x86/resctrl: Don't stop walking closids when a locksetup group is found
authorJames Morse <james.morse@arm.com>
Mon, 3 Jun 2019 17:25:31 +0000 (18:25 +0100)
committerThomas Gleixner <tglx@linutronix.de>
Wed, 12 Jun 2019 08:31:50 +0000 (10:31 +0200)
commit87d3aa28f345bea77c396855fa5d5fec4c24461f
tree05b3b4ee5e8c5f579d1ad0067d6306eb0d811c13
parentaab8445c4e1cceeb3f739352041ec1c2586bc923
x86/resctrl: Don't stop walking closids when a locksetup group is found

When a new control group is created __init_one_rdt_domain() walks all
the other closids to calculate the sets of used and unused bits.

If it discovers a pseudo_locksetup group, it breaks out of the loop.  This
means any later closid doesn't get its used bits added to used_b.  These
bits will then get set in unused_b, and added to the new control group's
configuration, even if they were marked as exclusive for a later closid.

When encountering a pseudo_locksetup group, we should continue. This is
because "a resource group enters 'pseudo-locked' mode after the schemata is
written while the resource group is in 'pseudo-locksetup' mode." When we
find a pseudo_locksetup group, its configuration is expected to be
overwritten, we can skip it.

Fixes: dfe9674b04ff6 ("x86/intel_rdt: Enable entering of pseudo-locksetup mode")
Signed-off-by: James Morse <james.morse@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Reinette Chatre <reinette.chatre@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: H Peter Avin <hpa@zytor.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20190603172531.178830-1-james.morse@arm.com
arch/x86/kernel/cpu/resctrl/rdtgroup.c