RE: [PATCH] scsi: pm8001: Fix phys_to_virt() usage on dma_addr_t

[Ajish.Koshy]

Hi John,

Was testing the patch on arm server. Didn't see crash there but observing timeouts and error 
handling getting triggered for drives. But the same code works fine on x86.

At your end do you still face similar situation on arm server ?

Thanks,
Ajish

-----Original Message-----
From: John Garry <john.garry@xxxxxxxxxx> 
Sent: Tuesday, December 7, 2021 07:24 PM
To: Ajish Koshy - I30923 <Ajish.Koshy@xxxxxxxxxxxxx>
Cc: jinpu.wang@xxxxxxxxxxxxxxx; jejb@xxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx; Viswas G - I30667 <Viswas.G@xxxxxxxxxxxxx>; linux-scsi@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; damien.lemoal@xxxxxxxxxxxxxxxxxx; Niklas.Cassel@xxxxxxx; Vasanthalakshmi Tharmarajan - I30664 <Vasanthalakshmi.Tharmarajan@xxxxxxxxxxxxx>
Subject: Re: [PATCH] scsi: pm8001: Fix phys_to_virt() usage on dma_addr_t

EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe

On 07/12/2021 10:36, Ajish.Koshy@xxxxxxxxxxxxx wrote:
> Well I could see this kernel panic [  126.843958] RIP: 0010:pm80xx_chip_smp_req+0x1d6/0x2e0 [pm80xx] with respect to existing driver on my current system x86 64bit after enabling the following 2 kernel boot arguments:
> -intel_iommu=on
> -iommu.passthrough=0
>
OK, so it seems that it was the kernel which was just not enabling the IOMMU previously, which would be consistent with what Niklas mentioned.

Anyway, please supply reviewed-by and/or tested-by tags so that the SCSI maintainers can pick it up.

I suppose that we should also have:

Fixes: f5860992db55 ("[SCSI] pm80xx: Added SPCv/ve specific hardware functionalities and relevant changes in common files")

Thanks for testing,
John