Skip to contentSkip to conversation

openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized in this function [-Werror=maybe-uninitialized]

Bug 1874073 Applied Martin Liska 5 participants 13 messages Apr 21, 2020 – Aug 20, 2020

lore
Share & tools
lore
#0 lore raw

Public bug reported:

I see the warning since gcc10:

static void openrisc_sim_init(MachineState *machine):
...

    qemu_irq *cpu_irqs[2];

...

    serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                   115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

I would initialize cpu_irqs[2] with {}.

** Affects: qemu
Importance: Undecided
Status: New

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#1 lore raw

Suggested patch:

hw/openrisc/openrisc_sim.c +1 -1
index 79e7049..724dcd0 100644
@@ -129,7 +129,7 @@ static void openrisc_sim_init(MachineState *machine)
     const char *kernel_filename = machine->kernel_filename;
     OpenRISCCPU *cpu = NULL;
     MemoryRegion *ram;
-    qemu_irq *cpu_irqs[2];
+    qemu_irq *cpu_irqs[2] = {};
     qemu_irq serial_irq;
     int n;
     unsigned int smp_cpus = machine->smp.cpus;
Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#2 lore raw
In reply to Martin Liska

I'm not sure why the compiler thinks it might be uninitialized here -- is it just that it's not aware that smp_cpus can't be zero and so the loop will always initialize cpu_irqs[0] ?

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#3 lore raw
In reply to Martin Liska

Note that our -Wmaybe-uninitialized warnings tends to report false positives. We have a rich meta bug for it:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=24639

That's why it can't prove the variable is initialized in all possible execution paths.

** Bug watch added: GCC Bugzilla #24639
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=24639

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#4 lore raw
In reply to Martin Liska

Does adding "assert(smp_cpus >= 1 && smp_cpus <= 2);" after the assignment to smp_cpus give the compiler enough information to know there's no use of uninitialized data?

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#5 lore raw
In reply to Martin Liska

Confirming Peter's suggestion does silent the warning.

-- >8 --

hw/openrisc/openrisc_sim.c +1 -0
index d08ce61811..02f5259e5e 100644
@@ -134,6 +134,7 @@ static void openrisc_sim_init(MachineState *machine)
     int n;
     unsigned int smp_cpus = machine->smp.cpus;
 
+    assert(smp_cpus >= 1 && smp_cpus <= 2);
     for (n = 0; n < smp_cpus; n++) {
         cpu = OPENRISC_CPU(cpu_create(machine->cpu_type));
         if (cpu == NULL) {

---

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#6 lore raw
In reply to Martin Liska

Hello. The assert approach is preferred.

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  New

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#7 lore raw
In reply to Martin Liska

** Changed in: qemu
Status: New => Confirmed

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  Confirmed

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#8 lore raw
In reply to Martin Liska

Merged in commit 1db889c71f37d5bad411b2ef83a69739d9d598f9.

** Changed in: qemu
Status: Confirmed => Fix Committed

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  Fix Committed

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#9 lore raw
In reply to Martin Liska

** Changed in: qemu
Status: Fix Committed => Fix Released

Signature · 27 lines
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1874073

Title:
  openrisc_sim.c:87:42: error: 'cpu_irqs[0]' may be used uninitialized
  in this function [-Werror=maybe-uninitialized]

Status in QEMU:
  Fix Released

Bug description:
  I see the warning since gcc10:

  static void openrisc_sim_init(MachineState *machine):
  ...
      qemu_irq *cpu_irqs[2];
  ...

  
      serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq,
                     115200, serial_hd(0), DEVICE_NATIVE_ENDIAN);

  I would initialize cpu_irqs[2] with {}.

To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1874073/+subscriptions
#10 lore raw

PATCH v2 1/1 hw/openrisc/openrisc_sim: Add assertion to silence GCC warning

When compiling with GCC 10 (Fedora 32) using CFLAGS=-O2 we get:

CC or1k-softmmu/hw/openrisc/openrisc_sim.o
hw/openrisc/openrisc_sim.c: In function ‘openrisc_sim_init’:
hw/openrisc/openrisc_sim.c:87:42: error: ‘cpu_irqs[0]’ may be used uninitialized in this function [-Werror=maybe-uninitialized]

     87 |         sysbus_connect_irq(s, i, cpu_irqs[i][irq_pin]);
        |                                  ~~~~~~~~^~~

While humans can tell smp_cpus will always be in the [1, 2] range, (openrisc_sim_machine_init sets mc->max_cpus = 2), the compiler can't.

Add an assertion to give the compiler a hint there's no use of uninitialized data.

---
v2: Fixed typo in subject (eblake)
Supersedes: <[email protected]>

---
 hw/openrisc/openrisc_sim.c | 1 +
 1 file changed, 1 insertion(+)
hw/openrisc/openrisc_sim.c +1 -0
index d08ce61811..02f5259e5e 100644
@@ -134,6 +134,7 @@ static void openrisc_sim_init(MachineState *machine)
     int n;
     unsigned int smp_cpus = machine->smp.cpus;
 
+    assert(smp_cpus >= 1 && smp_cpus <= 2);
     for (n = 0; n < smp_cpus; n++) {
         cpu = OPENRISC_CPU(cpu_create(machine->cpu_type));
         if (cpu == NULL) {
Signature · 1 line
2.21.3
Reviewed-by on v2 1/1
#11 lore raw

Re: [PATCH v2] hw/openrisc/openrisc_sim: Add assertion to silence GCC warning

Quoted 22 lines · Philippe Mathieu-Daudé

When compiling with GCC 10 (Fedora 32) using CFLAGS=-O2 we get:

CC or1k-softmmu/hw/openrisc/openrisc_sim.o
hw/openrisc/openrisc_sim.c: In function ‘openrisc_sim_init’:
hw/openrisc/openrisc_sim.c:87:42: error: ‘cpu_irqs[0]’ may be used uninitialized in this function [-Werror=maybe-uninitialized]

     87 |         sysbus_connect_irq(s, i, cpu_irqs[i][irq_pin]);
        |                                  ~~~~~~~~^~~

While humans can tell smp_cpus will always be in the [1, 2] range, (openrisc_sim_machine_init sets mc->max_cpus = 2), the compiler can't.

Add an assertion to give the compiler a hint there's no use of uninitialized data.

Quoted 6 lines

---
v2: Fixed typo in subject (eblake)
Supersedes: <[email protected]>

---
 hw/openrisc/openrisc_sim.c | 1 +
 1 file changed, 1 insertion(+)

r~

#12 lore raw

[Bug 1874073] Re: [PATCH v2] hw/openrisc/openrisc_sim: Add assertion to silence GCC warning

In reply to Philippe Mathieu-Daudé

Le 08/06/2020 à 18:06, Philippe Mathieu-Daudé a écrit :

Quoted 35 lines

When compiling with GCC 10 (Fedora 32) using CFLAGS=-O2 we get:

CC or1k-softmmu/hw/openrisc/openrisc_sim.o
hw/openrisc/openrisc_sim.c: In function ‘openrisc_sim_init’:
hw/openrisc/openrisc_sim.c:87:42: error: ‘cpu_irqs[0]’ may be used uninitialized in this function [-Werror=maybe-uninitialized]

     87 |         sysbus_connect_irq(s, i, cpu_irqs[i][irq_pin]);
        |                                  ~~~~~~~~^~~

While humans can tell smp_cpus will always be in the [1, 2] range, (openrisc_sim_machine_init sets mc->max_cpus = 2), the compiler can't.

Add an assertion to give the compiler a hint there's no use of uninitialized data.

---
v2: Fixed typo in subject (eblake)
Supersedes: <[email protected]>

---
 hw/openrisc/openrisc_sim.c | 1 +
 1 file changed, 1 insertion(+)
hw/openrisc/openrisc_sim.c +0 -0
index d08ce61811..02f5259e5e 100644

hw/openrisc/openrisc_sim.c static void openrisc_sim_init(MachineState *machine) · line 140

Quoted 7 lines
hw/openrisc/openrisc_sim.c +1 -0
@@ -134,6 +134,7 @@ static void openrisc_sim_init(MachineState *machine)
     int n;
     unsigned int smp_cpus = machine->smp.cpus;
 
+    assert(smp_cpus >= 1 && smp_cpus <= 2);
     for (n = 0; n < smp_cpus; n++) {
         cpu = OPENRISC_CPU(cpu_create(machine->cpu_type));
         if (cpu == NULL) {

Applied to my trivial-patches branch.

Thanks,
Laurent