diff options
author | NIIBE Yutaka <gniibe@fsij.org> | 2017-04-28 09:27:00 +0900 |
---|---|---|
committer | NIIBE Yutaka <gniibe@fsij.org> | 2017-04-28 09:27:00 +0900 |
commit | 9b651fb632f3697e70685c9ee340ab0cb2274bdf (patch) | |
tree | f8d32662197433f86b33b9ed9a2063e7517e79f0 /mpi/alpha | |
parent | 719468e53133d3bdf12156c5bfdea2bf15f9f6f1 (diff) | |
download | libgcrypt-9b651fb632f3697e70685c9ee340ab0cb2274bdf.tar.gz |
Spelling fixes in docs and comments.
--
GnuPG-bug-id: 3120
Reported-by: ka7 (klemens)
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
Diffstat (limited to 'mpi/alpha')
-rw-r--r-- | mpi/alpha/README | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/mpi/alpha/README b/mpi/alpha/README index 55c0a291..00addfd3 100644 --- a/mpi/alpha/README +++ b/mpi/alpha/README @@ -5,7 +5,7 @@ RELEVANT OPTIMIZATION ISSUES EV4 1. This chip has very limited store bandwidth. The on-chip L1 cache is -write-through, and a cache line is transfered from the store buffer to the +write-through, and a cache line is transferred from the store buffer to the off-chip L2 in as much 15 cycles on most systems. This delay hurts mpn_add_n, mpn_sub_n, mpn_lshift, and mpn_rshift. @@ -20,7 +20,7 @@ EV5 1. The memory bandwidth of this chip seems excellent, both for loads and stores. Even when the working set is larger than the on-chip L1 and L2 -caches, the perfromance remain almost unaffected. +caches, the performance remain almost unaffected. 2. mulq has a measured latency of 13 cycles and an issue rate of 1 each 8th cycle. umulh has a measured latency of 15 cycles and an issue rate of 1 |