From: Samiullah Khawaja <skhawaja@google.com>
To: Pranjal Shrivastava <praan@google.com>
Cc: David Woodhouse <dwmw2@infradead.org>,
Lu Baolu <baolu.lu@linux.intel.com>,
Joerg Roedel <joro@8bytes.org>, Will Deacon <will@kernel.org>,
Jason Gunthorpe <jgg@ziepe.ca>,
Robin Murphy <robin.murphy@arm.com>,
Kevin Tian <kevin.tian@intel.com>,
Alex Williamson <alex@shazbot.org>,
Shuah Khan <shuah@kernel.org>,
iommu@lists.linux.dev, linux-kernel@vger.kernel.org,
kvm@vger.kernel.org, Saeed Mahameed <saeedm@nvidia.com>,
Adithya Jayachandran <ajayachandra@nvidia.com>,
Parav Pandit <parav@nvidia.com>,
Leon Romanovsky <leonro@nvidia.com>,
William Tu <witu@nvidia.com>,
Pratyush Yadav <pratyush@kernel.org>,
Pasha Tatashin <pasha.tatashin@soleen.com>,
David Matlack <dmatlack@google.com>,
Andrew Morton <akpm@linux-foundation.org>,
Chris Li <chrisl@kernel.org>, Vipin Sharma <vipinsh@google.com>,
YiFei Zhu <zhuyifei@google.com>
Subject: Re: [PATCH 09/14] iommu/vt-d: preserve PASID table of preserved device
Date: Mon, 23 Mar 2026 18:51:55 +0000 [thread overview]
Message-ID: <acGLfrpMz4JbxaqS@google.com> (raw)
In-Reply-To: <acGEJhAQKKknC3n8@google.com>
On Mon, Mar 23, 2026 at 06:19:18PM +0000, Pranjal Shrivastava wrote:
>On Tue, Feb 03, 2026 at 10:09:43PM +0000, Samiullah Khawaja wrote:
>> In scalable mode the PASID table is used to fetch the io page tables.
>> Preserve and restore the PASID table of the preserved devices.
>>
>> Signed-off-by: Samiullah Khawaja <skhawaja@google.com>
>> ---
>> drivers/iommu/intel/iommu.c | 4 +-
>> drivers/iommu/intel/iommu.h | 5 ++
>> drivers/iommu/intel/liveupdate.c | 130 +++++++++++++++++++++++++++++++
>> drivers/iommu/intel/pasid.c | 7 +-
>> drivers/iommu/intel/pasid.h | 9 +++
>> include/linux/kho/abi/iommu.h | 8 ++
>> 6 files changed, 160 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
>> index 83faad53f247..2d0dae57f5a2 100644
>> --- a/drivers/iommu/intel/iommu.c
>> +++ b/drivers/iommu/intel/iommu.c
>> @@ -2944,8 +2944,10 @@ static bool __maybe_clean_unpreserved_context_entries(struct intel_iommu *iommu)
>> if (info->iommu != iommu)
>> continue;
>>
>> - if (dev_iommu_preserved_state(&pdev->dev))
>> + if (dev_iommu_preserved_state(&pdev->dev)) {
>> + pasid_cleanup_preserved_table(&pdev->dev);
>> continue;
>> + }
>>
>> domain_context_clear(info);
>> }
>> diff --git a/drivers/iommu/intel/iommu.h b/drivers/iommu/intel/iommu.h
>> index 057bd6035d85..d24d6aeaacc0 100644
>> --- a/drivers/iommu/intel/iommu.h
>> +++ b/drivers/iommu/intel/iommu.h
>> @@ -1286,6 +1286,7 @@ int intel_iommu_preserve(struct iommu_device *iommu, struct iommu_ser *iommu_ser
>> void intel_iommu_unpreserve(struct iommu_device *iommu, struct iommu_ser *iommu_ser);
>> void intel_iommu_liveupdate_restore_root_table(struct intel_iommu *iommu,
>> struct iommu_ser *iommu_ser);
>> +void pasid_cleanup_preserved_table(struct device *dev);
>> #else
>> static inline int intel_iommu_preserve_device(struct device *dev, struct device_ser *device_ser)
>> {
>> @@ -1309,6 +1310,10 @@ static inline void intel_iommu_liveupdate_restore_root_table(struct intel_iommu
>> struct iommu_ser *iommu_ser)
>> {
>> }
>> +
>> +static inline void pasid_cleanup_preserved_table(struct device *dev)
>> +{
>> +}
>> #endif
>>
>> #ifdef CONFIG_INTEL_IOMMU_SVM
>> diff --git a/drivers/iommu/intel/liveupdate.c b/drivers/iommu/intel/liveupdate.c
>> index 6dcb5783d1db..53bb5fe3a764 100644
>> --- a/drivers/iommu/intel/liveupdate.c
>> +++ b/drivers/iommu/intel/liveupdate.c
>> @@ -14,6 +14,7 @@
>> #include <linux/pci.h>
>>
>> #include "iommu.h"
>> +#include "pasid.h"
>> #include "../iommu-pages.h"
>>
>> static void unpreserve_iommu_context(struct intel_iommu *iommu, int end)
>> @@ -113,9 +114,89 @@ void intel_iommu_liveupdate_restore_root_table(struct intel_iommu *iommu,
>> iommu->reg_phys, iommu_ser->intel.root_table);
>> }
>>
>> +enum pasid_lu_op {
>> + PASID_LU_OP_PRESERVE = 1,
>> + PASID_LU_OP_UNPRESERVE,
>> + PASID_LU_OP_RESTORE,
>> + PASID_LU_OP_FREE,
>> +};
>> +
>> +static int pasid_lu_do_op(void *table, enum pasid_lu_op op)
>> +{
>> + int ret = 0;
>> +
>> + switch (op) {
>> + case PASID_LU_OP_PRESERVE:
>> + ret = iommu_preserve_page(table);
>> + break;
>> + case PASID_LU_OP_UNPRESERVE:
>> + iommu_unpreserve_page(table);
>> + break;
>> + case PASID_LU_OP_RESTORE:
>> + iommu_restore_page(virt_to_phys(table));
>> + break;
>> + case PASID_LU_OP_FREE:
>> + iommu_free_pages(table);
>> + break;
>> + }
>> +
>> + return ret;
>> +}
>> +
>> +static int pasid_lu_handle_pd(struct pasid_dir_entry *dir, enum pasid_lu_op op)
>> +{
>> + struct pasid_entry *table;
>> + int ret;
>> +
>> + /* Only preserve first table for NO_PASID. */
>> + table = get_pasid_table_from_pde(&dir[0]);
>> + if (!table)
>> + return -EINVAL;
>> +
>> + ret = pasid_lu_do_op(table, op);
>> + if (ret)
>> + return ret;
>> +
>> + ret = pasid_lu_do_op(dir, op);
>> + if (ret)
>> + goto err;
>> +
>> + return 0;
>> +err:
>> + if (op == PASID_LU_OP_PRESERVE)
>> + pasid_lu_do_op(table, PASID_LU_OP_UNPRESERVE);
>> +
>> + return ret;
>> +}
>> +
>> +void pasid_cleanup_preserved_table(struct device *dev)
>> +{
>> + struct pasid_table *pasid_table;
>> + struct pasid_dir_entry *dir;
>> + struct pasid_entry *table;
>> +
>> + pasid_table = intel_pasid_get_table(dev);
>> + if (!pasid_table)
>> + return;
>> +
>> + dir = pasid_table->table;
>> + table = get_pasid_table_from_pde(&dir[0]);
>> + if (!table)
>> + return;
>> +
>> + /* Cleanup everything except the first entry. */
>> + memset(&table[1], 0, SZ_4K - sizeof(*table));
>> + memset(&dir[1], 0, SZ_4K - sizeof(struct pasid_dir_entry));
>
>(Not too familiar with Intel IOMMU / VT-d)
>We seem to hardcode SZ_4K when clearing the directory entries. But in
>intel_pasid_alloc_table(), the allocation size seems to depend on
>max_pasid which could be larger than one page (order > 0)?
>
>If the directory is multi-page, won't we leave the trailing pages
>and full of stale PDE pointers that the HW could still walk?
Agreed. It should cleanup based on max_pasid. Will update.
>
>> +
>> + clflush_cache_range(&table[0], SZ_4K);
>> + clflush_cache_range(&dir[0], SZ_4K);
>> +}
>> +
>
>[ ------ >8 ------ ]
>
>Thanks,
>Praan
next prev parent reply other threads:[~2026-03-23 18:52 UTC|newest]
Thread overview: 95+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-03 22:09 [PATCH 00/14] iommu: Add live update state preservation Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 01/14] iommu: Implement IOMMU LU FLB callbacks Samiullah Khawaja
2026-03-11 21:07 ` Pranjal Shrivastava
2026-03-12 16:43 ` Samiullah Khawaja
2026-03-12 23:43 ` Pranjal Shrivastava
2026-03-13 16:47 ` Samiullah Khawaja
2026-03-13 15:36 ` Pranjal Shrivastava
2026-03-13 16:58 ` Samiullah Khawaja
2026-03-16 22:54 ` Vipin Sharma
2026-03-17 1:06 ` Samiullah Khawaja
2026-03-23 23:27 ` Vipin Sharma
2026-02-03 22:09 ` [PATCH 02/14] iommu: Implement IOMMU core liveupdate skeleton Samiullah Khawaja
2026-03-12 23:10 ` Pranjal Shrivastava
2026-03-13 18:42 ` Samiullah Khawaja
2026-03-17 20:09 ` Pranjal Shrivastava
2026-03-17 20:13 ` Samiullah Khawaja
2026-03-17 20:23 ` Pranjal Shrivastava
2026-03-17 21:03 ` Vipin Sharma
2026-03-18 18:51 ` Pranjal Shrivastava
2026-03-18 17:49 ` Samiullah Khawaja
2026-03-17 19:58 ` Vipin Sharma
2026-03-17 20:33 ` Samiullah Khawaja
2026-03-24 19:06 ` Vipin Sharma
2026-03-24 19:45 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 03/14] liveupdate: luo_file: Add internal APIs for file preservation Samiullah Khawaja
2026-03-18 10:00 ` Pranjal Shrivastava
2026-03-18 16:54 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 04/14] iommu/pages: Add APIs to preserve/unpreserve/restore iommu pages Samiullah Khawaja
2026-03-03 16:42 ` Ankit Soni
2026-03-03 18:41 ` Samiullah Khawaja
2026-03-20 17:27 ` Pranjal Shrivastava
2026-03-20 18:12 ` Samiullah Khawaja
2026-03-17 20:59 ` Vipin Sharma
2026-03-20 9:28 ` Pranjal Shrivastava
2026-03-20 18:27 ` Samiullah Khawaja
2026-03-20 11:01 ` Pranjal Shrivastava
2026-03-20 18:56 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 05/14] iommupt: Implement preserve/unpreserve/restore callbacks Samiullah Khawaja
2026-03-20 21:57 ` Pranjal Shrivastava
2026-03-23 16:41 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 06/14] iommu/vt-d: Implement device and iommu preserve/unpreserve ops Samiullah Khawaja
2026-03-19 16:04 ` Vipin Sharma
2026-03-19 16:27 ` Samiullah Khawaja
2026-03-20 23:01 ` Pranjal Shrivastava
2026-03-21 13:27 ` Pranjal Shrivastava
2026-03-23 18:32 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 07/14] iommu/vt-d: Restore IOMMU state and reclaimed domain ids Samiullah Khawaja
2026-03-19 20:54 ` Vipin Sharma
2026-03-20 1:05 ` Samiullah Khawaja
2026-03-22 19:51 ` Pranjal Shrivastava
2026-03-23 19:33 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 08/14] iommu: Restore and reattach preserved domains to devices Samiullah Khawaja
2026-03-10 5:16 ` Ankit Soni
2026-03-10 21:47 ` Samiullah Khawaja
2026-03-22 21:59 ` Pranjal Shrivastava
2026-03-23 18:02 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 09/14] iommu/vt-d: preserve PASID table of preserved device Samiullah Khawaja
2026-03-23 18:19 ` Pranjal Shrivastava
2026-03-23 18:51 ` Samiullah Khawaja [this message]
2026-02-03 22:09 ` [PATCH 10/14] iommufd-lu: Implement ioctl to let userspace mark an HWPT to be preserved Samiullah Khawaja
2026-03-19 23:35 ` Vipin Sharma
2026-03-20 0:40 ` Samiullah Khawaja
2026-03-20 23:34 ` Vipin Sharma
2026-03-23 16:24 ` Samiullah Khawaja
2026-03-25 14:37 ` Pranjal Shrivastava
2026-03-25 17:31 ` Samiullah Khawaja
2026-03-25 18:55 ` Pranjal Shrivastava
2026-03-25 20:19 ` Samiullah Khawaja
2026-03-25 20:36 ` Pranjal Shrivastava
2026-03-25 20:46 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 11/14] iommufd-lu: Persist iommu hardware pagetables for live update Samiullah Khawaja
2026-02-25 23:47 ` Samiullah Khawaja
2026-03-03 5:56 ` Ankit Soni
2026-03-03 18:51 ` Samiullah Khawaja
2026-03-23 20:28 ` Vipin Sharma
2026-03-23 21:34 ` Samiullah Khawaja
2026-03-25 20:08 ` Pranjal Shrivastava
2026-03-25 20:32 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 12/14] iommufd: Add APIs to preserve/unpreserve a vfio cdev Samiullah Khawaja
2026-03-23 20:59 ` Vipin Sharma
2026-03-23 21:38 ` Samiullah Khawaja
2026-03-25 20:24 ` Pranjal Shrivastava
2026-03-25 20:41 ` Samiullah Khawaja
2026-03-25 21:23 ` Pranjal Shrivastava
2026-03-26 0:16 ` Samiullah Khawaja
2026-02-03 22:09 ` [PATCH 13/14] vfio/pci: Preserve the iommufd state of the " Samiullah Khawaja
2026-02-17 4:18 ` Ankit Soni
2026-03-03 18:35 ` Samiullah Khawaja
2026-03-23 21:17 ` Vipin Sharma
2026-03-23 22:07 ` Samiullah Khawaja
2026-03-24 20:30 ` Vipin Sharma
2026-03-25 20:55 ` Pranjal Shrivastava
2026-02-03 22:09 ` [PATCH 14/14] iommufd/selftest: Add test to verify iommufd preservation Samiullah Khawaja
2026-03-23 22:18 ` Vipin Sharma
2026-03-25 21:05 ` Pranjal Shrivastava
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=acGLfrpMz4JbxaqS@google.com \
--to=skhawaja@google.com \
--cc=ajayachandra@nvidia.com \
--cc=akpm@linux-foundation.org \
--cc=alex@shazbot.org \
--cc=baolu.lu@linux.intel.com \
--cc=chrisl@kernel.org \
--cc=dmatlack@google.com \
--cc=dwmw2@infradead.org \
--cc=iommu@lists.linux.dev \
--cc=jgg@ziepe.ca \
--cc=joro@8bytes.org \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=leonro@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=parav@nvidia.com \
--cc=pasha.tatashin@soleen.com \
--cc=praan@google.com \
--cc=pratyush@kernel.org \
--cc=robin.murphy@arm.com \
--cc=saeedm@nvidia.com \
--cc=shuah@kernel.org \
--cc=vipinsh@google.com \
--cc=will@kernel.org \
--cc=witu@nvidia.com \
--cc=zhuyifei@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox