Am 7. November 2024 07:04:10 UTC schrieb Dmitry Frolov <fro...@swemel.ru>:
>If pcmc->pci_enabled is false, pcms->pcibus is NULL and is passed
>to pc_nic_init() where it is being dereferenced.
>
>Found making check with enabled sanitizers.
>
>Signed-off-by: Dmitry Frolov <fro...@swemel.ru>
>---
> hw/i386/pc_piix.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
>diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
>index 2bf6865d40..2a92d2dbb7 100644
>--- a/hw/i386/pc_piix.c
>+++ b/hw/i386/pc_piix.c
>@@ -313,9 +313,9 @@ static void pc_init1(MachineState *machine, const char
>*pci_type)
> /* init basic PC hardware */
> pc_basic_device_init(pcms, isa_bus, x86ms->gsi, x86ms->rtc,
> !MACHINE_CLASS(pcmc)->no_floppy, 0x4);
>-
>- pc_nic_init(pcmc, isa_bus, pcms->pcibus);
>-
>+ if (pcmc->pci_enabled) {
>+ pc_nic_init(pcmc, isa_bus, pcms->pcibus);
>+ }
Since pc_nic_init() is passed both an ISA and a PCI bus I think the NULL
dereference should be fixed there. Moreover, if pc_nic_init() is only invoked
when there is a PCI bus, the "isapc" machine won't have a nic at all.
How is this patch related to
https://patchew.org/QEMU/20241105171813.3031969-1-peter.mayd...@linaro.org ?
That is, do we need both patches to really fix the issue?
Best regards,
Bernhard
> #ifdef CONFIG_IDE_ISA
> if (!pcmc->pci_enabled) {
> DriveInfo *hd[MAX_IDE_BUS * MAX_IDE_DEVS];
