qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Prasad J Pandit <pjp@fedoraproject.org>
Subject: [Qemu-devel] [PULL 35/36] scsi: lsi: exit infinite loop while executing script (CVE-2019-12068)
Date: Tue, 20 Aug 2019 08:59:54 +0200	[thread overview]
Message-ID: <1566284395-30287-36-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1566284395-30287-1-git-send-email-pbonzini@redhat.com>

When executing script in lsi_execute_script(), the LSI scsi adapter
emulator advances 's->dsp' index to read next opcode. This can lead
to an infinite loop if the next opcode is empty. Move the existing
loop exit after 10k iterations so that it covers no-op opcodes as
well.

Reported-by: Bugs SysSec <bugs-syssec@rub.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi/lsi53c895a.c | 41 +++++++++++++++++++++++++++--------------
 1 file changed, 27 insertions(+), 14 deletions(-)

diff --git a/hw/scsi/lsi53c895a.c b/hw/scsi/lsi53c895a.c
index 10468c1..72f7b59 100644
--- a/hw/scsi/lsi53c895a.c
+++ b/hw/scsi/lsi53c895a.c
@@ -185,6 +185,9 @@ static const char *names[] = {
 /* Flag set if this is a tagged command.  */
 #define LSI_TAG_VALID     (1 << 16)
 
+/* Maximum instructions to process. */
+#define LSI_MAX_INSN    10000
+
 typedef struct lsi_request {
     SCSIRequest *req;
     uint32_t tag;
@@ -1132,7 +1135,21 @@ static void lsi_execute_script(LSIState *s)
 
     s->istat1 |= LSI_ISTAT1_SRUN;
 again:
-    insn_processed++;
+    if (++insn_processed > LSI_MAX_INSN) {
+        /* Some windows drivers make the device spin waiting for a memory
+           location to change.  If we have been executed a lot of code then
+           assume this is the case and force an unexpected device disconnect.
+           This is apparently sufficient to beat the drivers into submission.
+         */
+        if (!(s->sien0 & LSI_SIST0_UDC)) {
+            qemu_log_mask(LOG_GUEST_ERROR,
+                          "lsi_scsi: inf. loop with UDC masked");
+        }
+        lsi_script_scsi_interrupt(s, LSI_SIST0_UDC, 0);
+        lsi_disconnect(s);
+        trace_lsi_execute_script_stop();
+        return;
+    }
     insn = read_dword(s, s->dsp);
     if (!insn) {
         /* If we receive an empty opcode increment the DSP by 4 bytes
@@ -1569,19 +1586,7 @@ again:
             }
         }
     }
-    if (insn_processed > 10000 && s->waiting == LSI_NOWAIT) {
-        /* Some windows drivers make the device spin waiting for a memory
-           location to change.  If we have been executed a lot of code then
-           assume this is the case and force an unexpected device disconnect.
-           This is apparently sufficient to beat the drivers into submission.
-         */
-        if (!(s->sien0 & LSI_SIST0_UDC)) {
-            qemu_log_mask(LOG_GUEST_ERROR,
-                          "lsi_scsi: inf. loop with UDC masked");
-        }
-        lsi_script_scsi_interrupt(s, LSI_SIST0_UDC, 0);
-        lsi_disconnect(s);
-    } else if (s->istat1 & LSI_ISTAT1_SRUN && s->waiting == LSI_NOWAIT) {
+    if (s->istat1 & LSI_ISTAT1_SRUN && s->waiting == LSI_NOWAIT) {
         if (s->dcntl & LSI_DCNTL_SSM) {
             lsi_script_dma_interrupt(s, LSI_DSTAT_SSI);
         } else {
@@ -1969,6 +1974,10 @@ static void lsi_reg_writeb(LSIState *s, int offset, uint8_t val)
     case 0x2f: /* DSP[24:31] */
         s->dsp &= 0x00ffffff;
         s->dsp |= val << 24;
+        /*
+         * FIXME: if s->waiting != LSI_NOWAIT, this will only execute one
+         * instruction.  Is this correct?
+         */
         if ((s->dmode & LSI_DMODE_MAN) == 0
             && (s->istat1 & LSI_ISTAT1_SRUN) == 0)
             lsi_execute_script(s);
@@ -1987,6 +1996,10 @@ static void lsi_reg_writeb(LSIState *s, int offset, uint8_t val)
         break;
     case 0x3b: /* DCNTL */
         s->dcntl = val & ~(LSI_DCNTL_PFF | LSI_DCNTL_STD);
+        /*
+         * FIXME: if s->waiting != LSI_NOWAIT, this will only execute one
+         * instruction.  Is this correct?
+         */
         if ((val & LSI_DCNTL_STD) && (s->istat1 & LSI_ISTAT1_SRUN) == 0)
             lsi_execute_script(s);
         break;
-- 
1.8.3.1




  parent reply	other threads:[~2019-08-20  7:29 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-20  6:59 [Qemu-devel] [PULL 00/36] QEMU patches for 2018-08-20 Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 01/36] kvm: i386: halt poll control MSR support Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 02/36] target-i386: adds PV_SCHED_YIELD CPUID feature bit Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 03/36] loader: Handle memory-mapped ELFs Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 04/36] elf-ops.h: Map into memory the ELF to load Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 05/36] hw/i386/pc: Map into memory the initrd Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 06/36] memory: assert on out of scope notification Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 07/36] configure: Define target access alignment in configure Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 08/36] block: fix NetBSD qemu-iotests failure Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 09/36] 9p: simplify source file selection Paolo Bonzini
2020-11-03 20:31   ` Philippe Mathieu-Daudé
2019-08-20  6:59 ` [Qemu-devel] [PULL 10/36] target-i386: kvm: 'kvm_get_supported_msrs' cleanup Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 11/36] test-throttle: Fix uninitialized use of burst_length Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 12/36] tests: Fix uninitialized byte in test_visitor_in_fuzz Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 13/36] i386/kvm: initialize struct at full before ioctl call Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 14/36] target/i386: Return 'indefinite integer value' for invalid SSE fp->int conversions Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 15/36] memory: fix race between TCG and accesses to dirty bitmap Paolo Bonzini
2019-08-26 12:19   ` dovgaluk
2019-09-12  6:54     ` Pavel Dovgalyuk
2019-09-12 17:43       ` Richard Henderson
2019-09-12 22:16         ` Paolo Bonzini
2019-09-12 12:45     ` Paolo Bonzini
2022-08-02 16:17   ` Peter Maydell
2019-08-20  6:59 ` [Qemu-devel] [PULL 16/36] mc146818rtc: Remove reset notifiers Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 17/36] timer: " Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 18/36] replay: Remove host_clock_last Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 19/36] timer: last, remove last bits of last Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 20/36] kconfig: do not select VMMOUSE Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 21/36] replay: add missing fix for internal function Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 22/36] replay: document development rules Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 23/36] util/qemu-timer: refactor deadline calculation for external timers Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 24/36] replay: fix replay shutdown Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 25/36] replay: refine replay-time module Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 26/36] replay: rename step-related variables and functions Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 27/36] icount: clean up cpu_can_io at the entry to the block Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 28/36] icount: remove unnecessary gen_io_end calls Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 29/36] cpus-common: nuke finish_safe_work Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 30/36] cpus-common: assert BQL nesting within cpu-exclusive sections Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 31/36] kvm: vmxcap: Enhance with latest features Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 32/36] HACKING: Document 'struct' keyword usage Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 33/36] migration: do not rom_reset() during incoming migration Paolo Bonzini
2019-08-20  6:59 ` [Qemu-devel] [PULL 34/36] test-bitmap: test set 1 bit case for bitmap_set Paolo Bonzini
2019-08-20  6:59 ` Paolo Bonzini [this message]
2019-08-20  6:59 ` [Qemu-devel] [PULL 36/36] x86: Intel AVX512_BF16 feature enabling Paolo Bonzini
2019-08-20  7:42 ` [Qemu-devel] [PULL 00/36] QEMU patches for 2018-08-20 no-reply
2019-08-20  9:26 ` Peter Maydell
2019-08-20 23:42 ` 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=1566284395-30287-36-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=pjp@fedoraproject.org \
    --cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).