Hi,
On Mon, Jul 14, 2008 at 08:25:20AM -0700, Randy Dunlap wrote:
> On Mon, 14 Jul 2008 10:14:23 +0200 Dominik Brodowski wrote:
>
> > Don't be more zealous with memory than the firmware class core.
> >
> > Signed-off-by: Dominik Brodowski <[EMAIL PROTECTED]>
> > ---
> > drivers/pcmcia/ds.c | 4 ++--
> > 1 files changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c
> > index 461b8a9..4174d96 100644
> > --- a/drivers/pcmcia/ds.c
> > +++ b/drivers/pcmcia/ds.c
> > @@ -850,7 +850,7 @@ static int pcmcia_load_firmware(struct pcmcia_device
> > *dev, char * filename)
> > {
> > struct pcmcia_socket *s = dev->socket;
> > const struct firmware *fw;
> > - char path[20];
> > + char path[FIRMWARE_NAME_MAX];
> > int ret = -ENOMEM;
> > int no_funcs;
> > int old_funcs;
> > @@ -862,7 +862,7 @@ static int pcmcia_load_firmware(struct pcmcia_device
> > *dev, char * filename)
> >
> > ds_dbg(1, "trying to load CIS file %s\n", filename);
> >
> > - if (strlen(filename) > 14) {
> > + if (strlen(filename) > (FIRMWARE_NAME_MAX - 1)) {
> > printk(KERN_WARNING "pcmcia: CIS filename is too long [%s]\n",
> > filename);
> > return -EINVAL;
> > --
>
>
> http://bugzilla.kernel.org/show_bug.cgi?id=10975
>
> The submitter commented:
> This seems to actually make things worse...
> See http://puppylinux.com/blog/?viewDetailed=00177
> and http://puppylinux.com/blog/?viewDetailed=00178
Somehow I think that this doesn't relate to this patch, _but_: the card
in question, a "Dynacom Data/Fax PCMCIA" with manufactor id 0x13 and card
id 0x00, might simply not need this "GLOBETROTTER" CIS file. That's required
for card "Option Wireless Technology GSM/GPRS GlobeTrotter", which is a
really different card. Therefore...
@Michal,@Barry: could you send me a "lspcmcia -vvv" output, please?
Best,
Dominik
_______________________________________________
Linux PCMCIA reimplementation list
http://lists.infradead.org/mailman/listinfo/linux-pcmcia
