Messages in this thread | | | Subject | RE: [PATCH 01/01][retry 3] x86: L3 cache index disable for 2.6.26 | Date | Thu, 14 Aug 2008 09:02:34 -0500 | From | "Langsdorf, Mark" <> |
| |
> > + > > + ret = sscanf(buf, "%x", &val); > > + if (ret != 1) > > + return -EINVAL; > > Is it okay to strlen() on user-supplied data? Do they have to be > null-terminated? What about sscanf?
The strlen can go - it's no longer necessary.
If sscanf isn't safe in this context, there's a lot of code in drivers/cpufreq/cpufreq.c that I cribbed from that needs to be changed, too.
> > + val |= 0xc0000000; > > + pci_write_config_dword(dev, 0x1BC + index * 4, val > & ~0x40000000); > > + wbinvd(); > > + pci_write_config_dword(dev, 0x1BC + index * 4, val); > > Should it do if capable() test before doing such stuff?
I don't think so. If it got this far, it is an AMD processor from family 0x10 or later, so it has wbinvd().
-Mark Langsdorf Operating System Research Center AMD
| |