lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

arch/tile: catch up on various minor cleanups.

Browse Source 
None of these changes fix any actual bugs, but are just various cleanups
that fell out along the way.  In particular, some unused #defines and
includes are removed, PREFETCH_STRIDE is added (the default is right for
our shipping chips, but wrong for our next generation), our tile-specific
prefetching code is removed so the (identical) generic prefetching code
can be used instead, a comment is fixed to be proper GPL and not just a
"paste GPL here" token, a "//" comment is converted to "/* */", etc.

Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
This commit is contained in:
Chris Metcalf
2010-07-02 14:19:35 -04:00
parent bcd97c3f9a
commit ef06f55a5c
7 changed files with 16 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
arch/tile/include/asm/processor.h
View File

@ -267,32 +267,20 @@ extern int hash_default;
/* Should kernel stack pages be hash-for-home? */
extern int kstack_hash;
/* Does MAP_ANONYMOUS return hash-for-home pages by default? */
#define uheap_hash hash_default
#else
#define hash_default 0
#define kstack_hash 0
#define uheap_hash 0
#endif
/* Are we using huge pages in the TLB for kernel data? */
extern int kdata_huge;
/*
* Note that with OLOC the prefetch will return an unused read word to
* the issuing tile, which will cause some MDN traffic. Benchmarking
* should be done to see whether this outweighs prefetching.
*/
#define ARCH_HAS_PREFETCH
#define ARCH_HAS_PREFETCHW
#define ARCH_HAS_SPINLOCK_PREFETCH
#define prefetch(ptr) __builtin_prefetch((ptr), 0, 3)
#define prefetchw(ptr) __builtin_prefetch((ptr), 1, 3)
#ifdef CONFIG_SMP
#define spin_lock_prefetch(ptr) prefetchw(ptr)
#else
/* Nothing to prefetch. */
#define spin_lock_prefetch(lock) do { } while (0)
#endif
#define PREFETCH_STRIDE CHIP_L2_LINE_SIZE()
#else /* __ASSEMBLY__ */