freakazoid12345 has quit [Ping timeout: 240 seconds]
<SpaceCoaster>
Make a token investment,
<SpaceCoaster>
Then muddy the waters with proprietary changes and tying up the process. That amount of money might be sufficient to make others pause before they enter the market. Sorry to be such a cynic.
helium-3 is now known as dilfridge
[exa]_ has quit [Ping timeout: 240 seconds]
[exa] has joined #riscv
`join_su1line is now known as `join_subline
<geertu>
SpaceCoaster: Our competitor wanted to buy our CPU core provider, so let's switch to RISC-V, and switch to another competitor to bake our chips?
<geertu>
+ the foundry silo effect
<SpaceCoaster>
They encourage RISC-V on their terms / foundry / IP
<sorear>
this is intel's regularly scheduled attempt to enter a market other than windows PCs and servers
<sorear>
or possibly setup to spin off their manufacturing division like AMD did a few years ago; not following the financials close enough to be sure
<muurkha>
when did AMD spin off their manufacturing division? 01996?
<sorear>
wiki says globalfoundries started in 2009, haven't looked deeper
<muurkha>
interesting, apparently 02008 and 02009
freakazoid12345 has joined #riscv
freakazoid343 has quit [Ping timeout: 260 seconds]
<SpaceCoaster>
jimwilson: Do you think it will help SiFive ship chips?
<muurkha>
I wonder if it means SiFive will have to compete with Intel
<muurkha>
Intel looted the fastest ARM implementation in the world from the wreckage of DEC after torpedoing DEC with patent lawsuits
cwebber has quit [Remote host closed the connection]
mahmutov has joined #riscv
freakazoid12345 has joined #riscv
freakazoid343 has quit [Ping timeout: 250 seconds]
<gordonDrogon>
Hm. I have a 'muldiv' function to write in RV32IM - that's a function that takes 2 x 32-bit numbers, multiplies to give a 64-bit result then divides the result with a 32-bit number - anyone know if there are any libraries already out there that might do this, or something similar?
* gordonDrogon
finds MULH (doh!) but still need a 32/64 divide....
Andre_H has joined #riscv
<[exa]>
gordonDrogon: the output is 2 registers of the 64b result?
<gordonDrogon>
[exa], yes - if possible, avoiding using any RAM if possible...
<gordonDrogon>
it's a function used in my BCPL system to help with scaling for fied point arithmetic...
<gordonDrogon>
*fixed
pabs3 has quit [Read error: Connection reset by peer]
<[exa]>
how much instructions do you expect to use on it? there's one obvious 32-ish-instruction by just doing the division first on the upper part, then bitshifting the remainder by bits and adding that to the result
<gordonDrogon>
yes, a long-division style would work - it's not commonly used.
<gordonDrogon>
in my exsiting 65816 system I do it the long way - subtract and shift..
<gordonDrogon>
looks like gcc calls __divdi3 so I'll find the source for that...
riff-IRC has quit [Quit: PROTO-IRC v0.73a (C) 1988 NetSoft - Built on 11-13-1988 on AT&T System V]
jmdaemon has quit [Ping timeout: 240 seconds]
freakazoid12345 has quit [Ping timeout: 250 seconds]
riff-IRC has joined #riscv
ivii has joined #riscv
riff-IRC has quit [Quit: PROTO-IRC v0.73a (C) 1988 NetSoft - Built on 11-13-1988 on AT&T System V]
mahmutov has quit [Ping timeout: 240 seconds]
ivii has quit [Remote host closed the connection]
jmdaemon has joined #riscv
prabhakarlad has joined #riscv
jjido has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
epony has quit [Ping timeout: 240 seconds]
winterflaw has quit [Ping timeout: 276 seconds]
jjido has joined #riscv
epony has joined #riscv
Andre_H has quit [Quit: Leaving.]
jmdaemon has quit [Ping timeout: 250 seconds]
<The_Decryptor>
23:19 <sorear> or possibly setup to spin off their manufacturing division like AMD did a few years ago; < Yep, "Intel Foundry Services", they launched it last year
<JanC>
Intel are still using most of their own capacity today, but they have some leftover capacity that doesn't match what they need
<JanC>
it makes sense to not let that sit idle...
<JanC>
AMD split off their foundries because they needed the money from selling them