!xmLtiCaAJxfhURjrXl:matrix.org

NixOS RISC-V

199 Members
NixOS on RISC-V https://wiki.nixos.org/wiki/RISC-V https://pad.lassul.us/NixOS-riscv64-linux https://github.com/orgs/NixOS/teams/risc-v58 Servers

Load older messages


SenderMessageTime
23 Nov 2025
@rosscomputerguy:matrix.orgTristan RossBecause I thought that's what you mentioned it was when we were looking at this problem a couple weeks ago20:30:32
@dramforever:matrix.orgdramforever i said it's not a pie 20:30:45
@rosscomputerguy:matrix.orgTristan RossHuh20:30:59
@rosscomputerguy:matrix.orgTristan Ross So why else would mmap fail with 0xf000 when vm.mmap_min_addr is 65535? 20:31:29
@rosscomputerguy:matrix.orgTristan Rossmmap return EPERM20:31:49
@dramforever:matrix.orgdramforeverbecause it's not a pie20:31:50
@dramforever:matrix.orgdramforeverthat's why it fails20:31:54
@rosscomputerguy:matrix.orgTristan RossWhat?20:32:00
@dramforever:matrix.orgdramforever it fails because it's not position independent 20:32:06
@rosscomputerguy:matrix.orgTristan RossI thought you just said it's not a PIE problem20:32:06
@dramforever:matrix.orgdramforeverit's a problem because it's a PDE20:32:24
@rosscomputerguy:matrix.orgTristan RossYes, so it's the problem I was explaining. The executable should be a PIE but the kernel thinks its not.20:32:49
@dramforever:matrix.orgdramforeverthe executable has no signs of being a PIE20:33:04
@dramforever:matrix.orgdramforeverit has type ET_EXEC, first segment doesn't start at address 020:33:41
@rosscomputerguy:matrix.orgTristan RossYeah20:33:56
@rosscomputerguy:matrix.orgTristan RossAnd the good binaries have a different address for the PHDR20:34:08
@dramforever:matrix.orgdramforeverso why "should" it be a PIE?20:34:10
@rosscomputerguy:matrix.orgTristan RossSo mmap wouldn't EPERM?20:34:31
@dramforever:matrix.orgdramforeverthat can't be the general solution though20:37:05
@dramforever:matrix.orgdramforeverbecause random stuff just doesn't work built as PIE20:37:14
@dramforever:matrix.orgdramforeverlast time i checked this list includes gcc itself i think20:37:37
@rosscomputerguy:matrix.orgTristan RossOh, so how would GCC get fixed so it doesn't EPERM?20:38:05
@dramforever:matrix.orgdramforeveri have no idea why ubuntu doesn't have this problem20:53:02
@dramforever:matrix.orgdramforeveris their gcc a pie?20:53:07
@rosscomputerguy:matrix.orgTristan RossIdk20:54:12
@dramforever:matrix.orgdramforeverwhy is ubuntu not just a big monorepo that i can search anything i want in 🫠21:06:45
@rosscomputerguy:matrix.orgTristan Rosslol21:09:44
@dramforever:matrix.orgdramforevereh nevermind i think we build gcc as pie now21:12:19
@kwinz:matrix.orgkwinz https://cache.ztier.in riscv binary cache is only for unstable, correct?
is there a binary cache for stable?
22:01:05
@rosscomputerguy:matrix.orgTristan RossI've only known a few exist but idk the reliability22:16:02

Show newer messages


Back to Room ListRoom Version: 10