From: Tobias Röhmel <quic_trohmel@quicinc.com>

Thanks again for all the help!

Here is v4:
2. Made patch cleaner
3. Changed commit message
4. Replaced V8_R flag with ARM_FEATURE_PMSA|ARM_FEATURE_V8
5.
Reworked the code to use existing pmsav7 variables
Added migration support
Added VSCTLR:
I didn't add any functionality for it because I think
Qemu doesn't model the behaviour it influences.
6.
Lots of cleanup. I think I overcomplicated this a bit because
of a misunderstanding. I thought HCR_VM is independent of enabling
the different MPUs, but I see now that it doesn't make sense to enable
HCR_VM when the MPUs are not enabled. I also think that there is an Error
in the armv8-r manual supplement in Figure C1-3. With all that figured out
the code for pmsav8r doesn't look that different from pmsav7 :)

Tobias Röhmel (7):
  target/arm: Don't add all MIDR aliases for cores that immplement PMSA
  target/arm: Make RVBAR available for all ARMv8 CPUs
  target/arm: Make stage_2_format for cache attributes optional
  target/arm: Enable TTBCR_EAE for ARMv8-R AArch32
  target/arm: Add PMSAv8r registers
  target/arm: Add PMSAv8r functionality
  target/arm: Add ARM Cortex-R52 CPU

 target/arm/cpu.c          |  32 +++-
 target/arm/cpu.h          |  12 ++
 target/arm/cpu_tcg.c      |  42 +++++
 target/arm/debug_helper.c |   3 +
 target/arm/helper.c       | 327 ++++++++++++++++++++++++++++++++++++--
 target/arm/internals.h    |   4 +
 target/arm/machine.c      |  28 ++++
 target/arm/ptw.c          | 148 ++++++++++++++---
 target/arm/tlb_helper.c   |   3 +
 9 files changed, 562 insertions(+), 37 deletions(-)

-- 
2.34.1

On Sun, 23 Oct 2022 at 16:37, <tobias.roehmel@rwth-aachen.de> wrote:
>
> From: Tobias Röhmel <quic_trohmel@quicinc.com>
>
> Thanks again for all the help!
>
> Here is v4:
> 2. Made patch cleaner
> 3. Changed commit message
> 4. Replaced V8_R flag with ARM_FEATURE_PMSA|ARM_FEATURE_V8
> 5.
> Reworked the code to use existing pmsav7 variables
> Added migration support
> Added VSCTLR:
> I didn't add any functionality for it because I think
> Qemu doesn't model the behaviour it influences.
> 6.
> Lots of cleanup. I think I overcomplicated this a bit because
> of a misunderstanding. I thought HCR_VM is independent of enabling
> the different MPUs, but I see now that it doesn't make sense to enable
> HCR_VM when the MPUs are not enabled. I also think that there is an Error
> in the armv8-r manual supplement in Figure C1-3. With all that figured out
> the code for pmsav8r doesn't look that different from pmsav7 :)

I've now reviewed this version of the patchset; it's getting
quite close now, I think. Sorry it took me so long to get back to it.

-- PMM