Hi Neil,
On 11/28/2023 5:59 PM, neil.armstrong@xxxxxxxxxx wrote:
Hi,
On 23/11/2023 09:46, Can Guo wrote:
On SM8550, two sets of UFS PHY settings are provided, one set is to
support
HS-G5, another set is to support HS-G4 and lower gears. The two sets
of PHY
settings are programming different values to different registers, mixing
the two sets and/or overwriting one set with another set is definitely
not
blessed by UFS PHY designers.
To add HS-G5 support for SM8550, split the two sets of PHY settings into
their dedicated overlay tables, only the common parts of the two sets of
PHY settings are left in the .tbls.
Consider we are going to add even higher gear support in future, to avoid
adding more tables with different names, rename the .tbls_hs_g4 and
make it
an array, a size of 2 is enough as of now.
In this case, .tbls alone is not a complete set of PHY settings, so
either
tbls_hs_overlay[0] or tbls_hs_overlay[1] must be applied on top of the
.tbls to become a complete set of PHY settings.
Signed-off-by: Can Guo <quic_cang@xxxxxxxxxxx>
Since Vinod has already merged SM8650 UFS PHY support, I'm afraid this
serie will break UFS if the SM8650 UFS tables aren't updated aswell.
Could you confirm if this will be the case ?
Because this change modifies data struct, it would be caught during
compiliation stage even if this change applies cleanly, hence it won't
break UFS.
Thanks,
Can Guo.
Thanks,
Neil
