Messages in this thread | | | Date | Wed, 13 Aug 2008 22:38:31 +0200 | From | Pavel Machek <> | Subject | Re: [PATCH 01/01][retry 2] x86: L3 cache index disable for 2.6.26 |
| |
Hi! > diff -r e683983d4dd0 Documentation/ABI/testing/sysfs-devices-cache_disable > --- /dev/null Thu Jan 01 00:00:00 1970 +0000 > +++ b/Documentation/ABI/testing/sysfs-devices-cache_disable Wed Aug 13 09:06:52 2008 -0500 > @@ -0,0 +1,18 @@ > +What: /sys/devices/system/cpu/cpu*/cache/index*/cache_disable_X > +Date: Augsust 2008 > +KernelVersion: 2.6.27 > +Contact: osrc-kernel@elbe.amd.com > +Description: These files exist in every cpu's cache index directories. > + There are currently 2 cache_disable_# files in each > + directory. Reading from these files on a supported > + processor will return that cache disable index value > + for that processor and node. Writing to one of these > + files will cause the specificed cache index to be disable.
disabled.
> +#if defined(CONFIG_PCI) && defined(CONFIG_K8_NB) > +#include <linux/pci.h> > +#include <asm/k8.h> > +static struct pci_dev *get_k8_northbridge(int node) > +{ > + return k8_northbridges[node]; > +} > +#else > +static inline int pci_write_config_dword(struct pci_dev *dev, int where, > + u32 val) > +{ > + return 0; > +}
Spaces vs. tabs problem visible here. ...
> +static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf, > + unsigned int index) > +{ > + int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map)); > + struct pci_dev *dev = get_k8_northbridge(node); > + unsigned int reg = 0; > + > + if (!this_leaf->can_disable) > + return 0; > + > + pci_read_config_dword(dev, 0x1BC + index * 4, ®); > + return sprintf(buf, "%x\n", reg); > +}
...and getting serious here.
> +#define SHOW_CACHE_DISABLE(index) \ > +static ssize_t \ > +show_cache_disable_##index(struct _cpuid4_info *this_leaf, char *buf) \ > +{ \ > + return show_cache_disable(this_leaf, buf, index); \ > +} > + > +static ssize_t > +store_cache_disable(struct _cpuid4_info *this_leaf, const char *buf, > + size_t count, unsigned int index) > +{ > + int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map)); > + struct pci_dev *dev = get_k8_northbridge(node); > + ssize_t ret = 0; > + unsigned int val; > + > + if (!this_leaf->can_disable) > + return 0;
So if someone asks you to disable part of cache you can't disable, you return success?
> + if (strlen(buf) > 10) > + return -EINVAL; > + > + ret = sscanf(buf, "%x\n", &val); > + if (ret != 1) > + return -EINVAL; .... > + return strlen(buf);
Is this safe to do? What if buf is not null-terminated?
return count?
Actually, the sscanf is problematic, too, right?
Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
| |