diff options
Diffstat (limited to 'comm/third_party/libgcrypt/mpi/hppa/README')
| -rw-r--r-- | comm/third_party/libgcrypt/mpi/hppa/README | 84 |
1 files changed, 84 insertions, 0 deletions
diff --git a/comm/third_party/libgcrypt/mpi/hppa/README b/comm/third_party/libgcrypt/mpi/hppa/README new file mode 100644 index 0000000000..5a2d5fd970 --- /dev/null +++ b/comm/third_party/libgcrypt/mpi/hppa/README @@ -0,0 +1,84 @@ +This directory contains mpn functions for various HP PA-RISC chips. Code +that runs faster on the PA7100 and later implementations, is in the pa7100 +directory. + +RELEVANT OPTIMIZATION ISSUES + + Load and Store timing + +On the PA7000 no memory instructions can issue the two cycles after a store. +For the PA7100, this is reduced to one cycle. + +The PA7100 has a lookup-free cache, so it helps to schedule loads and the +dependent instruction really far from each other. + +STATUS + +1. mpn_mul_1 could be improved to 6.5 cycles/limb on the PA7100, using the + instructions bwlow (but some sw pipelining is needed to avoid the + xmpyu-fstds delay): + + fldds s1_ptr + + xmpyu + fstds N(%r30) + xmpyu + fstds N(%r30) + + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + + addc + stws res_ptr + addc + stws res_ptr + + addib Loop + +2. mpn_addmul_1 could be improved from the current 10 to 7.5 cycles/limb + (asymptotically) on the PA7100, using the instructions below. With proper + sw pipelining and the unrolling level below, the speed becomes 8 + cycles/limb. + + fldds s1_ptr + fldds s1_ptr + + xmpyu + fstds N(%r30) + xmpyu + fstds N(%r30) + xmpyu + fstds N(%r30) + xmpyu + fstds N(%r30) + + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + ldws N(%r30) + addc + addc + addc + addc + addc %r0,%r0,cy-limb + + ldws res_ptr + ldws res_ptr + ldws res_ptr + ldws res_ptr + add + stws res_ptr + addc + stws res_ptr + addc + stws res_ptr + addc + stws res_ptr + + addib |