diff options
| author | Lu Baolu <baolu.lu@linux.intel.com> | 2021-08-18 21:48:49 +0800 |
|---|---|---|
| committer | Joerg Roedel <jroedel@suse.de> | 2021-08-19 10:41:08 +0200 |
| commit | 48811c44349ffbb778d3e36b53beb03ad43a979c (patch) | |
| tree | 81a5adae10070e61997237c315217b4253a1b4a4 /include/linux/intel-svm.h | |
| parent | 289b3b005cb9d9dd6b30297b52c2b4596bc878b2 (diff) | |
| download | lwn-48811c44349ffbb778d3e36b53beb03ad43a979c.tar.gz lwn-48811c44349ffbb778d3e36b53beb03ad43a979c.zip | |
iommu/vt-d: Allow devices to have more than 32 outstanding PRs
The minimum per-IOMMU PRQ queue size is one 4K page, this is more entries
than the hardcoded limit of 32 in the current VT-d code. Some devices can
support up to 512 outstanding PRQs but underutilized by this limit of 32.
Although, 32 gives some rough fairness when multiple devices share the same
IOMMU PRQ queue, but far from optimal for customized use case. This extends
the per-IOMMU PRQ queue size to four 4K pages and let the devices have as
many outstanding page requests as they can.
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20210720013856.4143880-1-baolu.lu@linux.intel.com
Link: https://lore.kernel.org/r/20210818134852.1847070-7-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Diffstat (limited to 'include/linux/intel-svm.h')
| -rw-r--r-- | include/linux/intel-svm.h | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/include/linux/intel-svm.h b/include/linux/intel-svm.h index 10fa80eef13a..57cceecbe37f 100644 --- a/include/linux/intel-svm.h +++ b/include/linux/intel-svm.h @@ -14,6 +14,11 @@ #define SVM_REQ_EXEC (1<<1) #define SVM_REQ_PRIV (1<<0) +/* Page Request Queue depth */ +#define PRQ_ORDER 2 +#define PRQ_RING_MASK ((0x1000 << PRQ_ORDER) - 0x20) +#define PRQ_DEPTH ((0x1000 << PRQ_ORDER) >> 5) + /* * The SVM_FLAG_SUPERVISOR_MODE flag requests a PASID which can be used only * for access to kernel addresses. No IOTLB flushes are automatically done |