# Performance is >75% better than 64-bit code generated by Sun C and
# over 2x than 32-bit code. X[16] resides on stack, but access to it
# is scheduled for L2 latency and staged through 32 least significant
-# bits of %l0-%l7. The latter is done to achieve 32-/64-bit bit ABI
+# bits of %l0-%l7. The latter is done to achieve 32-/64-bit ABI
# duality. Nevetheless it's ~40% faster than SHA256, which is pretty
# good [optimal coefficient is 50%].
#
# SHA512 on UltraSPARC T1.
#
-# ...
+# It's not any faster than 64-bit code generated by Sun C 5.8. This is
+# because 64-bit code generator has the advantage of using 64-bit
+# loads(*) to access X[16], which I consciously traded for 32-/64-bit
+# ABI duality [as per above]. But it surpasses 32-bit Sun C generated
+# code by 60%, not to mention that it doesn't suffer from severe decay
+# when running 4 times physical cores threads and that it leaves gcc
+# [3.4] behind by over 4x factor! If compared to SHA256, single thread
+# performance is only 10% better, but overall throughput for maximum
+# amount of threads for given CPU exceeds corresponding one of SHA256
+# by 30% [again, optimal coefficient is 50%].
+#
+# (*) Unlike pre-T1 UltraSPARC loads on T1 are executed strictly
+# in-order, i.e. load instruction has to complete prior next
+# instruction in given thread is executed, even if the latter is
+# not dependent on load result! This means that on T1 two 32-bit
+# loads are always slower than one 64-bit load. Once again this
+# is unlike pre-T1 UltraSPARC, where, if scheduled appropriately,
+# 2x32-bit loads can be as fast as 1x64-bit ones.
$bits=32;
for (@ARGV) { $bits=64 if (/\-m64/ || /\-xarch\=v9/); }
srlx @X[(($i+9)/2)%8],32,$tmp1 ! X[i+9]
xor $tmp0,$tmp2,$tmp2 ! sigma1(X[i+14])
srl @X[($i/2)%8],0,$tmp0
+ add $tmp2,$tmp1,$tmp1
add $xi,$T1,$T1 ! +=X[i]
xor $tmp0,@X[($i/2)%8],@X[($i/2)%8]
- add $tmp2,$T1,$T1
add $tmp1,$T1,$T1
srl $T1,0,$T1
$code.=<<___;
srlx @X[($i/2)%8],32,$tmp1 ! X[i]
xor $tmp0,$tmp2,$tmp2 ! sigma1(X[i+14])
- srl @X[($i/2)%8],0,@X[($i/2)%8]
add $xi,$T1,$T1 ! +=X[i+9]
- add $tmp2,$T1,$T1
+ add $tmp2,$tmp1,$tmp1
+ srl @X[($i/2)%8],0,@X[($i/2)%8]
add $tmp1,$T1,$T1
sllx $T1,32,$tmp0
___
$code.=<<___;
.Lpic: call .+8
- sub %o7,.Lpic-K${label},$Ktbl
+ add %o7,K${label}-.Lpic,$Ktbl
$LD [$ctx+`0*$SZ`],$A
$LD [$ctx+`1*$SZ`],$B
.type sha${label}_block_data_order,#function
.size sha${label}_block_data_order,(.-sha${label}_block_data_order)
.asciz "SHA${label} block transform for SPARCv9, CRYPTOGAMS by <appro\@openssl.org>"
+.align 4
___
$code =~ s/\`([^\`]*)\`/eval $1/gem;