ACK/Cmnt: [PATCHv2 0/8][Hirsute][generic & riscv] Fixup old unmatched patches, thus fix l2 cache

[Tim Gardner]

On 4/1/21 8:57 AM, Dimitri John Ledkov wrote:
> Hi,
> 
> On Thu, 1 Apr 2021 at 15:53, Tim Gardner <tim.gardner at canonical.com> wrote:
>>
>> Acked-by: Tim Gardner <tim.gardner at canonical.com>
>>
>> I don't think the net diff you've included represents all of the patch
>> changes. For instance, I don't see any instance of
>> SIFIVE_L2_DIRECCFAIL_LOW in the net diff, yet that macro is created in
>> '[PATCH 5/8] RISC-V: sifive_l2_cache: Update L2 cache driver to support
>> SiFive FU740' and never removed. Shouldn't that change have shown up in
>> the net diff ?
>>
> 
> It is already in the linux-riscv tree, and remains unchanged.
> 
> Patch 0001-Revert-RISC-V-sifive_l2_cache-Update-L2-cache-driver.patch
> removes it.
> Patch 0005-RISC-V-sifive_l2_cache-Update-L2-cache-driver-to-sup.patch
> adds it back in.
> 
> Hence end state of this patch series, with the tagged release does not
> mention SIFIVE_L2_DIRECCFAIL_LOW as it remains unchanged.
> 
> I was told that submitting revert + fresh version of the patch is the
> way to rebase patches onto newer editions of them.
> 
> Thus if one compares the tagged riscv release versus
> reverts+recherry-pick, this is the correct net result of changes, i.e.
> whitespace changes in docs; reorder the enum; fix up l2
> implementation.
> 

You're correct. I forgot about the reverts.

rtg
-----------
Tim Gardner
Canonical, Inc