Date: Tue, 9 Feb 2016 17:45:38 +0530 From: Sreekanth Reddy <sreekanth.reddy@broadcom.com> To: John Baldwin <jhb@freebsd.org> Cc: freebsd-current@freebsd.org, ken@freebsd.org, scsi@freebsd.org, current@freebsd.org Subject: Re: Panic on reloading a driver with same DEVICE_PROBE() return value Message-ID: <CAK=zhgpSH73-AJKdiipHo9UAkTaGHs8=LPnKmYRgNs7-xAYFJQ@mail.gmail.com> In-Reply-To: <CAK=zhgpjD2aF-XNiSG6AHojJm1gxvARXTc2enrnoQzLHe=WksA@mail.gmail.com> References: <CAK=zhgoGwXSsD-mNF=jGov1FJFnpM7m_fZ0Jwsq4JR5yazB%2Bww@mail.gmail.com> <33145844.5EAIpev2G5@ralph.baldwin.cx> <CAK=zhgpjD2aF-XNiSG6AHojJm1gxvARXTc2enrnoQzLHe=WksA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi,
While debugging more, I got one more clue,
-----------------------------------------------------------------------------------------------
static driver_t mps_pci_driver = {
"mpr",
mps_methods,
sizeof(struct mps_softc)
};
static devclass_t mps_devclass;
DRIVER_MODULE(mpr, pci, mps_pci_driver, mps_devclass, 0, 0);
-------------------------------------------------------------------------------------------------
in the above code snip-set, if I changed "DRIVER_MODULE" line as
DRIVER_MODULE(mpr3, pci, mps_pci_driver, mps_devclass, 0, 0);
(i.e. from "mpr" to "mpr3") then I am not observing any panic and I
can load & unload the mpr driver multiple times.
~Sreekanth
On Tue, Feb 9, 2016 at 2:55 PM, Sreekanth Reddy
<sreekanth.reddy@broadcom.com> wrote:
> On Mon, Feb 8, 2016 at 11:58 PM, John Baldwin <jhb@freebsd.org> wrote:
>> On Monday, February 08, 2016 08:13:26 PM Sreekanth Reddy wrote:
>>> sc->facts = malloc(sizeof(MPI2_IOC_FACTS_REPLY), M_MPR, M_ZERO|M_NOWAIT);
>>
>> M_NOWAIT can fail with NULL. Normally in an attach routine you should use a
>> blocking malloc (M_WAITOK) which will not fail with NULL.
>
> Yes I agree that in an attach routine we should to use a blocking
> malloc (M_WAITOK) flag.
>
> But, here I am observing same page fault even with (M_WAITOK) flag.
>
> Here, page fault is occurring inside the malloc function and driver is
> not receiving any NULL or any other value.
>
>>
>>> * How can I reload the modified driver with return same device_probe()
>>> value as the one which has loaded during the OS boot time. I observed
>>> that whenever I try to load this modified driver using "kldload
>>> ./mpr.ko" command then I see that driver in the /boot/kernel/ path has
>>> executed.
>>
>> At runtime you can use the devctl(8) command on HEAD (it will be included
>> in 10.3) to force a device to detach/attach to force a re-probe.
>>
>> e.g.
>>
>> devctl detach mpr0
>>
>> # note that the device won't be named mpr0, you will need to use its PCI
>> # location for this next:
>>
>> devctl attach pci0:4:0:0
>
> I have started downloading the current kernel. I will verify this
> today/tomorrow.
>
>
>>
>> --
>> John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAK=zhgpSH73-AJKdiipHo9UAkTaGHs8=LPnKmYRgNs7-xAYFJQ>