Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 7fb06fc9 authored by Hidetoshi Seto's avatar Hidetoshi Seto Committed by H. Peter Anvin
Browse files

x86, mce: cleanup mce_start() 



Simplify interface of mce_start():

-       no_way_out = mce_start(no_way_out, &order);
+       order = mce_start(&no_way_out);

Now Monarch and Subjects share same exit(return) in usual path.

Signed-off-by: default avatarHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: default avatarH. Peter Anvin <hpa@zytor.com>
parent 33edbf02

arch/x86/kernel/cpu/mcheck/mce.c

+31 −35
Original line number Diff line number Diff line
@@ -691,23 +691,21 @@ static atomic_t global_nwo; 
 * in the entry order.

 * TBD double check parallel CPU hotunplug

 */

static int mce_start(int no_way_out, int *order)

static int mce_start(int *no_way_out)

{

	int nwo;

	int order;

	int cpus = num_online_cpus();

	u64 timeout = (u64)monarch_timeout * NSEC_PER_USEC;



	if (!timeout) {

		*order = -1;

		return no_way_out;

	}

	if (!timeout)

		return -1;



	atomic_add(no_way_out, &global_nwo);

	atomic_add(*no_way_out, &global_nwo);

	/*

	 * global_nwo should be updated before mce_callin

	 */

	smp_wmb();

	*order = atomic_add_return(1, &mce_callin);

	order = atomic_add_return(1, &mce_callin);



	/*

	 * Wait for everyone.
@@ -715,8 +713,7 @@ static int mce_start(int no_way_out, int *order) 
	while (atomic_read(&mce_callin) != cpus) {

		if (mce_timed_out(&timeout)) {

			atomic_set(&global_nwo, 0);

			*order = -1;

			return no_way_out;

			return -1;

		}

		ndelay(SPINUNIT);

	}
@@ -725,34 +722,34 @@ static int mce_start(int no_way_out, int *order) 
	 * mce_callin should be read before global_nwo

	 */

	smp_rmb();

	/*

	 * Cache the global no_way_out state.

	 */

	nwo = atomic_read(&global_nwo);



	if (order == 1) {

		/*

	 * Monarch starts executing now, the others wait.

		 * Monarch: Starts executing now, the others wait.

		 */

	if (*order == 1) {

		atomic_set(&mce_executing, 1);

		return nwo;

	}



	} else {

		/*

	 * Now start the scanning loop one by one

	 * in the original callin order.

	 * This way when there are any shared banks it will

	 * be only seen by one CPU before cleared, avoiding duplicates.

		 * Subject: Now start the scanning loop one by one in

		 * the original callin order.

		 * This way when there are any shared banks it will be

		 * only seen by one CPU before cleared, avoiding duplicates.

		 */

	while (atomic_read(&mce_executing) < *order) {

		while (atomic_read(&mce_executing) < order) {

			if (mce_timed_out(&timeout)) {

				atomic_set(&global_nwo, 0);

			*order = -1;

			return no_way_out;

				return -1;

			}

			ndelay(SPINUNIT);

		}

	return nwo;

	}



	/*

	 * Cache the global no_way_out state.

	 */

	*no_way_out = atomic_read(&global_nwo);



	return order;

}



/*
@@ -871,8 +868,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) 
	 * Establish sequential order between the CPUs entering the machine

	 * check handler.

	 */

	int order = -1;



	int order;

	/*

	 * If no_way_out gets set, there is no safe way to recover from this

	 * MCE.  If tolerant is cranked up, we'll try anyway.
@@ -917,7 +913,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) 
	 * This way we don't report duplicated events on shared banks

	 * because the first one to see it will clear it.

	 */

	no_way_out = mce_start(no_way_out, &order);

	order = mce_start(&no_way_out);

	for (i = 0; i < banks; i++) {

		__clear_bit(i, toclear);

		if (!bank[i])