Skip to content

Commit

Permalink
x86/pm: fix false positive kmemleak report in msr_build_context()
Browse files Browse the repository at this point in the history
Since commit e2a1256 ("x86/speculation: Restore speculation related MSRs during S3 resume"),
kmemleak reports this issue:

  unreferenced object 0xffff888009cedc00 (size 256):
    comm "swapper/0", pid 1, jiffies 4294693823 (age 73.764s)
    hex dump (first 32 bytes):
      00 00 00 00 00 00 00 00 48 00 00 00 00 00 00 00  ........H.......
      00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    backtrace:
      msr_build_context (include/linux/slab.h:621)
      pm_check_save_msr (arch/x86/power/cpu.c:520)
      do_one_initcall (init/main.c:1298)
      kernel_init_freeable (init/main.c:1370)
      kernel_init (init/main.c:1504)
      ret_from_fork (arch/x86/entry/entry_64.S:304)

It is easy to reproduce it on my side:

  - boot the VM with a debug kernel config [1]
  - wait ~1 minute
  - start a kmemleak scan

It seems kmemleak has an issue with the array allocated in
msr_build_context() and assigned to a pointer in a static structure
(saved_context.saved_msrs->array): there is no leak then.

It looks like this is a limitation from kmemleak but that's alright,
kmemleak_no_leak() can be used to avoid complaining about that.

Please note that it looks like this issue is not new, e.g.

  https://lore.kernel.org/all/[email protected]/
  https://lore.kernel.org/all/[email protected]/

But on my side, msr_build_context() is only used since:

  commit e2a1256 ("x86/speculation: Restore speculation related MSRs during S3 resume").

Depending on their CPUs, others have probably the same issue since:

  commit 7a9c2dd ("x86/pm: Introduce quirk framework to save/restore extra MSR registers around suspend/resume"),

hence the 'Fixes' tag here below to help with the backports. But I
understand if someone says the origin of this issue is more on
kmemleak's side. What is unclear to me is why this issue was not seen by
other people and CIs. Maybe the kernel config [1]?

[1] https://github.com/multipath-tcp/mptcp_net-next/files/8531660/kmemleak-cpu-sched-bisect.kconfig.txt

Fixes: 7a9c2dd ("x86/pm: Introduce quirk framework to save/restore extra MSR registers around suspend/resume")
Closes: #268
Signed-off-by: Matthieu Baerts <[email protected]>
  • Loading branch information
matttbe committed Apr 21, 2022
1 parent f1f573e commit 15a5ffa
Showing 1 changed file with 4 additions and 0 deletions.
4 changes: 4 additions & 0 deletions arch/x86/power/cpu.c
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,7 @@
#include <linux/tboot.h>
#include <linux/dmi.h>
#include <linux/pgtable.h>
#include <linux/kmemleak.h>

#include <asm/proto.h>
#include <asm/mtrr.h>
Expand Down Expand Up @@ -413,6 +414,9 @@ static int msr_build_context(const u32 *msr_id, const int num)
return -ENOMEM;
}

/* The pointer is going to be stored in static struct (saved_context) */
kmemleak_not_leak(msr_array);

if (saved_msrs->array) {
/*
* Multiple callbacks can invoke this function, so copy any
Expand Down

0 comments on commit 15a5ffa

Please sign in to comment.