Re: [UNTESTED PATCH] x86, mce: Avoid double entry of deferred errors into the genpool.

From: Borislav Petkov
Date: Sat Nov 21 2015 - 14:16:50 EST

On Wed, Nov 11, 2015 at 02:01:51PM -0800, Tony Luck wrote:
> We used to have a special ring buffer for deferred errors that
> was used to mark problem pages. We replaced that with a genpool.
> Then later converted mce_log() to also use the same genpool. As
> a result we end up adding all deferred errors to the genpool twice.
> Rearrange this code. Make sure to set the m.severity and m.usable_addr
> fields for deferred errors. Then if flags and mca_cfg.dont_log_ce mean
> we call mce_log() we are done, because that will add this entry to the
> genpool.
> If we skipped mce_log(), then we still want to take action for the
> deferred error, so add to the genpool.
> Changed the name of the boolean "error_logged" to "error_seen", we
> should set it whether of not we logged an error because the return
> value from machine_check_poll() is used to decide whether storms
> have subsided or not.
> Reported-by: Chen, Gong <>
> Signed-off-by: Tony Luck <tony.luck@xxxxxxxxx>
> ---
> arch/x86/kernel/cpu/mcheck/mce.c | 24 +++++++++++++-----------
> 1 file changed, 13 insertions(+), 11 deletions(-)


> @@ -626,9 +621,16 @@ bool machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
> * Don't get the IP here because it's unlikely to
> * have anything to do with the actual error location.
> */
> - if (!(flags & MCP_DONTLOG) && !mca_cfg.dont_log_ce) {
> - error_logged = true;
> + if (!(flags & MCP_DONTLOG) && !mca_cfg.dont_log_ce)
> mce_log(&m);
> + else if (m.usable_addr) {
> + /*
> + * Although we skipped logging this, we still want
> + * to take action. Add to the pool so the registered
> + * notifiers will see it.
> + */
> + if (!mce_gen_pool_add(&m))
> + mce_schedule_work();

Right, this still causes the error to come out on AMD because the
notifier calls amd_decode_mce().

I guess we can extend the "if (m.usable_addr)" check above with "if
error is not CE" too and only add it to the generic pool when its
severity is anything stronger than MCE_KEEP_SEVERITY...

Also, two more fixes I've done while injecting in a kvm guest I'm
sending as a reply to this message. Will inject on a real box too.



ECO tip #101: Trim your mails when you reply.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at