When compiling core.c, it may report error like:
"error: implicit declaration of function ‘in_nmi’"
Adding header file in_nmi defined could avoid this.
Signed-off-by: Jincheng Miao <jincheng.miao@gmail.com>
Put funcs, num_funcs, and mod in their own struct called kpatch_module.
This allows us to keep patch module specific variables in one place (and
we'll have more of these variables soon).
Support live patching on NMI handlers. This adds checks for
possible inconsistency of live patching on NMI handlers.
The inconsistency problem means that any concurrent execution
of old function and new function, which can lead unexpected results.
Current kpatch checks possible inconsistency problem with
stop_machine, which can cover only threads and normal interrupts.
However, beacuse NMI can not stop with it, stop_machine is not
enough for live patching on NMI handlers or sub-functions which are
invoked in the NMI context.
To check for possible inconsistency of live patching on those
functions, add an atomic flag to count patching target functions
invoked in NMI context while updating kpatch hash table. If the
flag is set by the target functions in NMI, we can not ensure
there is no concurrent execution on it.
This fixes the issue #65.
Changes from v5:
- Fix to add a NULL check in kpatch_get_committed_func().
Changes from v4:
- Change kpatch_operation to atomic_t.
- Use smp_rmb/wmb barriers between kpatch_operation and kpatch_status.
- Check in_nmi() first and if true, access kpatch_operation.
Changes from v3:
- Fix kpatch_apply/remove_patch to return 0 if succeeded.
Changes from v2:
- Clean up kpatch_get_committed_func as same style of kpatch_get_func.
- Rename opr to op in kpatch_ftrace_handler.
- Consolidate in_nmi() and kpatch_operation check into one condition.
- Fix UNPATCH/PATCH mistype in kpatch_register.
Changes from v1:
- Rename inconsistent_flag to kpatch_status.
- Introduce new enums and helper functions for kpatch_status.
- Use hash_del_rcu instead of hlist_del_rcu.
- Rename get_committed_func to kpatch_get_committed_func.
- Use ACCESS_ONCE for kpatch_operation to prevent compiler optimization.
- Fix to remove (!func || func->updating) condition from NMI check.
- Add more precise comments.
- Fix setting order of kpatch_status and kpatch_operation.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Handle registering error to unroll the ftrace filter.
This also introduces get_kpatch_func() and
kpatch_remove_funcs_from_filter() for holding up
redundant loops.
Changes from v2:
- Rebased on the latest kpatch.
Changes from v1:
- Rename get_kpatch_func to kpatch_get_func.
- Fix function definition style issue.
- Do not jump to a label in "if" block.
- Rollback the ftrace user counter if we hit an error.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
The kpatch utility is now user friendly enough that it can be used
instead of direct insmods. We should encourage people to use it, since
we will soon be adding user space functionality above and beyond
insmod/rmmod when loading and unloading.
Allow "kpatch load" to find the core module when kpatch is run directly
from the git tree. This gives the user the option to use the kpatch
utility directly without having to do a "make install".
While debugging the code for the bug table logic, I found it useful to
know which rela section and entry the error occurred on.
Signed-off-by: Seth Jennings <sjenning@redhat.com>
This commit adds a new function to properly handle the bug table.
It works by going through .rela__bug_table, after the changed
function symbols have already been marked, and rewrites the section
including only the relocations pertaining to bug entries for
changed functions.
The __bug_table section itself is not modified resulting in
"blank" bug entries: ones whose IP and filename pointers will
not be relocated and, therefore, will be zero. While a waste
of space, it simplifies the code not to remove these blank
entries. They do no harm.
Signed-off-by: Seth Jennings <sjenning@redhat.com>
The section header size is calculated at output time by libelf
and we use it as a read-only value from read files.
With the next patch we are changing the size of the .rela__bug_table
section. Lets use d_size instead since it is the value that tells
libelf how to calculate sh_size at output time.
Signed-off-by: Seth Jennings <sjenning@redhat.com>
Allow bundling of .bss.* sections that are the result of -fdata-sections
so that rela sections referencing data in bss sections by section symbol
can be replaced with the object symbol so it can be linked to the existing
data object in the kernel.
Signed-off-by: Seth Jennings <sjenning@redhat.com>
For log_normal and DIFF_FATAL messages, prefix them with the object name
to give more context, which is useful for patches which change multiple
objects. Also, no need to add the function and line number to
DIFF_FATAL messages, as the error strings already give enough
information.
Example messages:
meminfo.o: changed function: meminfo_proc_show
cmdline.o: no changed functions were found
A user may want to load a module without having to install it to the
initrd. In fact, 99% of the time I think that will be the typical usage
of "kpatch load", with a given file as an argument rather than a DB
module.
Have "kpatch load" and other commands take a file as input. If the file
is not found, then check the DB.
Try to be more consistent with the terminology. In various places we
call it a "hotpatch module", "hot patch module", or "patch module". How
about we just call it a "patch module" everywhere?
For now, taint with TAINT_USER when loading a patch module so that the
user can always detect when a kpatch module has been previously loaded.
Eventually we will want a dedicated TAINT_KPATCH flag in the kernel.
Unforunately the dracut module doesn't work if installed in
/usr/local/lib/dracut. It must always be installed in
/usr/lib/dracut regardless of the install prefix.
This reverts commit ab29b1ff59.
Reverting this commit because it causes kpatch-build to ignore any
errors reported by create-diff-object, treating all errors as meaning
that no changes occurred, which is a dangerous assumption to make.
