On 10/31/2019 6:29 AM, Will Deacon wrote:
[+Jeffrey]
On Tue, Oct 29, 2019 at 04:27:38PM -0700, Bjorn Andersson wrote:
The Kryo cores share errata 1009 with Falkor, so add their model
definitions and enable it for them as well.
Signed-off-by: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx>
---
Changes since v1:
- Use is_kryo_midr(), rather than listing each individual model.
Cheers, I've queued this up as a fix.
I also updated the E1009 entry in silicon-errata.rst but, in doing so, I
noticed that E1041 is listed there which apparently also affects
Kry^H^H^HHydra [1].
At which point, maybe we should rename both Kryo and Falkor in the tree
so that we consistently refer to Hydra as the underlying micro-architecture.
Obviously not something for 5.4, but it would sure help me to understand
what's doing on here.
Thoughts?
Unfortunately, Falkor is also an underlying micro-architecture, it just
happens to be strongly related to Hydra so a fair amount of the errata
affect both.
I don't want to be difficult. For ultimate "correctness", Falkor and
Hydra should probably be separate, however the Falkor architecture is
not widespread and unlikely to have much churn going forward. So I
think if it makes life easier for you, all the Falkor stuff can probably
be scrubbed and just merged into Hydra.
Will
[1] https://lore.kernel.org/kvmarm/20171115010505.GO11955@xxxxxxxxxxxxxx/
--
Jeffrey Hugo
Qualcomm Technologies, Inc. is a member of the
Code Aurora Forum, a Linux Foundation Collaborative Project.
