Re: [PATCH v2,RESEND] MIPS: Scan the DMI system information

[Tiezhu Yang]

On 2/3/20 10:46 PM, Jonas Gorski wrote:
> Hi,
>
> On Thu, 16 Jan 2020 at 14:28, Tiezhu Yang <yangtiezhu@xxxxxxxxxxx> wrote:
> > Enable DMI scanning on the MIPS architecture, this setups DMI identifiers
> > (dmi_system_id) for printing it out on task dumps and prepares DIMM entry
> > information (dmi_memdev_info) from the SMBIOS table. With this patch, the
> > driver can easily match various of mainboards.
> >
> > In the SMBIOS reference specification, the table anchor string "_SM_" is
> > present in the address range 0xF0000 to 0xFFFFF on a 16-byte boundary,
> > but there exists a special case for Loongson platform, when call function
> > dmi_early_remap, it should specify the start address to 0xFFFE000 due to
> > it is reserved for SMBIOS and can be normally access in the BIOS.
> >
> > This patch works fine on the Loongson 3A3000 platform which belongs to
> > MIPS architecture and has no influence on the other architectures such
> > as x86 and ARM.
> >
> > Co-developed-by: Yinglu Yang <yangyinglu@xxxxxxxxxxx>
> > Signed-off-by: Yinglu Yang <yangyinglu@xxxxxxxxxxx>
> > [jiaxun.yang@xxxxxxxxxxx: Refine definitions and Kconfig]
> > Signed-off-by: Jiaxun Yang <jiaxun.yang@xxxxxxxxxxx>
> > Signed-off-by: Tiezhu Yang <yangtiezhu@xxxxxxxxxxx>
> > Reviewed-by: Huacai Chen <chenhc@xxxxxxxxxx>
> > ---
> >
> > v2:
> >    - add SMBIOS_ENTRY_POINT_SCAN_START suggested by Jean
> >    - refine definitions and Kconfig by Jiaxun
> >
> >   arch/mips/Kconfig           | 10 ++++++++++
> >   arch/mips/include/asm/dmi.h | 20 ++++++++++++++++++++
> >   arch/mips/kernel/setup.c    |  2 ++
> >   drivers/firmware/dmi_scan.c |  6 +++++-
> >   4 files changed, 37 insertions(+), 1 deletion(-)
> >   create mode 100644 arch/mips/include/asm/dmi.h
> >
> > diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
> > index 4b83507..c097f78 100644
> > --- a/arch/mips/Kconfig
> > +++ b/arch/mips/Kconfig
> > @@ -2773,6 +2773,16 @@ config HW_PERF_EVENTS
> >            Enable hardware performance counter support for perf events. If
> >            disabled, perf events will use software events only.
> >
> > +config DMI
> > +       default y if MACH_LOONGSON64
> > +       select DMI_SCAN_MACHINE_NON_EFI_FALLBACK
> > +       bool "Enable DMI scanning"
> Is this option harmless to enable for other MIPS platforms? What
> happens if it is enabled on devices where PHYS_OFFSET isn't 0, so
> 0xF0000 is likely not backed by anything and accessing it might hang
> the system?
>
> This probably should depend on MACH_LOONGSON64.


Hi Jonas,

Thanks for your reply and suggestion.

With the current patch, on the MIPS architecture, the DMI option is
default y only if MACH_LOONGSON64 is set, otherwise it is not set.
When enable DMI manually on the MIPS platform which is not LOONGSON64,
the SMBIOS_ENTRY_POINT_SCAN_START is 0xF0000, dmi_early_remap() will
return NULL and goto err to print "DMI not present or invalid."

static void __init dmi_scan_machine(void)
{
...
ÂÂÂ ÂÂÂ p = dmi_early_remap(SMBIOS_ENTRY_POINT_SCAN_START, 0x10000);
ÂÂÂ ÂÂÂ if (p == NULL)
ÂÂÂ ÂÂÂ ÂÂÂ goto error;
...
Âerror:
ÂÂÂ pr_info("DMI not present or invalid.\n");
}

It seems harmless and the system can boot successful. However, in order
to avoid the unknown risks on the mips platform which is not LOONGSON64,
I think it is better to depend on MACH_LOONGSON64.

config DMI
ÂÂÂ bool "Enable DMI scanning"
ÂÂÂ depends on MACH_LOONGSON64
ÂÂÂ select DMI_SCAN_MACHINE_NON_EFI_FALLBACK
ÂÂÂ default y

If other mips platform also needs this DMI feature in the future,
the "depends on" conditon can be modified.

I will do it in the v3 patch.


Thanks,


Tiezhu Yang


> Regards
>
> Jonas