diff options
| author | David S. Miller <davem@sunset.davemloft.net> | 2006-01-31 23:13:29 -0800 |
|---|---|---|
| committer | David S. Miller <davem@sunset.davemloft.net> | 2006-03-20 01:11:31 -0800 |
| commit | b0fd4e49aea8a460afab7bc67cd618e2d19291d4 (patch) | |
| tree | a596c793cbc918bdcea462bcfe2f2f41fe8afeb2 | |
| parent | 30a6ecad9670d97c9d0fbfa7d80970aeb339bdec (diff) | |
[SPARC64]: Kill out-of-date commentary in asm-sparc64/tsb.h
Signed-off-by: David S. Miller <davem@davemloft.net>
| -rw-r--r-- | include/asm-sparc64/tsb.h | 8 |
1 files changed, 0 insertions, 8 deletions
diff --git a/include/asm-sparc64/tsb.h b/include/asm-sparc64/tsb.h index 1addd91d720..f384565212f 100644 --- a/include/asm-sparc64/tsb.h +++ b/include/asm-sparc64/tsb.h @@ -37,14 +37,6 @@ * choose to use bit 47 in the tag. Also, since we never map anything * at page zero in context zero, we use zero as an invalid tag entry. * When the lock bit is set, this forces a tag comparison failure. - * - * Currently, we allocate an 8K TSB per-process and we use it for both - * I-TLB and D-TLB misses. Perhaps at some point we'll add code that - * monitors the number of active pages in the process as we get - * major/minor faults, and grow the TSB in response. The only trick - * in implementing that is synchronizing the freeing of the old TSB - * wrt. parallel TSB updates occuring on other processors. On - * possible solution is to use RCU for the freeing of the TSB. */ #define TSB_TAG_LOCK_BIT 47 |