Commit graph

822 commits

Author SHA1 Message Date
Daniel Barlow
62c788eb86 add hook to run maintenance mode instead of rebooting 2023-11-22 00:05:55 +00:00
Daniel Barlow
bab6d346a8 add .../s6/bin to PATH for shutdownd 2023-11-22 00:05:03 +00:00
Daniel Barlow
a202ae476a extract console redirection stuff from "quit" function
so we can use it for scripts that don't reboot at their end
2023-11-21 23:32:37 +00:00
Daniel Barlow
7c9297f91d use shotdown instead of hpr in .s6-svscan/SIGFOO
this is to bring them into line with what more recent
s6-init-linux-maker creates
2023-11-21 23:19:00 +00:00
Daniel Barlow
a0bd250963 switch from getty to root shell on console
this just makes things marginally simpler
2023-11-21 23:09:48 +00:00
Daniel Barlow
c8b2d58dd3 exit 0 on service down even if no outputs to delete 2023-11-21 17:25:50 +00:00
Daniel Barlow
e5223f093f kernel.src may be a path not just a package
this makes it easier to hack the kernel locally and test things
2023-11-18 14:21:18 +00:00
Daniel Barlow
c563a6451f add missing param 2023-11-18 14:20:59 +00:00
Daniel Barlow
f45326b9d3 why we decided not to depend on kexec 2023-11-18 11:51:57 +00:00
Daniel Barlow
f9f4d97bb8 convert flash params to int 2023-11-12 20:39:06 +00:00
Daniel Barlow
abfb35a231 and entryPoint 2023-11-12 18:50:47 +00:00
Daniel Barlow
315907de98 convert hardware loadAddress to int 2023-11-12 18:47:31 +00:00
Daniel Barlow
185117843b convert tftp.loadAddress from string to int 2023-11-12 18:37:33 +00:00
Daniel Barlow
0131686661 use parseInt for hex values 2023-11-12 18:25:38 +00:00
Daniel Barlow
3da692f7ef don't import flashimage unconditionally, it breaks qemu 2023-11-12 18:11:13 +00:00
Daniel Barlow
f61e737b54 improve doc for outputs and hardware
Changed my mind about "installer" as a first-class concept, at least
in the current implementation. Not every documented output is an
installer
2023-11-12 17:15:58 +00:00
Daniel Barlow
262efaabe6 doc: put all the u-boot/serial stuff in one place to link from 2023-11-12 17:14:33 +00:00
Daniel Barlow
7cfb92e3ce more doc 2023-11-10 21:17:20 +00:00
Daniel Barlow
22882dabee think 2023-11-10 21:10:26 +00:00
Daniel Barlow
5e046490de support links from device pages to their installation methods 2023-11-09 23:02:35 +00:00
Daniel Barlow
a9760d239c basic doc for flashimage installer 2023-11-09 22:43:50 +00:00
Daniel Barlow
5729cfb4a7 document installation methods (only vmroot yet) 2023-11-09 22:14:31 +00:00
Daniel Barlow
7d5c7b9b44 export evaluation from default.nix and use it for docs 2023-11-09 22:14:31 +00:00
Daniel Barlow
23b3a2baef extract vmroot output into its own file 2023-11-08 23:19:11 +00:00
Daniel Barlow
4cb4f904f8 delete unused kconfig for arm qemus 2023-11-08 21:28:12 +00:00
Daniel Barlow
a9d847e2c0 add ext4 as rootfsType 2023-11-06 21:52:31 +00:00
Daniel Barlow
6bbff2f5b3 think think 2023-11-05 23:39:50 +00:00
Daniel Barlow
5c1f5fabe2 switch pppoe test to use armv7l 2023-11-05 23:19:40 +00:00
Daniel Barlow
6489a39424 qemu armv7 2023-11-05 23:19:11 +00:00
Daniel Barlow
c94d12934f remove direct use of run-liminix-vm from tests and doc 2023-11-05 20:37:23 +00:00
Daniel Barlow
c40eef25d6 qemu: use phram instead of block2mtd 2023-11-05 19:13:51 +00:00
Daniel Barlow
46991e2761 aarch64 ram starts at 0x40000000 2023-11-05 15:33:10 +00:00
Daniel Barlow
a135cb1217 introduce lim, the liminix library
so far we have lim.parseInt, which parses an integer from a string
with optional base-selecting-prefix (e.g. 0755, 0x12ab)
2023-11-05 15:13:06 +00:00
Daniel Barlow
863045b86b added hardware.ram.startAddress config
it's not 0 on arm32, so this will be useful for qemu
2023-11-05 15:11:58 +00:00
Daniel Barlow
629624bb25 replace multiway if with pkgs.stdenv.hostPlatform.qemuArch 2023-11-05 11:40:26 +00:00
Daniel Barlow
92b9bf959e options.system.outputs.initramfs -> initramfs module 2023-11-05 11:33:02 +00:00
Daniel Barlow
c5c5f1687a patch qemu to load uncompressed ARM kernels at correct offset 2023-11-05 11:32:47 +00:00
Daniel Barlow
80793aa694 belkin rt3200 is a ubifs device 2023-11-05 11:32:23 +00:00
Daniel Barlow
824536f9b3 in uimage FIT, honour ${arch} 2023-11-05 11:31:28 +00:00
Daniel Barlow
e6cb5e319b extract NETDEVICES kconfig to kernel.nix module 2023-11-05 11:31:23 +00:00
Daniel Barlow
c3ccee6506 preinit: print errno (in hex, it's easier) for failures 2023-11-05 11:27:57 +00:00
Daniel Barlow
6db982f25f preinit: pause before exiting
as explained in the comment, this is to give us a
chance to see error messages before the kernel panics
2023-11-05 11:27:57 +00:00
Daniel Barlow
86a5224f3c preinit: fix compiler warnings 2023-11-05 11:27:43 +00:00
Daniel Barlow
155a29d9b3 preinit: strip trailing newline(s) on /proc/cmdline 2023-11-05 11:27:34 +00:00
Daniel Barlow
e6ef4f78bb "ubimage" module contains ubifs image + instructions
Presently either you run this from U-Boot or you figure out for
yourself how to kexecboot into a recovery system :-)
2023-10-21 23:20:53 +01:00
Daniel Barlow
d2f517a4e9 preinit.c: reindent 2023-10-19 21:02:18 +01:00
Daniel Barlow
0f38ee0e9c remove PREINIT_USE_LIBC option as it is now the only option 2023-10-19 18:59:02 +01:00
Daniel Barlow
61dc5beca8 preinit: parse rootfstype from kernel command line 2023-10-19 18:56:09 +01:00
Daniel Barlow
f3225c2bd5 delete dup outputs.systemConfiguration
perhaps this should go into initramfs.nix not jffs2.nix
2023-10-19 10:09:08 +01:00
Daniel Barlow
8798ee9830 partial fix for timeout handling
1) "Unknown transfer id" message was because the local variable "tid"
is not a transfer id, it is a sequence number  - so the check was
actually comparing expected vs actual acknowledged sequence number,
not TID.  It's still a problem if we get the wrong one, but it
indicates a lost packet (so we should resend) not a packet that was
sent from somewhere else.

2) if the ACK packet has not been received, our retry should involve
_resending_ it, not just trying to wait for it again.

3) I have removed the timeout condition for terminating the resend
loop, because in practice (assuming both ends have the same timeout
setting) all it did was ensure that the loop only ran once. The
timeout is supposed to regulate how long we wait for before retrying
(it doesn't do this, we wait indefinitely), not how long we wait for
before giving up.
2023-10-18 23:35:23 +01:00