Commit Graph

105 Commits

Author SHA1 Message Date
Ciro Santilli 六四事件 法轮功
0deab8f8f7 user mode: use the nonexistent symlink workaround
I was considering setting --static by default to match gem5, but then
that breaks shared libraries like openblas... so let's just use the
ugly workaround for now as it seems to work...
2019-03-12 10:01:38 +00:00
Ciro Santilli 六四事件 法轮功
da900a579c LKMC v3.0
This is a squash commit, the unsquashed development went through many
unstable phases which would break bisects. The unsquashed branch is:
https://github.com/cirosantilli/linux-kernel-module-cheat/tree/v3.0-unsquash

The main improvement of this release was to greatly generalize the testing system.

The key addition was cli_function.py, which allows scripts such as ./run to
be transparently called either from Python or from the command line.

New tests scripts were created using this improved framework: test-baremetal
and test-user-mode.

We were lazy to port some of less important tests to the new setup, TODO's were
added, and we need comes they will be fixed. Getting started is however sacred
as usual and should work.

Other changes include:

-   gem5: update to 7fa4c946386e7207ad5859e8ade0bbfc14000d91

-   run: --tmux-args implies --tmux

-   run: add --userland-args to make userland arguments across QEMU and gem5

    Get rid of --userland-before as a consequence.

-   bring initrd and initramfs back to life

-   build-userland: create --static to make build a bit easier

-   gem5: --gem5-worktree also set --gem5-build-id

-   remove --gem5, use --emulator gem5 everywhere

    Allow passing --emulator multiple times for transparent tests selection
    just like --arch.

-   test-userland: allow selecting just a few tests

-   linux: update to v4.20

-   buildroot: update to 2018.08

    The main motivation for this was to fix the build for Ubuntu 18.10, which
    has glibc 2.28, which broke the 2018.05 build at the m4-host package with:

        #error "Please port gnulib fseeko.c to your platform!

-   getvar --type input

-   failed xen attempt, refactor timer, failed svc attempt, aarch64 use gicv3

-   build-doc: exit 1 on error, add to release testing

-   build: add --apt option to make things easier on other distros

-   build-linux: --no-modules-install
2019-01-22 00:00:00 +00:00
Enrico Tolotto
2a78702b71 rootfs_overlay: qemu_edu: fix kernel module name 2018-11-04 20:31:20 +01:00
Ciro Santilli 六四事件 法轮功
b82bd50888 kgdb: actually, module debug just works with automatic lx-symbols? What happened? 2018-11-02 22:00:07 +00:00
Ciro Santilli 六四事件 法轮功
98bada1e7c getting started: explain 9p module rebuild
split networking and filesystem toplevel secions

create closest overlayfs attempt so far and document it
2018-10-14 00:00:00 +00:00
Ciro Santilli 六四事件 法轮功
ddecb1bf5d make overlay scripts use kernel modules and executables relative to pwd in preparison for 9p 2018-10-13 00:00:00 +00:00
Ciro Santilli 六四事件 法轮功
21e223c563 TODO document minimal init 2018-10-12 09:30:33 +01:00
Ciro Santilli 六四事件 法轮功
cda40f4dfe I understand the shell environment 100% 2018-10-12 09:30:33 +01:00
Ciro Santilli 六四事件 法轮功
ce216152dd kernel cli: rw by default 2018-10-12 09:30:33 +01:00
Ciro Santilli 六四事件 法轮功
bc73cebff1 Build the Linux kernel independently from Buildroot
This will allow for other types of root filesystems that don't rely on Buildroot
to be added and used in the future.

Propagate --verbose on all build scripts to see full GCC commands.

build-all: allow for neat subsets

also 9p share rootfs_overlay. TODO document.
2018-10-12 09:30:33 +01:00
Ciro Santilli
56738a1c70 lkmc v2-rc
Unsquashed version at v2-rc-unsquashed, but that cannot be merged as it
breaks bisects at several points. All bugs will not bisect to this
humongous change.

It all started with a conversion of the Bash scripts to Python, mainly
because I couldn't stand not being able to properly use --options for
run which has a million options.

Then since that required a full testing, I decided to do all the
refactorings that I had in mind at once, and so I did and it became
v2-rc.

This is the largest patch I have ever done! OMG a few weeks of extra time.
I'm never writing a Bash script for anything that starts getting big again.

Some of the features are:

* separate build-qemu and build-gem5 commands
* common: convert scripts to python. Add --option for everything
* rename build to build-buildroot now that we are splitting all the build
  commands, Linux kernel to follow
* move all git submodules to submodules/ and all buildroot packages to
  packages/
* refactor the out/ structure. Keep projects on toplevel, because guest
  projects separate archs and host ones don't, making a toplevel arch wrong
* do-release: rename to just release
  https://stackoverflow.com/questions/16174992/cant-get-argparse-to-read-quoted-string-with-dashes-in-it
* run: add --terminal and explain gem5 pdb
* just track the lvimrc
* store CLI kernel config fragment inside buildlroot to avoid conflicts
* gem5: document m5 initparam
* readme: make a bunch of things awesomer
* readme: fix broken refs
* parsec-benchmark: update to 75d55ac446a43c47efb1044844a108c6c330184c
  Could not fetch otherwise.
* gem5: M5_OVERRIDE_PY_SOURCE
2018-09-14 07:42:31 +01:00
Ciro Santilli
eb4f4ba1d7 gem5.sh: simplify, and do m5 resetstats and m5 exit
This covers the most common use case of running a benchmark after restore.
2018-08-25 06:13:09 +01:00
Ciro Santilli
6045b9fa3d uio: move docs for this disgraceful failed example to README 2018-07-09 17:54:17 +01:00
Ciro Santilli
5d9418b276 bring CONFIG_FORTIFY_SOURCE back from accidental removal...
Notice that it is not working anymore.

Rename pci.c into qemu_edu.c

Organize kernel_module readmes further
2018-07-06 10:23:35 +01:00
Ciro Santilli
e05eb91add netlink: move docs to README 2018-07-06 08:56:22 +01:00
Ciro Santilli
237b27869e kstrto: move doc to README 2018-07-05 02:42:28 +01:00
Ciro Santilli
6c241279a6 split kernel module api docs to README 2018-07-04 10:03:02 +01:00
Ciro Santilli
50cb67e2e2 mmap: move doc to README 2018-07-01 20:38:16 +01:00
Ciro Santilli
2075fbaf5b Move poll, ktrhead and kthreads docs to README 2018-07-01 20:01:49 +01:00
Ciro Santilli
d840b0cb65 ioctl: move doc to README 2018-07-01 17:44:38 +01:00
Ciro Santilli
084e3faf5a seq_file: move doc to README 2018-07-01 16:46:15 +01:00
Ciro Santilli
923f655811 anonymous inode: move doc to readme 2018-07-01 16:10:29 +01:00
Ciro Santilli
9a4eae50c8 Move debugfs, rootfs and procfs documentation to README 2018-06-29 09:56:41 +01:00
Ciro Santilli
0cd1a2b602 kernel module: fix fops... and move its documentation into README
Sometimes I wonder if anyone has ever run this tutorial, otherwise how
can such basic bugs persist for so long?

test_all.sh: crete
2018-06-29 08:47:32 +01:00
Ciro Santilli
7f3671894f kernel_module: move character device example doc to README 2018-06-29 07:28:09 +01:00
Ciro Santilli
1721f184d6 gem5: track build and run variants separately with -M and -N
Otherwise, checking out branches is too insane, as it does not
update the worktrees, even though the gem5/gem5 module was updated.

gem5: expose build types, document debug builds.

simultaneous runs: store stdout and stderr on a file to allow running
all from a single terminal on the background cleanly.
2018-06-27 16:31:24 +01:00
Ciro Santilli
57dbe81f33 lkmc_many_files: create package to test a rootfs with many files
Automatically pick up packages from under packages/ into BR2_EXTERNAL.

Move many subdirectory READMEs into the toplevel and link to toplevel from those subreadmes instead.
2018-06-17 11:36:56 +01:00
Ciro Santilli
4484e866cb terminal: managed to spawn getty on TTYs manually from init 2018-06-14 12:05:56 +01:00
Ciro Santilli
6e495b0625 fb.sh: create shortcute for fb test 2018-06-03 22:03:39 +01:00
Ciro Santilli
dbecf15f8f inittab: spawn -/bin/sh instead of getty 2018-05-23 02:14:32 +01:00
Ciro Santilli
207306fa31 gpio: move documentation to README, document that it broke on -M virt 2018-05-22 08:09:06 +01:00
Ciro Santilli
37743c40c3 conf.sh: base insensitive for even less typing 2018-05-06 11:28:54 +01:00
Ciro Santilli
22a7b9bf2b config.sh: builtin grep! 2018-05-05 19:21:18 +01:00
Ciro Santilli
2c084f5fb2 run: create -F option to run base64 encoded command after busybox init
Fix ./run -h which was showing the build help instead.
2018-04-19 08:57:35 +01:00
Ciro Santilli
51e31cdc29 x86_64: rdstc
arm: attempt pmccntr, but didn't work, TODO why: no QEMU support?
2018-04-15 14:10:21 +01:00
Ciro Santilli
ae780f6750 Two working methods for module_init GDB step debugging!
Not perfect, but doable.
2018-04-15 11:05:03 +01:00
Ciro Santilli
2427a322fc kstrto 2018-04-13 08:37:00 +01:00
Ciro Santilli
42e634ffd1 debugfs.c: extend example with toplevel file, fop and kstrtoull_from_user 2018-04-13 07:58:58 +01:00
Ciro Santilli
1c3f16797a readme: more failed attempts at ttys! 2018-03-27 14:00:32 +01:00
Ciro Santilli
ac67ce3643 Login without asking for the username
We have wanted to do this since forever, but the last straw was gem5 aarch64,
which runs too fast, and makes it very hard to write "root" withing 60 seconds!

Another possibility to solve that would have been to find an answer for:
https://unix.stackexchange.com/questions/340333/how-can-i-get-bin-login-to-not-timeout
which we didn't, but not typing root at all is even better.

Take this opportunity to document how to login into user0.

sshd: automate and document further

gdbserver: automatically startup internet on /gdbserver.sh
2018-03-26 13:19:54 +01:00
Ciro Santilli
2656202dff init: disable all userland processes started from init
In order to make the system easier to understand.

These include:

* networking
* klogd and syslogd. TODO what are those for? I could not see anything useful that they do.

Also get rid of the useless S20random thing while we are at it.
2018-03-25 19:06:30 +01:00
Ciro Santilli
29908dffb2 Disable networking at runtime with lkmc_nonet cli instead of build
This is necessary because qemu and gem5 now use the same build folder.

Separate ./run -e and -f for kernel options before and after the ' - '.
This was already the better thing to do when -E was introduced,
but lkmc_nonet prompted me to do it nicer now.

Use the common script to find the out_dir on every toplevel script.

Include usage man pages on README.
2018-03-22 02:47:14 +00:00
Ciro Santilli
98ed442883 overlayfs 2018-03-17 12:08:52 +00:00
Ciro Santilli
451c9b4014 overlay: use /bin/sh for conf.sh instead of bash 2018-03-17 10:57:59 +00:00
Ciro Santilli
e8f09a76e6 Add a shortcute for /eval.sh
Create /eval_base64.sh to overcome quoting and newline limitations.

Documente poweroff's failure on readme.
2018-03-14 21:13:45 +00:00
Ciro Santilli
7b08433840 9p mount the output directory.
This is in preparation for moving parsec to 9p once we get it working on gem5.

We'll document it then when everything is working.
2018-03-11 22:00:55 +00:00
Ciro Santilli
6a7b1b436c One less tab 2018-03-10 23:14:41 +00:00
Ciro Santilli
efa8d2075c bak 2018-03-10 21:22:16 +00:00
Ciro Santilli
2b7d274949 config.sh kernel config helper 2018-03-09 18:06:44 +00:00
Ciro Santilli
04db718441 qemu: 9p host guest directory sharing 2018-03-09 17:14:10 +00:00