| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, load_url took a char * argument, from which it parsed a
newly allocated URL, and freed the URL before returning.
Commit 5be946c changed load_url (then load_file) to accept a parsed URL
instead of a char *, but didn't remove the free. Any URLs passed to
load_url are currently being unintionally free()ed.
This change removes the invalid free.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
We may see initrds specified on the kernel argument line, or as their
own configuration directive.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
Abort the pair parse if we don't have both a name and a value.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change the Makfile.am relocatable output files from automake _LIBRARIES
to automake _PROGRAMS. Also, change the output file name extension
from .o to .ro to better show these are relocatable files.
Fixes automake warnings like these:
discover/Makefile.am: `libparser.o' is not a standard library name
discover/Makefile.am: did you mean `libparser.a'?
Signed-off-by: Geoff Levand <geoff@infradead.org>
|
|
|
|
|
|
|
| |
By exiting with status == 2, boot hooks can update boot data by
printing name=value to stdout.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
We're not accounting for the action in the event header (only the device
string) when we pass the length of param data to event_parse_params.
This means we walk past the end of the event data while parsing params.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
Add a method of running pre-boot hooks. Executable files in
/etc/petitboot/boot.d/ are run (in order) before we start the boot
process.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
This keeps all the boot-specific details in one place.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
We'll need a sysconf dir to store the boot hooks.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
Updates & fixes by Jeremy Kerr <jk@ozlabs.org>.
Signed-off-by: Geoff Levand <geoff@infradead.org>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
| |
Add a library for (name, value) configuration.
Different storage backends are allowed (although currently hardcoded to
powerpc nvram), and config is read-only at present.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
|
| |
Useful for specifying defaults:
pb-event add@defaults name='Netboot' \
image=http://192.168.0.1/vmlinuz \
default
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
Allow the default boot process to be cancelled, via a message with
action PB_PROTOCOL_ACTION_CANCEL_DEFAULT.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
When we see a boot option with is_default set, store it in the handler
and register a timeout waiter.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
If we're booting from a default option, we don't have a boot command.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
Send the boot options to the client in the order discovered. This
change is in follow up to inverting the order that the server
saves the options internally.
Signed-off-by: Geoff Levand <geoff@infradead.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
When a client disconnects, the read from the client's fd will return
EOF. We should destroy the client in this situation.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
Now that we dynamically add parsers, we can use a list to
hold them. Also simplifies the test_run_parser() routine.
Signed-off-by: Geoff Levand <geoff@infradead.org>
|
|
|
|
|
|
| |
Ubuntu uses the linux16 symbol in thier conf files for memory test entries.
Signed-off-by: Geoff Levand <geoff@infradead.org>
|
|
|
|
|
|
| |
Grub2 menuentry entry text can use double or single quotes.
Signed-off-by: Geoff Levand <geoff@infradead.org>
|
|
|
|
|
|
| |
Add a resource type for grub, allowing us to parse search parameters.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
Any other implementation of resources will need to resolve againsst
particular devices, so make resolve_devpath_against_device
publically-accessible, and rename to not be devpath-specific.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
Remove some of the more noisy log messages, and add some information
pertinent to device resolution events.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
|
|
| |
This change moves some of the device-handler code into an #ifdef-ed
section, so we can easily drop the stuff that's not required for
testing.
Although the change is quite large, most of it is moving entire
functions around.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
Devices that have been added via the user path may not have a device
path. In this case, don't segfault in device_match_path, and break out
of mount_device early.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
|
|
| |
We should always be tallocing resources to a boot option context;
anything else (for example, the discover context) may have a different
lifetime.
In order to enforce this, we change the void *ctx argument to the
context_create functions to a struct discover_boot_option.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
conf_get_pair will read one-byte past the end of the conf buffer, so
always NUL-terminate.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
We're only picking up every second boot option.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
devnames are unique, and much shorter.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
No need for parsers to populate (or forget to populate, in the case of
most parsers) opt->device_id, as we should do it on finalise.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
| |
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
| |
This cleans up the log output a little.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
| |
Rather than creating boot options pre-emptively, in two paths, just do
it once when we see the menuentry option.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
| |
Keep options in the order that we discovered them in; this makes testing
a little easier, as the options appear in the list in the same order
as the config file.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|
|
|
|
|
|
|
|
|
|
| |
Currently, when a new UI client connects, we send all boot options to
all clients. This results in existing clients getting duplicate add
events.
Instead, we only want to send existing boot options to the new client.
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
|