[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <36bac0a9-89a6-ff7b-1870-a795a4f57700@redhat.com>
Date: Fri, 6 Jan 2017 12:45:54 +0100
From: Auger Eric <eric.auger@...hat.com>
To: Joerg Roedel <joro@...tes.org>
Cc: eric.auger.pro@...il.com, christoffer.dall@...aro.org,
marc.zyngier@....com, robin.murphy@....com,
alex.williamson@...hat.com, will.deacon@....com,
tglx@...utronix.de, jason@...edaemon.net,
linux-arm-kernel@...ts.infradead.org, kvm@...r.kernel.org,
drjones@...hat.com, linux-kernel@...r.kernel.org,
pranav.sawargaonkar@...il.com, iommu@...ts.linux-foundation.org,
punit.agrawal@....com, diana.craciun@....com, gpkulkarni@...il.com,
shankerd@...eaurora.org, bharat.bhushan@....com,
geethasowjanya.akula@...il.com
Subject: Re: [PATCH v6 08/18] iommu/vt-d: Implement reserved region get/put
callbacks
Hi Joerg,
On 06/01/2017 12:01, Joerg Roedel wrote:
> On Thu, Jan 05, 2017 at 07:04:36PM +0000, Eric Auger wrote:
>> +static void intel_iommu_get_resv_regions(struct device *device,
>> + struct list_head *head)
>> +{
>> + struct iommu_resv_region *reg;
>> +
>> + reg = iommu_alloc_resv_region(IOAPIC_RANGE_START,
>> + IOAPIC_RANGE_END - IOAPIC_RANGE_START + 1,
>> + 0, IOMMU_RESV_NOMAP);
>> + if (!reg)
>> + return;
>> + list_add_tail(®->list, head);
>> +}
>
> That is different from what AMD does, can you also report the RMRR
> regions for the device here (as direct-map regions)?
if I return RMRR regions as direct mapped regions,
iommu_group_create_direct_mappings will perform the 1-1 mapping.
I am not familiar with the intel-iommu code but I guess this job
currently is done in the intel driver:
iommu_prepare_rmrr_dev -> iommu_prepare_identity_map
->domain_prepare_identity_map -> iommu_domain_identity_map?
What is your feeling?
Thanks
Eric
>
>
>
> Joerg
>
Powered by blists - more mailing lists