From: lkcl Date: Tue, 29 Oct 2019 22:45:03 +0000 (+0000) Subject: (no commit message) X-Git-Tag: convert-csv-opcode-to-binary~3839 X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=e8e7ee9aeb457b1eae757e6ae8cff08ffc1e1aa6;p=libreriscv.git --- diff --git a/openpower.mdwn b/openpower.mdwn index dc1db37a6..ccf8b1d28 100644 --- a/openpower.mdwn +++ b/openpower.mdwn @@ -4,7 +4,7 @@ * FP16 needed * FCVT between 16/32/64 needed * c++11 atomics not very efficient -* no 16/48/64 opcodes, needs a shuffle of opcodes +* no 16/48/64 opcodes, needs a shuffle of opcodes. TODO investigate Power VLE * needs escape sequencing (ISAMUX/NS) # atomics @@ -13,9 +13,11 @@ Single instruction on RV, and x86, but multiple on Power. Needs investigation, p https://www.cl.cam.ac.uk/~pes20/cpp/cpp0xmappings.html +Hot loops contain significant instruction count, really need new c++11 atomics. To be proposed as new extension because other OpenPower members will need them too + # FP16 -Doesn't exist in Power, need to work out suitable opcodes, basically means duplicating the entire range of FP32/64 ops, symmetrically. +Doesn't exist in Power (does - as VLE?), need to work out suitable opcodes, basically means duplicating the entire range of FP32/64 ops, symmetrically. Usually done with a fmt field, 2 bit, last one is FP128 @@ -23,8 +25,12 @@ Usually done with a fmt field, 2 bit, last one is FP128 Absolutely critical, also to have official endorsement from OpenPower Foundation. +This will allow extending ISA (see ISAMUX/NS) in a clean fashion + # Compressed, 48, 64, VBLOCK +TODO investigate Power VLE + Under Esc Seq, move mulli, twi, tdi out of major OP000 then use the entire row, 2 bits instead of 3. * OP 000-000 and 000-001 for 16 bit compressed, 11 bit instructions