<bzimage>
hello osdev people , i am trying to rationalize something here , i have 3 intels and they all boot my usb iso as expected , but when i stick that thing to and amd 5 pro , it does pause at a point during boot , giving me a little warning about gpt header not at the end of the disk then waits until i press something
<bzimage>
then when i press to something i find myself in the sh of initrafms
<bzimage>
then when i sayd exec init it keeps booting normal...
<bzimage>
and i can normally install my linux distro to the internal hdd and everything normal then ,
<bzimage>
but this during boot of usb , why it does pause and that i need to say exec init , i have 0 idea
<bzimage>
this seems like something AMD didnt like with my boot config and scripts
simjnd has joined #osdev
simjnd has quit [Ping timeout: 252 seconds]
simjnd has joined #osdev
GeDaMo has joined #osdev
simjnd has quit [Ping timeout: 276 seconds]
Left_Turn has joined #osdev
simjnd has joined #osdev
simjnd has quit [Remote host closed the connection]
<geist>
so yeah AFAIK it doesnt. args are passed via the stack, though i guess for a more powerful machine addressing mode cisc machine like that it's not so bad
the_oz has quit [Remote host closed the connection]
<bslsk05>
community.nxp.com: Coldfire register ABI documentation - NXP Community
<GeDaMo>
"The register calling convention puts the first 3 integer arguments in D0, D1, and D2, and the first 2 pointer arguments in A0 and A1. I've never done floating point in coldfire, but I would guess the first 3 double arguments would go to FP0, FP1, and FP2. "
levitating has quit [Quit: Computer exploded]
<nikolar>
geist: well the supported addressing modes don't matter if you have to touch memory on every function call heh
<bslsk05>
wiki.freepascal.org: m68k - Lazarus wiki
<geist>
well, sure but i mean you can directly address stuff on the stack in something other than a load/store
<geist>
like 'add what is on the stack to this thing' sort of instructions, so it's not necessarily *more* instructions than what you would get with a pure load/store architecture and using a stack calling convention
<nikolar>
sure, it's not more instructions
<nikolar>
but it's more memory accesses
<nikolar>
which don't even need to touch memory if you were allowed to pass args in registers
<nikolar>
which is my point
<zid`>
doesn't sound like there's an official one, but gcc will inline / jmp to code fragments using pre-set registers etc sometimes
<zid`>
throw LTO at it and you should get most of the benefit
Turn_Left has quit [Ping timeout: 248 seconds]
<nikolar>
yeah i know about lto
<nikolar>
i was wondering in general
<zid`>
I know you know, I'm saying the loss is minor
<nikolar>
sure
Turn_Left has joined #osdev
Teukka has quit [Read error: Connection reset by peer]
netbsduser has joined #osdev
Teukka has joined #osdev
lanodan has quit [Quit: WeeChat 4.4.3]
[Kalisto] has quit [Quit: Ping timeout (120 seconds)]
[Kalisto] has joined #osdev
lanodan has joined #osdev
karenw has quit [Ping timeout: 248 seconds]
TkTech has quit [Remote host closed the connection]
TkTech has joined #osdev
simjnd has quit [Ping timeout: 276 seconds]
Celelibi has quit [Read error: Connection reset by peer]
Gooberpatrol66 has quit [Quit: Konversation terminated!]
Celelibi has joined #osdev
<geist>
but yeah does seem like you could pass the two first integer args in d0-d1, and two first pointer args in a0-a1 and still keep the same general ABI, since those are already trashed
Gooberpatrol66 has joined #osdev
Celelibi has quit [Read error: Connection reset by peer]
Celelibi has joined #osdev
ZipCPU has quit [Ping timeout: 244 seconds]
ZipCPU has joined #osdev
the_oz has quit [Ping timeout: 252 seconds]
netbsduser has quit [Ping timeout: 244 seconds]
netbsduser has joined #osdev
the_oz has joined #osdev
GeDaMo has quit [Quit: 0wt 0f v0w3ls.]
Turn_Left has quit [Remote host closed the connection]
Turn_Left has joined #osdev
eddof13 has quit [Quit: eddof13]
eddof13 has joined #osdev
simjnd has joined #osdev
simjnd has quit [Ping timeout: 248 seconds]
Celelibi has quit [Read error: Connection reset by peer]
<heat>
throwing LTO at originally shitty code/codegen is very rustpilled and C++pilled of you
<nikolar>
not even the same class of issues lol
<the_oz>
foundational
izabera is now known as isabella
<the_oz>
one lives but singujarly
* kof673
awards oz openstep points...you can trade them in for gnustep
<the_oz>
yey points
Celelibi has joined #osdev
Turn_Left has quit [Read error: Connection reset by peer]
simjnd has joined #osdev
simjnd has quit [Ping timeout: 252 seconds]
mxs9 has joined #osdev
<netbsduser>
lto is problematic
<netbsduser>
dangerous to dtrace, destructive to debugging, odious to observability
karenw has joined #osdev
eddof13 has quit [Quit: eddof13]
eddof13 has joined #osdev
eddof13 has quit [Quit: eddof13]
goliath has quit [Quit: SIGSEGV]
netbsduser has quit [Ping timeout: 260 seconds]
Lucretia has quit [Remote host closed the connection]