sorear changed the topic of #riscv to: RISC-V instruction set architecture | https://riscv.org | Logs: https://libera.irclog.whitequark.org/riscv | Backup if libera.chat and freenode fall over: irc.oftc.net
wolfshappen has quit [Quit: later]
jedix has quit [Ping timeout: 252 seconds]
freakazoid333 has quit [Read error: Connection reset by peer]
freakazoid333 has joined #riscv
vagrantc has joined #riscv
jedix has joined #riscv
hendursaga has quit [Ping timeout: 244 seconds]
TMM_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
TMM_ has joined #riscv
hendursaga has joined #riscv
ssb has quit [Ping timeout: 272 seconds]
ssb has joined #riscv
jedix has quit [Ping timeout: 252 seconds]
davidlt has joined #riscv
vagrantc has quit [Quit: leaving]
riff-IRC has quit [Remote host closed the connection]
FL4SHK has quit [Ping timeout: 252 seconds]
FL4SHK has joined #riscv
jedix has joined #riscv
geertu has quit [*.net *.split]
leah2 has quit [*.net *.split]
josuah has quit [*.net *.split]
wgrant has quit [*.net *.split]
rvalles has quit [*.net *.split]
bgamari has quit [*.net *.split]
oaken-source has quit [*.net *.split]
scruffyfurn has quit [*.net *.split]
aquijoule__ has quit [*.net *.split]
palmer has quit [*.net *.split]
dobson has quit [*.net *.split]
alexfanqi has quit [*.net *.split]
ahs3 has quit [*.net *.split]
geertu has joined #riscv
aquijoule__ has joined #riscv
oaken-source has joined #riscv
palmer has joined #riscv
dobson has joined #riscv
leah2 has joined #riscv
rvalles has joined #riscv
josuah has joined #riscv
wgrant has joined #riscv
ahs3 has joined #riscv
scruffyfurn has joined #riscv
alexfanqi has joined #riscv
bgamari has joined #riscv
winterflaw has joined #riscv
davidlt has quit [Ping timeout: 252 seconds]
adjtm has quit [Ping timeout: 265 seconds]
elastic_dog has joined #riscv
hendursa1 has joined #riscv
hendursaga has quit [Ping timeout: 244 seconds]
mahmutov has quit [Ping timeout: 268 seconds]
elastic_dog has quit [Ping timeout: 252 seconds]
elastic_dog has joined #riscv
jamtorus has joined #riscv
jellydonut has quit [Read error: Connection reset by peer]
mahmutov has joined #riscv
adjtm has joined #riscv
choozy has joined #riscv
Andre_H has joined #riscv
adjtm has quit [Remote host closed the connection]
adjtm has joined #riscv
riff-IRC has joined #riscv
frost has quit [Quit: Connection closed]
jwillikers has joined #riscv
hendursa1 has quit [Quit: hendursa1]
hendursaga has joined #riscv
adjtm has quit [Ping timeout: 252 seconds]
geertu has quit [Ping timeout: 258 seconds]
mahmutov_ has joined #riscv
mahmutov has quit [Ping timeout: 272 seconds]
choozy has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
davidlt has joined #riscv
mahmutov_ has quit [Ping timeout: 258 seconds]
geertu has joined #riscv
mahmutov_ has joined #riscv
adjtm has joined #riscv
choozy has joined #riscv
freakazoid333 has quit [Read error: Connection reset by peer]
freakazoid333 has joined #riscv
jedix has quit [Ping timeout: 272 seconds]
freakazoid333 has quit [Read error: Connection reset by peer]
loki_val has joined #riscv
crabbedhaloablut has quit [Ping timeout: 244 seconds]
pecastro has joined #riscv
pecastro_ has joined #riscv
compscipunk has joined #riscv
jedix has joined #riscv
choozy has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
GenTooMan has quit [Ping timeout: 256 seconds]
GenTooMan has joined #riscv
freakazoid333 has joined #riscv
geist has quit [Quit: leaving]
geist has joined #riscv
pecastro has quit [Quit: Lost terminal]
pecastro_ has quit [Quit: Lost terminal]
pecastro has joined #riscv
geist has quit [Quit: leaving]
TMM_ has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
TMM_ has joined #riscv
davidlt has quit [Ping timeout: 252 seconds]
jwillikers has quit [Remote host closed the connection]
geist has joined #riscv
mahmutov has joined #riscv
mahmutov_ has quit [Ping timeout: 252 seconds]
mahmutov has quit [Ping timeout: 258 seconds]
mahmutov has joined #riscv
mahmutov has quit [Ping timeout: 252 seconds]
mahmutov has joined #riscv
adjtm has quit [Ping timeout: 272 seconds]
<meowray> jrtc27: http://lists.infradead.org/pipermail/opensbi/2021-July/001325.html hmm, i think --target=riscv64 for bare-metal stuff is fine to emphasize that the compile mode follows the "generic ELF" (whatever this is defined in the compiler). not using linux-gnu should be fine for such stuff.
<jrtc27> the main thing I want to do is make sure people don't blindly add on -linux-gnu to the riscv64
<jrtc27> and instead ensure they see a full triple
<jrtc27> to stop the insanity that is using OS triples for firmware
<jrtc27> at least for LLVM; for GNU stuff they can do what they like to work around the user-unfriendly nature of GNU tools only supporting one OS per build
<meowray> i saw a linkg from Bin Meng. i don't know what opensbi is. just looking at these threads from a non-gnu toolchain perspective. "error: unknown emulation: elf64lriscv_lp64" :(
<meowray> this is the https://reviews.llvm.org/D95755 thing i dislike
<jrtc27> GCC+LLD is the combo I care least about
<jrtc27> even getting GCC to pick up LLD is non-trivial
<jrtc27> collect2 is crap and often picks up bfd despite -fuse-ld=lld
<jrtc27> I just want pure LLVM to work and not be handicapped by irrelevant GNU stupidity
<jrtc27> (ie triple abuse and lack of PIE support)
<meowray> -z text will trigger this GNU ld warning: relocation against `_fw_start' in read-only section `.entry' ; but -N can weirdly suppress it.
<jrtc27> yeah
<jrtc27> one of -N's side-effects is making the output text section writable
<jrtc27> GNU ld treats that as meaning text relocations are allowed, LLD doesn't because the input section is still read-only as that's what it checks
<meowray> is opensbi for u-boot (bootloader) stuff?
<jrtc27> IMO the input is wrong and LLD is justified to do that
<jrtc27> it's the TF-A equivalent (ish)
<jrtc27> runs in M-mode (EL3)
<meowray> in linux, a command like `b4 am http://lore.kernel.org/r/20210604165819.7947-1-jarmo.tiitto@gmail.com` can download the git commit. how to download the patch for opensbi?
<meowray> i can get your patches' message ids from the Gzip'd Text on http://lists.infradead.org/pipermail/opensbi/
<jrtc27> no clue, I hate mailing list workflows for anything other than fire-and-forget patches
<meowray> nit: --target= is slightly preferred to -target
<jrtc27> hmm, all the tests use -target though :P
<jrtc27> (and is what I habitually use...)
<jrtc27> I can switch it over though; I think the Linux kernel uses --target= so it'd be consistent with that
Tranmi has joined #riscv
Tranmi has quit [Remote host closed the connection]
<meowray> (i habitually use -target b/c it is slightly easy to type, but space separated options are just uncommon for driver options)
<jrtc27> ah is this a driver vs cc1 thing?
<meowray> clang-built-linux there is recent discussion on removing CROSS_COMPILE= for LLVM=1 but i guess it may still be needed for riscv64-linux-gnu-* users
<jrtc27> when would you ever want it to not be --target=riscvXLEN-linux-gnu?
<meowray> many traditional clang cc1 options use Separate (space-separated) but most driver options are Joined
<meowray> --target=riscvXLEN when it absolutely doesn't need executables/libraries found from an existing riscvXLEN_linux-gnu-gcc installation and you don't need compile mode defaults from linux-gnu
<meowray> the executables mainly include as (normally not needed because of -fintegrated-as default), objcopy (you don't use split dwarf|some gpu quirk), ld (if you have ld.lld)
adjtm has joined #riscv
<meowray> the -pie -N mode is indeed weird. don't know what it means
<jrtc27> I think "whatever that combination of internal options happens to do" ...
<jrtc27> an ET_EXEC with relocations is surely not something anyone ever thought of and intended to happen...
<jrtc27> that's what -static-pie (or maybe -static -pie?) is for
<jrtc27> (the fact that -static-pie is the GNU option name makes me sad...)
<meowray> (i made me sad too) -static-pie should have been -static -pie; but (1) GNU ld has a default dynamic linker so they need --no-dynamic-linker to cancel it (2) GNU ld normally tries to guess dynamic linking vs static linking and makes some weird decisions), and --no-dynamic-linker can be re-used to encode the "static linking" intention
<meowray> then there is the unfortunate gcc driver->ld translation code. hjl probably didn't want to test the combination of -static -pie so just added a new mode
<jrtc27> one of these years we'll tear this whole thing down and replace it with something saner that isn't filled with 30 years of legacy crud...
<jrtc27> maybe 2030
vagrantc has joined #riscv
freakazoid333 has quit [Read error: Connection reset by peer]
freakazoid333 has joined #riscv
winterflaw has quit [Ping timeout: 244 seconds]
mhorne has quit [Ping timeout: 252 seconds]
mahmutov has quit [Ping timeout: 268 seconds]
pecastro has quit [Ping timeout: 268 seconds]
mifr has quit [Ping timeout: 265 seconds]
mifr has joined #riscv
mhorne has joined #riscv