qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Wu Xiang <willx8@gmail.com>
Subject: [Qemu-devel] [PULL 04/42] target/i386: fix interrupt CPL error when using ist in x86-64
Date: Wed,  5 Jul 2017 09:14:07 +0200	[thread overview]
Message-ID: <1499238885-26161-5-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1499238885-26161-1-git-send-email-pbonzini@redhat.com>

From: Wu Xiang <willx8@gmail.com>

In do_interrupt64(), when interrupt stack table(ist) is enabled
and the the target code segment is conforming(e2 & DESC_C_MASK), the
old implementation always set new CPL to 0, and SS.RPL to 0.

This is incorrect for when CPL3 code access a CPL0 conforming code
segment, the CPL should remain unchanged. Otherwise higher privileged
code can be compromised.

The patch fix this for always set dpl = cpl when the target code segment
is conforming, and modify the last parameter `flags`, which contains
correct new CPL, in cpu_x86_load_seg_cache().

Signed-off-by: Wu Xiang <willx8@gmail.com>
Message-Id: <20170621142152.GA18094@wxdeubuntu.ipads-lab.se.sjtu.edu.cn>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 target/i386/seg_helper.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/target/i386/seg_helper.c b/target/i386/seg_helper.c
index 0374031..9af69c2 100644
--- a/target/i386/seg_helper.c
+++ b/target/i386/seg_helper.c
@@ -931,12 +931,14 @@ static void do_interrupt64(CPUX86State *env, int intno, int is_int,
         }
         new_stack = 0;
         esp = env->regs[R_ESP];
-        dpl = cpl;
     } else {
         raise_exception_err(env, EXCP0D_GPF, selector & 0xfffc);
         new_stack = 0; /* avoid warning */
         esp = 0; /* avoid warning */
     }
+    if (e2 & DESC_C_MASK) {
+        dpl = cpl;
+    }
     esp &= ~0xfLL; /* align stack */
 
     PUSHQ(esp, env->segs[R_SS].selector);
@@ -956,7 +958,7 @@ static void do_interrupt64(CPUX86State *env, int intno, int is_int,
 
     if (new_stack) {
         ss = 0 | dpl;
-        cpu_x86_load_seg_cache(env, R_SS, ss, 0, 0, 0);
+        cpu_x86_load_seg_cache(env, R_SS, ss, 0, 0, dpl << DESC_DPL_SHIFT);
     }
     env->regs[R_ESP] = esp;
 
-- 
1.8.3.1

  parent reply	other threads:[~2017-07-05  7:14 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-05  7:14 [Qemu-devel] [PULL 00/42] Misc changes for 2017-07-05 Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 01/42] vcpu_dirty: share the same field in CPUState for all accelerators Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 02/42] qemu-doc: Add missing "@c man end" statements Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 03/42] nbd: fix NBD over TLS Paolo Bonzini
2017-07-05  7:14 ` Paolo Bonzini [this message]
2017-07-05  7:14 ` [Qemu-devel] [PULL 05/42] target/i386: simplify handling of conforming code segments on interrupt Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 06/42] include/exec/poison: Add some more missing TARGET and CONFIG defines Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 07/42] Move CONFIG_KVM related definitions to kvm_i386.h Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 08/42] include/exec/poison: Mark CONFIG_KVM as poisoned, too Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 09/42] cpu: Introduce a wrapper for tlb_flush() that can be used in common code Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 10/42] include/exec/poison: Mark CONFIG_SOFTMMU as poisoned Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 11/42] Makefile: Move bootdevice.o to common-obj-y Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 12/42] hw/misc/edu: Compile the edu device as common object Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 13/42] sockets: avoid formatting buffer that may not be NUL terminated Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 14/42] xsave_helper: pull xsave and xrstor out of kvm.c into helper function Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 15/42] util/oslib-win32: Remove if conditional Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 16/42] tests/test-char.c: Don't use main_loop_wait()'s return value Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 17/42] main_loop: Make main_loop_wait() return void Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 18/42] checkpatch: should not use signal except for SIG_DFL or SIG_IGN Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 19/42] virtio-scsi: finalize IOMMU support Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 20/42] qemu-thread: Assert locks are initialized before using Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 21/42] qemu-doc: do not refer to years-old version numbers Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 22/42] configure: factor out list of supported Xen/KVM/HAX targets Paolo Bonzini
2017-07-10 15:49   ` Peter Maydell
2017-07-10 16:14     ` Paolo Bonzini
2017-07-10 16:24       ` Peter Maydell
2017-07-10 16:28         ` Paolo Bonzini
2017-07-14 15:26           ` Philippe Mathieu-Daudé
2017-07-14 16:58             ` Paolo Bonzini
2017-07-12  0:24         ` Stefano Stabellini
2017-07-10 16:29     ` Anthony PERARD
2017-07-05  7:14 ` [Qemu-devel] [PULL 23/42] configure: early test for supported targets Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 24/42] configure: add --disable-tcg configure option Paolo Bonzini
2017-07-14  5:24   ` Philippe Mathieu-Daudé
2017-07-14  9:02     ` Paolo Bonzini
2017-07-14 15:33       ` Philippe Mathieu-Daudé
2017-07-05  7:14 ` [Qemu-devel] [PULL 25/42] vl: convert -tb-size to qemu_strtoul Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 26/42] vl: add tcg_enabled() for tcg related code Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 27/42] tcg: move page_size_init() function Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 28/42] cpu: move interrupt handling out of translate-common.c Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 29/42] tcg: make tcg_allowed global Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 30/42] monitor: disable "info jit" and "info opcount" if !TCG Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 31/42] vapic: use tcg_enabled Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 32/42] tcg: add the tcg-stub.c file into accel/stubs/ Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 33/42] tcg: move tb_lock out of translate-all.h Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 34/42] exec: elide calls to tb_lock and tb_unlock Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 35/42] tcg: add CONFIG_TCG guards in headers Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 36/42] tcg: add the CONFIG_TCG into Makefiles Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 37/42] target/i386: move cpu_sync_bndcs_hflags() function Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 38/42] target/i386: make cpu_get_fp80()/cpu_set_fp80() static Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 39/42] target/i386: split cpu_set_mxcsr() and make cpu_set_fpuc() inline Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 40/42] target/i386: move TLB refill function out of helper.c Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 41/42] target/i386: add the tcg_enabled() in target/i386/ Paolo Bonzini
2017-07-05  7:14 ` [Qemu-devel] [PULL 42/42] target/i386: add the CONFIG_TCG into Makefiles Paolo Bonzini
2017-07-06 10:42 ` [Qemu-devel] [PULL 00/42] Misc changes for 2017-07-05 Peter Maydell
2017-07-06 14:01 ` no-reply

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=1499238885-26161-5-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=willx8@gmail.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;
as well as URLs for NNTP newsgroup(s).