);
+
 	copy_kernel_to_fpregs(&vcpu->arch.user_fpu->state);
 
 	fpregs_mark_activate();
Signature · 1 line
2.24.1
#2 lore raw

PATCH v2 2/4 KVM: x86: Ensure guest's FPU state is loaded when accessing for emulation

In reply to Sean Christopherson

Lock the FPU regs and reload the current thread's FPU state, which holds the guest's FPU state, to the CPU registers if necessary prior to accessing guest FPU state as part of emulation. kernel_fpu_begin() can be called from softirq context, therefore KVM must ensure softirqs are disabled (locking the FPU regs disables softirqs) when touching CPU FPU state.

Note, for all intents and purposes this reverts commit 6ab0b9feb82a7 ("x86,kvm: remove KVM emulator get_fpu / put_fpu"), but at the time it was applied, removing get/put_fpu() was correct. The re-introduction of {get,put}_fpu() is necessitated by the deferring of FPU state load.

Fixes 5f409e20b7945 ("x86/fpu: Defer FPU state load until return to userspace")
---
 arch/x86/kvm/emulate.c | 39 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)
arch/x86/kvm/emulate.c +39 -0
index 952d1a4f4d7e..2a5bed60ce50 100644
@@ -22,6 +22,7 @@
 #include "kvm_cache_regs.h"
 #include <asm/kvm_emulate.h>
 #include <linux/stringify.h>
+#include <asm/fpu/api.h>
 #include <asm/debugreg.h>
 #include <asm/nospec-branch.h>
 
@@ -1075,8 +1076,23 @@ static void fetch_register_operand(struct operand *op)
 	}
 }
 
+static void emulator_get_fpu(void)
+{
+	fpregs_lock();
+
+	fpregs_assert_state_consistent();
+	if (test_thread_flag(TIF_NEED_FPU_LOAD))
+		switch_fpu_return();
+}
+
+static void emulator_put_fpu(void)
+{
+	fpregs_unlock();
+}
+
 static void read_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, int reg)
 {
+	emulator_get_fpu();
 	switch (reg) {
 	case 0: asm("movdqa %%xmm0, %0" : "=m"(*data)); break;
 	case 1: asm("movdqa %%xmm1, %0" : "=m"(*data)); break;
@@ -1098,11 +1114,13 @@ static void read_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, int reg)
 #endif
 	default: BUG();
 	}
+	emulator_put_fpu();
 }
 
 static void write_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data,
 			  int reg)
 {
+	emulator_get_fpu();
 	switch (reg) {
 	case 0: asm("movdqa %0, %%xmm0" : : "m"(*data)); break;
 	case 1: asm("movdqa %0, %%xmm1" : : "m"(*data)); break;
@@ -1124,10 +1142,12 @@ static void write_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data,
 #endif
 	default: BUG();
 	}
+	emulator_put_fpu();
 }
 
 static void read_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
 {
+	emulator_get_fpu();
 	switch (reg) {
 	case 0: asm("movq %%mm0, %0" : "=m"(*data)); break;
 	case 1: asm("movq %%mm1, %0" : "=m"(*data)); break;
@@ -1139,10 +1159,12 @@ static void read_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
 	case 7: asm("movq %%mm7, %0" : "=m"(*data)); break;
 	default: BUG();
 	}
+	emulator_put_fpu();
 }
 
 static void write_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
 {
+	emulator_get_fpu();
 	switch (reg) {
 	case 0: asm("movq %0, %%mm0" : : "m"(*data)); break;
 	case 1: asm("movq %0, %%mm1" : : "m"(*data)); break;
@@ -1154,6 +1176,7 @@ static void write_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
 	case 7: asm("movq %0, %%mm7" : : "m"(*data)); break;
 	default: BUG();
 	}
+	emulator_put_fpu();
 }
 
 static int em_fninit(struct x86_emulate_ctxt *ctxt)
@@ -1161,7 +1184,9 @@ static int em_fninit(struct x86_emulate_ctxt *ctxt)
 	if (ctxt->ops->get_cr(ctxt, 0) & (X86_CR0_TS | X86_CR0_EM))
 		return emulate_nm(ctxt);
 
+	emulator_get_fpu();
 	asm volatile("fninit");
+	emulator_put_fpu();
 	return X86EMUL_CONTINUE;
 }
 
@@ -1172,7 +1197,9 @@ static int em_fnstcw(struct x86_emulate_ctxt *ctxt)
 	if (ctxt->ops->get_cr(ctxt, 0) & (X86_CR0_TS | X86_CR0_EM))
 		return emulate_nm(ctxt);
 
+	emulator_get_fpu();
 	asm volatile("fnstcw %0": "+m"(fcw));
+	emulator_put_fpu();
 
 	ctxt->dst.val = fcw;
 
@@ -1186,7 +1213,9 @@ static int em_fnstsw(struct x86_emulate_ctxt *ctxt)
 	if (ctxt->ops->get_cr(ctxt, 0) & (X86_CR0_TS | X86_CR0_EM))
 		return emulate_nm(ctxt);
 
+	emulator_get_fpu();
 	asm volatile("fnstsw %0": "+m"(fsw));
+	emulator_put_fpu();
 
 	ctxt->dst.val = fsw;
 
@@ -4092,8 +4121,12 @@ static int em_fxsave(struct x86_emulate_ctxt *ctxt)
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
+	emulator_get_fpu();
+
 	rc = asm_safe("fxsave %[fx]", , [fx] "+m"(fx_state));
 
+	emulator_put_fpu();
+
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
@@ -4136,6 +4169,8 @@ static int em_fxrstor(struct x86_emulate_ctxt *ctxt)
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
+	emulator_get_fpu();
+
 	if (size < __fxstate_size(16)) {
 		rc = fxregs_fixup(&fx_state, size);
 		if (rc != X86EMUL_CONTINUE)
@@ -4151,6 +4186,8 @@ static int em_fxrstor(struct x86_emulate_ctxt *ctxt)
 		rc = asm_safe("fxrstor %[fx]", : [fx] "m"(fx_state));
 
 out:
+	emulator_put_fpu();
+
 	return rc;
 }
 
@@ -5448,7 +5485,9 @@ static int flush_pending_x87_faults(struct x86_emulate_ctxt *ctxt)
 {
 	int rc;
 
+	emulator_get_fpu();
 	rc = asm_safe("fwait");
+	emulator_put_fpu();
 
 	if (unlikely(rc != X86EMUL_CONTINUE))
 		return emulate_exception(ctxt, MF_VECTOR, 0, false);
Signature · 1 line
2.24.1
#3 lore raw

PATCH v2 3/4 KVM: x86: Revert "KVM: X86: Fix fpu state crash in kvm guest"

In reply to Sean Christopherson

Reload the current thread's FPU state, which contains the guest's FPU state, to the CPU registers if necessary during vcpu_enter_guest(). TIF_NEED_FPU_LOAD can be set any time control is transferred out of KVM, e.g. if I/O is triggered during a KVM call to get_user_pages() or if a softirq occurs while KVM is scheduled in.

Moving the handling of TIF_NEED_FPU_LOAD from vcpu_enter_guest() to kvm_arch_vcpu_load(), effectively kvm_sched_in(), papered over a bug where kvm_put_guest_fpu() failed to account for TIF_NEED_FPU_LOAD. The easiest way to the kvm_put_guest_fpu() bug was to run with involuntary preemption enable, thus handling TIF_NEED_FPU_LOAD during kvm_sched_in() made the bug go away. But, removing the handling in vcpu_enter_guest() exposed KVM to the rare case of a softirq triggering kernel_fpu_begin() between vcpu_load() and vcpu_enter_guest().

Now that kvm_{load,put}_guest_fpu() correctly handle TIF_NEED_FPU_LOAD, revert the commit to both restore the vcpu_enter_guest() behavior and eliminate the superfluous switch_fpu_return() in kvm_arch_vcpu_load().

Note, leaving the handling in kvm_arch_vcpu_load() isn't wrong per se, but it is unnecessary, and most critically, makes it extremely difficult to find bugs such as the kvm_put_guest_fpu() issue due to shrinking the window where a softirq can corrupt state.

A sample trace triggered by warning if TIF_NEED_FPU_LOAD is set while vcpu state is loaded:

 <IRQ>
  gcmaes_crypt_by_sg.constprop.12+0x26e/0x660
  ? 0xffffffffc024547d
  ? __qdisc_run+0x83/0x510
  ? __dev_queue_xmit+0x45e/0x990
  ? ip_finish_output2+0x1a8/0x570
  ? fib4_rule_action+0x61/0x70
  ? fib4_rule_action+0x70/0x70
  ? fib_rules_lookup+0x13f/0x1c0
  ? helper_rfc4106_decrypt+0x82/0xa0
  ? crypto_aead_decrypt+0x40/0x70
  ? crypto_aead_decrypt+0x40/0x70
  ? crypto_aead_decrypt+0x40/0x70
  ? esp_output_tail+0x8f4/0xa5a [esp4]
  ? skb_ext_add+0xd3/0x170
  ? xfrm_input+0x7a6/0x12c0
  ? xfrm4_rcv_encap+0xae/0xd0
  ? xfrm4_transport_finish+0x200/0x200
  ? udp_queue_rcv_one_skb+0x1ba/0x460
  ? udp_unicast_rcv_skb.isra.63+0x72/0x90
  ? __udp4_lib_rcv+0x51b/0xb00
  ? ip_protocol_deliver_rcu+0xd2/0x1c0
  ? ip_local_deliver_finish+0x44/0x50
  ? ip_local_deliver+0xe0/0xf0
  ? ip_protocol_deliver_rcu+0x1c0/0x1c0
  ? ip_rcv+0xbc/0xd0
  ? ip_rcv_finish_core.isra.19+0x380/0x380
  ? __netif_receive_skb_one_core+0x7e/0x90
  ? netif_receive_skb_internal+0x3d/0xb0
  ? napi_gro_receive+0xed/0x150
  ? 0xffffffffc0243c77
  ? net_rx_action+0x149/0x3b0
  ? __do_softirq+0xe4/0x2f8
  ? handle_irq_event_percpu+0x6a/0x80
  ? irq_exit+0xe6/0xf0
  ? do_IRQ+0x7f/0xd0
  ? common_interrupt+0xf/0xf
  </IRQ>
  ? irq_entries_start+0x20/0x660
  ? vmx_get_interrupt_shadow+0x2f0/0x710 [kvm_intel]
  ? kvm_set_msr_common+0xfc7/0x2380 [kvm]
  ? recalibrate_cpu_khz+0x10/0x10
  ? ktime_get+0x3a/0xa0
  ? kvm_arch_vcpu_ioctl_run+0x107/0x560 [kvm]
  ? kvm_init+0x6bf/0xd00 [kvm]
  ? __seccomp_filter+0x7a/0x680
  ? do_vfs_ioctl+0xa4/0x630
  ? security_file_ioctl+0x32/0x50
  ? ksys_ioctl+0x60/0x90
  ? __x64_sys_ioctl+0x16/0x20
  ? do_syscall_64+0x5f/0x1a0
  ? entry_SYSCALL_64_after_hwframe+0x44/0xa9
---[ end trace 9564a1ccad733a90 ]---

This reverts commit e751732486eb3f159089a64d1901992b1357e7cc.

Fixes e751732486eb3 ("KVM: X86: Fix fpu state crash in kvm guest")
Reported-by [email protected]
Cc Wanpeng Li <[email protected]>
Cc Thomas Lambertz <[email protected]>
Cc Rik van Riel <[email protected]>
Cc Sebastian Andrzej Siewior <[email protected]>
Cc Borislav Petkov <[email protected]>
Cc Dave Hansen <[email protected]>
Cc Thomas Gleixner <[email protected]>
---
 arch/x86/kvm/x86.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)
arch/x86/kvm/x86.c +3 -6
index 0bbf0110bef7..815c93cf551b 100644
@@ -3458,10 +3458,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 
 	kvm_x86_ops->vcpu_load(vcpu, cpu);
 
-	fpregs_assert_state_consistent();
-	if (test_thread_flag(TIF_NEED_FPU_LOAD))
-		switch_fpu_return();
-
 	/* Apply any externally detected TSC adjustments (due to suspend) */
 	if (unlikely(vcpu->arch.tsc_offset_adjustment)) {
 		adjust_tsc_offset_host(vcpu, vcpu->arch.tsc_offset_adjustment);
@@ -8198,8 +8194,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 	trace_kvm_entry(vcpu->vcpu_id);
 	guest_enter_irqoff();
 
-	/* The preempt notifier should have taken care of the FPU already.  */
-	WARN_ON_ONCE(test_thread_flag(TIF_NEED_FPU_LOAD));
+	fpregs_assert_state_consistent();
+	if (test_thread_flag(TIF_NEED_FPU_LOAD))
+		switch_fpu_return();
 
 	if (unlikely(vcpu->arch.switch_db_regs)) {
 		set_debugreg(0, 7);
Signature · 1 line
2.24.1
#4 lore raw

PATCH v2 4/4 KVM: x86: Remove unused ctxt param from emulator's FPU accessors

In reply to Sean Christopherson

Remove an unused struct x86_emulate_ctxt * param from low level helpers used to access guest FPU state. The unused param was left behind by commit 6ab0b9feb82a ("x86,kvm: remove KVM emulator get_fpu / put_fpu").

No functional change intended.

---
 arch/x86/kvm/emulate.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)
arch/x86/kvm/emulate.c +13 -15
index 2a5bed60ce50..3e3b3cd60cce 100644
@@ -1090,7 +1090,7 @@ static void emulator_put_fpu(void)
 	fpregs_unlock();
 }
 
-static void read_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, int reg)
+static void read_sse_reg(sse128_t *data, int reg)
 {
 	emulator_get_fpu();
 	switch (reg) {
@@ -1117,8 +1117,7 @@ static void read_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, int reg)
 	emulator_put_fpu();
 }
 
-static void write_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data,
-			  int reg)
+static void write_sse_reg(sse128_t *data, int reg)
 {
 	emulator_get_fpu();
 	switch (reg) {
@@ -1145,7 +1144,7 @@ static void write_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data,
 	emulator_put_fpu();
 }
 
-static void read_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
+static void read_mmx_reg(u64 *data, int reg)
 {
 	emulator_get_fpu();
 	switch (reg) {
@@ -1162,7 +1161,7 @@ static void read_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
 	emulator_put_fpu();
 }
 
-static void write_mmx_reg(struct x86_emulate_ctxt *ctxt, u64 *data, int reg)
+static void write_mmx_reg(u64 *data, int reg)
 {
 	emulator_get_fpu();
 	switch (reg) {
@@ -1234,7 +1233,7 @@ static void decode_register_operand(struct x86_emulate_ctxt *ctxt,
 		op->type = OP_XMM;
 		op->bytes = 16;
 		op->addr.xmm = reg;
-		read_sse_reg(ctxt, &op->vec_val, reg);
+		read_sse_reg(&op->vec_val, reg);
 		return;
 	}
 	if (ctxt->d & Mmx) {
@@ -1285,7 +1284,7 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt,
 			op->type = OP_XMM;
 			op->bytes = 16;
 			op->addr.xmm = ctxt->modrm_rm;
-			read_sse_reg(ctxt, &op->vec_val, ctxt->modrm_rm);
+			read_sse_reg(&op->vec_val, ctxt->modrm_rm);
 			return rc;
 		}
 		if (ctxt->d & Mmx) {
@@ -1862,10 +1861,10 @@ static int writeback(struct x86_emulate_ctxt *ctxt, struct operand *op)
 				       op->bytes * op->count);
 		break;
 	case OP_XMM:
-		write_sse_reg(ctxt, &op->vec_val, op->addr.xmm);
+		write_sse_reg(&op->vec_val, op->addr.xmm);
 		break;
 	case OP_MM:
-		write_mmx_reg(ctxt, &op->mm_val, op->addr.mm);
+		write_mmx_reg(&op->mm_val, op->addr.mm);
 		break;
 	case OP_NONE:
 		/* no writeback */
@@ -5495,11 +5494,10 @@ static int flush_pending_x87_faults(struct x86_emulate_ctxt *ctxt)
 	return X86EMUL_CONTINUE;
 }
 
-static void fetch_possible_mmx_operand(struct x86_emulate_ctxt *ctxt,
-				       struct operand *op)
+static void fetch_possible_mmx_operand(struct operand *op)
 {
 	if (op->type == OP_MM)
-		read_mmx_reg(ctxt, &op->mm_val, op->addr.mm);
+		read_mmx_reg(&op->mm_val, op->addr.mm);
 }
 
 static int fastop(struct x86_emulate_ctxt *ctxt, void (*fop)(struct fastop *))
@@ -5578,10 +5576,10 @@ int x86_emulate_insn(struct x86_emulate_ctxt *ctxt)
 			 * Now that we know the fpu is exception safe, we can fetch
 			 * operands from it.
 			 */
-			fetch_possible_mmx_operand(ctxt, &ctxt->src);
-			fetch_possible_mmx_operand(ctxt, &ctxt->src2);
+			fetch_possible_mmx_operand(&ctxt->src);
+			fetch_possible_mmx_operand(&ctxt->src2);
 			if (!(ctxt->d & Mov))
-				fetch_possible_mmx_operand(ctxt, &ctxt->dst);
+				fetch_possible_mmx_operand(&ctxt->dst);
 		}
 
 		if (unlikely(emul_flags & X86EMUL_GUEST_MASK) && ctxt->intercept) {
Signature · 1 line
2.24.1
maintainer Applied
#5 lore raw

Re: [PATCH v2 4/4] KVM: x86: Remove unused ctxt param from emulator's FPU accessors

Quoted 5 lines · Sean Christopherson

Remove an unused struct x86_emulate_ctxt * param from low level helpers used to access guest FPU state. The unused param was left behind by commit 6ab0b9feb82a ("x86,kvm: remove KVM emulator get_fpu / put_fpu").

No functional change intended.

Makes sense since the new implementation of emulator_get/put_fpu can be considered generic, and therefore it's okay not to tie it to ctxt->ops.

Queued, thanks.

Paolo