Hi Martin, On Wed, Aug 9, 2023 at 6:56 AM Martin K. Petersen <martin.petersen@xxxxxxxxxx> wrote: > > > Hi Ranjan, > > > But for few registers zero may be a valid value and we don’t want > > those registers to get penalized with 30 read retries where 1/3 reads > > would have sufficed. > > If 0 is a valid register value you'll end up always doing 3 retries > before returning. Even if the first register reads were "successful". > Peculiar! >Yes, the registers that has “0” valid value will be read 3 times. Since “3” retries logic for all registers is already there since years, we don’t want to touch it to avoid any regression. Now when the retry count needs to be increased further to 30, we want to limit it to impacted registers only. Thanks, Ranjan > -- > Martin K. Petersen Oracle Linux Engineering
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature