On Thu, Apr 3, 2014 at 11:03 AM, Moore, Robert <robert.moore@xxxxxxxxx> wrote:
> On second thought, did you notice that the caller always deletes the list, regardless of the return status?
Hmm, no, I didn't notice that, and obviously Coverity didn't either.
So I guess there's no actual bug, as long as every caller deletes the
list regardless of status. And there's only one caller.
I wouldn't hold it up as good style, but I guess it's safe :)
> /* Build a list of PCI devices, from PciRegion up to RootPciDevice */
>
> Status = AcpiHwBuildPciList (RootPciDevice, PciRegion, &ListHead);
> if (ACPI_SUCCESS (Status))
> {
> /* Walk the list, updating the PCI device/function/bus numbers */
>
> Status = AcpiHwProcessPciList (PciId, ListHead);
> }
>
> /* Always delete the list */
>
> AcpiHwDeletePciList (ListHead);
> return_ACPI_STATUS (Status);
>
>> -----Original Message-----
>> From: Bjorn Helgaas [mailto:bhelgaas@xxxxxxxxxx]
>> Sent: Thursday, April 03, 2014 9:22 AM
>> To: Moore, Robert; Zheng, Lv; Wysocki, Rafael J
>> Cc: linux-acpi@xxxxxxxxxxxxxxx; devel@xxxxxxxxxx
>> Subject: Coverity CID 134632: acpi_hw_build_pci_list() memory leak
>>
>> Coverity complains that acpi_hw_build_pci_list() leaks memory when it
>> returns here:
>>
>> 201 if (!list_element) {
>>
>> CID 134632 (#1 of 2): Resource leak (RESOURCE_LEAK) 13. leaked_storage:
>> Variable list_head going out of scope leaks the storage it points to.
>> 202 return (AE_NO_MEMORY);
>> 203 }
>>
>> We're building up a list, and if we fail to allocate another element, we
>> return failure without deallocating the already built-up list.
>>
>> This is CID 134632.
>>
>> Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
