Re: [yocto] [PATCH][linux-yocto-3.0] drivers/misc/pch_phub.c: don't oops if dmi_get_system_info returns NULL

[Khem Raj]

On Thursday, July 28, 2011 04:51:41 PM Darren Hart wrote:
> Bruce,
> 
> Please apply to yocto/base. Fixes a boot issue for a
> tunnel creek development board.
> 
> --
> 
> commit 2b934c6236983392d01bef22e43af3051cac16f5
> 
> If dmi_get_system_info() returns NULL, pch_phub_probe() will dereferencea
> a zero pointer.
> 
> This oops was observed on an Atom based board which has no BIOS, but a
> bootloder which doesn't privde DMI data.
> 
> Signed-off-by: Alexander Stein 
> Cc: Tomoya MORINAGA 
> Cc: Greg KH 
> Signed-off-by: Andrew Morton 
> Signed-off-by: Linus Torvalds 
> ---
>  drivers/misc/pch_phub.c |5 -
>  1 files changed, 4 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/misc/pch_phub.c b/drivers/misc/pch_phub.c
> index 5fe79df..01eb67b 100644
> --- a/drivers/misc/pch_phub.c
> +++ b/drivers/misc/pch_phub.c
> @@ -686,6 +686,8 @@ static int __devinit pch_phub_probe(struct pci_dev
> *pdev, }
> 
>   if (id->driver_data == 1) { /* EG20T PCH */
> + const char *board_name;
> +
>   retval = sysfs_create_file(&pdev->dev.kobj,
>  &dev_attr_pch_mac.attr);
>   if (retval)
> @@ -701,7 +703,8 @@ static int __devinit pch_phub_probe(struct pci_dev
> *pdev, CLKCFG_CANCLK_MASK);
> 
>   /* quirk for CM-iTC board */
> - if (strstr(dmi_get_system_info(DMI_BOARD_NAME), "CM-iTC"))
> + board_name = dmi_get_system_info(DMI_BOARD_NAME);
> + if (board_name && strstr(board_name, "CM-iTC"))

May be it could be just if ( dmi_get_system_info(DMI_BOARD_NAME) && 
strstr(board_name, "CM-iTC"))

although I think compiler will already do it internally

>   pch_phub_read_modify_write_reg(chip,
>   (unsigned int)CLKCFG_REG_OFFSET,
>   CLKCFG_UART_48MHZ | 
> CLKCFG_BAUDDIV |
-- 
Khem Raj
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto

Re: [yocto] [PATCH][linux-yocto-3.0] drivers/misc/pch_phub.c: don't oops if dmi_get_system_info returns NULL

[Bruce Ashfield]

On 11-07-30 2:34 PM, Khem Raj wrote:

On Thursday, July 28, 2011 04:51:41 PM Darren Hart wrote:

Bruce,

Please apply to yocto/base. Fixes a boot issue for a
tunnel creek development board.

--

commit 2b934c6236983392d01bef22e43af3051cac16f5

If dmi_get_system_info() returns NULL, pch_phub_probe() will dereferencea
a zero pointer.

This oops was observed on an Atom based board which has no BIOS, but a
bootloder which doesn't privde DMI data.

Signed-off-by: Alexander Stein
Cc: Tomoya MORINAGA
Cc: Greg KH
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
---
  drivers/misc/pch_phub.c |5 -
  1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/drivers/misc/pch_phub.c b/drivers/misc/pch_phub.c
index 5fe79df..01eb67b 100644
--- a/drivers/misc/pch_phub.c
+++ b/drivers/misc/pch_phub.c
@@ -686,6 +686,8 @@ static int __devinit pch_phub_probe(struct pci_dev
*pdev, }

if (id->driver_data == 1) { /* EG20T PCH */
+   const char *board_name;
+
retval = sysfs_create_file(&pdev->dev.kobj,
&dev_attr_pch_mac.attr);
if (retval)
@@ -701,7 +703,8 @@ static int __devinit pch_phub_probe(struct pci_dev
*pdev, CLKCFG_CANCLK_MASK);

/* quirk for CM-iTC board */
-   if (strstr(dmi_get_system_info(DMI_BOARD_NAME), "CM-iTC"))
+   board_name = dmi_get_system_info(DMI_BOARD_NAME);
+   if (board_name&&  strstr(board_name, "CM-iTC"))


May be it could be just if ( dmi_get_system_info(DMI_BOARD_NAME)&&
strstr(board_name, "CM-iTC"))

although I think compiler will already do it internally


Could very well be. I grabbed this one directly from linus' tree,
so I've got it as-is for now, but I've tagged it to be considered
later if we do decide that we want to send a tweak of this upstream

Thanks!

Bruce




pch_phub_read_modify_write_reg(chip,
(unsigned int)CLKCFG_REG_OFFSET,
CLKCFG_UART_48MHZ | 
CLKCFG_BAUDDIV |


___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto