RE: [PATCH v2 1/2] scsi: pm80xx: mask and unmask upper interrupt vectors 32-63

[<Ajish.Koshy@microchip.com>]

Thank you Damien for your comments below. Will make changes in v3.

> On 4/5/22 18:28, Ajish Koshy wrote:
> > When upper inbound and outbound queues 32-63 are enabled, we see
> upper
> > vectors 32-63 in interrupt service routine. We need corresponding
> > registers to handle masking and unmasking of these upper interrupts.
> >
> > To achieve this, we use registers MSGU_ODMR_U(0x34) to mask and
> > MSGU_ODMR_CLR_U(0x3C) to unmask the interrupts. In these registers bit
> > 0-31 represents interrupt vectors 32-63.
> >
> > Signed-off-by: Ajish Koshy <Ajish.Koshy@microchip.com>
> > Signed-off-by: Viswas G <Viswas.G@microchip.com>
> > Fixes: 05c6c029a44d ("scsi: pm80xx: Increase number of supported
> > queues")
> > ---
> >   drivers/scsi/pm8001/pm80xx_hwi.c | 31 +++++++++++++++++++++++++----
> --
> >   1 file changed, 25 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/scsi/pm8001/pm80xx_hwi.c
> > b/drivers/scsi/pm8001/pm80xx_hwi.c
> > index 9bb31f66db85..3e6413e21bfe 100644
> > --- a/drivers/scsi/pm8001/pm80xx_hwi.c
> > +++ b/drivers/scsi/pm8001/pm80xx_hwi.c
> > @@ -1728,9 +1728,17 @@ pm80xx_chip_interrupt_enable(struct
> pm8001_hba_info *pm8001_ha, u8 vec)
> >   {
> >   #ifdef PM8001_USE_MSIX
> >       u32 mask;
> > -     mask = (u32)(1 << vec);
> >
> > -     pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_CLR, (u32)(mask &
> 0xFFFFFFFF));
> > +     if (vec < 32) {
> > +             mask = 1U << vec;
> 
> Nit: Drop this...
> 
> > +             /*vectors 0 - 31*/
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_CLR, mask);
> 
> ...and do:
> 
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_CLR, 1U << vec);

OK

> 
> > +     } else {
> > +             vec = vec - 32;
> > +             mask = 1U << vec;
> > +             /*vectors 32 - 63*/
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_CLR_U, mask);
> 
> And here:
> 
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_CLR_U,
>                             1U << (vec - 32));
> 
> Then you do not need the mask variable.

OK

> 
> > +     }
> >       return;
> >   #endif
> >       pm80xx_chip_intx_interrupt_enable(pm8001_ha);
> > @@ -1747,11 +1755,22 @@ pm80xx_chip_interrupt_disable(struct
> pm8001_hba_info *pm8001_ha, u8 vec)
> >   {
> >   #ifdef PM8001_USE_MSIX
> >       u32 mask;
> > -     if (vec == 0xFF)
> > +
> > +     if (vec == 0xFF) {
> 
> This is not symmetric with pm80xx_chip_interrupt_enable(). Does
> pm80xx_chip_interrupt_enable() need the same case too ?

I believe it's not needed right now. 

For interrupt disable I could to see the following entry point
PM8001_CHIP_DISP->interrupt_disable(pm8001_ha, 0xFF)
is used in pm8001_pci_remove, pm8001_pci_suspend,
pm8001_pci_resume

But same is not the case with interrupt enable function. I don't
see interrupt enable being called with 0xFF.

> 
> >               mask = 0xFFFFFFFF;
> > -     else
> > -             mask = (u32)(1 << vec);
> > -     pm8001_cw32(pm8001_ha, 0, MSGU_ODMR, (u32)(mask &
> 0xFFFFFFFF));
> > +             /* disable all vectors 0-31, 32-63*/
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR, mask);
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_U, mask);
> 
> Similar here, no need for the mask variable.

OK.

> 
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR, 0xFFFFFFFF);
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_U, 0xFFFFFFFF);
> 
> > +     } else if (vec < 32) {
> > +             mask = 1U << vec;
> > +             /*vectors 0 - 31*/
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR, mask);
> 
> And here.
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR, 1U << vec);

OK

> 
> > +     } else {
> > +             vec = vec - 32;
> > +             mask = 1U << vec;
> > +             /*vectors 32 - 63*/
> > +             pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_U, mask);
> 
> And here.
>                 pm8001_cw32(pm8001_ha, 0, MSGU_ODMR_U,
>                             1U << (vec - 32));

OK
> 
> > +     }
> >       return;
> >   #endif
> >       pm80xx_chip_intx_interrupt_disable(pm8001_ha);
> 
> 
> --
> Damien Le Moal
> Western Digital Research