generic-library/openssl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

aes-s390x.pl: revisit buffer allocation and add performance data.

Browse Source
This commit is contained in:
Andy Polyakov 2010-07-14 08:43:38 +00:00
parent c981086d40
commit 26064d7f77
1 changed files with 21 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
crypto/aes/asm/aes-s390x.pl
View File

@ -44,7 +44,7 @@
# Unlike previous version hardware support detection takes place only # Unlike previous version hardware support detection takes place only
# at the moment of key schedule setup, which is denoted in key->rounds. # at the moment of key schedule setup, which is denoted in key->rounds.
# This is done, because deferred key setup can't be made MT-safe, not # This is done, because deferred key setup can't be made MT-safe, not
# for key lengthes longer than 128 bits. # for keys longer than 128 bits.
# #
# Add AES_cbc_encrypt, which gives incredible performance improvement, # Add AES_cbc_encrypt, which gives incredible performance improvement,
# it was measured to be ~6.6x. It's less than previously mentioned 8x, # it was measured to be ~6.6x. It's less than previously mentioned 8x,
@ -52,7 +52,13 @@
# May 2010. # May 2010.
# #
# Add AES_ctr32_encrypt. # Add AES_ctr32_encrypt. If hardware-assisted, it provides up to 4.3x
# performance improvement over "generic" counter mode routine relying
# on single-block, also hardware-assisted, AES_encrypt. "Up to" refers
# to the fact that exact throughput value depends on current stack
# frame alignment within 4KB page. In worst case you get ~75% of the
# maximum, but *on average* it would be as much as ~98%. Meaning that
# worst case is unlike, it's like hitting ravine on plateau.
while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {} while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {}
open STDOUT,">$output"; open STDOUT,">$output";
@ -1367,24 +1373,27 @@ $code.=<<___ if (!$softonly);
lg $iv0,0($ivp) # load ivec lg $iv0,0($ivp) # load ivec
lg $ivp,8($ivp) lg $ivp,8($ivp)
# prepare and allocate stack frame # prepare and allocate stack frame at the top of 4K page
lghi $s0,-272 # guarantee at least 256-bytes buffer # with 1K reserved for eventual signal handling
lghi $s0,-1024-256-16# guarantee at least 256-bytes buffer
lghi $s1,-4096 lghi $s1,-4096
lgr $fp,$sp
algr $s0,$sp algr $s0,$sp
lgr $fp,$sp
ngr $s0,$s1 # align at page boundary ngr $s0,$s1 # align at page boundary
la $sp,0($s0) # alloca slgr $fp,$s0 # total buffer size
stg $fp,0($s0) # back-chain lgr $s2,$sp
lghi $s1,1024+16 # sl[g]fi is extended-immediate facility
slgr $fp,$s1 # deduct reservation to get usable buffer size
# buffer size is at lest 256 and at most 3072+256-16
# calculate resultant buffer size la $sp,1024($s0) # alloca
la $s0,16($s0) # buffer starts at offset of 16 srlg $fp,$fp,4 # convert bytes to blocks, minimum 16
slgr $fp,$s0 stg $s2,0($sp) # back-chain
srlg $fp,$fp,4 # $fp is buffer length in blocks, minimum 16
stg $fp,8($sp) stg $fp,8($sp)
slgr $len,$fp slgr $len,$fp
brc 1,.Lctr32_hw_loop # not zero, no borrow brc 1,.Lctr32_hw_loop # not zero, no borrow
algr $fp,$len algr $fp,$len # input is shorter than allocated buffer
lghi $len,0 lghi $len,0
stg $fp,8($sp) stg $fp,8($sp)