Re: [PATCH] hw/misc/max78000_gcr: Do not ignore address_space_write() errors

["Philippe Mathieu-Daudé" <philmd@linaro.org>]

On 7/10/25 15:48, Philippe Mathieu-Daudé wrote:
> On 7/10/25 10:27, Peter Maydell wrote:
>> On Tue, 7 Oct 2025 at 03:40, Philippe Mathieu-Daudé 
>> <philmd@linaro.org> wrote:
>>>
>>> Convert max78000_gcr_ops[] to take MemTxAttrs and return MemTxResult,
>>> allowing the write() path to return error on failure.
>>
>> *Should* it return a MEMTX error on this failure, though?
>> This is a question of what the hardware behaviour is,
>> and there's no reference to the datasheet in this
>> commit message...
> 
> Right. Thanks!

Looking at "MAX78000 User Guide (UG7456; Rev 1; 3/2024)",
chapter "4.7.2 RAM Zeroization" and table "4-67: Memory
Zeroize Control Register", IIUC failure can not happen.

Would that change be OK?

-      address_space_write(&s->sram_as, SYSRAM0_START,
-                          MEMTXATTRS_UNSPECIFIED, zero, 0x8000);
+      /* RAM Zeroization can not fail */
+      (void)address_space_write(&s->sram_as, SYSRAM0_START,
+                                MEMTXATTRS_UNSPECIFIED, zero, 0x8000);

Otherwise, what would be the recommended way to express return value
has to be ignored? Assertion doesn't seem right, since this is not a
programming error.

Thanks,

Phil.