ACPI: OSL: Prevent acpi_release_memory() from returning too early
PATCH 1/1 ACPI: OSL: Prevent acpi_release_memory() from returning too early
From: Rafael J. Wysocki <[email protected]>
After commit 1757659d022b ("ACPI: OSL: Implement deferred unmapping of ACPI memory") in some cases acpi_release_memory() may return before the target memory mappings actually go away, because they are released asynchronously now.
Prevent it from returning prematurely by making it wait for the next RCU grace period to elapse, for all of the RCU callbacks to complete and for all of the scheduled work items to be flushed before returning.
--- drivers/acpi/osl.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)
drivers/acpi/osl.c +17 -2
@@ -1575,11 +1575,26 @@ static acpi_status acpi_deactivate_mem_racpi_status acpi_release_memory (acpi_handle handle,struct resource *res,u32 level) { +acpi_status status; +if (!(res->flags &IORESOURCE_MEM ))return AE_TYPE ; -return acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, - acpi_deactivate_mem_region,NULL , res,NULL ); + status =acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, + acpi_deactivate_mem_region,NULL , + res,NULL ); +if (ACPI_FAILURE (status)) +return status; + +/* +* Wait for all of the mappings queued up for removal by +* acpi_deactivate_mem_region() to actually go away. +*/ +synchronize_rcu (); +rcu_barrier (); +flush_scheduled_work (); + +return AE_OK ; }EXPORT_SYMBOL_GPL (acpi_release_memory);
Works here for me:
----
$ l /sys/class/typec total 0 10012 0 drwxr-xr-x 2 root root 0 Aug 21 11:25 ./ 10 0 drwxr-xr-x 67 root root 0 Aug 21 11:25 ../ 27771 0 lrwxrwxrwx 1 root root 0 Aug 21 11:25 port0 -> ../../devices/platform/USBC000:00/typec/port0/ 35601 0 lrwxrwxrwx 1 root root 0 Aug 21 11:25 port0-partner -> ../../devices/platform/USBC000:00/typec/port0/port0-partner/ 36686 0 lrwxrwxrwx 1 root root 0 Aug 21 11:25 port1 -> ../../devices/platform/USBC000:00/typec/port1/ $ ----
Quoted 56 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After commit
1757659d022b("ACPI: OSL: Implement deferred unmapping of ACPI memory") in some casesacpi_release_memory()may return before the target memory mappings actually go away, because they are released asynchronously now.Prevent it from returning prematurely by making it wait for the next RCU grace period to elapse, for all of the RCU callbacks to complete and for all of the scheduled work items to be flushed before returning.
Fixes 1757659d022b ("ACPI: OSL: Implement deferred unmapping of ACPI memory")Reported-by Kenneth R. Crudup <[email protected]>Signed-off-by Rafael J. Wysocki <[email protected]>--- drivers/acpi/osl.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)Index:
linux-pm/drivers/acpi/osl.c
===================================================================drivers/acpi/osl.c +17
-2@@ -1575,11 +1575,26 @@ static acpi_status acpi_deactivate_mem_racpi_status acpi_release_memory (acpi_handle handle,struct resource *res,u32 level) { +acpi_status status; +if (!(res->flags &IORESOURCE_MEM ))return AE_TYPE ; -return acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, - acpi_deactivate_mem_region,NULL , res,NULL ); + status =acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, + acpi_deactivate_mem_region,NULL , + res,NULL ); +if (ACPI_FAILURE (status)) +return status; + +/* +* Wait for all of the mappings queued up for removal by +* acpi_deactivate_mem_region() to actually go away. +*/ +synchronize_rcu (); +rcu_barrier (); +flush_scheduled_work (); + +return AE_OK ; }EXPORT_SYMBOL_GPL (acpi_release_memory);
Signature · 1 line
Kenneth R. Crudup Sr. SW Engineer, Scott County Consulting, Orange County CA
Quoted 7 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After commit
1757659d022b("ACPI: OSL: Implement deferred unmapping of ACPI memory") in some casesacpi_release_memory()may return before the target memory mappings actually go away, because they are released asynchronously now.
Quoted 8 lines
Prevent it from returning prematurely by making it wait for the next RCU grace period to elapse, for all of the RCU callbacks to complete and for all of the scheduled work items to be flushed before returning.
Fixes 1757659d022b ("ACPI: OSL: Implement deferred unmapping of ACPI memory")Reported-by Kenneth R. Crudup <[email protected]>Signed-off-by Rafael J. Wysocki <[email protected]>
FWIW:
Quoted 26 lines
--- drivers/acpi/osl.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)Index:
linux-pm/drivers/acpi/osl.c
===================================================================drivers/acpi/osl.c +9
-2@@ -1575,11 +1575,26 @@ static acpi_status acpi_deactivate_mem_racpi_status acpi_release_memory (acpi_handle handle,struct resource *res,u32 level) { +acpi_status status; +if (!(res->flags &IORESOURCE_MEM ))return AE_TYPE ; -return acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, - acpi_deactivate_mem_region,NULL , res,NULL ); + status =acpi_walk_namespace (ACPI_TYPE_REGION , handle, level, + acpi_deactivate_mem_region,NULL , + res,NULL ); +if (ACPI_FAILURE (status)) +return status; + +/*
drivers/acpi/osl.c static acpi_status acpi_deactivate_mem_r · line 1599
Quoted 10 lines
drivers/acpi/osl.c +8
-0+* Wait for all of the mappings queued up for removal by +* acpi_deactivate_mem_region() to actually go away. +*/ +synchronize_rcu (); +rcu_barrier (); +flush_scheduled_work (); + +return AE_OK ; }EXPORT_SYMBOL_GPL (acpi_release_memory);
thanks,
Signature · 1 line
heikki
Quoted 7 lines · Rafael J. Wysocki
From: Rafael J. Wysocki <[email protected]>
After commit
1757659d022b("ACPI: OSL: Implement deferred unmapping of ACPI memory") in some casesacpi_release_memory()may return before the target memory mappings actually go away, because they are released asynchronously now.
Quoted 8 lines
Prevent it from returning prematurely by making it wait for the next RCU grace period to elapse, for all of the RCU callbacks to complete and for all of the scheduled work items to be flushed before returning.
Fixes 1757659d022b ("ACPI: OSL: Implement deferred unmapping of ACPI memory")Reported-by Kenneth R. Crudup <[email protected]>Signed-off-by Rafael J. Wysocki <[email protected]>