mbox series

[v5,0/4] arm64: Dereference CPU operations indirectly

Message ID 20200318230145.72097-1-gshan@redhat.com
Headers show
Series arm64: Dereference CPU operations indirectly | expand

Message

Gavin Shan March 18, 2020, 11:01 p.m. UTC
The subject perhaps can't precisely indicate what this series does, but
keep it same as before for consistency.

In current implementation, an array (@cpu_ops[NR_CPUS]) is maintained
to dereference the CPU operations. 2KB memory are consumed when NR_CPUS
is configured to 256. It's too much than what I expected. This series
reworks the implementation to dereference the CPU operations by using
two CPU operations pointer with assumption - all secondary CPUs should
have unified CPU operations. With this, 16-bytes memory will be used
for same purpose.

PATCH[1/4] isn't too much relevant, to declare ACPI parking protocol only
when CONFIG_ARM64_ACPI_PARKING_PROTOCOL has been enabled. PATCH[2/4]
renames cpu_read_ops() to init_cpu_ops(), which is obviously more precise
because it's initializing the CPU operations. PATCH[3/4] introduces
get_cpu_ops(), preparing for droping the array of CPU operations.
PATCH[4/4] removes the CPU operations deferencing array and replaces
it with two pointers with the assumption: all secondary CPUs should have
same enablement method.

Changelog
=========
v5:
   * Rebase to 5.6.rc6 and retest                        (Gavin Shan)
   * Improved commit log for PATCH[1]                    (Gavin Shan)
   * Add helper function __cpu_try_die()                 (Mark Rutland)
   * Two pointers to track the operations for boot CPU
     and the secondary CPUs separately                   (Mark Rutland)
   * Drop PATCH[5] because @cpu parameter is still
     needed by get_cpu_ops()                             (Gavin Shan)
v4:
   * Rebase to 5.6.rc3 and retest                        (Gavin Shan)
   * Improved commit log for PATCH[4/5] with link tag    (Lorenzo Pieralisi)
   * Using pointer instead of index to dereference the
     unified CPU operations                              (Lorenzo Pieralisi)
   * Merge logic of cpu_get_ops() to get_cpu_method()    (Gavin Shan)
v3:
   * Assume all CPUs have same enablement method. With this, the used
     memory is further squeezed from 64 bytes to 4 bytes (Lorenzo Pieralisi)
   * Add PATCH[5/5] to remove argument of get_cpu_ops()  (Gavin Shan)
v2:
   * Pack 4 CPUs' indexes into one byte. 64 bytes are consumed in order
     to get the CPU operations                            (Robin Murphy)
   * Use ARRAY_SIZE() to iterate @cpu_ops[]               (Robin Murphy)
   * Make index-0 valid                                   (Robin Murphy)

Gavin Shan (4):
  arm64: Declare ACPI parking protocol CPU operation if needed
  arm64: Rename cpu_read_ops() to init_cpu_ops()
  arm64: Introduce get_cpu_ops() helper function
  arm64: Remove CPU operations dereferencing array

 arch/arm64/include/asm/cpu_ops.h |  8 +--
 arch/arm64/kernel/cpu_ops.c      | 84 +++++++++++++++++---------------
 arch/arm64/kernel/cpuidle.c      |  9 ++--
 arch/arm64/kernel/setup.c        |  8 +--
 arch/arm64/kernel/smp.c          | 72 +++++++++++++++++----------
 5 files changed, 107 insertions(+), 74 deletions(-)

Comments

Catalin Marinas March 24, 2020, 5:29 p.m. UTC | #1
On Thu, Mar 19, 2020 at 10:01:41AM +1100, Gavin Shan wrote:
> Gavin Shan (4):
>   arm64: Declare ACPI parking protocol CPU operation if needed
>   arm64: Rename cpu_read_ops() to init_cpu_ops()
>   arm64: Introduce get_cpu_ops() helper function
>   arm64: Remove CPU operations dereferencing array

I queued the first 3 patches for 5.7, they are useful on their own as a
code clean-up. I'll wait for the debate with Mark to settle on the last
patch.

Thanks.
Gavin Shan March 25, 2020, 11:49 a.m. UTC | #2
On 3/25/20 4:29 AM, Catalin Marinas wrote:
> On Thu, Mar 19, 2020 at 10:01:41AM +1100, Gavin Shan wrote:
>> Gavin Shan (4):
>>    arm64: Declare ACPI parking protocol CPU operation if needed
>>    arm64: Rename cpu_read_ops() to init_cpu_ops()
>>    arm64: Introduce get_cpu_ops() helper function
>>    arm64: Remove CPU operations dereferencing array
> 
> I queued the first 3 patches for 5.7, they are useful on their own as a
> code clean-up. I'll wait for the debate with Mark to settle on the last
> patch.
> 
> Thanks.
> 

Thanks, Catalin. I'm also waiting for further comments on the last patch
from Mark. The patch can also be dropped if your guys think it's not needed.
Anyway, please let me know :)

Thanks,
Gavin