Skip to contentSkip to conversation

Miscellaneous error handling fixes

PATCH Superseded Markus Armbruster 8 participants 23 messages Apr 20, 2020 – Apr 21, 2020 Block layer

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover Miscellaneous error handling fixes

Maintainers decide whether any of these are serious enough to still go into 5.0. I doubt it.

Markus Armbruster (11):
cryptodev: Fix cryptodev_builtin_cleanup() error API violation block/file-posix: Fix check_cache_dropped() error handling
cpus: Fix configure_icount() error API violation
cpus: Proper range-checking for -icount shift=N arm/virt: Fix virt_machine_device_plug_cb() error API violation
fdc: Fix fallback=auto error handling
bochs-display: Fix vgamem=SIZE error handling
virtio-net: Fix duplex=... and speed=... error handling xen/pt: Fix flawed conversion to realize()
io: Fix qio_channel_socket_close() error handling migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

 backends/cryptodev-builtin.c | 10 ++-----
 block/file-posix.c           |  5 +++-
 cpus.c                       | 52 +++++++++++++++++++++---------------
 hw/arm/virt.c                |  4 +--
 hw/block/fdc.c               |  1 +
 hw/display/bochs-display.c   |  6 +++--
 hw/net/virtio-net.c          |  5 +++-
 hw/xen/xen_pt.c              | 12 ++++-----
 io/channel-socket.c          |  5 ++--
 migration/colo.c             |  8 +++++-
 10 files changed, 63 insertions(+), 45 deletions(-)
Signature · 1 line
2.21.1
#1 lore raw

PATCH 1/11 cryptodev: Fix cryptodev_builtin_cleanup() error API violation

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

cryptodev_builtin_cleanup() passes @errp to cryptodev_builtin_sym_close_session() in a loop. Harmless, because cryptodev_builtin_sym_close_session() can't actually fail. Fix it anyway.

---
 backends/cryptodev-builtin.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)
backends/cryptodev-builtin.c +2 -8
index c8ae3b9742..14316333fe 100644
@@ -282,12 +282,7 @@ static int cryptodev_builtin_sym_close_session(
     CryptoDevBackendBuiltin *builtin =
                       CRYPTODEV_BACKEND_BUILTIN(backend);
 
-    if (session_id >= MAX_NUM_SESSIONS ||
-              builtin->sessions[session_id] == NULL) {
-        error_setg(errp, "Cannot find a valid session id: %" PRIu64 "",
-                      session_id);
-        return -1;
-    }
+    assert(session_id < MAX_NUM_SESSIONS && builtin->sessions[session_id]);
 
     qcrypto_cipher_free(builtin->sessions[session_id]->cipher);
     g_free(builtin->sessions[session_id]);
@@ -356,8 +351,7 @@ static void cryptodev_builtin_cleanup(
 
     for (i = 0; i < MAX_NUM_SESSIONS; i++) {
         if (builtin->sessions[i] != NULL) {
-            cryptodev_builtin_sym_close_session(
-                    backend, i, 0, errp);
+            cryptodev_builtin_sym_close_session(backend, i, 0, &error_abort);
         }
     }
 
Signature · 1 line
2.21.1
#2 lore raw

PATCH 2/11 block/file-posix: Fix check_cache_dropped() error handling

In reply to Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

check_cache_dropped() calls error_setg() in a loop. It fails to break the loop in one instance. If a subsequent iteration error_setg()s again, it trips error_setv()'s assertion.

Fix it to break the loop.

Fixes 31be8a2a97ecba7d31a82932286489cac318e9e9
Cc Stefan Hajnoczi <[email protected]>
---
 block/file-posix.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
block/file-posix.c +4 -1
index 7e19bbff5f..094e3b0212 100644
@@ -2691,10 +2691,13 @@ static void check_cache_dropped(BlockDriverState *bs, Error **errp)
         vec_end = DIV_ROUND_UP(length, page_size);
         for (i = 0; i < vec_end; i++) {
             if (vec[i] & 0x1) {
-                error_setg(errp, "page cache still in use!");
                 break;
             }
         }
+        if (i < vec_end) {
+            error_setg(errp, "page cache still in use!");
+            break;
+        }
     }
 
     if (window) {
Signature · 1 line
2.21.1
Reviewed-by on v1 2/11
#3 lore raw

Re: [PATCH 02/11] block/file-posix: Fix check_cache_dropped() error handling

Quoted 10 lines · Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

check_cache_dropped() calls error_setg() in a loop. It fails to break the loop in one instance. If a subsequent iteration error_setg()s again, it trips error_setv()'s assertion.

Quoted 8 lines

Fix it to break the loop.

Fixes 31be8a2a97ecba7d31a82932286489cac318e9e9
Cc Stefan Hajnoczi <[email protected]>
---
  block/file-posix.c | 5 ++++-
  1 file changed, 4 insertions(+), 1 deletion(-)

Doesn't seem to be a fresh regression in 5.0 (present since 3.0), so no need to rush it in before 5.1.

Signature · 3 lines
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org
#4 lore raw

PATCH 3/11 cpus: Fix configure_icount() error API violation

In reply to Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

configure_icount() is wrong that way. Harmless, because its @errp is always &error_abort or &error_fatal.

Just as wrong (and just as harmless): when it fails, it can still update global state.

Fix all that.

---
 cpus.c | 51 ++++++++++++++++++++++++++++++---------------------
 1 file changed, 30 insertions(+), 21 deletions(-)
cpus.c +30 -21
index ef441bdf62..1b542b37f9 100644
@@ -797,40 +797,49 @@ void cpu_ticks_init(void)
 
 void configure_icount(QemuOpts *opts, Error **errp)
 {
-    const char *option;
+    const char *option = qemu_opt_get(opts, "shift");
+    bool sleep = qemu_opt_get_bool(opts, "sleep", true);
+    bool align = qemu_opt_get_bool(opts, "align", false);
+    long time_shift = -1;
     char *rem_str = NULL;
 
-    option = qemu_opt_get(opts, "shift");
-    if (!option) {
-        if (qemu_opt_get(opts, "align") != NULL) {
-            error_setg(errp, "Please specify shift option when using align");
-        }
+    if (!option && qemu_opt_get(opts, "align")) {
+        error_setg(errp, "Please specify shift option when using align");
         return;
     }
 
-    icount_sleep = qemu_opt_get_bool(opts, "sleep", true);
+    if (align && !sleep) {
+        error_setg(errp, "align=on and sleep=off are incompatible");
+        return;
+    }
+
+    if (strcmp(option, "auto") != 0) {
+        errno = 0;
+        time_shift = strtol(option, &rem_str, 0);
+        if (errno != 0 || *rem_str != '\0' || !strlen(option)) {
+            error_setg(errp, "icount: Invalid shift value");
+            return;
+        }
+    } else if (icount_align_option) {
+        error_setg(errp, "shift=auto and align=on are incompatible");
+        return;
+    } else if (!icount_sleep) {
+        error_setg(errp, "shift=auto and sleep=off are incompatible");
+        return;
+    }
+
+    icount_sleep = sleep;
     if (icount_sleep) {
         timers_state.icount_warp_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL_RT,
                                          icount_timer_cb, NULL);
     }
 
-    icount_align_option = qemu_opt_get_bool(opts, "align", false);
+    icount_align_option = align;
 
-    if (icount_align_option && !icount_sleep) {
-        error_setg(errp, "align=on and sleep=off are incompatible");
-    }
-    if (strcmp(option, "auto") != 0) {
-        errno = 0;
-        timers_state.icount_time_shift = strtol(option, &rem_str, 0);
-        if (errno != 0 || *rem_str != '\0' || !strlen(option)) {
-            error_setg(errp, "icount: Invalid shift value");
-        }
+    if (time_shift >= 0) {
+        timers_state.icount_time_shift = time_shift;
         use_icount = 1;
         return;
-    } else if (icount_align_option) {
-        error_setg(errp, "shift=auto and align=on are incompatible");
-    } else if (!icount_sleep) {
-        error_setg(errp, "shift=auto and sleep=off are incompatible");
     }
 
     use_icount = 2;
Signature · 1 line
2.21.1
#5 lore raw

PATCH 4/11 cpus: Proper range-checking for -icount shift=N

In reply to Markus Armbruster

timers_state.icount_time_shift must be in [0,63] to avoid undefined behavior when shifting by it, e.g. in cpu_icount_to_ns(). icount_adjust() clamps it to [0,MAX_ICOUNT_SHIFT], with MAX_ICOUNT_SHIFT = 10. configure_icount() doesn't. Fix that.

Fixes a8bfac37085c3372366d722f131a7e18d664ee4d
Cc Paolo Bonzini <[email protected]>
---
 cpus.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
cpus.c +3 -4
index 1b542b37f9..5670c96bcf 100644
@@ -25,6 +25,7 @@
 #include "qemu/osdep.h"
 #include "qemu-common.h"
 #include "qemu/config-file.h"
+#include "qemu/cutils.h"
 #include "migration/vmstate.h"
 #include "monitor/monitor.h"
 #include "qapi/error.h"
@@ -801,7 +802,6 @@ void configure_icount(QemuOpts *opts, Error **errp)
     bool sleep = qemu_opt_get_bool(opts, "sleep", true);
     bool align = qemu_opt_get_bool(opts, "align", false);
     long time_shift = -1;
-    char *rem_str = NULL;
 
     if (!option && qemu_opt_get(opts, "align")) {
         error_setg(errp, "Please specify shift option when using align");
@@ -814,9 +814,8 @@ void configure_icount(QemuOpts *opts, Error **errp)
     }
 
     if (strcmp(option, "auto") != 0) {
-        errno = 0;
-        time_shift = strtol(option, &rem_str, 0);
-        if (errno != 0 || *rem_str != '\0' || !strlen(option)) {
+        if (qemu_strtol(option, NULL, 0, &time_shift) < 0
+            || time_shift < 0 || time_shift > MAX_ICOUNT_SHIFT) {
             error_setg(errp, "icount: Invalid shift value");
             return;
         }
Signature · 1 line
2.21.1
#6 lore raw

PATCH 5/11 arm/virt: Fix virt_machine_device_plug_cb() error API violation

In reply to Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

virt_machine_device_plug_cb() passes @errp to cryptodev_builtin_sym_close_session() in a loop. Harmless, because cryptodev_builtin_sym_close_session() can't actually fail. Fix by dropping its Error ** parameter.

---
 hw/arm/virt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
hw/arm/virt.c +2 -2
index 7dc96abf72..cca5316256 100644
@@ -1186,7 +1186,7 @@ static void create_smmu(const VirtMachineState *vms,
     g_free(node);
 }
 
-static void create_virtio_iommu_dt_bindings(VirtMachineState *vms, Error **errp)
+static void create_virtio_iommu_dt_bindings(VirtMachineState *vms)
 {
     const char compat[] = "virtio,pci-iommu";
     uint16_t bdf = vms->virtio_iommu_bdf;
@@ -2118,7 +2118,7 @@ static void virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
 
         vms->iommu = VIRT_IOMMU_VIRTIO;
         vms->virtio_iommu_bdf = pci_get_bdf(pdev);
-        create_virtio_iommu_dt_bindings(vms, errp);
+        create_virtio_iommu_dt_bindings(vms);
     }
 }
 
Signature · 1 line
2.21.1
Reviewed-by on v1 5/11
#7 lore raw

Re: [PATCH 05/11] arm/virt: Fix virt_machine_device_plug_cb() error API violation

Quoted 6 lines · Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

Quoted 8 lines

virt_machine_device_plug_cb() passes @errp to cryptodev_builtin_sym_close_session() in a loop. Harmless, because cryptodev_builtin_sym_close_session() can't actually fail. Fix by dropping its Error ** parameter.

Quoted 27 lines
---
  hw/arm/virt.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
hw/arm/virt.c +2 -2
index 7dc96abf72..cca5316256 100644
@@ -1186,7 +1186,7 @@ static void create_smmu(const VirtMachineState *vms,
      g_free(node);
  }
 
-static void create_virtio_iommu_dt_bindings(VirtMachineState *vms, Error **errp)
+static void create_virtio_iommu_dt_bindings(VirtMachineState *vms)
  {
      const char compat[] = "virtio,pci-iommu";
      uint16_t bdf = vms->virtio_iommu_bdf;
@@ -2118,7 +2118,7 @@ static void virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
 
          vms->iommu = VIRT_IOMMU_VIRTIO;
          vms->virtio_iommu_bdf = pci_get_bdf(pdev);
-        create_virtio_iommu_dt_bindings(vms, errp);
+        create_virtio_iommu_dt_bindings(vms);
      }
  }
 
#8 lore raw

PATCH 6/11 fdc: Fix fallback=auto error handling

In reply to Markus Armbruster

fdctrl_realize_common() rejects fallback=auto. Used by devices "isa-fdc", "sysbus-fdc", "SUNW,fdtwo". The error handling is broken:

    $ qemu-system-x86_64 -nodefaults -device isa-fdc,fallback=auto,driveA=fd0 -drive if=none,id=fd0
    **
    ERROR:/work/armbru/qemu/hw/block/fdc.c:434:pick_drive_type: assertion failed: (drv->drive != FLOPPY_DRIVE_TYPE_AUTO)
    Aborted (core dumped)

Cause: fdctrl_realize_common() neglects to bail out after setting the error. Fix that.

Fixes a73275dd6fc3bfda33165bebc28e0c33c20cb0a0
Cc John Snow <[email protected]>
---
 hw/block/fdc.c | 1 +
 1 file changed, 1 insertion(+)
hw/block/fdc.c +1 -0
index 33bc9e2f92..9628cc171e 100644
@@ -2615,6 +2615,7 @@ static void fdctrl_realize_common(DeviceState *dev, FDCtrl *fdctrl,
 
     if (fdctrl->fallback == FLOPPY_DRIVE_TYPE_AUTO) {
         error_setg(errp, "Cannot choose a fallback FDrive type of 'auto'");
+        return;
     }
 
     /* Fill 'command_to_handler' lookup table */
Signature · 1 line
2.21.1
Reviewed-by on v1 6/11
#9 lore raw

Re: [PATCH 06/11] fdc: Fix fallback=auto error handling

Quoted 24 lines · Markus Armbruster

fdctrl_realize_common() rejects fallback=auto. Used by devices "isa-fdc", "sysbus-fdc", "SUNW,fdtwo". The error handling is broken:

     $ qemu-system-x86_64 -nodefaults -device isa-fdc,fallback=auto,driveA=fd0 -drive if=none,id=fd0
     **
     ERROR:/work/armbru/qemu/hw/block/fdc.c:434:pick_drive_type: assertion failed: (drv->drive != FLOPPY_DRIVE_TYPE_AUTO)
     Aborted (core dumped)

Cause: fdctrl_realize_common() neglects to bail out after setting the error. Fix that.

Fixes a73275dd6fc3bfda33165bebc28e0c33c20cb0a0
Cc John Snow <[email protected]>
---
  hw/block/fdc.c | 1 +
  1 file changed, 1 insertion(+)
hw/block/fdc.c +0 -0
index 33bc9e2f92..9628cc171e 100644

hw/block/fdc.c static void fdctrl_realize_common(DeviceState *dev, FDCtrl *fdctrl, · line 2621

Quoted 7 lines
hw/block/fdc.c +1 -0
@@ -2615,6 +2615,7 @@ static void fdctrl_realize_common(DeviceState *dev, FDCtrl *fdctrl,
 
      if (fdctrl->fallback == FLOPPY_DRIVE_TYPE_AUTO) {
          error_setg(errp, "Cannot choose a fallback FDrive type of 'auto'");
+        return;
      }
 
      /* Fill 'command_to_handler' lookup table */
#10 lore raw

PATCH 7/11 bochs-display: Fix vgamem=SIZE error handling

In reply to Markus Armbruster

bochs_display_realize() rejects out-of-range vgamem. The error handling is broken:

    $ qemu-system-x86_64 -S -display none -monitor stdio
    QEMU 4.2.93 monitor - type 'help' for more information
    (qemu) device_add bochs-display,vgamem=1
    Error: bochs-display: video memory too small
    (qemu) device_add bochs-display,vgamem=1
    RAMBlock "0000:00:04.0/bochs-display-vram" already registered, abort!
    Aborted (core dumped)

Cause: bochs_display_realize() neglects to bail out after setting the error. Fix that.

Fixes 765c94290863eef1fc4a67819d452cc13b7854a1
Cc Gerd Hoffmann <[email protected]>
---
 hw/display/bochs-display.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
hw/display/bochs-display.c +4 -2
index 70eb619ef4..e763a0a72d 100644
@@ -267,16 +267,18 @@ static void bochs_display_realize(PCIDevice *dev, Error **errp)
     Object *obj = OBJECT(dev);
     int ret;
 
-    s->con = graphic_console_init(DEVICE(dev), 0, &bochs_display_gfx_ops, s);
-
     if (s->vgamem < 4 * MiB) {
         error_setg(errp, "bochs-display: video memory too small");
+        return;
     }
     if (s->vgamem > 256 * MiB) {
         error_setg(errp, "bochs-display: video memory too big");
+        return;
     }
     s->vgamem = pow2ceil(s->vgamem);
 
+    s->con = graphic_console_init(DEVICE(dev), 0, &bochs_display_gfx_ops, s);
+
     memory_region_init_ram(&s->vram, obj, "bochs-display-vram", s->vgamem,
                            &error_fatal);
     memory_region_init_io(&s->vbe, obj, &bochs_display_vbe_ops, s,
Signature · 1 line
2.21.1
Reviewed-by on v1 7/11
#11 lore raw

Re: [PATCH 07/11] bochs-display: Fix vgamem=SIZE error handling

Quoted 37 lines · Markus Armbruster

bochs_display_realize() rejects out-of-range vgamem. The error handling is broken:

     $ qemu-system-x86_64 -S -display none -monitor stdio
     QEMU 4.2.93 monitor - type 'help' for more information
     (qemu) device_add bochs-display,vgamem=1
     Error: bochs-display: video memory too small
     (qemu) device_add bochs-display,vgamem=1
     RAMBlock "0000:00:04.0/bochs-display-vram" already registered, abort!
     Aborted (core dumped)

Cause: bochs_display_realize() neglects to bail out after setting the error. Fix that.

Fixes 765c94290863eef1fc4a67819d452cc13b7854a1
Cc Gerd Hoffmann <[email protected]>
---
  hw/display/bochs-display.c | 6 ++++--
  1 file changed, 4 insertions(+), 2 deletions(-)
hw/display/bochs-display.c +1 -2
index 70eb619ef4..e763a0a72d 100644
@@ -267,16 +267,18 @@ static void bochs_display_realize(PCIDevice *dev, Error **errp)
      Object *obj = OBJECT(dev);
      int ret;
 
-    s->con = graphic_console_init(DEVICE(dev), 0, &bochs_display_gfx_ops, s);
-
      if (s->vgamem < 4 * MiB) {
          error_setg(errp, "bochs-display: video memory too small");
+        return;
      }
      if (s->vgamem > 256 * MiB) {

hw/display/bochs-display.c static void bochs_display_realize(PCIDevice *dev, Error **errp) · line 284

Quoted 10 lines
hw/display/bochs-display.c +3 -0
          error_setg(errp, "bochs-display: video memory too big");
+        return;
      }
      s->vgamem = pow2ceil(s->vgamem);
 
+    s->con = graphic_console_init(DEVICE(dev), 0, &bochs_display_gfx_ops, s);
+
      memory_region_init_ram(&s->vram, obj, "bochs-display-vram", s->vgamem,
                             &error_fatal);
      memory_region_init_io(&s->vbe, obj, &bochs_display_vbe_ops, s,
#12 lore raw

PATCH 8/11 virtio-net: Fix duplex=... and speed=... error handling

In reply to Markus Armbruster

virtio_net_device_realize() rejects invalid duplex and speed values. The error handling is broken:

    $ ../qemu/bld-sani/x86_64-softmmu/qemu-system-x86_64 -S -display none -monitor stdio
    QEMU 4.2.93 monitor - type 'help' for more information
    (qemu) device_add virtio-net,duplex=x
    Error: 'duplex' must be 'half' or 'full'
    (qemu) c
    =================================================================
    ==15654==ERROR: AddressSanitizer: heap-use-after-free on address 0x62e000014590 at pc 0x560b75c8dc13 bp 0x7fffdf1a6950 sp 0x7fffdf1a6940
    READ of size 8 at 0x62e000014590 thread T0
	#0 0x560b75c8dc12 in object_dynamic_cast_assert /work/armbru/qemu/qom/object.c:826
	#1 0x560b74c38ac0 in virtio_vmstate_change /work/armbru/qemu/hw/virtio/virtio.c:3210
	#2 0x560b74d9765e in vm_state_notify /work/armbru/qemu/softmmu/vl.c:1271
	#3 0x560b7494ba72 in vm_prepare_start /work/armbru/qemu/cpus.c:2156
	#4 0x560b7494bacd in vm_start /work/armbru/qemu/cpus.c:2162
	#5 0x560b75a7d890 in qmp_cont /work/armbru/qemu/monitor/qmp-cmds.c:160
	#6 0x560b75a8d70a in hmp_cont /work/armbru/qemu/monitor/hmp-cmds.c:1043
	#7 0x560b75a799f2 in handle_hmp_command /work/armbru/qemu/monitor/hmp.c:1082
    [...]
    0x62e000014590 is located 33168 bytes inside of 42288-byte region [0x62e00000c400,0x62e000016930)
    freed by thread T1 here:
	#0 0x7feadd39491f in __interceptor_free (/lib64/libasan.so.5+0x10d91f)
	#1 0x7feadcebcd7c in g_free (/lib64/libglib-2.0.so.0+0x55d7c)
	#2 0x560b75c8fd40 in object_unref /work/armbru/qemu/qom/object.c:1128
	#3 0x560b7498a625 in memory_region_unref /work/armbru/qemu/memory.c:1762
	#4 0x560b74999fa4 in do_address_space_destroy /work/armbru/qemu/memory.c:2788
	#5 0x560b762362fc in call_rcu_thread /work/armbru/qemu/util/rcu.c:283
	#6 0x560b761c8884 in qemu_thread_start /work/armbru/qemu/util/qemu-thread-posix.c:519
	#7 0x7fead9be34bf in start_thread (/lib64/libpthread.so.0+0x84bf)

    previously allocated by thread T0 here:
	#0 0x7feadd394d18 in __interceptor_malloc (/lib64/libasan.so.5+0x10dd18)
	#1 0x7feadcebcc88 in g_malloc (/lib64/libglib-2.0.so.0+0x55c88)
	#2 0x560b75c8cf8a in object_new /work/armbru/qemu/qom/object.c:699
	#3 0x560b75010ad9 in qdev_device_add /work/armbru/qemu/qdev-monitor.c:654
	#4 0x560b750120c2 in qmp_device_add /work/armbru/qemu/qdev-monitor.c:805
	#5 0x560b75012c1b in hmp_device_add /work/armbru/qemu/qdev-monitor.c:905
    [...]
    ==15654==ABORTING

Cause: virtio_net_device_realize() neglects to bail out after setting the error. Fix that.

Fixes 9473939ed7addcaaeb8fde5c093918fb7fa0919c
Cc "Michael S. Tsirkin" <[email protected]>
Cc Jason Wang <[email protected]>
---
 hw/net/virtio-net.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
hw/net/virtio-net.c +4 -1
index a46e3b37a7..b52ff4ab63 100644
@@ -2947,6 +2947,7 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
             n->net_conf.duplex = DUPLEX_FULL;
         } else {
             error_setg(errp, "'duplex' must be 'half' or 'full'");
+            return;
         }
         n->host_features |= (1ULL << VIRTIO_NET_F_SPEED_DUPLEX);
     } else {
@@ -2955,7 +2956,9 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
 
     if (n->net_conf.speed < SPEED_UNKNOWN) {
         error_setg(errp, "'speed' must be between 0 and INT_MAX");
-    } else if (n->net_conf.speed >= 0) {
+        return;
+    }
+    if (n->net_conf.speed >= 0) {
         n->host_features |= (1ULL << VIRTIO_NET_F_SPEED_DUPLEX);
     }
 
Signature · 1 line
2.21.1
Reviewed-by on v1 8/11
#13 lore raw

Re: [PATCH 08/11] virtio-net: Fix duplex=... and speed=... error handling

Quoted 67 lines · Markus Armbruster

virtio_net_device_realize() rejects invalid duplex and speed values. The error handling is broken:

     $ ../qemu/bld-sani/x86_64-softmmu/qemu-system-x86_64 -S -display none -monitor stdio
     QEMU 4.2.93 monitor - type 'help' for more information
     (qemu) device_add virtio-net,duplex=x
     Error: 'duplex' must be 'half' or 'full'
     (qemu) c
     =================================================================
     ==15654==ERROR: AddressSanitizer: heap-use-after-free on address 0x62e000014590 at pc 0x560b75c8dc13 bp 0x7fffdf1a6950 sp 0x7fffdf1a6940
     READ of size 8 at 0x62e000014590 thread T0
	#0 0x560b75c8dc12 in object_dynamic_cast_assert /work/armbru/qemu/qom/object.c:826
	#1 0x560b74c38ac0 in virtio_vmstate_change /work/armbru/qemu/hw/virtio/virtio.c:3210
	#2 0x560b74d9765e in vm_state_notify /work/armbru/qemu/softmmu/vl.c:1271
	#3 0x560b7494ba72 in vm_prepare_start /work/armbru/qemu/cpus.c:2156
	#4 0x560b7494bacd in vm_start /work/armbru/qemu/cpus.c:2162
	#5 0x560b75a7d890 in qmp_cont /work/armbru/qemu/monitor/qmp-cmds.c:160
	#6 0x560b75a8d70a in hmp_cont /work/armbru/qemu/monitor/hmp-cmds.c:1043
	#7 0x560b75a799f2 in handle_hmp_command /work/armbru/qemu/monitor/hmp.c:1082
     [...]
     0x62e000014590 is located 33168 bytes inside of 42288-byte region [0x62e00000c400,0x62e000016930)
     freed by thread T1 here:
	#0 0x7feadd39491f in __interceptor_free (/lib64/libasan.so.5+0x10d91f)
	#1 0x7feadcebcd7c in g_free (/lib64/libglib-2.0.so.0+0x55d7c)
	#2 0x560b75c8fd40 in object_unref /work/armbru/qemu/qom/object.c:1128
	#3 0x560b7498a625 in memory_region_unref /work/armbru/qemu/memory.c:1762
	#4 0x560b74999fa4 in do_address_space_destroy /work/armbru/qemu/memory.c:2788
	#5 0x560b762362fc in call_rcu_thread /work/armbru/qemu/util/rcu.c:283
	#6 0x560b761c8884 in qemu_thread_start /work/armbru/qemu/util/qemu-thread-posix.c:519
	#7 0x7fead9be34bf in start_thread (/lib64/libpthread.so.0+0x84bf)

     previously allocated by thread T0 here:
	#0 0x7feadd394d18 in __interceptor_malloc (/lib64/libasan.so.5+0x10dd18)
	#1 0x7feadcebcc88 in g_malloc (/lib64/libglib-2.0.so.0+0x55c88)
	#2 0x560b75c8cf8a in object_new /work/armbru/qemu/qom/object.c:699
	#3 0x560b75010ad9 in qdev_device_add /work/armbru/qemu/qdev-monitor.c:654
	#4 0x560b750120c2 in qmp_device_add /work/armbru/qemu/qdev-monitor.c:805
	#5 0x560b75012c1b in hmp_device_add /work/armbru/qemu/qdev-monitor.c:905
     [...]
     ==15654==ABORTING

Cause: virtio_net_device_realize() neglects to bail out after setting the error. Fix that.

Fixes 9473939ed7addcaaeb8fde5c093918fb7fa0919c
Cc "Michael S. Tsirkin" <[email protected]>
Cc Jason Wang <[email protected]>
---
  hw/net/virtio-net.c | 5 ++++-
  1 file changed, 4 insertions(+), 1 deletion(-)
hw/net/virtio-net.c +1 -0
index a46e3b37a7..b52ff4ab63 100644
@@ -2947,6 +2947,7 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
              n->net_conf.duplex = DUPLEX_FULL;
          } else {
              error_setg(errp, "'duplex' must be 'half' or 'full'");
+            return;
          }
          n->host_features |= (1ULL << VIRTIO_NET_F_SPEED_DUPLEX);
      } else {

hw/net/virtio-net.c static void virtio_net_device_realize(DeviceState *dev, Error **errp) · line 2964

Quoted 10 lines
hw/net/virtio-net.c +3 -1
@@ -2955,7 +2956,9 @@ static void virtio_net_device_realize(DeviceState *dev, Error **errp)
 
      if (n->net_conf.speed < SPEED_UNKNOWN) {
          error_setg(errp, "'speed' must be between 0 and INT_MAX");
-    } else if (n->net_conf.speed >= 0) {
+        return;
+    }
+    if (n->net_conf.speed >= 0) {
          n->host_features |= (1ULL << VIRTIO_NET_F_SPEED_DUPLEX);
      }
 
#14 lore raw

PATCH 9/11 xen/pt: Fix flawed conversion to realize()

In reply to Markus Armbruster

The conversion of xen_pt_initfn() to xen_pt_realize() blindly replaced XEN_PT_ERR() by error_setg(). Several error conditions that did not fail xen_pt_initfn() now fail xen_pt_realize(). Unsurprisingly, the cleanup on these errors looks highly suspicious.

Revert the inappropriate replacements.

Fixes 5a11d0f7549e24a10e178a9dc8ff5e698031d9a6
Cc Stefano Stabellini <[email protected]>
Cc Anthony Perard <[email protected]>
Cc Paul Durrant <[email protected]>
---
 hw/xen/xen_pt.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
hw/xen/xen_pt.c +6 -6
index b91082cb8b..81d5ad8da7 100644
@@ -858,8 +858,8 @@ static void xen_pt_realize(PCIDevice *d, Error **errp)
 
     rc = xc_physdev_map_pirq(xen_xc, xen_domid, machine_irq, &pirq);
     if (rc < 0) {
-        error_setg_errno(errp, errno, "Mapping machine irq %u to"
-                         " pirq %i failed", machine_irq, pirq);
+        XEN_PT_ERR(d, "Mapping machine irq %u to pirq %i failed, (err: %d)\n",
+                   machine_irq, pirq, errno);
 
         /* Disable PCI intx assertion (turn on bit10 of devctl) */
         cmd |= PCI_COMMAND_INTX_DISABLE;
@@ -880,8 +880,8 @@ static void xen_pt_realize(PCIDevice *d, Error **errp)
                                        PCI_SLOT(d->devfn),
                                        e_intx);
         if (rc < 0) {
-            error_setg_errno(errp, errno, "Binding of interrupt %u failed",
-                             e_intx);
+            XEN_PT_ERR(d, "Binding of interrupt %i failed! (err: %d)\n",
+                       e_intx, errno);
 
             /* Disable PCI intx assertion (turn on bit10 of devctl) */
             cmd |= PCI_COMMAND_INTX_DISABLE;
@@ -889,8 +889,8 @@ static void xen_pt_realize(PCIDevice *d, Error **errp)
 
             if (xen_pt_mapped_machine_irq[machine_irq] == 0) {
                 if (xc_physdev_unmap_pirq(xen_xc, xen_domid, machine_irq)) {
-                    error_setg_errno(errp, errno, "Unmapping of machine"
-                            " interrupt %u failed", machine_irq);
+                    XEN_PT_ERR(d, "Unmapping of machine interrupt %i failed!"
+                               " (err: %d)\n", machine_irq, errno);
                 }
             }
             s->machine_irq = 0;
Signature · 1 line
2.21.1
Reviewed-by on v1 9/11
#15 lore raw

RE: [PATCH 09/11] xen/pt: Fix flawed conversion to realize()

Quoted 13 lines

-----Original Message-----
From: Markus Armbruster <[email protected]>
Sent: 20 April 2020 09:33
To: [email protected]
Cc: Stefano Stabellini <[email protected]>; Anthony Perard <[email protected]>; Paul Durrant <[email protected]>; [email protected]
Subject: [PATCH 09/11] xen/pt: Fix flawed conversion to realize()

The conversion of xen_pt_initfn() to xen_pt_realize() blindly replaced XEN_PT_ERR() by error_setg(). Several error conditions that did not fail xen_pt_initfn() now fail xen_pt_realize(). Unsurprisingly, the cleanup on these errors looks highly suspicious.

Quoted 8 lines

Revert the inappropriate replacements.

Fixes 5a11d0f7549e24a10e178a9dc8ff5e698031d9a6
Cc Stefano Stabellini <[email protected]>
Cc Anthony Perard <[email protected]>
Cc Paul Durrant <[email protected]>
#16 lore raw

PATCH 10/11 io: Fix qio_channel_socket_close() error handling

In reply to Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qio_channel_socket_close() passes @errp first to socket_listen_cleanup(), and then, if closesocket() fails, to error_setg_errno(). If socket_listen_cleanup() failed, this will trip the assertion in error_setv().

Fix by ignoring a second error.

Fixes 73564c407caedf992a1c688b5fea776a8b56ba2a
Cc Daniel P. Berrangé <[email protected]>
---
 io/channel-socket.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
io/channel-socket.c +3 -2
index b74f5b92a0..e1b4667087 100644
@@ -704,6 +704,7 @@ qio_channel_socket_close(QIOChannel *ioc,
 {
     QIOChannelSocket *sioc = QIO_CHANNEL_SOCKET(ioc);
     int rc = 0;
+    Error *err = NULL;
 
     if (sioc->fd != -1) {
 #ifdef WIN32
@@ -715,8 +716,8 @@ qio_channel_socket_close(QIOChannel *ioc,
 
         if (closesocket(sioc->fd) < 0) {
             sioc->fd = -1;
-            error_setg_errno(errp, errno,
-                             "Unable to close socket");
+            error_setg_errno(&err, errno, "Unable to close socket");
+            error_propagate(errp, err);
             return -1;
         }
         sioc->fd = -1;
Signature · 1 line
2.21.1
Reviewed-by on v1 10/11
#17 lore raw

Re: [PATCH 10/11] io: Fix qio_channel_socket_close() error handling

Quoted 11 lines · Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qio_channel_socket_close() passes @errp first to socket_listen_cleanup(), and then, if closesocket() fails, to error_setg_errno(). If socket_listen_cleanup() failed, this will trip the assertion in error_setv().

Quoted 8 lines

Fix by ignoring a second error.

Fixes 73564c407caedf992a1c688b5fea776a8b56ba2a
Cc Daniel P. Berrangé <[email protected]>
---
 io/channel-socket.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

Regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#18 lore raw

PATCH 11/11 migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

In reply to Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qmp_xen_colo_do_checkpoint() passes @errp first to replication_do_checkpoint_all(), and then to colo_notify_filters_event(). If both fail, this will trip the assertion in error_setv().

Similar code in secondary_vm_do_failover() calls colo_notify_filters_event() only after replication_do_checkpoint_all() succeeded. Do the same here.

Fixes 0e8818f023616677416840d6ddc880db8de3c967
Cc Zhang Chen <[email protected]>
Cc zhanghailiang <[email protected]>
---
 migration/colo.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
migration/colo.c +7 -1
index a54ac84f41..1b3493729b 100644
@@ -263,7 +263,13 @@ ReplicationStatus *qmp_query_xen_replication_status(Error **errp)
 
 void qmp_xen_colo_do_checkpoint(Error **errp)
 {
-    replication_do_checkpoint_all(errp);
+    Error *err = NULL;
+
+    replication_do_checkpoint_all(&err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
     /* Notify all filters of all NIC to do checkpoint */
     colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);
 }
Signature · 1 line
2.21.1
Reviewed-by on v1 11/11
#19 lore raw

RE: [PATCH 11/11] migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

Quoted 54 lines

-----Original Message-----
From: Markus Armbruster [mailto:[email protected]]
Sent: Monday, April 20, 2020 4:33 PM
To: [email protected]
Cc: Zhang Chen <[email protected]>; Zhanghailiang <[email protected]>
Subject: [PATCH 11/11] migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qmp_xen_colo_do_checkpoint() passes @errp first to replication_do_checkpoint_all(), and then to colo_notify_filters_event(). If both fail, this will trip the assertion in error_setv().

Similar code in secondary_vm_do_failover() calls colo_notify_filters_event() only after replication_do_checkpoint_all() succeeded. Do the same here.

Fixes 0e8818f023616677416840d6ddc880db8de3c967
Cc Zhang Chen <[email protected]>
Cc zhanghailiang <[email protected]>
---
 migration/colo.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
migration/colo.c +7 -1
index a54ac84f41..1b3493729b 100644
@@ -263,7 +263,13 @@ ReplicationStatus
 *qmp_query_xen_replication_status(Error **errp)
 
 void qmp_xen_colo_do_checkpoint(Error **errp)
 {
-    replication_do_checkpoint_all(errp);
+    Error *err = NULL;
+
+    replication_do_checkpoint_all(&err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
     /* Notify all filters of all NIC to do checkpoint */
     colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);

}
--
2.21.1

Reviewed-by on v1 11/11
#20 lore raw

Re: [PATCH 11/11] migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

In reply to Markus Armbruster
Quoted 33 lines · Markus Armbruster

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qmp_xen_colo_do_checkpoint() passes @errp first to replication_do_checkpoint_all(), and then to colo_notify_filters_event(). If both fail, this will trip the assertion in error_setv().

Similar code in secondary_vm_do_failover() calls colo_notify_filters_event() only after replication_do_checkpoint_all() succeeded. Do the same here.

Fixes 0e8818f023616677416840d6ddc880db8de3c967
Cc Zhang Chen <[email protected]>
Cc zhanghailiang <[email protected]>
---
  migration/colo.c | 8 +++++++-
  1 file changed, 7 insertions(+), 1 deletion(-)
migration/colo.c +0 -1
index a54ac84f41..1b3493729b 100644
@@ -263,7 +263,13 @@ ReplicationStatus *qmp_query_xen_replication_status(Error **errp)
 
  void qmp_xen_colo_do_checkpoint(Error **errp)
  {
-    replication_do_checkpoint_all(errp);

migration/colo.c ReplicationStatus *qmp_query_xen_replication_status(Error **errp) · line 275

Quoted 10 lines
migration/colo.c +7 -0
+    Error *err = NULL;
+
+    replication_do_checkpoint_all(&err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
      /* Notify all filters of all NIC to do checkpoint */
      colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);
  }
Reviewed-by on v1 11/11
#21 lore raw

RE: [PATCH 11/11] migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

In reply to Markus Armbruster
Quoted 20 lines

-----Original Message-----
From: Markus Armbruster <[email protected]>
Sent: Monday, April 20, 2020 4:33 PM
To: [email protected]
Cc: Zhang, Chen <[email protected]>; zhanghailiang <[email protected]>
Subject: [PATCH 11/11] migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling

The Error ** argument must be NULL, &error_abort, &error_fatal, or a pointer to a variable containing NULL. Passing an argument of the latter kind twice without clearing it in between is wrong: if the first call sets an error, it no longer points to NULL for the second call.

qmp_xen_colo_do_checkpoint() passes @errp first to replication_do_checkpoint_all(), and then to colo_notify_filters_event(). If both fail, this will trip the assertion in error_setv().

Quoted 8 lines

Similar code in secondary_vm_do_failover() calls colo_notify_filters_event() only after replication_do_checkpoint_all() succeeded. Do the same here.

Fixes 0e8818f023616677416840d6ddc880db8de3c967
Cc Zhang Chen <[email protected]>
Cc zhanghailiang <[email protected]>

Looks good for me.

Thanks
Zhang Chen

Quoted 24 lines
---
 migration/colo.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
migration/colo.c index a54ac84f41..1b3493729b renamed from migration/colo.c +0 -0

100644

migration/colo.c +7 -2
@@ -263,7 +263,13 @@ ReplicationStatus
 *qmp_query_xen_replication_status(Error **errp)
 
 void qmp_xen_colo_do_checkpoint(Error **errp)  {
-    replication_do_checkpoint_all(errp);
+    Error *err = NULL;
+
+    replication_do_checkpoint_all(&err);
+    if (err) {
+        error_propagate(errp, err);
+        return;
+    }
     /* Notify all filters of all NIC to do checkpoint */
     colo_notify_filters_event(COLO_EVENT_CHECKPOINT, errp);  }
--

2.21.1

#22 lore raw
In reply to Markus Armbruster
Patchew URL: https://patchew.org/QEMU/[email protected]/

Patchew URL: https://patchew.org/QEMU/[email protected]/

Hi,

This series seems to have some coding style problems. See output below for more information:

Subject: [PATCH 00/11] Miscellaneous error handling fixes
Message-id: [email protected]
Type: series

=== TEST SCRIPT BEGIN ===
#!/bin/bash
git rev-parse base > /dev/null || exit 0
git config --local diff.renamelimit 0
git config --local diff.renames True
git config --local diff.algorithm histogram
./scripts/checkpatch.pl --mailback base..
=== TEST SCRIPT END ===

Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384
Switched to a new branch 'test'
942181b migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling a8c2f29 io: Fix qio_channel_socket_close() error handling cedb56e xen/pt: Fix flawed conversion to realize()
7173a0e virtio-net: Fix duplex=... and speed=... error handling 9c7afc3 bochs-display: Fix vgamem=SIZE error handling f3ad7d4 fdc: Fix fallback=auto error handling
e484e15 arm/virt: Fix virt_machine_device_plug_cb() error API violation 1c54f10 cpus: Proper range-checking for -icount shift=N 0a69dc4 cpus: Fix configure_icount() error API violation bafe5f5 block/file-posix: Fix check_cache_dropped() error handling a690fa1 cryptodev: Fix cryptodev_builtin_cleanup() error API violation

=== OUTPUT BEGIN ===
1/11 Checking commit a690fa1dc425 (cryptodev: Fix cryptodev_builtin_cleanup() error API violation) 2/11 Checking commit bafe5f5e9888 (block/file-posix: Fix check_cache_dropped() error handling) 3/11 Checking commit 0a69dc45007c (cpus: Fix configure_icount() error API violation)
ERROR: consider using qemu_strtol in preference to strtol
#58: FILE: cpus.c:818:
+ time_shift = strtol(option, &rem_str, 0);

total: 1 errors, 0 warnings, 70 lines checked

Patch 3/11 has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS.

4/11 Checking commit 1c54f102aed3 (cpus: Proper range-checking for -icount shift=N) 5/11 Checking commit e484e1536ed7 (arm/virt: Fix virt_machine_device_plug_cb() error API violation) 6/11 Checking commit f3ad7d444a66 (fdc: Fix fallback=auto error handling)
7/11 Checking commit 9c7afc3c57a4 (bochs-display: Fix vgamem=SIZE error handling) 8/11 Checking commit 7173a0e399fd (virtio-net: Fix duplex=... and speed=... error handling) 9/11 Checking commit cedb56ec3da3 (xen/pt: Fix flawed conversion to realize())
10/11 Checking commit a8c2f294abb2 (io: Fix qio_channel_socket_close() error handling) 11/11 Checking commit 942181b72505 (migration/colo: Fix qmp_xen_colo_do_checkpoint() error handling) === OUTPUT END ===

Test command exited with code: 1

The full log is available at
http://patchew.org/logs/[email protected]/testing.checkpatch/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]