Lucas De Marchi [Fri, 31 Aug 2012 09:59:26 +0000 (06:59 -0300)]
Remove test directory
All tests should be in testsuite. The remaining tests in this directory
are not relevant enough to be ported. git log can be consulted if in
future we decide to put them in testsuite.
Lucas De Marchi [Thu, 30 Aug 2012 17:10:47 +0000 (14:10 -0300)]
TODO: deprecate use of rmmod -w
As discussed with Rusty Russel, it would be nice to remove the related
code from kernel. Deprecate its use on kmod, so people know they
shouldn't be using it.
Lucas De Marchi [Fri, 17 Aug 2012 12:42:47 +0000 (09:42 -0300)]
modprobe: Unconditionally use KMOD_PROBE_APPLY_BLACKLIST_ALIAS_ONLY
This fixes a change in behavior regarding kmod and module-init-tools:
when trying to load a module by alias, we should check if it's
blacklisted, regardless of the command line arguments passed.
This was reported by "Dmitry V. Levin <ldv@altlinux.org>".
Lucas De Marchi [Fri, 17 Aug 2012 12:38:05 +0000 (09:38 -0300)]
libkmod-module: Add KMOD_PROBE_APPLY_BLACKLIST_ALIAS_ONLY flag
With this flag kmod_module_probe_insert_module() check if module is
blacklisted only if it's also an alias. This is needed in order to allow
blacklisting a module by name and effectively blacklisting all its
aliases as module-init-tools was doing.
Before this patch we could load pcspkr module as follows:
/etc/modprobe.d/test.conf:
alias yay pcspkr
blacklist pcspkr
$ modprobe yay
Now libkmod has support to blacklist "yay" because "pcspkr" is blacklisted.
Martin Pitt [Wed, 1 Aug 2012 21:35:34 +0000 (23:35 +0200)]
testsuite preload: Factorize into macros, add more stat and open variants
Instead of replicating the same code several times, define and use macros for
the various types of wrapped functions in the testsuite's path.c LD_PRELOAD
wrapper.
Add various __xstat() variants and open64(), which are being used when enabling
large file support.
Colin Walters [Mon, 30 Jul 2012 16:27:10 +0000 (12:27 -0400)]
build-sys: Add --disable-manpages option
1) Embedded systems often don't want man pages on the
target; rather than pointlessly building them, then ignoring
the result, allow just not building them at all
2) When bootstrapping an operating systems, documentation is the
source of many cyclical dependencies, and allowing it to
be explicitly disabled is useful for earlier build passes.
Dave Reisner [Wed, 25 Jul 2012 23:29:40 +0000 (19:29 -0400)]
man/modinfo: document --basedir option
This was never documented in the manpage of module-init-tools either.
The flag is identical in function to modprobe's --dirname option, so use
the same language to describe it.
Dave Reisner [Wed, 25 Jul 2012 23:30:19 +0000 (19:30 -0400)]
man/modprobe: clarify --dirname option
The documentation for this flag leads one to believe that the full path
to the module directory is needed. In reality, this flag specifies only
the root of the module path.
Lucas De Marchi [Thu, 21 Jun 2012 14:30:56 +0000 (11:30 -0300)]
testsuite: use right offset for module name
We need to cope with the case in which a 32 bits machine is opening a 64
bits kernel module and vice-versa. The offset in `struct module' are
different and do not depend on the architecture we are running, but
rather on the architecture they were created for.
This fixes `make check' in 32 bits machines (since we are shipping 64
bits modules for testing)
Lucas De Marchi [Tue, 19 Jun 2012 16:09:23 +0000 (13:09 -0300)]
build-sys: Make dirs writable on rootfs creation
Autofoo make the dist dir as readonly. If we copy it, tools needing to
create sysfs entries will not be able to do so, because they can't
create the needed directories/files.
It would be much better if autofoo allowed to let the files as is
instead of converting them to read-only.
Lucas De Marchi [Fri, 15 Jun 2012 23:13:31 +0000 (20:13 -0300)]
depmod: return error when index is truncated due to ENOSPC
Before:
=======
[lucas@vader kmod]$ sudo depmod
[lucas@vader kmod]$ echo $?
0
[lucas@vader kmod]$ ls -l /lib/modules/$(uname -r)
total 12
drwxr-xr-x 8 root root 160 Jun 13 11:05 kernel
-rw-r--r-- 1 root root 12288 Jun 15 21:29 modules.alias
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.alias.bin
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.dep
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.dep.bin
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.devname
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.softdep
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.symbols
-rw-r--r-- 1 root root 0 Jun 15 21:29 modules.symbols.bin
Note that modules.alias is truncated and the other have size == 0
After:
======
[lucas@vader kmod]$ sudo ./tools/depmod
WARNING: could not open /lib/modules/3.5.0-rc2-demarchi-00028-g94fa83c/modules.order: No such file or directory
ERROR: Could not create index: output truncated: No space left on device
[lucas@vader kmod]$ echo $?
1
Lucas De Marchi [Tue, 12 Jun 2012 04:43:46 +0000 (01:43 -0300)]
libkmod-config: refactor functions to get config
It makes more sense to have libkmod-config.c deal with the configuration
directly and the others get the config from ctx. As a bonus point we get
a smaller binary. Following numbers are for x86-64, libkmod + kmod:
Before:
text data bss dec hex filename
128840 1496 104 130440 1fd88 tools/modprobe
After:
text data bss dec hex filename
128392 1496 104 129992 1fbc8 tools/modprobe
Lucas De Marchi [Wed, 6 Jun 2012 04:44:23 +0000 (01:44 -0300)]
testsuite: add test for install-commands loop
This loop is similar to the one that comes with install rules of
alsa-utils package. It can be easily verified by reverting commit abd5557 and running the testsuite.
Lucas De Marchi [Wed, 6 Jun 2012 04:03:15 +0000 (01:03 -0300)]
testsuite: add timeout for each test
Each test must run under 2 seconds. Ideally they should run in much less
than this; just give an arbitrary number so we don't wait forever in
case we reached an infinite loop somewhere.
Lucas De Marchi [Thu, 24 May 2012 04:31:36 +0000 (01:31 -0300)]
testsuite: check if rootfs dir is dirty before running
Keep around a stamp-rootfs file that is generated together with the
rootfs. testsuite checks each test directory if its mtime is greater
than stamp's mtime, deciding if rootfs should be re-generated.
Mike Frysinger [Tue, 15 May 2012 22:29:44 +0000 (19:29 -0300)]
libkmod: move function to the only file using it
If we don't have --gc-sections support, linking kmod fails:
libkmod/.libs/libkmod-util.a(libkmod-util.o): In function 'underscores':
libkmod/libkmod-util.c:117: undefined reference to 'kmod_log'
This is because libkmod-util.la uses kmod_log(), that is in libkmod.la.
Move the function so we don't have a dependency loop while building the
libraries and it works with compilers with no support for --gc-sections.
Dan McGee [Thu, 23 Feb 2012 05:56:58 +0000 (23:56 -0600)]
test-conversion: convert test-blacklist to new infrastructure
Add a modprobe.conf with some blacklist entries in a test rootfs, and
then ensure our blacklist function actually cuts out the two listed
entries (and doesn't cut out the others).
Lucas De Marchi [Fri, 11 May 2012 03:04:59 +0000 (00:04 -0300)]
build-sys: copy rootfs to another directory
We can't use the rootfs directory because it breaks out-of-tree build
and in future we want to make modifications to the fake filesystem such
as adding and removing files.
We need to call "chmod -R +w" in the resulting directory because when we
distribute the source with make dist all files will be readonly.
Dan McGee [Fri, 3 Feb 2012 19:24:18 +0000 (13:24 -0600)]
testsuite: ship testsuite/rootfs unzipped
The current configuration is dumb in any number of ways:
1) If the rationale was for space savings, it works the opposite- the
git repo gets more bloated because we are adding binary compressed
blobs that share little in common with their parent, and anyone that
wants to run the test suite has to unzip it anyway.
2) It is a pain in the butt to add new tests, and not accidentally lose
any new rootfs you built in the directory.
3) `git status` won't help you if you are tweaking files in the rootfs
and don't know they have been changed, or if some test did that and
you couldn't detect it.
4) `git log` won't help you find out what is changing in the rootfs test
directory itself when changes are made to the binary blob, such as
new files being added or even existing files being tweaked.
5) The files just aren't that big anyway- 2.7MB unzipped.
Lucas De Marchi [Fri, 11 May 2012 00:11:27 +0000 (21:11 -0300)]
build: remove private symbol from linker script
There's no reason kmod_log should be exported, remove it from linker
script. This doesn't break the API/ABI because we are luck: since the
function had visibility=hidden it was not getting exported as a global
symbol.
Dave Reisner [Mon, 7 May 2012 23:41:42 +0000 (19:41 -0400)]
depmod: report failures in loading symbols
Previously, depmod would relegate failures of kmod_module_get_symbols()
to debug output, assuming the "error" was simply a lack of symbols.
Leave the ENOENT return to debug output, but report anything else as a
real error.
Dave Reisner [Mon, 7 May 2012 23:41:41 +0000 (19:41 -0400)]
libkmod-file: gracefully handle errors from zlib
zlib won't necessarily set the system errno, and this is particularly
evident on corrupted data (which results in a double free). Use zlib's
gzerror to detect the failure, returning a generic EINVAL when zlib
doesn't provide us with an errno.
Mike Frysinger [Sat, 21 Apr 2012 06:53:23 +0000 (02:53 -0400)]
libkmod: make sure to export kmod_log
If we don't have --gc-sections support, linking kmod fails:
libkmod/.libs/libkmod-util.a(libkmod-util.o): In function 'underscores':
libkmod/libkmod-util.c:117: undefined reference to 'kmod_log'
This is because kmod_log is missing the export define, even though it's
already listed in the exported symbol list.
Dave Reisner [Mon, 16 Apr 2012 14:37:32 +0000 (10:37 -0400)]
modprobe: handle -ENOENT return from init_module
init_module returns -ENOENT when the module failed to load because of a
bad parameter or unknown symbol. Throw a more descriptive error message
than the generic "No such file or directory" to alert the user.
Lucas De Marchi [Sat, 17 Mar 2012 18:12:44 +0000 (15:12 -0300)]
build-sys: don't set CFLAGS and LDFLAGS
These variables are supposed to be set by user. What we can do in
configure is to set another variable and AC_SUBST() it. Then in
Makefile.am we assign it to AM_{CFLAGS,LDFLAGS}. This way user can
always override their values, in configure or make phase.
Randy Witt [Fri, 16 Mar 2012 05:01:18 +0000 (01:01 -0400)]
configure.ac: Move link only flags out of CFLAGS and into LDFLAGS.
On the ARM gcc 4.2.2 I'm using I get many
"--as-needed: linker input file unused because linking not done"
style errors when libtool is using compile mode.
In order to silence these warnings and be "correct", the flags that
only make sense for linking were moved into LDFLAGS.
Randy Witt [Fri, 16 Mar 2012 05:01:17 +0000 (01:01 -0400)]
Add CC_CHECK_LDFLAGS_APPEND m4 macro.
This is mostly just preparation for the next patch. But this macro is
used for flags that are only needed during linking but that don't
make sense for normal compilation.
I saw this exact type of patch online for systemd, but it never seems to
have actually been incorporated into their attributes.m4.
Lucas De Marchi [Thu, 15 Mar 2012 12:19:34 +0000 (09:19 -0300)]
config: use order /etc, /run, /lib
This matches the change in systemd and udev. Log message on udev's
change by Kay Sievers:
After long consideration we came to the conclusion that user
configuration in /etc should always override the (generally
computer generated) configuration in /run. User configuration
should always be what matters over anything else. Hence rearrange
the search orders accordingly. In general this should change
very little as overriding like this is seldomn done so far,
and the order between /etc and /usr stays the same.
Lucas De Marchi [Thu, 15 Mar 2012 03:14:35 +0000 (00:14 -0300)]
modprobe: don't check if module builtin to decide if it's builtin
More or less confusing message, but if module is builtin in the live
system, it doesn't mean it's builtin in the target kernel.
Instead we now check if module has a path. It don't have a path only if
it's builtin in the target or if it doesn't exist at all. The latter
should not be a problem since this code is being called from inside the
library. Anyway, put an assert to make sure we get bug reports if any
case slipped in here.
Dave Reisner [Thu, 15 Mar 2012 02:15:21 +0000 (22:15 -0400)]
modprobe: fix error path in removing modules
We really haven't paid this code much attention, and it's somewhat
evident in our divergence in behavior from module-init-tools. This patch
asserts the following behavior on exit: