[Devel] [PATCH RHEL7 COMMIT] ms/KVM: x86: introduce linear_{read, write}_system

Konstantin Khorenko khorenko at virtuozzo.com
Mon Jun 18 18:43:37 MSK 2018


The commit is pushed to "branch-rh7-3.10.0-862.3.2.vz7.61.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-862.3.2.vz7.61.4
------>
commit de7f169175e6f69c723063d67fc6a2028e01541a
Author: Vasily Averin <vvs at virtuozzo.com>
Date:   Mon Jun 18 18:43:37 2018 +0300

    ms/KVM: x86: introduce linear_{read,write}_system
    
    Wrap the common invocation of ctxt->ops->read_std and ctxt->ops->write_std, so
    as to have a smaller patch when the functions grow another argument.
    
    Fixes: 129a72a0d3c8 ("KVM: x86: Introduce segmented_write_std", 2017-01-12)
    Cc: stable at vger.kernel.org
    Signed-off-by: Paolo Bonzini <pbonzini at redhat.com>
    (cherry picked from commit 79367a65743975e5cac8d24d08eccc7fdae832b0)
    Signed-off-by: Vasily Averin <vvs at virtuozzo.com>
    
    =========================
    Patchset description:
    kvm: guest userspace to guest kernel write
    
    This patch set fixes CVE-2018-10853 kvm: guest userspace to guest kernel write
    
    https://bugzilla.redhat.com/show_bug.cgi?id=1589890
    A flaw was found in Linux Kernel KVM versions greater than and including 4.10.
    In which certain instructions such as sgdt/sidt call segmented_write_std
    doesn't propagate access correctly. As such, during userspace induced exception,
    the guest can incorrectly assume that the exception happened in the kernel and panic.
    
    https://jira.sw.ru/browse/PSBM-85796
    
    Paolo Bonzini (3):
      KVM: x86: introduce linear_{read,write}_system
      KVM: x86: pass kvm_vcpu to kvm_read_guest_virt and
        kvm_write_guest_virt_system
      kvm: x86: use correct privilege level for sgdt/sidt/fxsave/fxrstor
        access
---
 arch/x86/kvm/emulate.c | 64 +++++++++++++++++++++++++-------------------------
 1 file changed, 32 insertions(+), 32 deletions(-)

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 97490a846f64..4e69bf739a29 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -827,6 +827,19 @@ static inline int jmp_rel(struct x86_emulate_ctxt *ctxt, int rel)
 	return assign_eip_near(ctxt, ctxt->_eip + rel);
 }
 
+static int linear_read_system(struct x86_emulate_ctxt *ctxt, ulong linear,
+			      void *data, unsigned size)
+{
+	return ctxt->ops->read_std(ctxt, linear, data, size, &ctxt->exception);
+}
+
+static int linear_write_system(struct x86_emulate_ctxt *ctxt,
+			       ulong linear, void *data,
+			       unsigned int size)
+{
+	return ctxt->ops->write_std(ctxt, linear, data, size, &ctxt->exception);
+}
+
 static int segmented_read_std(struct x86_emulate_ctxt *ctxt,
 			      struct segmented_address addr,
 			      void *data,
@@ -1525,8 +1538,7 @@ static int read_interrupt_descriptor(struct x86_emulate_ctxt *ctxt,
 		return emulate_gp(ctxt, index << 3 | 0x2);
 
 	addr = dt.address + index * 8;
-	return ctxt->ops->read_std(ctxt, addr, desc, sizeof *desc,
-				   &ctxt->exception);
+	return linear_read_system(ctxt, addr, desc, sizeof *desc);
 }
 
 static void get_descriptor_table_ptr(struct x86_emulate_ctxt *ctxt,
@@ -1589,8 +1601,7 @@ static int read_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	return ctxt->ops->read_std(ctxt, *desc_addr_p, desc, sizeof(*desc),
-				   &ctxt->exception);
+	return linear_read_system(ctxt, *desc_addr_p, desc, sizeof(*desc));
 }
 
 /* allowed just for 8 bytes segments */
@@ -1604,8 +1615,7 @@ static int write_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	return ctxt->ops->write_std(ctxt, addr, desc, sizeof *desc,
-				    &ctxt->exception);
+	return linear_write_system(ctxt, addr, desc, sizeof *desc);
 }
 
 static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
@@ -1766,8 +1776,7 @@ static int __load_segment_descriptor(struct x86_emulate_ctxt *ctxt,
 				return ret;
 		}
 	} else if (ctxt->mode == X86EMUL_MODE_PROT64) {
-		ret = ctxt->ops->read_std(ctxt, desc_addr+8, &base3,
-				sizeof(base3), &ctxt->exception);
+		ret = linear_read_system(ctxt, desc_addr+8, &base3, sizeof(base3));
 		if (ret != X86EMUL_CONTINUE)
 			return ret;
 		if (is_noncanonical_address(get_desc_base(&seg_desc) |
@@ -2080,11 +2089,11 @@ static int __emulate_int_real(struct x86_emulate_ctxt *ctxt, int irq)
 	eip_addr = dt.address + (irq << 2);
 	cs_addr = dt.address + (irq << 2) + 2;
 
-	rc = ops->read_std(ctxt, cs_addr, &cs, 2, &ctxt->exception);
+	rc = linear_read_system(ctxt, cs_addr, &cs, 2);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
-	rc = ops->read_std(ctxt, eip_addr, &eip, 2, &ctxt->exception);
+	rc = linear_read_system(ctxt, eip_addr, &eip, 2);
 	if (rc != X86EMUL_CONTINUE)
 		return rc;
 
@@ -3082,35 +3091,30 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt,
 			  u16 tss_selector, u16 old_tss_sel,
 			  ulong old_tss_base, struct desc_struct *new_desc)
 {
-	const struct x86_emulate_ops *ops = ctxt->ops;
 	struct tss_segment_16 tss_seg;
 	int ret;
 	u32 new_tss_base = get_desc_base(new_desc);
 
-	ret = ops->read_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg,
-			    &ctxt->exception);
+	ret = linear_read_system(ctxt, old_tss_base, &tss_seg, sizeof tss_seg);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
 	save_state_to_tss16(ctxt, &tss_seg);
 
-	ret = ops->write_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg,
-			     &ctxt->exception);
+	ret = linear_write_system(ctxt, old_tss_base, &tss_seg, sizeof tss_seg);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
-	ret = ops->read_std(ctxt, new_tss_base, &tss_seg, sizeof tss_seg,
-			    &ctxt->exception);
+	ret = linear_read_system(ctxt, new_tss_base, &tss_seg, sizeof tss_seg);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
 	if (old_tss_sel != 0xffff) {
 		tss_seg.prev_task_link = old_tss_sel;
 
-		ret = ops->write_std(ctxt, new_tss_base,
-				     &tss_seg.prev_task_link,
-				     sizeof tss_seg.prev_task_link,
-				     &ctxt->exception);
+		ret = linear_write_system(ctxt, new_tss_base,
+					  &tss_seg.prev_task_link,
+					  sizeof tss_seg.prev_task_link);
 		if (ret != X86EMUL_CONTINUE)
 			return ret;
 	}
@@ -3226,38 +3230,34 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt,
 			  u16 tss_selector, u16 old_tss_sel,
 			  ulong old_tss_base, struct desc_struct *new_desc)
 {
-	const struct x86_emulate_ops *ops = ctxt->ops;
 	struct tss_segment_32 tss_seg;
 	int ret;
 	u32 new_tss_base = get_desc_base(new_desc);
 	u32 eip_offset = offsetof(struct tss_segment_32, eip);
 	u32 ldt_sel_offset = offsetof(struct tss_segment_32, ldt_selector);
 
-	ret = ops->read_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg,
-			    &ctxt->exception);
+	ret = linear_read_system(ctxt, old_tss_base, &tss_seg, sizeof tss_seg);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
 	save_state_to_tss32(ctxt, &tss_seg);
 
 	/* Only GP registers and segment selectors are saved */
-	ret = ops->write_std(ctxt, old_tss_base + eip_offset, &tss_seg.eip,
-			     ldt_sel_offset - eip_offset, &ctxt->exception);
+	ret = linear_write_system(ctxt, old_tss_base + eip_offset, &tss_seg.eip,
+				  ldt_sel_offset - eip_offset);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
-	ret = ops->read_std(ctxt, new_tss_base, &tss_seg, sizeof tss_seg,
-			    &ctxt->exception);
+	ret = linear_read_system(ctxt, new_tss_base, &tss_seg, sizeof tss_seg);
 	if (ret != X86EMUL_CONTINUE)
 		return ret;
 
 	if (old_tss_sel != 0xffff) {
 		tss_seg.prev_task_link = old_tss_sel;
 
-		ret = ops->write_std(ctxt, new_tss_base,
-				     &tss_seg.prev_task_link,
-				     sizeof tss_seg.prev_task_link,
-				     &ctxt->exception);
+		ret = linear_write_system(ctxt, new_tss_base,
+					  &tss_seg.prev_task_link,
+					  sizeof tss_seg.prev_task_link);
 		if (ret != X86EMUL_CONTINUE)
 			return ret;
 	}


More information about the Devel mailing list