Skip to contentSkip to conversation

paravirtual IOMMU interface

PATCH Superseded Paul Durrant 5 participants 47 messages Jul 7, 2018 – Jul 16, 2018

v2 of 14 Superseded Latest is v14 →

lore
Share & tools
lore
#0 lore raw

v2 cover paravirtual IOMMU interface

The idea of a paravirtual IOMMU interface was last discussed on xen-devel several years ago and narrowed down on a draft specification [1]. There was also an RFC patch series posted with an implementation, however this was never followed through.

In this patch series I have tried to simplify the interface and therefore have moved away from the draft specification. There is not yet any new specification but hopefully the interface in the introduced iommu_op header file will be understandable without such a specification.

For testing purposes, I have implemented patches to a Linux PV dom0 to set up a 1:1 BFN:GFN mapping for normal RAM at start of day and implement a new set of DMA ops based on direct map and standard swiotlb. 'Foreign' GFNs (i.e. grant maps) are then mapped and unmapped dynamically. See [2] for the code.

[1] https://lists.xenproject.org/archives/html/xen-devel/2016-02/msg01428.html
[2] http://xenbits.xen.org/gitweb/?p=people/pauldu/linux.git;a=shortlog;h=refs/heads/pv-iommu9

Paul Durrant (13):
grant_table: use term 'mfn' for machine frame numbers...
iommu: introduce the concept of BFN...
iommu: make use of type-safe BFN and MFN in exported functions
iommu: push use of type-safe BFN and MFN into iommu_ops
iommu: don't domain_crash() inside iommu_map/unmap_page() public / x86: introduce __HYPERCALL_iommu_op
iommu: track reserved ranges using a rangeset
x86: add iommu_op to query reserved ranges
vtd: add lookup_page method to iommu_ops
x86: add iommu_op to enable modification of IOMMU mappings
memory: add get_paged_gfn() as a wrapper...
x86: add iommu_ops to modify and flush IOMMU mappings
x86: extend the map and unmap iommu_ops to support grant references

 tools/flask/policy/modules/xen.if             |   1 +
 xen/arch/arm/p2m.c                            |   8 +-
 xen/arch/x86/Makefile                         |   1 +
 xen/arch/x86/hvm/emulate.c                    |  32 +-
 xen/arch/x86/hvm/hvm.c                        |  16 +-
 xen/arch/x86/hvm/hypercall.c                  |   1 +
 xen/arch/x86/hypercall.c                      |   1 +
 xen/arch/x86/iommu_op.c                       | 545 ++++++++++++++++++++++++++
 xen/arch/x86/mm.c                             |  14 +-
 xen/arch/x86/mm/p2m-ept.c                     |  18 +-
 xen/arch/x86/mm/p2m-pt.c                      |  52 ++-
 xen/arch/x86/mm/p2m.c                         |  42 +-
 xen/arch/x86/pv/hypercall.c                   |   1 +
 xen/arch/x86/x86_64/mm.c                      |   5 +-
 xen/common/grant_table.c                      | 335 +++++++++++-----
 xen/common/memory.c                           |  52 ++-
 xen/drivers/passthrough/amd/iommu_cmd.c       |  18 +-
 xen/drivers/passthrough/amd/iommu_map.c       |  86 ++--
 xen/drivers/passthrough/amd/pci_amd_iommu.c   |   4 +-
 xen/drivers/passthrough/arm/smmu.c            |  23 +-
 xen/drivers/passthrough/iommu.c               |  66 ++--
 xen/drivers/passthrough/vtd/iommu.c           |  95 +++--
 xen/drivers/passthrough/vtd/iommu.h           |   1 +
 xen/drivers/passthrough/vtd/x86/vtd.c         |  23 +-
 xen/drivers/passthrough/x86/iommu.c           |   7 +-
 xen/include/Makefile                          |   2 +
 xen/include/asm-arm/p2m.h                     |   3 +
 xen/include/asm-x86/hvm/svm/amd-iommu-proto.h |   8 +-
 xen/include/asm-x86/p2m.h                     |   2 +
 xen/include/public/iommu_op.h                 | 171 ++++++++
 xen/include/public/xen.h                      |   1 +
 xen/include/xen/grant_table.h                 |   7 +
 xen/include/xen/hypercall.h                   |  12 +
 xen/include/xen/iommu.h                       |  66 +++-
 xen/include/xen/mm.h                          |  10 +-
 xen/include/xlat.lst                          |   6 +
 xen/include/xsm/dummy.h                       |   6 +
 xen/include/xsm/xsm.h                         |   6 +
 xen/xsm/dummy.c                               |   1 +
 xen/xsm/flask/hooks.c                         |   6 +
 xen/xsm/flask/policy/access_vectors           |   2 +
 41 files changed, 1440 insertions(+), 316 deletions(-)
 create mode 100644 xen/arch/x86/iommu_op.c
 create mode 100644 xen/include/public/iommu_op.h

---
Cc: Andrew Cooper <[email protected]>
Cc: Daniel De Graaf <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Suravee Suthikulpanit <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#1 lore raw

PATCH v2 1/13 grant_table: use term 'mfn' for machine frame numbers...

...rather than more ambiguous term 'frame'.

There are many places in the grant table code that use a variable or field name '.*frame' to refer to a quantity that is strictly an MFN, and even has type mfn_t.
This patch is a purely cosmetic patch that substitutes 'frame' with 'mfn' in those places to make the purpose of the variable or field name more obvious.

---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- New in v2.

---
 xen/common/grant_table.c | 142 ++++++++++++++++++++++++-----------------------
 1 file changed, 72 insertions(+), 70 deletions(-)
xen/common/grant_table.c +72 -70
index 51e1f5ca4c..d2610e320c 100644
@@ -167,7 +167,7 @@ struct gnttab_unmap_common {
 
     /* Shared state beteen *_unmap and *_unmap_complete */
     uint16_t done;
-    mfn_t frame;
+    mfn_t mfn;
     struct domain *rd;
     grant_ref_t ref;
 };
@@ -266,7 +266,7 @@ struct active_grant_entry {
                                 grant.                                */
     grant_ref_t   trans_gref;
     struct domain *trans_domain;
-    mfn_t         frame;  /* Frame being granted.                     */
+    mfn_t         mfn;    /* Machine frame being granted.             */
 #ifndef NDEBUG
     gfn_t         gfn;    /* Guest's idea of the frame being granted. */
 #endif
@@ -370,15 +370,15 @@ static inline unsigned int grant_to_status_frames(unsigned int grant_frames)
 /* Check if the page has been paged out, or needs unsharing.
    If rc == GNTST_okay, *page contains the page struct with a ref taken.
    Caller must do put_page(*page).
-   If any error, *page = NULL, *frame = INVALID_MFN, no ref taken. */
-static int get_paged_frame(unsigned long gfn, mfn_t *frame,
+   If any error, *page = NULL, *mfn = INVALID_MFN, no ref taken. */
+static int get_paged_frame(unsigned long gfn, mfn_t *mfn,
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
     int rc = GNTST_okay;
     p2m_type_t p2mt;
 
-    *frame = INVALID_MFN;
+    *mfn = INVALID_MFN;
     *page = get_page_from_gfn(rd, gfn, &p2mt,
                               readonly ? P2M_ALLOC : P2M_UNSHARE);
     if ( !*page )
@@ -405,7 +405,7 @@ static int get_paged_frame(unsigned long gfn, mfn_t *frame,
         return GNTST_bad_page;
     }
 
-    *frame = page_to_mfn(*page);
+    *mfn = page_to_mfn(*page);
 
     return rc;
 }
@@ -843,7 +843,7 @@ static struct active_grant_entry *grant_map_exists(const struct domain *ld,
         struct active_grant_entry *act = active_entry_acquire(rgt, ref);
 
         if ( act->pin && act->domid == ld->domain_id &&
-             mfn_eq(act->frame, mfn) )
+             mfn_eq(act->mfn, mfn) )
             return act;
         active_entry_release(act);
     }
@@ -885,7 +885,7 @@ static unsigned int mapkind(
         if ( !(map->flags & (GNTMAP_device_map|GNTMAP_host_map)) ||
              map->domid != rd->domain_id )
             continue;
-        if ( mfn_eq(_active_entry(rd->grant_table, map->ref).frame, mfn) )
+        if ( mfn_eq(_active_entry(rd->grant_table, map->ref).mfn, mfn) )
             kind |= map->flags & GNTMAP_readonly ?
                     MAPKIND_READ : MAPKIND_WRITE;
     }
@@ -908,7 +908,7 @@ map_grant_ref(
     struct grant_table *lgt, *rgt;
     struct vcpu   *led;
     grant_handle_t handle;
-    mfn_t frame;
+    mfn_t mfn;
     struct page_info *pg = NULL;
     int            rc = GNTST_okay;
     u32            old_pin;
@@ -1001,13 +1001,13 @@ map_grant_ref(
                                 shared_entry_v1(rgt, op->ref).frame :
                                 shared_entry_v2(rgt, op->ref).full_page.frame;
 
-            rc = get_paged_frame(gfn, &frame, &pg,
+            rc = get_paged_frame(gfn, &mfn, &pg,
                                  op->flags & GNTMAP_readonly, rd);
             if ( rc != GNTST_okay )
                 goto unlock_out_clear;
             act_set_gfn(act, _gfn(gfn));
             act->domid = ld->domain_id;
-            act->frame = frame;
+            act->mfn = mfn;
             act->start = 0;
             act->length = PAGE_SIZE;
             act->is_sub_page = false;
@@ -1024,7 +1024,7 @@ map_grant_ref(
         act->pin += (op->flags & GNTMAP_readonly) ?
             GNTPIN_hstr_inc : GNTPIN_hstw_inc;
 
-    frame = act->frame;
+    mfn = act->mfn;
     act_pin = act->pin;
 
     cache_flags = (shah->flags & (GTF_PAT | GTF_PWT | GTF_PCD) );
@@ -1035,7 +1035,7 @@ map_grant_ref(
     /* pg may be set, with a refcount included, from get_paged_frame(). */
     if ( !pg )
     {
-        pg = mfn_valid(frame) ? mfn_to_page(frame) : NULL;
+        pg = mfn_valid(mfn) ? mfn_to_page(mfn) : NULL;
         if ( pg )
             owner = page_get_owner_and_reference(pg);
     }
@@ -1061,18 +1061,18 @@ map_grant_ref(
             goto undo_out;
         }
 
-        if ( !iomem_access_permitted(rd, mfn_x(frame), mfn_x(frame)) )
+        if ( !iomem_access_permitted(rd, mfn_x(mfn), mfn_x(mfn)) )
         {
             gdprintk(XENLOG_WARNING,
                      "Iomem mapping not permitted %#"PRI_mfn" (domain %d)\n",
-                     mfn_x(frame), rd->domain_id);
+                     mfn_x(mfn), rd->domain_id);
             rc = GNTST_general_error;
             goto undo_out;
         }
 
         if ( op->flags & GNTMAP_host_map )
         {
-            rc = create_grant_host_mapping(op->host_addr, frame, op->flags,
+            rc = create_grant_host_mapping(op->host_addr, mfn, op->flags,
                                            cache_flags);
             if ( rc != GNTST_okay )
                 goto undo_out;
@@ -1112,7 +1112,7 @@ map_grant_ref(
                 typecnt++;
             }
 
-            rc = create_grant_host_mapping(op->host_addr, frame, op->flags, 0);
+            rc = create_grant_host_mapping(op->host_addr, mfn, op->flags, 0);
             if ( rc != GNTST_okay )
                 goto undo_out;
 
@@ -1124,7 +1124,7 @@ map_grant_ref(
     could_not_pin:
         if ( !rd->is_dying )
             gdprintk(XENLOG_WARNING, "Could not pin grant frame %#"PRI_mfn"\n",
-                     mfn_x(frame));
+                     mfn_x(mfn));
         rc = GNTST_general_error;
         goto undo_out;
     }
@@ -1139,18 +1139,18 @@ map_grant_ref(
 
         /* We're not translated, so we know that gmfns and mfns are
            the same things, so the IOMMU entry is always 1-to-1. */
-        kind = mapkind(lgt, rd, frame);
+        kind = mapkind(lgt, rd, mfn);
         if ( (act_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) &&
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(frame), mfn_x(frame),
+                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
                                      IOMMUF_readable|IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(frame), mfn_x(frame),
+                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
                                      IOMMUF_readable);
         }
         if ( err )
@@ -1180,7 +1180,7 @@ map_grant_ref(
     if ( need_iommu )
         double_gt_unlock(lgt, rgt);
 
-    op->dev_bus_addr = mfn_to_maddr(frame);
+    op->dev_bus_addr = mfn_to_maddr(mfn);
     op->handle       = handle;
     op->status       = GNTST_okay;
 
@@ -1190,7 +1190,7 @@ map_grant_ref(
  undo_out:
     if ( host_map_created )
     {
-        replace_grant_host_mapping(op->host_addr, frame, 0, op->flags);
+        replace_grant_host_mapping(op->host_addr, mfn, 0, op->flags);
         gnttab_flush_tlb(ld);
     }
 
@@ -1365,18 +1365,18 @@ unmap_common(
         goto act_release_out;
     }
 
-    op->frame = act->frame;
+    op->mfn = act->mfn;
 
     if ( op->dev_bus_addr &&
-         unlikely(op->dev_bus_addr != mfn_to_maddr(act->frame)) )
+         unlikely(op->dev_bus_addr != mfn_to_maddr(act->mfn)) )
         PIN_FAIL(act_release_out, GNTST_general_error,
                  "Bus address doesn't match gntref (%"PRIx64" != %"PRIpaddr")\n",
-                 op->dev_bus_addr, mfn_to_maddr(act->frame));
+                 op->dev_bus_addr, mfn_to_maddr(act->mfn));
 
     if ( op->host_addr && (flags & GNTMAP_host_map) )
     {
         if ( (rc = replace_grant_host_mapping(op->host_addr,
-                                              op->frame, op->new_addr,
+                                              op->mfn, op->new_addr,
                                               flags)) < 0 )
             goto act_release_out;
 
@@ -1411,12 +1411,12 @@ unmap_common(
 
         double_gt_lock(lgt, rgt);
 
-        kind = mapkind(lgt, rd, op->frame);
+        kind = mapkind(lgt, rd, op->mfn);
         if ( !kind )
-            err = iommu_unmap_page(ld, mfn_x(op->frame));
+            err = iommu_unmap_page(ld, mfn_x(op->mfn));
         else if ( !(kind & MAPKIND_WRITE) )
-            err = iommu_map_page(ld, mfn_x(op->frame),
-                                 mfn_x(op->frame), IOMMUF_readable);
+            err = iommu_map_page(ld, mfn_x(op->mfn),
+                                 mfn_x(op->mfn), IOMMUF_readable);
 
         double_gt_unlock(lgt, rgt);
 
@@ -1426,7 +1426,7 @@ unmap_common(
 
     /* If just unmapped a writable mapping, mark as dirtied */
     if ( rc == GNTST_okay && !(flags & GNTMAP_readonly) )
-         gnttab_mark_dirty(rd, op->frame);
+         gnttab_mark_dirty(rd, op->mfn);
 
     op->status = rc;
     rcu_unlock_domain(rd);
@@ -1463,11 +1463,11 @@ unmap_common_complete(struct gnttab_unmap_common *op)
     else
         status = &status_entry(rgt, op->ref);
 
-    pg = mfn_to_page(op->frame);
+    pg = mfn_to_page(op->mfn);
 
     if ( op->done & GNTMAP_device_map )
     {
-        if ( !is_iomem_page(act->frame) )
+        if ( !is_iomem_page(act->mfn) )
         {
             if ( op->done & GNTMAP_readonly )
                 put_page(pg);
@@ -1484,7 +1484,7 @@ unmap_common_complete(struct gnttab_unmap_common *op)
 
     if ( op->done & GNTMAP_host_map )
     {
-        if ( !is_iomem_page(op->frame) )
+        if ( !is_iomem_page(op->mfn) )
         {
             if ( gnttab_host_mapping_get_page_type(op->done & GNTMAP_readonly,
                                                    ld, rd) )
@@ -1525,7 +1525,7 @@ unmap_grant_ref(
     common->done = 0;
     common->new_addr = 0;
     common->rd = NULL;
-    common->frame = INVALID_MFN;
+    common->mfn = INVALID_MFN;
 
     unmap_common(common);
     op->status = common->status;
@@ -1591,7 +1591,7 @@ unmap_and_replace(
     common->done = 0;
     common->dev_bus_addr = 0;
     common->rd = NULL;
-    common->frame = INVALID_MFN;
+    common->mfn = INVALID_MFN;
 
     unmap_common(common);
     op->status = common->status;
@@ -2320,7 +2320,7 @@ release_grant_for_copy(
     struct grant_table *rgt = rd->grant_table;
     grant_entry_header_t *sha;
     struct active_grant_entry *act;
-    mfn_t r_frame;
+    mfn_t mfn;
     uint16_t *status;
     grant_ref_t trans_gref;
     struct domain *td;
@@ -2329,7 +2329,7 @@ release_grant_for_copy(
 
     act = active_entry_acquire(rgt, gref);
     sha = shared_entry_header(rgt, gref);
-    r_frame = act->frame;
+    mfn = act->mfn;
 
     if ( rgt->gt_version == 1 )
     {
@@ -2350,7 +2350,7 @@ release_grant_for_copy(
     }
     else
     {
-        gnttab_mark_dirty(rd, r_frame);
+        gnttab_mark_dirty(rd, mfn);
 
         act->pin -= GNTPIN_hstw_inc;
         if ( !(act->pin & (GNTPIN_devw_mask|GNTPIN_hstw_mask)) )
@@ -2390,15 +2390,17 @@ static void fixup_status_for_copy_pin(const struct active_grant_entry *act,
         gnttab_clear_flag(_GTF_reading, status);
 }
 
-/* Grab a frame number from a grant entry and update the flags and pin
-   count as appropriate. If rc == GNTST_okay, note that this *does*
-   take one ref count on the target page, stored in *page.
-   If there is any error, *page = NULL, no ref taken. */
+/*
+ * Grab a machine frame number from a grant entry and update the flags
+ * and pin count as appropriate. If rc == GNTST_okay, note that this *does*
+ * take one ref count on the target page, stored in *page.
+ * If there is any error, *page = NULL, no ref taken.
+ */
 static int
 acquire_grant_for_copy(
     struct domain *rd, grant_ref_t gref, domid_t ldom, bool readonly,
-    mfn_t *frame, struct page_info **page,
-    uint16_t *page_off, uint16_t *length, bool allow_transitive)
+    mfn_t *mfn, struct page_info **page, uint16_t *page_off,
+    uint16_t *length, bool allow_transitive)
 {
     struct grant_table *rgt = rd->grant_table;
     grant_entry_v2_t *sha2;
@@ -2409,7 +2411,7 @@ acquire_grant_for_copy(
     domid_t trans_domid;
     grant_ref_t trans_gref;
     struct domain *td;
-    mfn_t grant_frame;
+    mfn_t grant_mfn;
     uint16_t trans_page_off;
     uint16_t trans_length;
     bool is_sub_page;
@@ -2487,7 +2489,7 @@ acquire_grant_for_copy(
         grant_read_unlock(rgt);
 
         rc = acquire_grant_for_copy(td, trans_gref, rd->domain_id,
-                                    readonly, &grant_frame, page,
+                                    readonly, &grant_mfn, page,
                                     &trans_page_off, &trans_length,
                                     false);
 
@@ -2511,7 +2513,7 @@ acquire_grant_for_copy(
         if ( rgt->gt_version != 2 ||
              act->pin != old_pin ||
              (old_pin && (act->domid != ldom ||
-                          !mfn_eq(act->frame, grant_frame) ||
+                          !mfn_eq(act->mfn, grant_mfn) ||
                           act->start != trans_page_off ||
                           act->length != trans_length ||
                           act->trans_domain != td ||
@@ -2535,7 +2537,7 @@ acquire_grant_for_copy(
             act->length = trans_length;
             act->trans_domain = td;
             act->trans_gref = trans_gref;
-            act->frame = grant_frame;
+            act->mfn = grant_mfn;
             act_set_gfn(act, INVALID_GFN);
             /*
              * The actual remote remote grant may or may not be a sub-page,
@@ -2559,7 +2561,7 @@ acquire_grant_for_copy(
         {
             unsigned long gfn = shared_entry_v1(rgt, gref).frame;
 
-            rc = get_paged_frame(gfn, &grant_frame, page, readonly, rd);
+            rc = get_paged_frame(gfn, &grant_mfn, page, readonly, rd);
             if ( rc != GNTST_okay )
                 goto unlock_out_clear;
             act_set_gfn(act, _gfn(gfn));
@@ -2569,7 +2571,7 @@ acquire_grant_for_copy(
         }
         else if ( !(sha2->hdr.flags & GTF_sub_page) )
         {
-            rc = get_paged_frame(sha2->full_page.frame, &grant_frame, page,
+            rc = get_paged_frame(sha2->full_page.frame, &grant_mfn, page,
                                  readonly, rd);
             if ( rc != GNTST_okay )
                 goto unlock_out_clear;
@@ -2580,7 +2582,7 @@ acquire_grant_for_copy(
         }
         else
         {
-            rc = get_paged_frame(sha2->sub_page.frame, &grant_frame, page,
+            rc = get_paged_frame(sha2->sub_page.frame, &grant_mfn, page,
                                  readonly, rd);
             if ( rc != GNTST_okay )
                 goto unlock_out_clear;
@@ -2598,13 +2600,13 @@ acquire_grant_for_copy(
             act->length = trans_length;
             act->trans_domain = td;
             act->trans_gref = trans_gref;
-            act->frame = grant_frame;
+            act->mfn = grant_mfn;
         }
     }
     else
     {
-        ASSERT(mfn_valid(act->frame));
-        *page = mfn_to_page(act->frame);
+        ASSERT(mfn_valid(act->mfn));
+        *page = mfn_to_page(act->mfn);
         td = page_get_owner_and_reference(*page);
         /*
          * act->pin being non-zero should guarantee the page to have a
@@ -2628,7 +2630,7 @@ acquire_grant_for_copy(
 
     *page_off = act->start;
     *length = act->length;
-    *frame = act->frame;
+    *mfn = act->mfn;
 
     active_entry_release(act);
     grant_read_unlock(rgt);
@@ -2658,7 +2660,7 @@ struct gnttab_copy_buf {
 
     /* Mapped etc. */
     struct domain *domain;
-    mfn_t frame;
+    mfn_t mfn;
     struct page_info *page;
     void *virt;
     bool_t read_only;
@@ -2764,7 +2766,7 @@ static int gnttab_copy_claim_buf(const struct gnttab_copy *op,
         rc = acquire_grant_for_copy(buf->domain, ptr->u.ref,
                                     current->domain->domain_id,
                                     buf->read_only,
-                                    &buf->frame, &buf->page,
+                                    &buf->mfn, &buf->page,
                                     &buf->ptr.offset, &buf->len,
                                     opt_transitive_grants);
         if ( rc != GNTST_okay )
@@ -2774,7 +2776,7 @@ static int gnttab_copy_claim_buf(const struct gnttab_copy *op,
     }
     else
     {
-        rc = get_paged_frame(ptr->u.gmfn, &buf->frame, &buf->page,
+        rc = get_paged_frame(ptr->u.gmfn, &buf->mfn, &buf->page,
                              buf->read_only, buf->domain);
         if ( rc != GNTST_okay )
             PIN_FAIL(out, rc,
@@ -2792,14 +2794,14 @@ static int gnttab_copy_claim_buf(const struct gnttab_copy *op,
             if ( !buf->domain->is_dying )
                 gdprintk(XENLOG_WARNING,
                          "Could not get writable frame %#"PRI_mfn"\n",
-                         mfn_x(buf->frame));
+                         mfn_x(buf->mfn));
             rc = GNTST_general_error;
             goto out;
         }
         buf->have_type = 1;
     }
 
-    buf->virt = map_domain_page(buf->frame);
+    buf->virt = map_domain_page(buf->mfn);
     rc = GNTST_okay;
 
  out:
@@ -2843,7 +2845,7 @@ static int gnttab_copy_buf(const struct gnttab_copy *op,
 
     memcpy(dest->virt + op->dest.offset, src->virt + op->source.offset,
            op->len);
-    gnttab_mark_dirty(dest->domain, dest->frame);
+    gnttab_mark_dirty(dest->domain, dest->mfn);
     rc = GNTST_okay;
  out:
     return rc;
@@ -3661,7 +3663,7 @@ gnttab_release_mappings(
         else
             status = &status_entry(rgt, ref);
 
-        pg = mfn_to_page(act->frame);
+        pg = mfn_to_page(act->mfn);
 
         if ( map->flags & GNTMAP_readonly )
         {
@@ -3669,7 +3671,7 @@ gnttab_release_mappings(
             {
                 BUG_ON(!(act->pin & GNTPIN_devr_mask));
                 act->pin -= GNTPIN_devr_inc;
-                if ( !is_iomem_page(act->frame) )
+                if ( !is_iomem_page(act->mfn) )
                     put_page(pg);
             }
 
@@ -3678,7 +3680,7 @@ gnttab_release_mappings(
                 BUG_ON(!(act->pin & GNTPIN_hstr_mask));
                 act->pin -= GNTPIN_hstr_inc;
                 if ( gnttab_release_host_mappings(d) &&
-                     !is_iomem_page(act->frame) )
+                     !is_iomem_page(act->mfn) )
                     put_page(pg);
             }
         }
@@ -3688,7 +3690,7 @@ gnttab_release_mappings(
             {
                 BUG_ON(!(act->pin & GNTPIN_devw_mask));
                 act->pin -= GNTPIN_devw_inc;
-                if ( !is_iomem_page(act->frame) )
+                if ( !is_iomem_page(act->mfn) )
                     put_page_and_type(pg);
             }
 
@@ -3697,7 +3699,7 @@ gnttab_release_mappings(
                 BUG_ON(!(act->pin & GNTPIN_hstw_mask));
                 act->pin -= GNTPIN_hstw_inc;
                 if ( gnttab_release_host_mappings(d) &&
-                     !is_iomem_page(act->frame) )
+                     !is_iomem_page(act->mfn) )
                 {
                     if ( gnttab_host_mapping_get_page_type((map->flags &
                                                             GNTMAP_readonly),
@@ -3754,7 +3756,7 @@ void grant_table_warn_active_grants(struct domain *d)
 #ifndef NDEBUG
                    gfn_x(act->gfn),
 #endif
-                   mfn_x(act->frame));
+                   mfn_x(act->mfn));
         active_entry_release(act);
     }
 
@@ -3963,7 +3965,7 @@ static void gnttab_usage_print(struct domain *rd)
 
         /*      [0xXXX]  ddddd 0xXXXXX 0xXXXXXXXX      ddddd 0xXXXXXX 0xXX */
         printk("[0x%03x]  %5d 0x%"PRI_mfn" 0x%08x      %5d 0x%06"PRIx64" 0x%02x\n",
-               ref, act->domid, mfn_x(act->frame), act->pin,
+               ref, act->domid, mfn_x(act->mfn), act->pin,
                sha->domid, frame, status);
         active_entry_release(act);
     }
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Reviewed-by on v2 1/13
#2 lore raw

Re: [PATCH v2 01/13] grant_table: use term 'mfn' for machine frame numbers...

Quoted 10 lines · Paul Durrant

...rather than more ambiguous term 'frame'.

There are many places in the grant table code that use a variable or field name '.*frame' to refer to a quantity that is strictly an MFN, and even has type mfn_t.
This patch is a purely cosmetic patch that substitutes 'frame' with 'mfn' in those places to make the purpose of the variable or field name more obvious.

Looks good. One minor comment...

xen/common/grant_table.c release_grant_for_copy( · line 2326

Quoted 9 lines
xen/common/grant_table.c +1 -1
@@ -2320,7 +2320,7 @@ release_grant_for_copy(
     struct grant_table *rgt = rd->grant_table;
     grant_entry_header_t *sha;
     struct active_grant_entry *act;
-    mfn_t r_frame;
+    mfn_t mfn;
     uint16_t *status;
     grant_ref_t trans_gref;
     struct domain *td;

If you end up respinning this, you might consider making this 'rmfn' or somethign, just because everything in this function seems to have the 'r' prefix.

But either way:

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#3 lore raw

Re: [PATCH v2 01/13] grant_table: use term 'mfn' for machine frame numbers...

Quoted 27 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 14:20
To: Paul Durrant <[email protected]>; [email protected]
Cc: Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Jan Beulich <[email protected]>; Julien Grall <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH v2 01/13] grant_table: use term 'mfn' for machine frame numbers...

Quoted 10 lines · Paul Durrant

...rather than more ambiguous term 'frame'.

There are many places in the grant table code that use a variable or field name '.*frame' to refer to a quantity that is strictly an MFN, and even has type mfn_t.
This patch is a purely cosmetic patch that substitutes 'frame' with 'mfn' in those places to make the purpose of the variable or field name more obvious.

Looks good. One minor comment...

Quoted 14 lines

xen/common/grant_table.c release_grant_for_copy( · line 2326

Quoted 9 lines
xen/common/grant_table.c +1 -1
@@ -2320,7 +2320,7 @@ release_grant_for_copy(
     struct grant_table *rgt = rd->grant_table;
     grant_entry_header_t *sha;
     struct active_grant_entry *act;
-    mfn_t r_frame;
+    mfn_t mfn;
     uint16_t *status;
     grant_ref_t trans_gref;
     struct domain *td;

If you end up respinning this, you might consider making this 'rmfn' or somethign, just because everything in this function seems to have the 'r' prefix.

I looked again and TBH I'd prefer to just stick with mfn.

Quoted 3 lines

But either way:

Thanks,

Paul
_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#4 lore raw

PATCH v2 2/13 iommu: introduce the concept of BFN...

In reply to Paul Durrant

...meaning 'bus frame number' i.e. a frame number mapped in the IOMMU rather than the MMU.

This patch is a largely cosmetic change that substitutes the terms 'gfn' and 'gaddr' for 'bfn' and 'baddr' in all the places where the frame number or address relate to the IOMMU rather than the MMU.

The parts that are not purely cosmetic are:

- the introduction of a type-safe declaration of bfn_t and definition of INVALID_BFN to make the substitution of gfn_x(INVALID_GFN) mechanical.
- the introduction of __bfn_to_baddr and __baddr_to_bfn (and type-safe variants without the leading __) with some use of the former.

Subsequent patches will convert code to make use of type-safe BFNs.

---
Cc: Suravee Suthikulpanit <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Kevin Tian <[email protected]>

v2:
- Addressed comments from Jan.

---
 xen/drivers/passthrough/amd/iommu_cmd.c     | 18 +++----
 xen/drivers/passthrough/amd/iommu_map.c     | 76 ++++++++++++++---------------
 xen/drivers/passthrough/amd/pci_amd_iommu.c |  2 +-
 xen/drivers/passthrough/arm/smmu.c          | 19 +++++---
 xen/drivers/passthrough/iommu.c             | 29 +++++------
 xen/drivers/passthrough/vtd/iommu.c         | 30 ++++++------
 xen/include/xen/iommu.h                     | 38 ++++++++++++---
 xen/include/xen/mm.h                        | 10 +++-
 8 files changed, 131 insertions(+), 91 deletions(-)
xen/drivers/passthrough/amd/iommu_cmd.c +9 -9
index 08247fa354..f93becd6e1 100644
@@ -284,7 +284,7 @@ void invalidate_iommu_all(struct amd_iommu *iommu)
 }
 
 void amd_iommu_flush_iotlb(u8 devfn, const struct pci_dev *pdev,
-                           uint64_t gaddr, unsigned int order)
+                           baddr_t baddr, unsigned int order)
 {
     unsigned long flags;
     struct amd_iommu *iommu;
@@ -315,12 +315,12 @@ void amd_iommu_flush_iotlb(u8 devfn, const struct pci_dev *pdev,
 
     /* send INVALIDATE_IOTLB_PAGES command */
     spin_lock_irqsave(&iommu->lock, flags);
-    invalidate_iotlb_pages(iommu, maxpend, 0, queueid, gaddr, req_id, order);
+    invalidate_iotlb_pages(iommu, maxpend, 0, queueid, baddr, req_id, order);
     flush_command_buffer(iommu);
     spin_unlock_irqrestore(&iommu->lock, flags);
 }
 
-static void amd_iommu_flush_all_iotlbs(struct domain *d, uint64_t gaddr,
+static void amd_iommu_flush_all_iotlbs(struct domain *d, baddr_t baddr,
                                        unsigned int order)
 {
     struct pci_dev *pdev;
@@ -333,7 +333,7 @@ static void amd_iommu_flush_all_iotlbs(struct domain *d, uint64_t gaddr,
         u8 devfn = pdev->devfn;
 
         do {
-            amd_iommu_flush_iotlb(devfn, pdev, gaddr, order);
+            amd_iommu_flush_iotlb(devfn, pdev, baddr, order);
             devfn += pdev->phantom_stride;
         } while ( devfn != pdev->devfn &&
                   PCI_SLOT(devfn) == PCI_SLOT(pdev->devfn) );
@@ -342,7 +342,7 @@ static void amd_iommu_flush_all_iotlbs(struct domain *d, uint64_t gaddr,
 
 /* Flush iommu cache after p2m changes. */
 static void _amd_iommu_flush_pages(struct domain *d,
-                                   uint64_t gaddr, unsigned int order)
+                                   baddr_t baddr, unsigned int order)
 {
     unsigned long flags;
     struct amd_iommu *iommu;
@@ -352,13 +352,13 @@ static void _amd_iommu_flush_pages(struct domain *d,
     for_each_amd_iommu ( iommu )
     {
         spin_lock_irqsave(&iommu->lock, flags);
-        invalidate_iommu_pages(iommu, gaddr, dom_id, order);
+        invalidate_iommu_pages(iommu, baddr, dom_id, order);
         flush_command_buffer(iommu);
         spin_unlock_irqrestore(&iommu->lock, flags);
     }
 
     if ( ats_enabled )
-        amd_iommu_flush_all_iotlbs(d, gaddr, order);
+        amd_iommu_flush_all_iotlbs(d, baddr, order);
 }
 
 void amd_iommu_flush_all_pages(struct domain *d)
@@ -367,9 +367,9 @@ void amd_iommu_flush_all_pages(struct domain *d)
 }
 
 void amd_iommu_flush_pages(struct domain *d,
-                           unsigned long gfn, unsigned int order)
+                           unsigned long bfn, unsigned int order)
 {
-    _amd_iommu_flush_pages(d, (uint64_t) gfn << PAGE_SHIFT, order);
+    _amd_iommu_flush_pages(d, __bfn_to_baddr(bfn), order);
 }
 
 void amd_iommu_flush_device(struct amd_iommu *iommu, uint16_t bdf)
xen/drivers/passthrough/amd/iommu_map.c +38 -38

251 lines not on this page · raw message

xen/drivers/passthrough/amd/pci_amd_iommu.c +1 -1
index 12d2695b89..d608631e6e 100644
@@ -578,7 +578,7 @@ static void amd_dump_p2m_table_level(struct page_info* pg, int level,
                 maddr_to_page(next_table_maddr), next_level,
                 address, indent + 1);
         else
-            printk("%*sgfn: %08lx  mfn: %08lx\n",
+            printk("%*sbfn: %08lx  mfn: %08lx\n",
                    indent, "",
                    (unsigned long)PFN_DOWN(address),
                    (unsigned long)PFN_DOWN(next_table_maddr));
xen/drivers/passthrough/arm/smmu.c +13 -6
index 74c09b0991..f9f2e8dca4 100644
@@ -2551,7 +2551,7 @@ static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
 }
 
 static int __must_check arm_smmu_iotlb_flush(struct domain *d,
-                                             unsigned long gfn,
+                                             unsigned long bfn,
                                              unsigned int page_count)
 {
 	/* ARM SMMU v1 doesn't have flush by VMA and VMID */
@@ -2737,10 +2737,12 @@ static void arm_smmu_iommu_domain_teardown(struct domain *d)
 	xfree(xen_domain);
 }
 
-static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
+static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
 			unsigned long mfn, unsigned int flags)
 {
 	p2m_type_t t;
+	unsigned long frame = bfn;
+	unsigned long gfn;
 
 	/*
 	 * Grant mappings can be used for DMA requests. The dev_bus_addr
@@ -2748,10 +2750,11 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
 	 * protected by an IOMMU, Xen needs to add a 1:1 mapping in the domain
 	 * p2m to allow DMA request to work.
 	 * This is only valid when the domain is directed mapped. Hence this
-	 * function should only be used by gnttab code with gfn == mfn.
+	 * function should only be used by gnttab code with gfn == mfn == bfn.
 	 */
 	BUG_ON(!is_domain_direct_mapped(d));
-	BUG_ON(mfn != gfn);
+	BUG_ON(mfn != frame);
+	gfn = frame;
 
 	/* We only support readable and writable flags */
 	if (!(flags & (IOMMUF_readable | IOMMUF_writable)))
@@ -2766,8 +2769,12 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
 	return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
 }
 
-static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long gfn)
+static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long bfn)
 {
+	unsigned long frame = bfn;
+	unsigned long gfn = frame;
+	unsigned long mfn = frame;
+
 	/*
 	 * This function should only be used by gnttab code when the domain
 	 * is direct mapped
@@ -2775,7 +2782,7 @@ static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long gfn)
 	if ( !is_domain_direct_mapped(d) )
 		return -EINVAL;
 
-	return guest_physmap_remove_page(d, _gfn(gfn), _mfn(gfn), 0);
+	return guest_physmap_remove_page(d, _gfn(gfn), _mfn(mfn), 0);
 }
 
 static const struct iommu_ops arm_smmu_iommu_ops = {
xen/drivers/passthrough/iommu.c +15 -14
index 2c44fabf99..9a3bb6a43e 100644
@@ -185,7 +185,8 @@ void __hwdom_init iommu_hwdom_init(struct domain *d)
         page_list_for_each ( page, &d->page_list )
         {
             unsigned long mfn = mfn_x(page_to_mfn(page));
-            unsigned long gfn = mfn_to_gmfn(d, mfn);
+            unsigned long frame = mfn_to_gmfn(d, mfn);
+            unsigned long bfn = frame;
             unsigned int mapping = IOMMUF_readable;
             int ret;
 
@@ -194,7 +195,7 @@ void __hwdom_init iommu_hwdom_init(struct domain *d)
                   == PGT_writable_page) )
                 mapping |= IOMMUF_writable;
 
-            ret = hd->platform_ops->map_page(d, gfn, mfn, mapping);
+            ret = hd->platform_ops->map_page(d, bfn, mfn, mapping);
             if ( !rc )
                 rc = ret;
 
@@ -255,7 +256,7 @@ void iommu_domain_destroy(struct domain *d)
     arch_iommu_domain_destroy(d);
 }
 
-int iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn,
+int iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
                    unsigned int flags)
 {
     const struct domain_iommu *hd = dom_iommu(d);
@@ -264,13 +265,13 @@ int iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn,
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->map_page(d, gfn, mfn, flags);
+    rc = hd->platform_ops->map_page(d, bfn, mfn, flags);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU mapping gfn %#lx to mfn %#lx failed: %d\n",
-                   d->domain_id, gfn, mfn, rc);
+                   "d%d: IOMMU mapping bfn %#lx to mfn %#lx failed: %d\n",
+                   d->domain_id, bfn, mfn, rc);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
@@ -279,7 +280,7 @@ int iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn,
     return rc;
 }
 
-int iommu_unmap_page(struct domain *d, unsigned long gfn)
+int iommu_unmap_page(struct domain *d, unsigned long bfn)
 {
     const struct domain_iommu *hd = dom_iommu(d);
     int rc;
@@ -287,13 +288,13 @@ int iommu_unmap_page(struct domain *d, unsigned long gfn)
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->unmap_page(d, gfn);
+    rc = hd->platform_ops->unmap_page(d, bfn);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU unmapping gfn %#lx failed: %d\n",
-                   d->domain_id, gfn, rc);
+                   "d%d: IOMMU unmapping bfn %#lx failed: %d\n",
+                   d->domain_id, bfn, rc);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
@@ -319,7 +320,7 @@ static void iommu_free_pagetables(unsigned long unused)
                             cpumask_cycle(smp_processor_id(), &cpu_online_map));
 }
 
-int iommu_iotlb_flush(struct domain *d, unsigned long gfn,
+int iommu_iotlb_flush(struct domain *d, unsigned long bfn,
                       unsigned int page_count)
 {
     const struct domain_iommu *hd = dom_iommu(d);
@@ -328,13 +329,13 @@ int iommu_iotlb_flush(struct domain *d, unsigned long gfn,
     if ( !iommu_enabled || !hd->platform_ops || !hd->platform_ops->iotlb_flush )
         return 0;
 
-    rc = hd->platform_ops->iotlb_flush(d, gfn, page_count);
+    rc = hd->platform_ops->iotlb_flush(d, bfn, page_count);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU IOTLB flush failed: %d, gfn %#lx, page count %u\n",
-                   d->domain_id, rc, gfn, page_count);
+                   "d%d: IOMMU IOTLB flush failed: %d, bfn %#lx, page count %u\n",
+                   d->domain_id, rc, bfn, page_count);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
xen/drivers/passthrough/vtd/iommu.c +15 -15
index 1710256823..48f62e0e8d 100644
@@ -585,7 +585,7 @@ static int __must_check iommu_flush_all(void)
 }
 
 static int __must_check iommu_flush_iotlb(struct domain *d,
-                                          unsigned long gfn,
+                                          unsigned long bfn,
                                           bool_t dma_old_pte_present,
                                           unsigned int page_count)
 {
@@ -612,12 +612,12 @@ static int __must_check iommu_flush_iotlb(struct domain *d,
         if ( iommu_domid == -1 )
             continue;
 
-        if ( page_count != 1 || gfn == gfn_x(INVALID_GFN) )
+        if ( page_count != 1 || bfn == bfn_x(INVALID_BFN) )
             rc = iommu_flush_iotlb_dsi(iommu, iommu_domid,
                                        0, flush_dev_iotlb);
         else
             rc = iommu_flush_iotlb_psi(iommu, iommu_domid,
-                                       (paddr_t)gfn << PAGE_SHIFT_4K,
+                                       __bfn_to_baddr(bfn),
                                        PAGE_ORDER_4K,
                                        !dma_old_pte_present,
                                        flush_dev_iotlb);
@@ -633,15 +633,15 @@ static int __must_check iommu_flush_iotlb(struct domain *d,
 }
 
 static int __must_check iommu_flush_iotlb_pages(struct domain *d,
-                                                unsigned long gfn,
+                                                unsigned long bfn,
                                                 unsigned int page_count)
 {
-    return iommu_flush_iotlb(d, gfn, 1, page_count);
+    return iommu_flush_iotlb(d, bfn, 1, page_count);
 }
 
 static int __must_check iommu_flush_iotlb_all(struct domain *d)
 {
-    return iommu_flush_iotlb(d, gfn_x(INVALID_GFN), 0, 0);
+    return iommu_flush_iotlb(d, bfn_x(INVALID_BFN), 0, 0);
 }
 
 /* clear one page's page table */
@@ -1767,7 +1767,7 @@ static void iommu_domain_teardown(struct domain *d)
 }
 
 static int __must_check intel_iommu_map_page(struct domain *d,
-                                             unsigned long gfn,
+                                             unsigned long bfn,
                                              unsigned long mfn,
                                              unsigned int flags)
 {
@@ -1786,14 +1786,14 @@ static int __must_check intel_iommu_map_page(struct domain *d,
 
     spin_lock(&hd->arch.mapping_lock);
 
-    pg_maddr = addr_to_dma_page_maddr(d, (paddr_t)gfn << PAGE_SHIFT_4K, 1);
+    pg_maddr = addr_to_dma_page_maddr(d, __bfn_to_baddr(bfn), 1);
     if ( pg_maddr == 0 )
     {
         spin_unlock(&hd->arch.mapping_lock);
         return -ENOMEM;
     }
     page = (struct dma_pte *)map_vtd_domain_page(pg_maddr);
-    pte = page + (gfn & LEVEL_MASK);
+    pte = page + (bfn & LEVEL_MASK);
     old = *pte;
     dma_set_pte_addr(new, (paddr_t)mfn << PAGE_SHIFT_4K);
     dma_set_pte_prot(new,
@@ -1817,22 +1817,22 @@ static int __must_check intel_iommu_map_page(struct domain *d,
     unmap_vtd_domain_page(page);
 
     if ( !this_cpu(iommu_dont_flush_iotlb) )
-        rc = iommu_flush_iotlb(d, gfn, dma_pte_present(old), 1);
+        rc = iommu_flush_iotlb(d, bfn, dma_pte_present(old), 1);
 
     return rc;
 }
 
 static int __must_check intel_iommu_unmap_page(struct domain *d,
-                                               unsigned long gfn)
+                                               unsigned long bfn)
 {
     /* Do nothing if hardware domain and iommu supports pass thru. */
     if ( iommu_passthrough && is_hardware_domain(d) )
         return 0;
 
-    return dma_pte_clear_one(d, (paddr_t)gfn << PAGE_SHIFT_4K);
+    return dma_pte_clear_one(d, __bfn_to_baddr(bfn));
 }
 
-int iommu_pte_flush(struct domain *d, u64 gfn, u64 *pte,
+int iommu_pte_flush(struct domain *d, uint64_t bfn, uint64_t *pte,
                     int order, int present)
 {
     struct acpi_drhd_unit *drhd;
@@ -1856,7 +1856,7 @@ int iommu_pte_flush(struct domain *d, u64 gfn, u64 *pte,
             continue;
 
         rc = iommu_flush_iotlb_psi(iommu, iommu_domid,
-                                   (paddr_t)gfn << PAGE_SHIFT_4K,
+                                   __bfn_to_baddr(bfn),
                                    order, !present, flush_dev_iotlb);
         if ( rc > 0 )
         {
@@ -2626,7 +2626,7 @@ static void vtd_dump_p2m_table_level(paddr_t pt_maddr, int level, paddr_t gpa,
             vtd_dump_p2m_table_level(dma_pte_addr(*pte), next_level, 
                                      address, indent + 1);
         else
-            printk("%*sgfn: %08lx mfn: %08lx\n",
+            printk("%*sbfn: %08lx mfn: %08lx\n",
                    indent, "",
                    (unsigned long)(address >> PAGE_SHIFT_4K),
                    (unsigned long)(dma_pte_addr(*pte) >> PAGE_SHIFT_4K));
xen/include/xen/iommu.h +32 -6
index 6b42e3b876..14ada0c14e 100644
@@ -23,11 +23,37 @@
 #include <xen/page-defs.h>
 #include <xen/spinlock.h>
 #include <xen/pci.h>
+#include <xen/typesafe.h>
 #include <public/hvm/ioreq.h>
 #include <public/domctl.h>
 #include <asm/device.h>
 #include <asm/iommu.h>
 
+TYPE_SAFE(uint64_t, bfn);
+#define PRI_bfn     PRIx64
+#define INVALID_BFN _bfn(~0UL)
+
+#ifndef bfn_t
+#define bfn_t /* Grep fodder: bfn_t, _bfn() and bfn_x() are defined above */
+#define _bfn
+#define bfn_x
+#undef bfn_t
+#undef _bfn
+#undef bfn_x
+#endif
+
+#define IOMMU_PAGE_SHIFT 12
+#define IOMMU_PAGE_SIZE  (_AC(1,L) << IOMMU_PAGE_SHIFT)
+#define IOMMU_PAGE_MASK  (~(IOMMU_PAGE_SIZE - 1))
+
+typedef uint64_t baddr_t;
+
+#define __bfn_to_baddr(bfn) ((baddr_t)(bfn) << IOMMU_PAGE_SHIFT)
+#define __baddr_to_bfn(baddr) ((uint64_t)(baddr >> IOMMU_PAGE_SHIFT))
+
+#define bfn_to_baddr(bfn) __bfn_to_baddr(bfn_x(bfn))
+#define baddr_to_bfn(baddr) _bfn(__baddr_to_bfn(baddr))
+
 extern bool_t iommu_enable, iommu_enabled;
 extern bool_t force_iommu, iommu_dom0_strict, iommu_verbose;
 extern bool_t iommu_workaround_bios_bug, iommu_igfx, iommu_passthrough;
@@ -60,9 +86,9 @@ void iommu_teardown(struct domain *d);
 #define IOMMUF_readable  (1u<<_IOMMUF_readable)
 #define _IOMMUF_writable 1
 #define IOMMUF_writable  (1u<<_IOMMUF_writable)
-int __must_check iommu_map_page(struct domain *d, unsigned long gfn,
+int __must_check iommu_map_page(struct domain *d, unsigned long bfn,
                                 unsigned long mfn, unsigned int flags);
-int __must_check iommu_unmap_page(struct domain *d, unsigned long gfn);
+int __must_check iommu_unmap_page(struct domain *d, unsigned long bfn);
 
 enum iommu_feature
 {
@@ -152,9 +178,9 @@ struct iommu_ops {
 #endif /* HAS_PCI */
 
     void (*teardown)(struct domain *d);
-    int __must_check (*map_page)(struct domain *d, unsigned long gfn,
+    int __must_check (*map_page)(struct domain *d, unsigned long bfn,
                                  unsigned long mfn, unsigned int flags);
-    int __must_check (*unmap_page)(struct domain *d, unsigned long gfn);
+    int __must_check (*unmap_page)(struct domain *d, unsigned long bfn);
     void (*free_page_table)(struct page_info *);
 #ifdef CONFIG_X86
     void (*update_ire_from_apic)(unsigned int apic, unsigned int reg, unsigned int value);
@@ -165,7 +191,7 @@ struct iommu_ops {
     void (*resume)(void);
     void (*share_p2m)(struct domain *d);
     void (*crash_shutdown)(void);
-    int __must_check (*iotlb_flush)(struct domain *d, unsigned long gfn,
+    int __must_check (*iotlb_flush)(struct domain *d, unsigned long bfn,
                                     unsigned int page_count);
     int __must_check (*iotlb_flush_all)(struct domain *d);
     int (*get_reserved_device_memory)(iommu_grdm_t *, void *);
@@ -187,7 +213,7 @@ int iommu_do_pci_domctl(struct xen_domctl *, struct domain *d,
 int iommu_do_domctl(struct xen_domctl *, struct domain *d,
                     XEN_GUEST_HANDLE_PARAM(xen_domctl_t));
 
-int __must_check iommu_iotlb_flush(struct domain *d, unsigned long gfn,
+int __must_check iommu_iotlb_flush(struct domain *d, unsigned long bfn,
                                    unsigned int page_count);
 int __must_check iommu_iotlb_flush_all(struct domain *d);
 
xen/include/xen/mm.h +8 -2
index e928551c91..e0387ff175 100644
@@ -14,8 +14,9 @@
  * should be adhered to.
  *
  * mfn: Machine Frame Number
- *   The values Xen puts into its own pagetables.  This is the host physical
- *   memory address space with RAM, MMIO etc.
+ *   The values Xen puts into its own pagetables, 2nd stage pagetables (where
+ *   hardware assisted 2nd stage translation is used) or IOMMU page tables.
+ *   This is the host physical memory address space with RAM, MMIO etc.
  *
  * gfn: Guest Frame Number
  *   The values a guest puts in its own pagetables.  For an auto-translated
@@ -26,6 +27,11 @@
  *   A linear idea of a guest physical address space. For an auto-translated
  *   guest, pfn == gfn while for a non-translated guest, pfn != gfn.
  *
+ * bfn: Bus Frame Number (definitions in include/xen/iommu.h)
+ *   The linear frame numbers of IOMMU address space. All initiators for a
+ *   given domain share a single IOMMU address space and, by default, Xen will
+ *   ensure bfn == pfn.
+ *
  * WARNING: Some of these terms have changed over time while others have been
  * used inconsistently, meaning that a lot of existing code does not match the
  * definitions above.  New code should use these terms as described here, and
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#5 lore raw

Re: [PATCH v2 02/13] iommu: introduce the concept of BFN...

xen/drivers/passthrough/amd/iommu_map.c int amd_iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn, · line 660

Quoted 13 lines · Paul Durrant
xen/drivers/passthrough/amd/iommu_map.c +2 -2
@@ -651,34 +651,34 @@ int amd_iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn,
     if ( rc )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Root table alloc failed, gfn = %lx\n", gfn);
+        AMD_IOMMU_DEBUG("Root table alloc failed, bfn = %lx\n", bfn);
         domain_crash(d);
         return rc;
     }
 
     /* Since HVM domain is initialized with 2 level IO page table,
-     * we might need a deeper page table for lager gfn now */
+     * we might need a deeper page table for wider bfn now */

Oh, come on, no lager gfns? ;-)

xen/drivers/passthrough/arm/smmu.c static void arm_smmu_iommu_domain_teardown(struct domain *d) · line 2745

Quoted 11 lines
xen/drivers/passthrough/arm/smmu.c +3 -1
@@ -2737,10 +2737,12 @@ static void arm_smmu_iommu_domain_teardown(struct domain *d)
        xfree(xen_domain);
 }
 
-static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
+static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
                        unsigned long mfn, unsigned int flags)
 {
        p2m_type_t t;
+       unsigned long frame = bfn;
+       unsigned long gfn;

I don't understand what the poind of these temporary variables are. They don't seem to really add any information at this point.

Quoted 18 lines
xen/drivers/passthrough/arm/smmu.c +3 -2
        /*
         * Grant mappings can be used for DMA requests. The dev_bus_addr
@@ -2748,10 +2750,11 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
         * protected by an IOMMU, Xen needs to add a 1:1 mapping in the domain
         * p2m to allow DMA request to work.
         * This is only valid when the domain is directed mapped. Hence this
-        * function should only be used by gnttab code with gfn == mfn.
+        * function should only be used by gnttab code with gfn == mfn == bfn.
         */
        BUG_ON(!is_domain_direct_mapped(d));
-       BUG_ON(mfn != gfn);
+       BUG_ON(mfn != frame);
+       gfn = frame;
 
        /* We only support readable and writable flags */
        if (!(flags & (IOMMUF_readable | IOMMUF_writable)))

xen/drivers/passthrough/arm/smmu.c static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn, · line 2776

Quoted 9 lines
xen/drivers/passthrough/arm/smmu.c +4 -1
@@ -2766,8 +2769,12 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,
        return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
 }
 
-static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long gfn)
+static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long bfn)
 {
+       unsigned long frame = bfn;
+       unsigned long gfn = frame;
+       unsigned long mfn = frame;

Same here.

xen/drivers/passthrough/iommu.c void __hwdom_init iommu_hwdom_init(struct domain *d) · line 189

Quoted 11 lines
xen/drivers/passthrough/iommu.c +2 -1
index 2c44fabf99..9a3bb6a43e 100644
@@ -185,7 +185,8 @@ void __hwdom_init iommu_hwdom_init(struct domain *d)
         page_list_for_each ( page, &d->page_list )
         {
             unsigned long mfn = mfn_x(page_to_mfn(page));
-            unsigned long gfn = mfn_to_gmfn(d, mfn);
+            unsigned long frame = mfn_to_gmfn(d, mfn);
+            unsigned long bfn = frame;

And here.

Everything else looks good.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#6 lore raw

Re: [PATCH v2 02/13] iommu: introduce the concept of BFN...

Quoted 41 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of George Dunlap
Sent: 10 July 2018 14:47
To: Paul Durrant <[email protected]>
Cc: xen-devel <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Suravee Suthikulpanit <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH v2 02/13] iommu: introduce the concept of BFN...

Review int amd_iommu_map_page(struct domain *d,

Quoted 1 line · Paul Durrant
(quoted diff) +0 -0
@@ -651,34 +651,34 @@ int amd_iommu_map_page(struct domain *d,

unsigned long gfn, unsigned long mfn,

xen/drivers/passthrough/amd/iommu_map.c int amd_iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn, · line 660

Quoted 12 lines
xen/drivers/passthrough/amd/iommu_map.c +2 -2
     if ( rc )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Root table alloc failed, gfn = %lx\n", gfn);
+        AMD_IOMMU_DEBUG("Root table alloc failed, bfn = %lx\n", bfn);
         domain_crash(d);
         return rc;
     }
 
     /* Since HVM domain is initialized with 2 level IO page table,
-     * we might need a deeper page table for lager gfn now */
+     * we might need a deeper page table for wider bfn now */

Oh, come on, no lager gfns? ;-)

Review static void

Quoted 1 line
(quoted diff) +0 -0
@@ -2737,10 +2737,12 @@ static void

arm_smmu_iommu_domain_teardown(struct domain *d)

xen/drivers/passthrough/arm/smmu.c static void arm_smmu_iommu_domain_teardown(struct domain *d)

Quoted 4 lines
xen/drivers/passthrough/arm/smmu.c +0 -1
        xfree(xen_domain);
 }
 
-static int __must_check arm_smmu_map_page(struct domain *d,

unsigned long gfn,

Review static void · line 2740

Quoted 1 line
(quoted diff) +1 -0
+static int __must_check arm_smmu_map_page(struct domain *d,

unsigned long bfn,

Quoted 8 lines

xen/drivers/passthrough/arm/smmu.c static void arm_smmu_iommu_domain_teardown(struct domain *d) · line 2745

Quoted 5 lines
xen/drivers/passthrough/arm/smmu.c +2 -0
                        unsigned long mfn, unsigned int flags)
 {
        p2m_type_t t;
+       unsigned long frame = bfn;
+       unsigned long gfn;

I don't understand what the poind of these temporary variables are. They don't seem to really add any information at this point.

I added these at Jan's request. I can defer introducing them to the later patch introducing typesafe bfns and gfns if that's preferable. The eventual code will be the same.

Paul

Quoted 51 lines

xen/drivers/passthrough/arm/smmu.c static int __must_check

Quoted 5 lines
xen/drivers/passthrough/arm/smmu.c +0 -0
        /*
         * Grant mappings can be used for DMA requests. The dev_bus_addr
@@ -2748,10 +2750,11 @@ static int __must_check

arm_smmu_map_page(struct domain *d, unsigned long gfn,

Review static int __must_check · line 2750

Quoted 1 line
(quoted diff) +0 -0
         * protected by an IOMMU, Xen needs to add a 1:1 mapping in the

domain

xen/drivers/passthrough/arm/smmu.c static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,

Quoted 4 lines
xen/drivers/passthrough/arm/smmu.c +1 -1
         * p2m to allow DMA request to work.
         * This is only valid when the domain is directed mapped. Hence this
-        * function should only be used by gnttab code with gfn == mfn.
+        * function should only be used by gnttab code with gfn == mfn ==

bfn.

xen/drivers/passthrough/arm/smmu.c static int __must_check

Quoted 9 lines
xen/drivers/passthrough/arm/smmu.c +2 -1
         */
        BUG_ON(!is_domain_direct_mapped(d));
-       BUG_ON(mfn != gfn);
+       BUG_ON(mfn != frame);
+       gfn = frame;
 
        /* We only support readable and writable flags */
        if (!(flags & (IOMMUF_readable | IOMMUF_writable)))
@@ -2766,8 +2769,12 @@ static int __must_check

arm_smmu_map_page(struct domain *d, unsigned long gfn,

xen/drivers/passthrough/arm/smmu.c static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn,

Quoted 4 lines
xen/drivers/passthrough/arm/smmu.c +0 -1
        return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
 }
 
-static int __must_check arm_smmu_unmap_page(struct domain *d,

unsigned long gfn)

Review static int __must_check · line 2772

Quoted 1 line
(quoted diff) +1 -0
+static int __must_check arm_smmu_unmap_page(struct domain *d,

unsigned long bfn)

xen/drivers/passthrough/arm/smmu.c static int __must_check arm_smmu_map_page(struct domain *d, unsigned long gfn, · line 2776

Quoted 4 lines
xen/drivers/passthrough/arm/smmu.c +3 -0
 {
+       unsigned long frame = bfn;
+       unsigned long gfn = frame;
+       unsigned long mfn = frame;

Same here.

Review

Quoted 1 line
(quoted diff) +0 -0

b/xen/drivers/passthrough/iommu.c

xen/drivers/passthrough/iommu.c void __hwdom_init iommu_hwdom_init(struct

Quoted 4 lines

index 2c44fabf99..9a3bb6a43e 100644

xen/drivers/passthrough/iommu.c +0 -0
@@ -185,7 +185,8 @@ void __hwdom_init iommu_hwdom_init(struct

domain *d)

xen/drivers/passthrough/iommu.c void __hwdom_init iommu_hwdom_init(struct domain *d) · line 189

Quoted 6 lines
xen/drivers/passthrough/iommu.c +2 -1
         page_list_for_each ( page, &d->page_list )
         {
             unsigned long mfn = mfn_x(page_to_mfn(page));
-            unsigned long gfn = mfn_to_gmfn(d, mfn);
+            unsigned long frame = mfn_to_gmfn(d, mfn);
+            unsigned long bfn = frame;
Quoted 5 lines

And here.

Everything else looks good.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#7 lore raw

Re: [PATCH v2 02/13] iommu: introduce the concept of BFN...

Quoted 48 lines
Quoted 48 lines
Quoted 1 line · On 10.07.18 at 16:08, <[email protected]> wrote:

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of George Dunlap
Sent: 10 July 2018 14:47
To: Paul Durrant <[email protected]>
Cc: xen-devel <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; Suravee Suthikulpanit <[email protected]>; Julien Grall <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH v2 02/13] iommu: introduce the concept of BFN...

Review int amd_iommu_map_page(struct domain *d,

Quoted 1 line · Paul Durrant
(quoted diff) +0 -0
@@ -651,34 +651,34 @@ int amd_iommu_map_page(struct domain *d,

unsigned long gfn, unsigned long mfn,

xen/drivers/passthrough/amd/iommu_map.c int amd_iommu_map_page(struct domain *d, unsigned long gfn, unsigned long mfn, · line 660

Quoted 12 lines
xen/drivers/passthrough/amd/iommu_map.c +2 -2
     if ( rc )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Root table alloc failed, gfn = %lx\n", gfn);
+        AMD_IOMMU_DEBUG("Root table alloc failed, bfn = %lx\n", bfn);
         domain_crash(d);
         return rc;
     }
 
     /* Since HVM domain is initialized with 2 level IO page table,
-     * we might need a deeper page table for lager gfn now */
+     * we might need a deeper page table for wider bfn now */

Oh, come on, no lager gfns? ;-)

Review static void

Quoted 1 line
(quoted diff) +0 -0
@@ -2737,10 +2737,12 @@ static void

arm_smmu_iommu_domain_teardown(struct domain *d)

xen/drivers/passthrough/arm/smmu.c static void arm_smmu_iommu_domain_teardown(struct domain *d)

Quoted 4 lines
xen/drivers/passthrough/arm/smmu.c +0 -1
        xfree(xen_domain);
 }
 
-static int __must_check arm_smmu_map_page(struct domain *d,

unsigned long gfn,

Review static void · line 2740

Quoted 1 line
(quoted diff) +1 -0
+static int __must_check arm_smmu_map_page(struct domain *d,

unsigned long bfn,

Quoted 5 lines
xen/drivers/passthrough/arm/smmu.c +2 -0
                        unsigned long mfn, unsigned int flags)
 {
        p2m_type_t t;
+       unsigned long frame = bfn;
+       unsigned long gfn;
Quoted 6 lines
Quoted 2 lines

I don't understand what the poind of these temporary variables are. They don't seem to really add any information at this point.

I added these at Jan's request. I can defer introducing them to the later patch introducing typesafe bfns and gfns if that's preferable. The eventual code will be the same.

Hmm, I've just gone back to my replies on v1, and I can't find any request like that. Is this perhaps a transformation of me having said "I'd prefer if such could at least be accompanied by a comment clarifying why this mix of address spaces is correct in the specific case"?

Jan

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#8 lore raw

PATCH v2 3/13 iommu: make use of type-safe BFN and MFN in exported functions

In reply to Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A subsequent patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

---
 xen/arch/arm/p2m.c                    |  6 ++++-
 xen/arch/x86/mm.c                     | 11 ++++----
 xen/arch/x86/mm/p2m-ept.c             | 15 ++++++++---
 xen/arch/x86/mm/p2m-pt.c              | 49 +++++++++++++++++++++--------------
 xen/arch/x86/mm/p2m.c                 | 24 ++++++++++++-----
 xen/arch/x86/x86_64/mm.c              |  5 ++--
 xen/common/grant_table.c              | 16 +++++++-----
 xen/common/memory.c                   |  7 +++--
 xen/drivers/passthrough/iommu.c       | 25 +++++++++---------
 xen/drivers/passthrough/vtd/x86/vtd.c |  3 ++-
 xen/include/xen/iommu.h               | 14 +++++++---
 11 files changed, 110 insertions(+), 65 deletions(-)
xen/arch/arm/p2m.c +5 -1
index d43c3aa896..42c16da4dd 100644
@@ -951,7 +951,11 @@ static int __p2m_set_entry(struct p2m_domain *p2m,
 
     if ( need_iommu(p2m->domain) &&
          (lpae_valid(orig_pte) || lpae_valid(*entry)) )
-        rc = iommu_iotlb_flush(p2m->domain, gfn_x(sgfn), 1UL << page_order);
+    {
+        unsigned long frame = gfn_x(sgfn);
+
+        rc = iommu_iotlb_flush(p2m->domain, _bfn(frame), 1UL << page_order);
+    }
     else
         rc = 0;
 
xen/arch/x86/mm.c +6 -5
index 4629bcaa47..4e66c6742d 100644
@@ -2729,14 +2729,15 @@ static int _get_page_type(struct page_info *page, unsigned long type,
         struct domain *d = page_get_owner(page);
         if ( d && is_pv_domain(d) && unlikely(need_iommu(d)) )
         {
-            gfn_t gfn = _gfn(mfn_to_gmfn(d, mfn_x(page_to_mfn(page))));
+            mfn_t mfn = page_to_mfn(page);
+            unsigned long frame = mfn_to_gmfn(d, mfn_x(mfn));
 
             if ( (x & PGT_type_mask) == PGT_writable_page )
-                iommu_ret = iommu_unmap_page(d, gfn_x(gfn));
+                iommu_ret = iommu_unmap_page(d, _bfn(frame));
             else if ( type == PGT_writable_page )
-                iommu_ret = iommu_map_page(d, gfn_x(gfn),
-                                           mfn_x(page_to_mfn(page)),
-                                           IOMMUF_readable|IOMMUF_writable);
+                iommu_ret = iommu_map_page(d, _bfn(frame), mfn,
+                                           IOMMUF_readable |
+                                           IOMMUF_writable);
         }
     }
 
xen/arch/x86/mm/p2m-ept.c +11 -4
index 14b593923b..4951234197 100644
@@ -866,19 +866,26 @@ out:
     if ( rc == 0 && p2m_is_hostp2m(p2m) && need_iommu(d) &&
          need_modify_vtd_table )
     {
+        unsigned long frame = gfn;
+
         if ( iommu_hap_pt_share )
-            rc = iommu_pte_flush(d, gfn, &ept_entry->epte, order, vtd_pte_present);
+            rc = iommu_pte_flush(d, frame, &ept_entry->epte, order,
+                                 vtd_pte_present);
         else
         {
+            bfn_t bfn = _bfn(frame);
+
             if ( iommu_flags )
                 for ( i = 0; i < (1 << order); i++ )
                 {
-                    rc = iommu_map_page(d, gfn + i, mfn_x(mfn) + i, iommu_flags);
+                    rc = iommu_map_page(d, bfn_add(bfn, i),
+                                        mfn_add(mfn, i), iommu_flags);
                     if ( unlikely(rc) )
                     {
                         while ( i-- )
                             /* If statement to satisfy __must_check. */
-                            if ( iommu_unmap_page(p2m->domain, gfn + i) )
+                            if ( iommu_unmap_page(p2m->domain,
+                                                  bfn_add(bfn, i)) )
                                 continue;
 
                         break;
@@ -887,7 +894,7 @@ out:
             else
                 for ( i = 0; i < (1 << order); i++ )
                 {
-                    ret = iommu_unmap_page(d, gfn + i);
+                    ret = iommu_unmap_page(d, bfn_add(bfn, i));
                     if ( !rc )
                         rc = ret;
                 }
xen/arch/x86/mm/p2m-pt.c +29 -20

65 lines not on this page · raw message

xen/arch/x86/mm/p2m.c +18 -6
index c53cab44d9..ce12bcff42 100644
@@ -714,9 +714,12 @@ p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn,
 
         if ( need_iommu(p2m->domain) )
         {
+            unsigned long frame = mfn;
+            bfn_t bfn = _bfn(frame);
+
             for ( i = 0; i < (1 << page_order); i++ )
             {
-                int ret = iommu_unmap_page(p2m->domain, mfn + i);
+                int ret = iommu_unmap_page(p2m->domain, bfn_add(bfn, i));
 
                 if ( !rc )
                     rc = ret;
@@ -773,16 +776,18 @@ guest_physmap_add_entry(struct domain *d, gfn_t gfn, mfn_t mfn,
     {
         if ( need_iommu(d) && t == p2m_ram_rw )
         {
+            unsigned long frame = mfn_x(mfn);
+            bfn_t bfn = _bfn(frame);
+
             for ( i = 0; i < (1 << page_order); i++ )
             {
-                rc = iommu_map_page(d, mfn_x(mfn_add(mfn, i)),
-                                    mfn_x(mfn_add(mfn, i)),
+                rc = iommu_map_page(d, bfn_add(bfn, i), mfn_add(mfn, i),
                                     IOMMUF_readable|IOMMUF_writable);
                 if ( rc != 0 )
                 {
                     while ( i-- > 0 )
                         /* If statement to satisfy __must_check. */
-                        if ( iommu_unmap_page(d, mfn_x(mfn_add(mfn, i))) )
+                        if ( iommu_unmap_page(d, bfn_add(bfn, i)) )
                             continue;
 
                     return rc;
@@ -1155,9 +1160,13 @@ int set_identity_p2m_entry(struct domain *d, unsigned long gfn_l,
 
     if ( !paging_mode_translate(p2m->domain) )
     {
+        unsigned long frame = gfn_l;
+
         if ( !need_iommu(d) )
             return 0;
-        return iommu_map_page(d, gfn_l, gfn_l, IOMMUF_readable|IOMMUF_writable);
+
+        return iommu_map_page(d, _bfn(frame), _mfn(frame),
+                              IOMMUF_readable | IOMMUF_writable);
     }
 
     gfn_lock(p2m, gfn, 0);
@@ -1245,9 +1254,12 @@ int clear_identity_p2m_entry(struct domain *d, unsigned long gfn_l)
 
     if ( !paging_mode_translate(d) )
     {
+        unsigned long frame = gfn_l;
+
         if ( !need_iommu(d) )
             return 0;
-        return iommu_unmap_page(d, gfn_l);
+
+        return iommu_unmap_page(d, _bfn(frame));
     }
 
     gfn_lock(p2m, gfn, 0);
xen/arch/x86/x86_64/mm.c +3 -2
index cca4ae926e..cc58e4cef4 100644
@@ -1429,13 +1429,14 @@ int memory_add(unsigned long spfn, unsigned long epfn, unsigned int pxm)
     if ( iommu_enabled && !iommu_passthrough && !need_iommu(hardware_domain) )
     {
         for ( i = spfn; i < epfn; i++ )
-            if ( iommu_map_page(hardware_domain, i, i, IOMMUF_readable|IOMMUF_writable) )
+            if ( iommu_map_page(hardware_domain, _bfn(i), _mfn(i),
+                                IOMMUF_readable | IOMMUF_writable) )
                 break;
         if ( i != epfn )
         {
             while (i-- > old_max)
                 /* If statement to satisfy __must_check. */
-                if ( iommu_unmap_page(hardware_domain, i) )
+                if ( iommu_unmap_page(hardware_domain, _bfn(i)) )
                     continue;
 
             goto destroy_m2p;
xen/common/grant_table.c +9 -7
index d2610e320c..d0926d13e0 100644
@@ -1132,6 +1132,8 @@ map_grant_ref(
     need_iommu = gnttab_need_iommu_mapping(ld);
     if ( need_iommu )
     {
+        unsigned long frame = mfn_x(mfn);
+        bfn_t bfn = _bfn(frame);
         unsigned int kind;
         int err = 0;
 
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);
         }
         if ( err )
         {
@@ -1406,6 +1407,8 @@ unmap_common(
 
     if ( rc == GNTST_okay && gnttab_need_iommu_mapping(ld) )
     {
+        unsigned long frame = mfn_x(op->mfn);
+        bfn_t bfn = _bfn(frame);
         unsigned int kind;
         int err = 0;
 
@@ -1413,10 +1416,9 @@ unmap_common(
 
         kind = mapkind(lgt, rd, op->mfn);
         if ( !kind )
-            err = iommu_unmap_page(ld, mfn_x(op->mfn));
+            err = iommu_unmap_page(ld, bfn);
         else if ( !(kind & MAPKIND_WRITE) )
-            err = iommu_map_page(ld, mfn_x(op->mfn),
-                                 mfn_x(op->mfn), IOMMUF_readable);
+            err = iommu_map_page(ld, bfn, op->mfn, IOMMUF_readable);
 
         double_gt_unlock(lgt, rgt);
 
xen/common/memory.c +5 -2
index 75010b78a5..e42da275c3 100644
@@ -829,15 +829,18 @@ int xenmem_add_to_physmap(struct domain *d, struct xen_add_to_physmap *xatp,
 #ifdef CONFIG_HAS_PASSTHROUGH
     if ( need_iommu(d) )
     {
+        unsigned long frame;
         int ret;
 
         this_cpu(iommu_dont_flush_iotlb) = 0;
 
-        ret = iommu_iotlb_flush(d, xatp->idx - done, done);
+        frame = xatp->idx - done;
+        ret = iommu_iotlb_flush(d, _bfn(frame), done);
         if ( unlikely(ret) && rc >= 0 )
             rc = ret;
 
-        ret = iommu_iotlb_flush(d, xatp->gpfn - done, done);
+        frame = xatp->gpfn - done;
+        ret = iommu_iotlb_flush(d, _bfn(frame), done);
         if ( unlikely(ret) && rc >= 0 )
             rc = ret;
     }
xen/drivers/passthrough/iommu.c +12 -13
index 9a3bb6a43e..2aa0c126d1 100644
@@ -256,7 +256,7 @@ void iommu_domain_destroy(struct domain *d)
     arch_iommu_domain_destroy(d);
 }
 
-int iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
+int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
                    unsigned int flags)
 {
     const struct domain_iommu *hd = dom_iommu(d);
@@ -265,13 +265,13 @@ int iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->map_page(d, bfn, mfn, flags);
+    rc = hd->platform_ops->map_page(d, bfn_x(bfn), mfn_x(mfn), flags);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU mapping bfn %#lx to mfn %#lx failed: %d\n",
-                   d->domain_id, bfn, mfn, rc);
+                   "d%d: IOMMU mapping bfn %"PRI_bfn" to mfn %"PRI_mfn" failed: %d\n",
+                   d->domain_id, bfn_x(bfn), mfn_x(mfn), rc);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
@@ -280,7 +280,7 @@ int iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
     return rc;
 }
 
-int iommu_unmap_page(struct domain *d, unsigned long bfn)
+int iommu_unmap_page(struct domain *d, bfn_t bfn)
 {
     const struct domain_iommu *hd = dom_iommu(d);
     int rc;
@@ -288,13 +288,13 @@ int iommu_unmap_page(struct domain *d, unsigned long bfn)
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->unmap_page(d, bfn);
+    rc = hd->platform_ops->unmap_page(d, bfn_x(bfn));
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU unmapping bfn %#lx failed: %d\n",
-                   d->domain_id, bfn, rc);
+                   "d%d: IOMMU unmapping bfn %"PRI_bfn" failed: %d\n",
+                   d->domain_id, bfn_x(bfn), rc);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
@@ -320,8 +320,7 @@ static void iommu_free_pagetables(unsigned long unused)
                             cpumask_cycle(smp_processor_id(), &cpu_online_map));
 }
 
-int iommu_iotlb_flush(struct domain *d, unsigned long bfn,
-                      unsigned int page_count)
+int iommu_iotlb_flush(struct domain *d, bfn_t bfn, unsigned int page_count)
 {
     const struct domain_iommu *hd = dom_iommu(d);
     int rc;
@@ -329,13 +328,13 @@ int iommu_iotlb_flush(struct domain *d, unsigned long bfn,
     if ( !iommu_enabled || !hd->platform_ops || !hd->platform_ops->iotlb_flush )
         return 0;
 
-    rc = hd->platform_ops->iotlb_flush(d, bfn, page_count);
+    rc = hd->platform_ops->iotlb_flush(d, bfn_x(bfn), page_count);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
             printk(XENLOG_ERR
-                   "d%d: IOMMU IOTLB flush failed: %d, bfn %#lx, page count %u\n",
-                   d->domain_id, rc, bfn, page_count);
+                   "d%d: IOMMU IOTLB flush failed: %d, bfn %"PRI_bfn", page count %u\n",
+                   d->domain_id, rc, bfn_x(bfn), page_count);
 
         if ( !is_hardware_domain(d) )
             domain_crash(d);
xen/drivers/passthrough/vtd/x86/vtd.c +2 -1
index cc2bfea162..dc37dce4b6 100644
@@ -155,7 +155,8 @@ void __hwdom_init vtd_set_hwdom_mapping(struct domain *d)
         tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
         for ( j = 0; j < tmp; j++ )
         {
-            int ret = iommu_map_page(d, pfn * tmp + j, pfn * tmp + j,
+            int ret = iommu_map_page(d, _bfn(pfn * tmp + j),
+                                     _mfn(pfn * tmp + j),
                                      IOMMUF_readable|IOMMUF_writable);
 
             if ( !rc )
xen/include/xen/iommu.h +10 -4
index 14ada0c14e..a3c36c1148 100644
@@ -24,6 +24,7 @@
 #include <xen/spinlock.h>
 #include <xen/pci.h>
 #include <xen/typesafe.h>
+#include <xen/mm.h>
 #include <public/hvm/ioreq.h>
 #include <public/domctl.h>
 #include <asm/device.h>
@@ -42,6 +43,11 @@ TYPE_SAFE(uint64_t, bfn);
 #undef bfn_x
 #endif
 
+static inline bfn_t bfn_add(bfn_t bfn, unsigned long i)
+{
+    return _bfn(bfn_x(bfn) + i);
+}
+
 #define IOMMU_PAGE_SHIFT 12
 #define IOMMU_PAGE_SIZE  (_AC(1,L) << IOMMU_PAGE_SHIFT)
 #define IOMMU_PAGE_MASK  (~(IOMMU_PAGE_SIZE - 1))
@@ -86,9 +92,9 @@ void iommu_teardown(struct domain *d);
 #define IOMMUF_readable  (1u<<_IOMMUF_readable)
 #define _IOMMUF_writable 1
 #define IOMMUF_writable  (1u<<_IOMMUF_writable)
-int __must_check iommu_map_page(struct domain *d, unsigned long bfn,
-                                unsigned long mfn, unsigned int flags);
-int __must_check iommu_unmap_page(struct domain *d, unsigned long bfn);
+int __must_check iommu_map_page(struct domain *d, bfn_t bfn,
+                                mfn_t mfn, unsigned int flags);
+int __must_check iommu_unmap_page(struct domain *d, bfn_t bfn);
 
 enum iommu_feature
 {
@@ -213,7 +219,7 @@ int iommu_do_pci_domctl(struct xen_domctl *, struct domain *d,
 int iommu_do_domctl(struct xen_domctl *, struct domain *d,
                     XEN_GUEST_HANDLE_PARAM(xen_domctl_t));
 
-int __must_check iommu_iotlb_flush(struct domain *d, unsigned long bfn,
+int __must_check iommu_iotlb_flush(struct domain *d, bfn_t bfn,
                                    unsigned int page_count);
 int __must_check iommu_iotlb_flush_all(struct domain *d);
 
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#9 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 19 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A subsequent patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

Quoted 4 lines

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

Explain this one to me? At the moment I don't see any value from having the extra variable in the middle.

Everything else looks good.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#10 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 35 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 15:01
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Julien Grall <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>
Subject: Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 2 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A

subsequent

Quoted 17 lines

patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

Quoted 7 lines
Quoted 4 lines

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

Explain this one to me? At the moment I don't see any value from having the extra variable in the middle.

This was something that Jan wanted.

Paul

Quoted 4 lines

Everything else looks good.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#11 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 41 lines
Quoted 41 lines
Quoted 1 line · On 10.07.18 at 16:10, <[email protected]> wrote:

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 15:01
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Julien Grall <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>
Subject: Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 2 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A

subsequent

Quoted 21 lines

patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

Quoted 4 lines
Quoted 2 lines

Explain this one to me? At the moment I don't see any value from having the extra variable in the middle.

This was something that Jan wanted.

Ah, yes, it was in a reply to this patch's v1 that I had suggested a neutrally named variable in case it can hold values from more than one space. In the particular case of _get_page_type(), where I had also given the v1 comment, I can't however see that it is now really obvious that a 1:1 mapping is being established. To me that would mean passing the same local variable (suitably type wrapped where needed) twice into iommu_map_page(). It is not clear to me what use a mfn_to_gmfn() invocation is inside a is_pv_domain() guarded block, now that we don't permit translated PV domains anymore (not that I would think that they had worked before the code was removed).

Jan

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#12 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 59 lines

-----Original Message-----
From: Jan Beulich [mailto:[email protected]]
Sent: 10 July 2018 15:29
To: Paul Durrant <[email protected]>
Cc: Julien Grall <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; xen- devel <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Tim (Xen.org) <[email protected]>
Subject: RE: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 5 lines
Quoted 5 lines
Quoted 1 line · On 10.07.18 at 16:10, <[email protected]> wrote:

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 15:01
To: Paul Durrant <[email protected]>; xen-

[email protected]

Quoted 3 lines
Quoted 3 lines

Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>;

Konrad

Quoted 2 lines
Quoted 2 lines

Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Julien Grall <[email protected]>; Tim

(Xen.org)

Quoted 3 lines
Quoted 3 lines

<[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>
Subject: Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and

MFN

Quoted 28 lines
Quoted 28 lines

in exported functions

Quoted 2 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A

subsequent

Quoted 21 lines

patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

Quoted 8 lines
Quoted 4 lines
Quoted 2 lines

Explain this one to me? At the moment I don't see any value from having the extra variable in the middle.

This was something that Jan wanted.

Ah, yes, it was in a reply to this patch's v1 that I had suggested a neutrally named variable in case it can hold values from more than one space.

Yes, it was:

" Along the lines of what I've said earlier about mixing address spaces, this would perhaps not so much need a comment (it's a 1:1 mapping after all), but rather making more obvious that it's a 1:1 mapping. This in particular would mean to me to latch page_to_mfn(page) into a (neutrally named, e.g. "frame") local variable, and use the result in a way that makes obviously especially on the "map" path that this really requests a 1:1 mapping. By implication from the 1:1 mapping it'll then (hopefully) be clear to the reader that which exact name space is used doesn't really matter."

Quoted 10 lines

In the particular case of _get_page_type(), where I had also given the v1 comment, I can't however see that it is now really obvious that a 1:1 mapping is being established. To me that would mean passing the same local variable (suitably type wrapped where needed) twice into iommu_map_page(). It is not clear to me what use a mfn_to_gmfn() invocation is inside a is_pv_domain() guarded block, now that we don't permit translated PV domains anymore (not that I would think that they had worked before the code was removed).

I can see about trying to clean up the mfn_to_gmfn() and re-phrase things to make it more obvious that the map is 1:1.

Paul

Quoted 2 lines

Jan

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#13 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

In reply to Paul Durrant
Quoted 19 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A subsequent patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

Quoted 4 lines

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

A couple of comments on particular instances...

Quoted 5 lines
xen/arch/x86/mm/p2m.c +0 -0
index c53cab44d9..ce12bcff42 100644

xen/arch/x86/mm/p2m.c p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn, · line 722

Quoted 10 lines
xen/arch/x86/mm/p2m.c +4 -1
@@ -714,9 +714,12 @@ p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn,
 
         if ( need_iommu(p2m->domain) )
         {
+            unsigned long frame = mfn;
+            bfn_t bfn = _bfn(frame);
+
             for ( i = 0; i < (1 << page_order); i++ )
             {
-                int ret = iommu_unmap_page(p2m->domain, mfn + i);
+                int ret = iommu_unmap_page(p2m->domain, bfn_add(bfn, i));

Having a 'bfn' variable here makes some sense, because otherwise, if mfn ever gets the mfn_t type, you'll have

  iommu_unmap_page(... _bfn(mfn_x(mfn)+i));

being able to use bfn_add() is much cleaner. I don't think the intermediate 'frame' variable in the case, really adds anything.

Quoted 18 lines
xen/common/grant_table.c +2 -1
index d2610e320c..d0926d13e0 100644
@@ -1132,6 +1132,8 @@ map_grant_ref(
     need_iommu = gnttab_need_iommu_mapping(ld);
     if ( need_iommu )
     {
+        unsigned long frame = mfn_x(mfn);
+        bfn_t bfn = _bfn(frame);
         unsigned int kind;
         int err = 0;
 
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),

xen/common/grant_table.c map_grant_ref( · line 1155

Quoted 10 lines
xen/common/grant_table.c +3 -3
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);

Here's an example where I think having an extra variable is somewhat dangerous. Before this change, it's obvious that you have a 1:1 mapping; now, looking just at this line, it's not obvious that bfn == mfn. Worse, there's a risk that there will be some sort of bug introduced which changes bfn, such that bfn != mfn anymore.

If you have to use an intermediate variable here, this should be

  iommu_map_page(..., _bfn(frame), _mfn(frame), ...);

But I really think

  iommu_map_page(..., _bfn(mfn_x(mfn)), mfn, ...);

makes the most sense here.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#14 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 104 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 17:13
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Julien Grall <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>
Subject: Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 2 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A

subsequent

Quoted 17 lines

patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

Quoted 4 lines

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

A couple of comments on particular instances...

xen/arch/x86/mm/p2m.c p2m_remove_page(struct p2m_domain *p2m,

Quoted 5 lines
xen/arch/x86/mm/p2m.c +0 -0
index c53cab44d9..ce12bcff42 100644
@@ -714,9 +714,12 @@ p2m_remove_page(struct p2m_domain *p2m,

unsigned long gfn_l, unsigned long mfn,

xen/arch/x86/mm/p2m.c p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn, · line 722

Quoted 10 lines
xen/arch/x86/mm/p2m.c +4 -1
         if ( need_iommu(p2m->domain) )
         {
+            unsigned long frame = mfn;
+            bfn_t bfn = _bfn(frame);
+
             for ( i = 0; i < (1 << page_order); i++ )
             {
-                int ret = iommu_unmap_page(p2m->domain, mfn + i);
+                int ret = iommu_unmap_page(p2m->domain, bfn_add(bfn, i));

Having a 'bfn' variable here makes some sense, because otherwise, if mfn ever gets the mfn_t type, you'll have

  iommu_unmap_page(... _bfn(mfn_x(mfn)+i));

being able to use bfn_add() is much cleaner. I don't think the intermediate 'frame' variable in the case, really adds anything.

Quoted 18 lines
xen/common/grant_table.c +2 -1
index d2610e320c..d0926d13e0 100644
@@ -1132,6 +1132,8 @@ map_grant_ref(
     need_iommu = gnttab_need_iommu_mapping(ld);
     if ( need_iommu )
     {
+        unsigned long frame = mfn_x(mfn);
+        bfn_t bfn = _bfn(frame);
         unsigned int kind;
         int err = 0;
 
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),

xen/common/grant_table.c map_grant_ref( · line 1155

Quoted 10 lines
xen/common/grant_table.c +3 -3
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);

Here's an example where I think having an extra variable is somewhat dangerous. Before this change, it's obvious that you have a 1:1 mapping; now, looking just at this line, it's not obvious that bfn == mfn. Worse, there's a risk that there will be some sort of bug introduced which changes bfn, such that bfn != mfn anymore.

If you have to use an intermediate variable here, this should be

Quoted 7 lines
  iommu_map_page(..., _bfn(frame), _mfn(frame), ...);

But I really think

  iommu_map_page(..., _bfn(mfn_x(mfn)), mfn, ...);

makes the most sense here.

How about:

#define mfn_to_bfn(mfn) (_bfn(mfn_x(mfn))

iommu_map_page(..., mfn_to_bfn(mfn), mfn, ...);

?

I can similarly define gfn_to_bfn() for places where it is needed.

Paul

Quoted 2 lines

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#15 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 114 lines · Paul Durrant
Quoted 104 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 17:13
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Julien Grall <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>
Subject: Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 2 lines · Paul Durrant

This patch modifies the declaration of the entry points to the IOMMU sub-system to use bfn_t and mfn_t in place of unsigned long. A

subsequent

Quoted 17 lines

patch will similarly modify the methods in the iommu_ops structure.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

Quoted 4 lines

v2:
- Addressed comments from Jan.
- Use intermediate 'frame' variable to avoid directly encapsulating mfn or gfn values as bfns.

A couple of comments on particular instances...

xen/arch/x86/mm/p2m.c p2m_remove_page(struct p2m_domain *p2m,

Quoted 5 lines
xen/arch/x86/mm/p2m.c +0 -0
index c53cab44d9..ce12bcff42 100644
@@ -714,9 +714,12 @@ p2m_remove_page(struct p2m_domain *p2m,

unsigned long gfn_l, unsigned long mfn,

xen/arch/x86/mm/p2m.c p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn, · line 722

Quoted 10 lines
xen/arch/x86/mm/p2m.c +4 -1
         if ( need_iommu(p2m->domain) )
         {
+            unsigned long frame = mfn;
+            bfn_t bfn = _bfn(frame);
+
             for ( i = 0; i < (1 << page_order); i++ )
             {
-                int ret = iommu_unmap_page(p2m->domain, mfn + i);
+                int ret = iommu_unmap_page(p2m->domain, bfn_add(bfn, i));

Having a 'bfn' variable here makes some sense, because otherwise, if mfn ever gets the mfn_t type, you'll have

  iommu_unmap_page(... _bfn(mfn_x(mfn)+i));

being able to use bfn_add() is much cleaner. I don't think the intermediate 'frame' variable in the case, really adds anything.

Quoted 18 lines
xen/common/grant_table.c +2 -1
index d2610e320c..d0926d13e0 100644
@@ -1132,6 +1132,8 @@ map_grant_ref(
     need_iommu = gnttab_need_iommu_mapping(ld);
     if ( need_iommu )
     {
+        unsigned long frame = mfn_x(mfn);
+        bfn_t bfn = _bfn(frame);
         unsigned int kind;
         int err = 0;
 
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),

xen/common/grant_table.c map_grant_ref( · line 1155

Quoted 10 lines
xen/common/grant_table.c +3 -3
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);

Here's an example where I think having an extra variable is somewhat dangerous. Before this change, it's obvious that you have a 1:1 mapping; now, looking just at this line, it's not obvious that bfn == mfn. Worse, there's a risk that there will be some sort of bug introduced which changes bfn, such that bfn != mfn anymore.

If you have to use an intermediate variable here, this should be

Quoted 7 lines
  iommu_map_page(..., _bfn(frame), _mfn(frame), ...);

But I really think

  iommu_map_page(..., _bfn(mfn_x(mfn)), mfn, ...);

makes the most sense here.

How about:

Quoted 7 lines

#define mfn_to_bfn(mfn) (_bfn(mfn_x(mfn))

iommu_map_page(..., mfn_to_bfn(mfn), mfn, ...);

?

I can similarly define gfn_to_bfn() for places where it is needed.

That works for me.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#16 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

In reply to Paul Durrant
Quoted 38 lines
Quoted 28 lines
Quoted 1 line · On 10.07.18 at 18:18, <[email protected]> wrote:

From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 17:13

Quoted 5 lines · Paul Durrant
xen/common/grant_table.c +0 -1
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),

xen/common/grant_table.c map_grant_ref( · line 1155

Quoted 10 lines
xen/common/grant_table.c +3 -3
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);

Here's an example where I think having an extra variable is somewhat dangerous. Before this change, it's obvious that you have a 1:1 mapping; now, looking just at this line, it's not obvious that bfn == mfn. Worse, there's a risk that there will be some sort of bug introduced which changes bfn, such that bfn != mfn anymore.

If you have to use an intermediate variable here, this should be

Quoted 7 lines
  iommu_map_page(..., _bfn(frame), _mfn(frame), ...);

But I really think

  iommu_map_page(..., _bfn(mfn_x(mfn)), mfn, ...);

makes the most sense here.

How about:

Quoted 7 lines

#define mfn_to_bfn(mfn) (_bfn(mfn_x(mfn))

iommu_map_page(..., mfn_to_bfn(mfn), mfn, ...);

?

I can similarly define gfn_to_bfn() for places where it is needed.

Please don't: Such macros (see others like mfn_to_gmfn() or pfn_to_pdx()) imply translation between address spaces, yet there's no translation here. If anything, the macro name would have to make clear there's no translation, e.g. cast_mfn_to_bfn(), but I think the open coding is then more clear and not significantly longer.

Jan

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#17 lore raw

Re: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 58 lines

-----Original Message-----
From: Jan Beulich [mailto:[email protected]]
Sent: 11 July 2018 08:57
To: Paul Durrant <[email protected]>
Cc: Julien Grall <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Wei Liu <[email protected]>; Jun Nakajima <[email protected]>; Kevin Tian <[email protected]>; Stefano Stabellini <[email protected]>; xen- devel <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Tim (Xen.org) <[email protected]>
Subject: RE: [PATCH v2 03/13] iommu: make use of type-safe BFN and MFN in exported functions

Quoted 44 lines
Quoted 28 lines
Quoted 1 line · On 10.07.18 at 18:18, <[email protected]> wrote:

From: George Dunlap [mailto:[email protected]]
Sent: 10 July 2018 17:13

Quoted 5 lines · Paul Durrant
xen/common/grant_table.c +0 -1
@@ -1144,14 +1146,13 @@ map_grant_ref(
              !(old_pin & (GNTPIN_hstw_mask|GNTPIN_devw_mask)) )
         {
             if ( !(kind & MAPKIND_WRITE) )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),

xen/common/grant_table.c map_grant_ref( · line 1155

Quoted 10 lines
xen/common/grant_table.c +3 -3
-                                     IOMMUF_readable|IOMMUF_writable);
+                err = iommu_map_page(ld, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
         }
         else if ( act_pin && !old_pin )
         {
             if ( !kind )
-                err = iommu_map_page(ld, mfn_x(mfn), mfn_x(mfn),
-                                     IOMMUF_readable);
+                err = iommu_map_page(ld, bfn, mfn, IOMMUF_readable);

Here's an example where I think having an extra variable is somewhat dangerous. Before this change, it's obvious that you have a 1:1 mapping; now, looking just at this line, it's not obvious that bfn == mfn. Worse, there's a risk that there will be some sort of bug introduced which changes bfn, such that bfn != mfn anymore.

If you have to use an intermediate variable here, this should be

Quoted 7 lines
  iommu_map_page(..., _bfn(frame), _mfn(frame), ...);

But I really think

  iommu_map_page(..., _bfn(mfn_x(mfn)), mfn, ...);

makes the most sense here.

How about:

#define mfn_to_bfn(mfn) (_bfn(mfn_x(mfn))

iommu_map_page(..., mfn_to_bfn(mfn), mfn, ...);

?

Quoted 8 lines
Quoted 1 line

I can similarly define gfn_to_bfn() for places where it is needed.

Please don't: Such macros (see others like mfn_to_gmfn() or pfn_to_pdx()) imply translation between address spaces, yet there's no translation here. If anything, the macro name would have to make clear there's no translation, e.g. cast_mfn_to_bfn(), but I think the open coding is then more clear and not significantly longer.

Ok. I will get rid of the intermediates again and open code as before.

Paul

Quoted 2 lines

Jan

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#18 lore raw

PATCH v2 4/13 iommu: push use of type-safe BFN and MFN into iommu_ops

In reply to Paul Durrant

This patch modifies the methods in struct iommu_ops to use type-safe BFN and MFN. This follows on from the prior patch that modified the functions exported in xen/iommu.h.

---
Cc: Suravee Suthikulpanit <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: Andrew Cooper <[email protected]>

v2:
- Addressed comments from Jan.
- Extend use of intermediate 'frame' variable to avoid directly encapsulating gfn values as bfns or vice versa.

---
 xen/drivers/passthrough/amd/iommu_map.c       | 46 ++++++++++++++++-----------
 xen/drivers/passthrough/amd/pci_amd_iommu.c   |  2 +-
 xen/drivers/passthrough/arm/smmu.c            | 26 +++++++--------
 xen/drivers/passthrough/iommu.c               |  9 +++---
 xen/drivers/passthrough/vtd/iommu.c           | 25 +++++++--------
 xen/drivers/passthrough/x86/iommu.c           |  2 +-
 xen/include/asm-x86/hvm/svm/amd-iommu-proto.h |  8 ++---
 xen/include/xen/iommu.h                       | 13 +++++---
 8 files changed, 72 insertions(+), 59 deletions(-)
xen/drivers/passthrough/amd/iommu_map.c +27 -19
index 4deab9cd2f..5a9a0af320 100644
@@ -631,7 +631,7 @@ static int update_paging_mode(struct domain *d, unsigned long bfn)
     return 0;
 }
 
-int amd_iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
+int amd_iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
                        unsigned int flags)
 {
     bool_t need_flush = 0;
@@ -651,7 +651,8 @@ int amd_iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
     if ( rc )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Root table alloc failed, bfn = %lx\n", bfn);
+        AMD_IOMMU_DEBUG("Root table alloc failed, bfn = %"PRI_bfn"\n",
+                        bfn_x(bfn));
         domain_crash(d);
         return rc;
     }
@@ -660,25 +661,27 @@ int amd_iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
      * we might need a deeper page table for wider bfn now */
     if ( is_hvm_domain(d) )
     {
-        if ( update_paging_mode(d, bfn) )
+        if ( update_paging_mode(d, bfn_x(bfn)) )
         {
             spin_unlock(&hd->arch.mapping_lock);
-            AMD_IOMMU_DEBUG("Update page mode failed bfn = %lx\n", bfn);
+            AMD_IOMMU_DEBUG("Update page mode failed bfn = %"PRI_bfn"\n",
+                            bfn_x(bfn));
             domain_crash(d);
             return -EFAULT;
         }
     }
 
-    if ( iommu_pde_from_bfn(d, bfn, pt_mfn) || (pt_mfn[1] == 0) )
+    if ( iommu_pde_from_bfn(d, bfn_x(bfn), pt_mfn) || (pt_mfn[1] == 0) )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Invalid IO pagetable entry bfn = %lx\n", bfn);
+        AMD_IOMMU_DEBUG("Invalid IO pagetable entry bfn = %"PRI_bfn"\n",
+                        bfn_x(bfn));
         domain_crash(d);
         return -EFAULT;
     }
 
     /* Install 4k mapping first */
-    need_flush = set_iommu_pte_present(pt_mfn[1], bfn, mfn,
+    need_flush = set_iommu_pte_present(pt_mfn[1], bfn_x(bfn), mfn_x(mfn),
                                        IOMMU_PAGING_MODE_LEVEL_1,
                                        !!(flags & IOMMUF_writable),
                                        !!(flags & IOMMUF_readable));
@@ -690,7 +693,7 @@ int amd_iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
     /* 4K mapping for PV guests never changes, 
      * no need to flush if we trust non-present bits */
     if ( is_hvm_domain(d) )
-        amd_iommu_flush_pages(d, bfn, 0);
+        amd_iommu_flush_pages(d, bfn_x(bfn), 0);
 
     for ( merge_level = IOMMU_PAGING_MODE_LEVEL_2;
           merge_level <= hd->arch.paging_mode; merge_level++ )
@@ -698,15 +701,16 @@ int amd_iommu_map_page(struct domain *d, unsigned long bfn, unsigned long mfn,
         if ( pt_mfn[merge_level] == 0 )
             break;
         if ( !iommu_update_pde_count(d, pt_mfn[merge_level],
-                                     bfn, mfn, merge_level) )
+                                     bfn_x(bfn), mfn_x(mfn), merge_level) )
             break;
 
-        if ( iommu_merge_pages(d, pt_mfn[merge_level], bfn,
+        if ( iommu_merge_pages(d, pt_mfn[merge_level], bfn_x(bfn),
                                flags, merge_level) )
         {
             spin_unlock(&hd->arch.mapping_lock);
             AMD_IOMMU_DEBUG("Merge iommu page failed at level %d, "
-                            "bfn = %lx mfn = %lx\n", merge_level, bfn, mfn);
+                            "bfn = %"PRI_bfn" mfn = %"PRI_mfn"\n",
+                            merge_level, bfn_x(bfn), mfn_x(mfn));
             domain_crash(d);
             return -EFAULT;
         }
@@ -720,7 +724,7 @@ out:
     return 0;
 }
 
-int amd_iommu_unmap_page(struct domain *d, unsigned long bfn)
+int amd_iommu_unmap_page(struct domain *d, bfn_t bfn)
 {
     unsigned long pt_mfn[7];
     struct domain_iommu *hd = dom_iommu(d);
@@ -742,31 +746,33 @@ int amd_iommu_unmap_page(struct domain *d, unsigned long bfn)
      * we might need a deeper page table for lager bfn now */
     if ( is_hvm_domain(d) )
     {
-        int rc = update_paging_mode(d, bfn);
+        int rc = update_paging_mode(d, bfn_x(bfn));
 
         if ( rc )
         {
             spin_unlock(&hd->arch.mapping_lock);
-            AMD_IOMMU_DEBUG("Update page mode failed bfn = %lx\n", bfn);
+            AMD_IOMMU_DEBUG("Update page mode failed bfn = %"PRI_bfn"\n",
+                            bfn_x(bfn));
             if ( rc != -EADDRNOTAVAIL )
                 domain_crash(d);
             return rc;
         }
     }
 
-    if ( iommu_pde_from_bfn(d, bfn, pt_mfn) || (pt_mfn[1] == 0) )
+    if ( iommu_pde_from_bfn(d, bfn_x(bfn), pt_mfn) || (pt_mfn[1] == 0) )
     {
         spin_unlock(&hd->arch.mapping_lock);
-        AMD_IOMMU_DEBUG("Invalid IO pagetable entry bfn = %lx\n", bfn);
+        AMD_IOMMU_DEBUG("Invalid IO pagetable entry bfn = %"PRI_bfn"\n",
+                        bfn_x(bfn));
         domain_crash(d);
         return -EFAULT;
     }
 
     /* mark PTE as 'page not present' */
-    clear_iommu_pte_present(pt_mfn[1], bfn);
+    clear_iommu_pte_present(pt_mfn[1], bfn_x(bfn));
     spin_unlock(&hd->arch.mapping_lock);
 
-    amd_iommu_flush_pages(d, bfn, 0);
+    amd_iommu_flush_pages(d, bfn_x(bfn), 0);
 
     return 0;
 }
@@ -787,7 +793,9 @@ int amd_iommu_reserve_domain_unity_map(struct domain *domain,
     gfn = phys_addr >> PAGE_SHIFT;
     for ( i = 0; i < npages; i++ )
     {
-        rt = amd_iommu_map_page(domain, gfn +i, gfn +i, flags);
+        unsigned long frame = gfn + i;
+
+        rt = amd_iommu_map_page(domain, _bfn(frame), _mfn(frame), flags);
         if ( rt != 0 )
             return rt;
     }
xen/drivers/passthrough/amd/pci_amd_iommu.c +1 -1
index d608631e6e..eea22c3d0d 100644
@@ -271,7 +271,7 @@ static void __hwdom_init amd_iommu_hwdom_init(struct domain *d)
              */
             if ( mfn_valid(_mfn(pfn)) )
             {
-                int ret = amd_iommu_map_page(d, pfn, pfn,
+                int ret = amd_iommu_map_page(d, _bfn(pfn), _mfn(pfn),
                                              IOMMUF_readable|IOMMUF_writable);
 
                 if ( !rc )
xen/drivers/passthrough/arm/smmu.c +13 -13
index f9f2e8dca4..7db493756d 100644
@@ -2551,7 +2551,7 @@ static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
 }
 
 static int __must_check arm_smmu_iotlb_flush(struct domain *d,
-                                             unsigned long bfn,
+                                             bfn_t bfn,
                                              unsigned int page_count)
 {
 	/* ARM SMMU v1 doesn't have flush by VMA and VMID */
@@ -2737,12 +2737,12 @@ static void arm_smmu_iommu_domain_teardown(struct domain *d)
 	xfree(xen_domain);
 }
 
-static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
-			unsigned long mfn, unsigned int flags)
+static int __must_check arm_smmu_map_page(struct domain *d, bfn_t bfn,
+			mfn_t mfn, unsigned int flags)
 {
+	unsigned long frame = bfn_x(bfn);
 	p2m_type_t t;
-	unsigned long frame = bfn;
-	unsigned long gfn;
+	gfn_t gfn;
 
 	/*
 	 * Grant mappings can be used for DMA requests. The dev_bus_addr
@@ -2753,8 +2753,8 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
 	 * function should only be used by gnttab code with gfn == mfn == bfn.
 	 */
 	BUG_ON(!is_domain_direct_mapped(d));
-	BUG_ON(mfn != frame);
-	gfn = frame;
+	BUG_ON(!mfn_eq(mfn, _mfn(frame));
+	gfn = _gfn(frame);
 
 	/* We only support readable and writable flags */
 	if (!(flags & (IOMMUF_readable | IOMMUF_writable)))
@@ -2766,14 +2766,14 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
 	 * The function guest_physmap_add_entry replaces the current mapping
 	 * if there is already one...
 	 */
-	return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
+	return guest_physmap_add_entry(d, gfn, mfn, 0, t);
 }
 
-static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long bfn)
+static int __must_check arm_smmu_unmap_page(struct domain *d, bfn_t bfn)
 {
-	unsigned long frame = bfn;
-	unsigned long gfn = frame;
-	unsigned long mfn = frame;
+	unsigned long frame = bfn_x(bfn);
+	gfn_t gfn = _gfn(frame);
+	mfn_t mfn = _mfn(frame);
 
 	/*
 	 * This function should only be used by gnttab code when the domain
@@ -2782,7 +2782,7 @@ static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long bfn)
 	if ( !is_domain_direct_mapped(d) )
 		return -EINVAL;
 
-	return guest_physmap_remove_page(d, _gfn(gfn), _mfn(mfn), 0);
+	return guest_physmap_remove_page(d, gfn, mfn, 0);
 }
 
 static const struct iommu_ops arm_smmu_iommu_ops = {
xen/drivers/passthrough/iommu.c +5 -4
index 2aa0c126d1..67a90a314f 100644
@@ -195,7 +195,8 @@ void __hwdom_init iommu_hwdom_init(struct domain *d)
                   == PGT_writable_page) )
                 mapping |= IOMMUF_writable;
 
-            ret = hd->platform_ops->map_page(d, bfn, mfn, mapping);
+            ret = hd->platform_ops->map_page(d, _bfn(bfn), _mfn(mfn),
+                                             mapping);
             if ( !rc )
                 rc = ret;
 
@@ -265,7 +266,7 @@ int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->map_page(d, bfn_x(bfn), mfn_x(mfn), flags);
+    rc = hd->platform_ops->map_page(d, bfn, mfn, flags);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
@@ -288,7 +289,7 @@ int iommu_unmap_page(struct domain *d, bfn_t bfn)
     if ( !iommu_enabled || !hd->platform_ops )
         return 0;
 
-    rc = hd->platform_ops->unmap_page(d, bfn_x(bfn));
+    rc = hd->platform_ops->unmap_page(d, bfn);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
@@ -328,7 +329,7 @@ int iommu_iotlb_flush(struct domain *d, bfn_t bfn, unsigned int page_count)
     if ( !iommu_enabled || !hd->platform_ops || !hd->platform_ops->iotlb_flush )
         return 0;
 
-    rc = hd->platform_ops->iotlb_flush(d, bfn_x(bfn), page_count);
+    rc = hd->platform_ops->iotlb_flush(d, bfn, page_count);
     if ( unlikely(rc) )
     {
         if ( !d->is_shutting_down && printk_ratelimit() )
xen/drivers/passthrough/vtd/iommu.c +12 -13
index 48f62e0e8d..2aa97c9efe 100644
@@ -585,7 +585,7 @@ static int __must_check iommu_flush_all(void)
 }
 
 static int __must_check iommu_flush_iotlb(struct domain *d,
-                                          unsigned long bfn,
+                                          bfn_t bfn,
                                           bool_t dma_old_pte_present,
                                           unsigned int page_count)
 {
@@ -612,12 +612,12 @@ static int __must_check iommu_flush_iotlb(struct domain *d,
         if ( iommu_domid == -1 )
             continue;
 
-        if ( page_count != 1 || bfn == bfn_x(INVALID_BFN) )
+        if ( page_count != 1 || bfn_eq(bfn, INVALID_BFN) )
             rc = iommu_flush_iotlb_dsi(iommu, iommu_domid,
                                        0, flush_dev_iotlb);
         else
             rc = iommu_flush_iotlb_psi(iommu, iommu_domid,
-                                       __bfn_to_baddr(bfn),
+                                       bfn_to_baddr(bfn),
                                        PAGE_ORDER_4K,
                                        !dma_old_pte_present,
                                        flush_dev_iotlb);
@@ -633,7 +633,7 @@ static int __must_check iommu_flush_iotlb(struct domain *d,
 }
 
 static int __must_check iommu_flush_iotlb_pages(struct domain *d,
-                                                unsigned long bfn,
+                                                bfn_t bfn,
                                                 unsigned int page_count)
 {
     return iommu_flush_iotlb(d, bfn, 1, page_count);
@@ -641,7 +641,7 @@ static int __must_check iommu_flush_iotlb_pages(struct domain *d,
 
 static int __must_check iommu_flush_iotlb_all(struct domain *d)
 {
-    return iommu_flush_iotlb(d, bfn_x(INVALID_BFN), 0, 0);
+    return iommu_flush_iotlb(d, INVALID_BFN, 0, 0);
 }
 
 /* clear one page's page table */
@@ -676,7 +676,7 @@ static int __must_check dma_pte_clear_one(struct domain *domain, u64 addr)
     iommu_flush_cache_entry(pte, sizeof(struct dma_pte));
 
     if ( !this_cpu(iommu_dont_flush_iotlb) )
-        rc = iommu_flush_iotlb_pages(domain, addr >> PAGE_SHIFT_4K, 1);
+        rc = iommu_flush_iotlb_pages(domain, baddr_to_bfn(addr), 1);
 
     unmap_vtd_domain_page(page);
 
@@ -1767,8 +1767,7 @@ static void iommu_domain_teardown(struct domain *d)
 }
 
 static int __must_check intel_iommu_map_page(struct domain *d,
-                                             unsigned long bfn,
-                                             unsigned long mfn,
+                                             bfn_t bfn, mfn_t mfn,
                                              unsigned int flags)
 {
     struct domain_iommu *hd = dom_iommu(d);
@@ -1786,16 +1785,16 @@ static int __must_check intel_iommu_map_page(struct domain *d,
 
     spin_lock(&hd->arch.mapping_lock);
 
-    pg_maddr = addr_to_dma_page_maddr(d, __bfn_to_baddr(bfn), 1);
+    pg_maddr = addr_to_dma_page_maddr(d, bfn_to_baddr(bfn), 1);
     if ( pg_maddr == 0 )
     {
         spin_unlock(&hd->arch.mapping_lock);
         return -ENOMEM;
     }
     page = (struct dma_pte *)map_vtd_domain_page(pg_maddr);
-    pte = page + (bfn & LEVEL_MASK);
+    pte = page + (bfn_x(bfn) & LEVEL_MASK);
     old = *pte;
-    dma_set_pte_addr(new, (paddr_t)mfn << PAGE_SHIFT_4K);
+    dma_set_pte_addr(new, mfn_to_maddr(mfn));
     dma_set_pte_prot(new,
                      ((flags & IOMMUF_readable) ? DMA_PTE_READ  : 0) |
                      ((flags & IOMMUF_writable) ? DMA_PTE_WRITE : 0));
@@ -1823,13 +1822,13 @@ static int __must_check intel_iommu_map_page(struct domain *d,
 }
 
 static int __must_check intel_iommu_unmap_page(struct domain *d,
-                                               unsigned long bfn)
+                                               bfn_t bfn)
 {
     /* Do nothing if hardware domain and iommu supports pass thru. */
     if ( iommu_passthrough && is_hardware_domain(d) )
         return 0;
 
-    return dma_pte_clear_one(d, __bfn_to_baddr(bfn));
+    return dma_pte_clear_one(d, bfn_to_baddr(bfn));
 }
 
 int iommu_pte_flush(struct domain *d, uint64_t bfn, uint64_t *pte,
xen/drivers/passthrough/x86/iommu.c +1 -1
index 68182afd91..379882c690 100644
@@ -65,7 +65,7 @@ int arch_iommu_populate_page_table(struct domain *d)
             {
                 ASSERT(!(gfn >> DEFAULT_DOMAIN_ADDRESS_WIDTH));
                 BUG_ON(SHARED_M2P(gfn));
-                rc = hd->platform_ops->map_page(d, gfn, mfn,
+                rc = hd->platform_ops->map_page(d, _bfn(gfn), _mfn(mfn),
                                                 IOMMUF_readable |
                                                 IOMMUF_writable);
             }
xen/include/asm-x86/hvm/svm/amd-iommu-proto.h +4 -4
index 99bc21c7b3..dce9ed6b83 100644
@@ -52,9 +52,9 @@ int amd_iommu_init(void);
 int amd_iommu_update_ivrs_mapping_acpi(void);
 
 /* mapping functions */
-int __must_check amd_iommu_map_page(struct domain *d, unsigned long gfn,
-                                    unsigned long mfn, unsigned int flags);
-int __must_check amd_iommu_unmap_page(struct domain *d, unsigned long gfn);
+int __must_check amd_iommu_map_page(struct domain *d, bfn_t bfn,
+                                    mfn_t mfn, unsigned int flags);
+int __must_check amd_iommu_unmap_page(struct domain *d, bfn_t bfn);
 u64 amd_iommu_get_next_table_from_pte(u32 *entry);
 int __must_check amd_iommu_alloc_root(struct domain_iommu *hd);
 int amd_iommu_reserve_domain_unity_map(struct domain *domain,
@@ -77,7 +77,7 @@ void iommu_dte_set_guest_cr3(u32 *dte, u16 dom_id, u64 gcr3,
 
 /* send cmd to iommu */
 void amd_iommu_flush_all_pages(struct domain *d);
-void amd_iommu_flush_pages(struct domain *d, unsigned long gfn,
+void amd_iommu_flush_pages(struct domain *d, unsigned long bfn,
                            unsigned int order);
 void amd_iommu_flush_iotlb(u8 devfn, const struct pci_dev *pdev,
                            uint64_t gaddr, unsigned int order);
xen/include/xen/iommu.h +9 -4
index a3c36c1148..624784fec8 100644
@@ -48,6 +48,11 @@ static inline bfn_t bfn_add(bfn_t bfn, unsigned long i)
     return _bfn(bfn_x(bfn) + i);
 }
 
+static inline bool_t bfn_eq(bfn_t x, bfn_t y)
+{
+    return bfn_x(x) == bfn_x(y);
+}
+
 #define IOMMU_PAGE_SHIFT 12
 #define IOMMU_PAGE_SIZE  (_AC(1,L) << IOMMU_PAGE_SHIFT)
 #define IOMMU_PAGE_MASK  (~(IOMMU_PAGE_SIZE - 1))
@@ -184,9 +189,9 @@ struct iommu_ops {
 #endif /* HAS_PCI */
 
     void (*teardown)(struct domain *d);
-    int __must_check (*map_page)(struct domain *d, unsigned long bfn,
-                                 unsigned long mfn, unsigned int flags);
-    int __must_check (*unmap_page)(struct domain *d, unsigned long bfn);
+    int __must_check (*map_page)(struct domain *d, bfn_t bfn, mfn_t mfn,
+                                 unsigned int flags);
+    int __must_check (*unmap_page)(struct domain *d, bfn_t bfn);
     void (*free_page_table)(struct page_info *);
 #ifdef CONFIG_X86
     void (*update_ire_from_apic)(unsigned int apic, unsigned int reg, unsigned int value);
@@ -197,7 +202,7 @@ struct iommu_ops {
     void (*resume)(void);
     void (*share_p2m)(struct domain *d);
     void (*crash_shutdown)(void);
-    int __must_check (*iotlb_flush)(struct domain *d, unsigned long bfn,
+    int __must_check (*iotlb_flush)(struct domain *d, bfn_t bfn,
                                     unsigned int page_count);
     int __must_check (*iotlb_flush_all)(struct domain *d);
     int (*get_reserved_device_memory)(iommu_grdm_t *, void *);
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#19 lore raw

Re: [PATCH v2 04/13] iommu: push use of type-safe BFN and MFN into iommu_ops

xen/drivers/passthrough/amd/iommu_map.c int amd_iommu_reserve_domain_unity_map(struct domain *domain, · line 798

Quoted 8 lines · Paul Durrant
xen/drivers/passthrough/amd/iommu_map.c +3 -1
@@ -787,7 +793,9 @@ int amd_iommu_reserve_domain_unity_map(struct domain *domain,
     gfn = phys_addr >> PAGE_SHIFT;
     for ( i = 0; i < npages; i++ )
     {
-        rt = amd_iommu_map_page(domain, gfn +i, gfn +i, flags);
+        unsigned long frame = gfn + i;
+
+        rt = amd_iommu_map_page(domain, _bfn(frame), _mfn(frame), flags);

Here's an example where the intermediate variable may actually improve things, if only by avoiding code duplication. :-)

Quoted 7 lines
xen/drivers/passthrough/arm/smmu.c +1 -1
@@ -2766,14 +2766,14 @@ static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn,
         * The function guest_physmap_add_entry replaces the current mapping
         * if there is already one...
         */
-       return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
+       return guest_physmap_add_entry(d, gfn, mfn, 0, t);
 }

xen/drivers/passthrough/arm/smmu.c static int __must_check arm_smmu_map_page(struct domain *d, unsigned long bfn, · line 2776

Quoted 10 lines
xen/drivers/passthrough/arm/smmu.c +4 -4
 
-static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned long bfn)
+static int __must_check arm_smmu_unmap_page(struct domain *d, bfn_t bfn)
 {
-       unsigned long frame = bfn;
-       unsigned long gfn = frame;
-       unsigned long mfn = frame;
+       unsigned long frame = bfn_x(bfn);
+       gfn_t gfn = _gfn(frame);
+       mfn_t mfn = _mfn(frame);

Even if we decide to use an intermediate variable in many other cases, this construct in particular seems excessive.

Everything else looks good, thanks.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#20 lore raw

PATCH v2 5/13 iommu: don't domain_crash() inside iommu_map/unmap_page()

In reply to Paul Durrant

Turn iommu_map/unmap_page() into straightforward wrappers that check the existence of the relevant iommu_op and call through to it. This makes them usable by PV IOMMU code to be delivered in future patches. Leave the decision on whether to invoke domain_crash() up to the caller. This has the added benefit that the (module/line number) message that domain_crash() spits out will be more indicative of where the problem lies.

NOTE: This patch includes one bit of clean-up in set_identity_p2m_entry()

      replacing use of p2m->domain with the domain pointer passed into the
      function.

---
Cc: Stefano Stabellini <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Jun Nakajima <[email protected]>
Cc: Kevin Tian <[email protected]>
Cc: George Dunlap <[email protected]>

v2:
- New in v2.

---
 xen/arch/arm/p2m.c                  |  2 ++
 xen/arch/x86/mm.c                   |  3 +++
 xen/arch/x86/mm/p2m-ept.c           |  3 +++
 xen/arch/x86/mm/p2m-pt.c            |  3 +++
 xen/arch/x86/mm/p2m.c               | 20 ++++++++++++++++++--
 xen/common/grant_table.c            |  8 ++++++++
 xen/common/memory.c                 |  3 +++
 xen/drivers/passthrough/iommu.c     | 12 ------------
 xen/drivers/passthrough/x86/iommu.c |  5 +++++
 9 files changed, 45 insertions(+), 14 deletions(-)
xen/arch/arm/p2m.c +2 -0
index 42c16da4dd..b99f4ff913 100644
@@ -955,6 +955,8 @@ static int __p2m_set_entry(struct p2m_domain *p2m,
         unsigned long frame = gfn_x(sgfn);
 
         rc = iommu_iotlb_flush(p2m->domain, _bfn(frame), 1UL << page_order);
+        if ( unlikely(rc) && !is_hardware_domain(p2m->domain) )
+            domain_crash(p2m->domain);
     }
     else
         rc = 0;
xen/arch/x86/mm.c +3 -0
index 4e66c6742d..32f09bac0a 100644
@@ -2738,6 +2738,9 @@ static int _get_page_type(struct page_info *page, unsigned long type,
                 iommu_ret = iommu_map_page(d, _bfn(frame), mfn,
                                            IOMMUF_readable |
                                            IOMMUF_writable);
+
+            if ( unlikely(iommu_ret) && !is_hardware_domain(d) )
+                domain_crash(d);
         }
     }
 
xen/arch/x86/mm/p2m-ept.c +3 -0
index 4951234197..5fc1e1a838 100644
@@ -898,6 +898,9 @@ out:
                     if ( !rc )
                         rc = ret;
                 }
+
+            if ( unlikely(rc) && !is_hardware_domain(d) )
+                domain_crash(d);
         }
     }
 
xen/arch/x86/mm/p2m-pt.c +3 -0
index 467ac8cb76..4e04525195 100644
@@ -719,6 +719,9 @@ p2m_pt_set_entry(struct p2m_domain *p2m, gfn_t gfn_, mfn_t mfn,
                         rc = ret;
                 }
         }
+
+        if ( unlikely(rc) && !is_hardware_domain(p2m->domain) )
+            domain_crash(p2m->domain);
     }
 
     /*
xen/arch/x86/mm/p2m.c +18 -2
index ce12bcff42..5ff1d51a0e 100644
@@ -724,6 +724,9 @@ p2m_remove_page(struct p2m_domain *p2m, unsigned long gfn_l, unsigned long mfn,
                 if ( !rc )
                     rc = ret;
             }
+
+            if ( unlikely(rc) && !is_hardware_domain(p2m->domain) )
+                domain_crash(p2m->domain);
         }
 
         return rc;
@@ -790,6 +793,9 @@ guest_physmap_add_entry(struct domain *d, gfn_t gfn, mfn_t mfn,
                         if ( iommu_unmap_page(d, bfn_add(bfn, i)) )
                             continue;
 
+                    if ( !is_hardware_domain(d) )
+                        domain_crash(d);
+
                     return rc;
                 }
             }
@@ -1158,7 +1164,7 @@ int set_identity_p2m_entry(struct domain *d, unsigned long gfn_l,
     struct p2m_domain *p2m = p2m_get_hostp2m(d);
     int ret;
 
-    if ( !paging_mode_translate(p2m->domain) )
+    if ( !paging_mode_translate(d) )
     {
         unsigned long frame = gfn_l;
 
@@ -1167,6 +1173,11 @@ int set_identity_p2m_entry(struct domain *d, unsigned long gfn_l,
 
         return iommu_map_page(d, _bfn(frame), _mfn(frame),
                               IOMMUF_readable | IOMMUF_writable);
+
+        if ( unlikely(ret) && !is_hardware_domain(d) )
+            domain_crash(d);
+
+        return ret;
     }
 
     gfn_lock(p2m, gfn, 0);
@@ -1259,7 +1270,12 @@ int clear_identity_p2m_entry(struct domain *d, unsigned long gfn_l)
         if ( !need_iommu(d) )
             return 0;
 
-        return iommu_unmap_page(d, _bfn(frame));
+        ret = iommu_unmap_page(d, _bfn(frame));
+
+        if ( unlikely(ret) && !is_hardware_domain(d) )
+            domain_crash(d);
+
+        return ret;
     }
 
     gfn_lock(p2m, gfn, 0);
xen/common/grant_table.c +8 -0
index d0926d13e0..c6b99c4116 100644
@@ -1156,6 +1156,9 @@ map_grant_ref(
         }
         if ( err )
         {
+            if ( !is_hardware_domain(ld) )
+                domain_crash(ld);
+
             double_gt_unlock(lgt, rgt);
             rc = GNTST_general_error;
             goto undo_out;
@@ -1423,7 +1426,12 @@ unmap_common(
         double_gt_unlock(lgt, rgt);
 
         if ( err )
+        {
+            if ( !is_hardware_domain(ld) )
+                domain_crash(ld);
+
             rc = GNTST_general_error;
+        }
     }
 
     /* If just unmapped a writable mapping, mark as dirtied */
xen/common/memory.c +3 -0
index e42da275c3..35da9ca80e 100644
@@ -843,6 +843,9 @@ int xenmem_add_to_physmap(struct domain *d, struct xen_add_to_physmap *xatp,
         ret = iommu_iotlb_flush(d, _bfn(frame), done);
         if ( unlikely(ret) && rc >= 0 )
             rc = ret;
+
+        if ( unlikely(rc < 0) && !is_hardware_domain(d) )
+            domain_crash(d);
     }
 #endif
 
xen/drivers/passthrough/iommu.c +0 -12
index 67a90a314f..7e59211a59 100644
@@ -273,9 +273,6 @@ int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
             printk(XENLOG_ERR
                    "d%d: IOMMU mapping bfn %"PRI_bfn" to mfn %"PRI_mfn" failed: %d\n",
                    d->domain_id, bfn_x(bfn), mfn_x(mfn), rc);
-
-        if ( !is_hardware_domain(d) )
-            domain_crash(d);
     }
 
     return rc;
@@ -296,9 +293,6 @@ int iommu_unmap_page(struct domain *d, bfn_t bfn)
             printk(XENLOG_ERR
                    "d%d: IOMMU unmapping bfn %"PRI_bfn" failed: %d\n",
                    d->domain_id, bfn_x(bfn), rc);
-
-        if ( !is_hardware_domain(d) )
-            domain_crash(d);
     }
 
     return rc;
@@ -336,9 +330,6 @@ int iommu_iotlb_flush(struct domain *d, bfn_t bfn, unsigned int page_count)
             printk(XENLOG_ERR
                    "d%d: IOMMU IOTLB flush failed: %d, bfn %"PRI_bfn", page count %u\n",
                    d->domain_id, rc, bfn_x(bfn), page_count);
-
-        if ( !is_hardware_domain(d) )
-            domain_crash(d);
     }
 
     return rc;
@@ -359,9 +350,6 @@ int iommu_iotlb_flush_all(struct domain *d)
             printk(XENLOG_ERR
                    "d%d: IOMMU IOTLB flush all failed: %d\n",
                    d->domain_id, rc);
-
-        if ( !is_hardware_domain(d) )
-            domain_crash(d);
     }
 
     return rc;
xen/drivers/passthrough/x86/iommu.c +5 -0
index 379882c690..9cd05d42d0 100644
@@ -104,8 +104,13 @@ int arch_iommu_populate_page_table(struct domain *d)
     this_cpu(iommu_dont_flush_iotlb) = 0;
 
     if ( !rc )
+    {
         rc = iommu_iotlb_flush_all(d);
 
+        if ( unlikely(rc) )
+            domain_crash(d);
+    }
+
     if ( rc && rc != -ERESTART )
         iommu_teardown(d);
 
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Reviewed-by on v2 5/13
#21 lore raw

Re: [PATCH v2 05/13] iommu: don't domain_crash() inside iommu_map/unmap_page()

Quoted 7 lines · Paul Durrant

Turn iommu_map/unmap_page() into straightforward wrappers that check the existence of the relevant iommu_op and call through to it. This makes them usable by PV IOMMU code to be delivered in future patches. Leave the decision on whether to invoke domain_crash() up to the caller. This has the added benefit that the (module/line number) message that domain_crash() spits out will be more indicative of where the problem lies.

Quoted 5 lines

NOTE: This patch includes one bit of clean-up in set_identity_p2m_entry()

      replacing use of p2m->domain with the domain pointer passed into the
      function.

I have some minor reservations about this, since it could leave open the possibility that a failed operation *doesn't* result in a domain crash when it should. But all the functions in question are __must_check, so the risk of that sort of bug creeping in is significantly lower.

So:

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

Reviewed-by on v2 5/13
#22 lore raw

Re: [PATCH v2 05/13] iommu: don't domain_crash() inside iommu_map/unmap_page()

In reply to Paul Durrant
Quoted 7 lines · Paul Durrant

Turn iommu_map/unmap_page() into straightforward wrappers that check the existence of the relevant iommu_op and call through to it. This makes them usable by PV IOMMU code to be delivered in future patches. Leave the decision on whether to invoke domain_crash() up to the caller. This has the added benefit that the (module/line number) message that domain_crash() spits out will be more indicative of where the problem lies.

Quoted 5 lines

NOTE: This patch includes one bit of clean-up in set_identity_p2m_entry()

      replacing use of p2m->domain with the domain pointer passed into the
      function.

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#23 lore raw

PATCH v2 6/13 public / x86: introduce __HYPERCALL_iommu_op

In reply to Paul Durrant

This patch introduces the boilerplate for a new hypercall to allow a domain to control IOMMU mappings for its own pages.
Whilst there is duplication of code between the native and compat entry points which appears ripe for some form of combination, I think it is better to maintain the separation as-is because the compat entry point will necessarily gain complexity in subsequent patches.

NOTE: This hypercall is only implemented for x86 and is currently
restricted by XSM to dom0. Its scope can be expanded in future if need be.

---
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Jan Beulich <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Daniel De Graaf <[email protected]>

v2:
- Get rid of the can_control_iommu() function, leaving this patch as pure boilerplate.
- Re-structure the hypercall to use a buffer array, similar to that used by __HYPERCALL_dm_op, to allow for future expansion of op structure without affecting binary compatibility.
- Drop use of __ in public header guard.

---
 tools/flask/policy/modules/xen.if   |   1 +
 xen/arch/x86/Makefile               |   1 +
 xen/arch/x86/hvm/hypercall.c        |   1 +
 xen/arch/x86/hypercall.c            |   1 +
 xen/arch/x86/iommu_op.c             | 182 ++++++++++++++++++++++++++++++++++++
 xen/arch/x86/pv/hypercall.c         |   1 +
 xen/include/Makefile                |   2 +
 xen/include/public/iommu_op.h       |  64 +++++++++++++
 xen/include/public/xen.h            |   1 +
 xen/include/xen/hypercall.h         |  12 +++
 xen/include/xlat.lst                |   2 +
 xen/include/xsm/dummy.h             |   6 ++
 xen/include/xsm/xsm.h               |   6 ++
 xen/xsm/dummy.c                     |   1 +
 xen/xsm/flask/hooks.c               |   6 ++
 xen/xsm/flask/policy/access_vectors |   2 +
 16 files changed, 289 insertions(+)
 create mode 100644 xen/arch/x86/iommu_op.c
 create mode 100644 xen/include/public/iommu_op.h
tools/flask/policy/modules/xen.if +1 -0
index 7aefd0061e..5161611471 100644
@@ -60,6 +60,7 @@ define(`create_domain_common', `
 	allow $1 $2:grant setup;
 	allow $1 $2:hvm { getparam hvmctl sethvmc
 			setparam nested altp2mhvm altp2mhvm_op dm };
+	allow $1 $2:resource control_iommu;
 ')
 
 # create_domain(priv, target)
xen/arch/x86/Makefile +1 -0
index 5563c813dd..c3cf364439 100644
@@ -33,6 +33,7 @@ obj-$(CONFIG_CRASH_DEBUG) += gdbstub.o
 obj-y += hypercall.o
 obj-y += i387.o
 obj-y += i8259.o
+obj-y += iommu_op.o
 obj-y += io_apic.o
 obj-$(CONFIG_LIVEPATCH) += alternative.o livepatch.o
 obj-y += msi.o
xen/arch/x86/hvm/hypercall.c +1 -0
index 85eacd7d33..3574966827 100644
@@ -137,6 +137,7 @@ static const hypercall_table_t hvm_hypercall_table[] = {
     COMPAT_CALL(mmuext_op),
     HYPERCALL(xenpmu_op),
     COMPAT_CALL(dm_op),
+    COMPAT_CALL(iommu_op),
     HYPERCALL(arch_1)
 };
 
xen/arch/x86/hypercall.c +1 -0
index 90e88c1d2c..045753e702 100644
@@ -68,6 +68,7 @@ const hypercall_args_t hypercall_args_table[NR_hypercalls] =
     ARGS(xenpmu_op, 2),
     ARGS(dm_op, 3),
     ARGS(mca, 1),
+    ARGS(iommu_op, 2),
     ARGS(arch_1, 1),
 };
 
xen/arch/x86/iommu_op.c new file +182 -0

182 lines not on this page · raw message

xen/arch/x86/pv/hypercall.c +1 -0
index bbc3011d1a..d23f9af42f 100644
@@ -80,6 +80,7 @@ const hypercall_table_t pv_hypercall_table[] = {
     HYPERCALL(xenpmu_op),
     COMPAT_CALL(dm_op),
     HYPERCALL(mca),
+    COMPAT_CALL(iommu_op),
     HYPERCALL(arch_1),
 };
 
xen/include/Makefile +2 -0
index 7c5034e6e0..3e6421c4b4 100644
@@ -11,6 +11,7 @@ headers-y := \
     compat/features.h \
     compat/grant_table.h \
     compat/kexec.h \
+    compat/iommu_op.h \
     compat/memory.h \
     compat/nmi.h \
     compat/physdev.h \
@@ -29,6 +30,7 @@ headers-$(CONFIG_X86)     += compat/arch-x86/xen-$(compat-arch-y).h
 headers-$(CONFIG_X86)     += compat/hvm/dm_op.h
 headers-$(CONFIG_X86)     += compat/hvm/hvm_op.h
 headers-$(CONFIG_X86)     += compat/hvm/hvm_vcpu.h
+headers-$(CONFIG_X86)     += compat/iommu_op.h
 headers-y                 += compat/arch-$(compat-arch-y).h compat/pmu.h compat/xlat.h
 headers-$(CONFIG_FLASK)   += compat/xsm/flask_op.h
 
xen/include/public/iommu_op.h new file +64 -0

64 lines not on this page · raw message

xen/include/public/xen.h +1 -0
index fb1df8f293..68b0968e7d 100644
@@ -121,6 +121,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_ulong_t);
 #define __HYPERVISOR_xc_reserved_op       39 /* reserved for XenClient */
 #define __HYPERVISOR_xenpmu_op            40
 #define __HYPERVISOR_dm_op                41
+#define __HYPERVISOR_iommu_op             42
 
 /* Architecture-specific hypercall definitions. */
 #define __HYPERVISOR_arch_0               48
xen/include/xen/hypercall.h +12 -0
index cc99aea57d..2ebc999f4b 100644
@@ -16,6 +16,7 @@
 #include <public/version.h>
 #include <public/pmu.h>
 #include <public/hvm/dm_op.h>
+#include <public/iommu_op.h>
 #include <asm/hypercall.h>
 #include <xsm/xsm.h>
 
@@ -148,6 +149,10 @@ do_dm_op(
     unsigned int nr_bufs,
     XEN_GUEST_HANDLE_PARAM(xen_dm_op_buf_t) bufs);
 
+extern long
+do_iommu_op(unsigned int nr_bufs,
+            XEN_GUEST_HANDLE_PARAM(xen_iommu_op_buf_t) bufs);
+
 #ifdef CONFIG_COMPAT
 
 extern int
@@ -205,6 +210,13 @@ compat_dm_op(
     unsigned int nr_bufs,
     XEN_GUEST_HANDLE_PARAM(void) bufs);
 
+#include <compat/iommu_op.h>
+
+DEFINE_XEN_GUEST_HANDLE(compat_iommu_op_buf_t);
+extern int
+compat_iommu_op(unsigned int nr_bufs,
+                XEN_GUEST_HANDLE_PARAM(compat_iommu_op_buf_t) bufs);
+
 #endif
 
 void arch_get_xen_caps(xen_capabilities_info_t *info);
xen/include/xlat.lst +2 -0
index 527332054a..3b15c18c4e 100644
@@ -77,6 +77,8 @@
 ?	vcpu_hvm_context		hvm/hvm_vcpu.h
 ?	vcpu_hvm_x86_32			hvm/hvm_vcpu.h
 ?	vcpu_hvm_x86_64			hvm/hvm_vcpu.h
+!	iommu_op			iommu_op.h
+!	iommu_op_buf			iommu_op.h
 ?	kexec_exec			kexec.h
 !	kexec_image			kexec.h
 !	kexec_range			kexec.h
xen/include/xsm/dummy.h +6 -0
index ff6b2dbf39..32784e7872 100644
@@ -715,6 +715,12 @@ static XSM_INLINE int xsm_dm_op(XSM_DEFAULT_ARG struct domain *d)
     return xsm_default_action(action, current->domain, d);
 }
 
+static XSM_INLINE int xsm_iommu_op(XSM_DEFAULT_ARG struct domain *d)
+{
+    XSM_ASSERT_ACTION(XSM_PRIV);
+    return xsm_default_action(action, current->domain, d);
+}
+
 #endif /* CONFIG_X86 */
 
 #include <public/version.h>
xen/include/xsm/xsm.h +6 -0
index f0c6fc7e29..b7394c7d76 100644
@@ -178,6 +178,7 @@ struct xsm_operations {
     int (*ioport_mapping) (struct domain *d, uint32_t s, uint32_t e, uint8_t allow);
     int (*pmu_op) (struct domain *d, unsigned int op);
     int (*dm_op) (struct domain *d);
+    int (*iommu_op) (struct domain *d);
 #endif
     int (*xen_version) (uint32_t cmd);
     int (*domain_resource_map) (struct domain *d);
@@ -686,6 +687,11 @@ static inline int xsm_dm_op(xsm_default_t def, struct domain *d)
     return xsm_ops->dm_op(d);
 }
 
+static inline int xsm_iommu_op(xsm_default_t def, struct domain *d)
+{
+    return xsm_ops->iommu_op(d);
+}
+
 #endif /* CONFIG_X86 */
 
 static inline int xsm_xen_version (xsm_default_t def, uint32_t op)
xen/xsm/dummy.c +1 -0
index 6e751199ee..02512a1566 100644
@@ -155,6 +155,7 @@ void __init xsm_fixup_ops (struct xsm_operations *ops)
     set_to_dummy_if_null(ops, ioport_mapping);
     set_to_dummy_if_null(ops, pmu_op);
     set_to_dummy_if_null(ops, dm_op);
+    set_to_dummy_if_null(ops, iommu_op);
 #endif
     set_to_dummy_if_null(ops, xen_version);
     set_to_dummy_if_null(ops, domain_resource_map);
xen/xsm/flask/hooks.c +6 -0
index 78bc32602e..a04786106f 100644
@@ -1672,6 +1672,11 @@ static int flask_dm_op(struct domain *d)
     return current_has_perm(d, SECCLASS_HVM, HVM__DM);
 }
 
+static int flask_iommu_op(struct domain *d)
+{
+    return current_has_perm(d, SECCLASS_RESOURCE, RESOURCE__CONTROL_IOMMU);
+}
+
 #endif /* CONFIG_X86 */
 
 static int flask_xen_version (uint32_t op)
@@ -1850,6 +1855,7 @@ static struct xsm_operations flask_ops = {
     .ioport_mapping = flask_ioport_mapping,
     .pmu_op = flask_pmu_op,
     .dm_op = flask_dm_op,
+    .iommu_op = flask_iommu_op,
 #endif
     .xen_version = flask_xen_version,
     .domain_resource_map = flask_domain_resource_map,
xen/xsm/flask/policy/access_vectors +2 -0
index c5d85486d0..0c894b733e 100644
@@ -473,6 +473,8 @@ class resource
 # checked for PHYSDEVOP_setup_gsi (target IRQ)
 # checked for PHYSDEVOP_pci_mmcfg_reserved (target xen_t)
     setup
+# checked for IOMMU_OP
+    control_iommu
 }
 
 # Class security describes the FLASK security server itself; these operations
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#24 lore raw

Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

xen/arch/x86/iommu_op.c line 64

Quoted 7 lines · Paul Durrant
xen/arch/x86/iommu_op.c new file +7 -0
+long do_iommu_op(unsigned int nr_bufs,
+                 XEN_GUEST_HANDLE_PARAM(xen_iommu_op_buf_t) bufs)
+{
+    unsigned int i;
+    long rc;
+
+    rc = xsm_iommu_op(XSM_PRIV, current->domain);

My only comment here is, doesn't this mean that XSM can only provide "yes/no" functionality for this entire hypercall, rather than being able to enable or disable individual operations?

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#25 lore raw

Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

Quoted 14 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 11 July 2018 10:10
To: Paul Durrant <[email protected]>; [email protected]
Cc: Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Jan Beulich <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Daniel De Graaf <[email protected]>
Subject: Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

On 07/07/2018 12:05 PM, Paul Durrant wrote:

Quoted 11 lines

xen/arch/x86/iommu_op.c line 64

Quoted 7 lines
xen/arch/x86/iommu_op.c new file +7 -0
+long do_iommu_op(unsigned int nr_bufs,
+                 XEN_GUEST_HANDLE_PARAM(xen_iommu_op_buf_t) bufs)
+{
+    unsigned int i;
+    long rc;
+
+    rc = xsm_iommu_op(XSM_PRIV, current->domain);

My only comment here is, doesn't this mean that XSM can only provide "yes/no" functionality for this entire hypercall, rather than being able to enable or disable individual operations?

That's true. I had not really considered the need to have finer grained control, but since you queried it, I'll re-work it so that it is possible to veto individual ops.

Paul

Quoted 2 lines

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#26 lore raw

Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

In reply to Paul Durrant
Quoted 7 lines · Paul Durrant

This patch introduces the boilerplate for a new hypercall to allow a domain to control IOMMU mappings for its own pages.
Whilst there is duplication of code between the native and compat entry points which appears ripe for some form of combination, I think it is better to maintain the separation as-is because the compat entry point will necessarily gain complexity in subsequent patches.

Since this is a new op, why is there compatibility issue? Can we do away with the compat entry point all together?

I haven't read the rest of this series, maybe it will become clearer later.

Wei.

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#27 lore raw

Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

Quoted 23 lines

-----Original Message-----
From: Wei Liu [mailto:[email protected]]
Sent: 16 July 2018 15:14
To: Paul Durrant <[email protected]>
Cc: [email protected]; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Jan Beulich <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>; Daniel De Graaf <[email protected]>
Subject: Re: [PATCH v2 06/13] public / x86: introduce __HYPERCALL_iommu_op

Quoted 7 lines · Paul Durrant

This patch introduces the boilerplate for a new hypercall to allow a domain to control IOMMU mappings for its own pages.
Whilst there is duplication of code between the native and compat entry points which appears ripe for some form of combination, I think it is better to maintain the separation as-is because the compat entry point will necessarily gain complexity in subsequent patches.

Quoted 5 lines

Since this is a new op, why is there compatibility issue? Can we do away with the compat entry point all together?

I haven't read the rest of this series, maybe it will become clearer later.

Look at the patch adding the query op... that's when it becomes necessary.

Paul

Quoted 2 lines

Wei.

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#28 lore raw

PATCH v2 7/13 iommu: track reserved ranges using a rangeset

In reply to Paul Durrant

Ranges that should be considered reserved in the IOMMU are not necessarily limited to RMRRs. If iommu_inclusive_mapping is set then any frame number falling within an E820 reserved region should also be considered as reserved in the IOMMU.
This patch adds a rangeset to the domain_iommu structure that is then used to track all reserved ranges. This will be needed by a subsequent patch to test whether it is safe to modify a particular IOMMU entry.

---
Cc: Jan Beulich <[email protected]>
Cc: Kevin Tian <[email protected]>

v2:
- New in v2.

---
 xen/drivers/passthrough/iommu.c       | 10 +++++++++-
 xen/drivers/passthrough/vtd/iommu.c   | 20 +++++++++++++-------
 xen/drivers/passthrough/vtd/x86/vtd.c | 24 ++++++++++++++++++++----
 xen/include/xen/iommu.h               |  6 ++++++
 4 files changed, 48 insertions(+), 12 deletions(-)
xen/drivers/passthrough/iommu.c +9 -1
index 7e59211a59..3fbd3ebaf6 100644
@@ -147,6 +147,10 @@ int iommu_domain_init(struct domain *d)
     if ( !iommu_enabled )
         return 0;
 
+    hd->reserved_ranges = rangeset_new(d, NULL, 0);
+    if ( !hd->reserved_ranges )
+        return -ENOMEM;
+
     hd->platform_ops = iommu_get_ops();
     return hd->platform_ops->init(d);
 }
@@ -249,12 +253,16 @@ int iommu_construct(struct domain *d)
 
 void iommu_domain_destroy(struct domain *d)
 {
-    if ( !iommu_enabled || !dom_iommu(d)->platform_ops )
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
         return;
 
     iommu_teardown(d);
 
     arch_iommu_domain_destroy(d);
+
+    rangeset_destroy(hd->reserved_ranges);
 }
 
 int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
xen/drivers/passthrough/vtd/iommu.c +13 -7
index 2aa97c9efe..7cd3813b9f 100644
@@ -1911,6 +1911,7 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
     unsigned long end_pfn = PAGE_ALIGN_4K(rmrr->end_address) >> PAGE_SHIFT_4K;
     struct mapped_rmrr *mrmrr;
     struct domain_iommu *hd = dom_iommu(d);
+    int err = 0;
 
     ASSERT(pcidevs_locked());
     ASSERT(rmrr->base_address < rmrr->end_address);
@@ -1924,8 +1925,6 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
         if ( mrmrr->base == rmrr->base_address &&
              mrmrr->end == rmrr->end_address )
         {
-            int ret = 0;
-
             if ( map )
             {
                 ++mrmrr->count;
@@ -1935,28 +1934,35 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
             if ( --mrmrr->count )
                 return 0;
 
-            while ( base_pfn < end_pfn )
+            err = rangeset_remove_range(hd->reserved_ranges,
+                                        base_pfn, end_pfn);
+            while ( !err && base_pfn < end_pfn )
             {
                 if ( clear_identity_p2m_entry(d, base_pfn) )
-                    ret = -ENXIO;
+                    err = -ENXIO;
+
                 base_pfn++;
             }
 
             list_del(&mrmrr->list);
             xfree(mrmrr);
-            return ret;
+            return err;
         }
     }
 
     if ( !map )
         return -ENOENT;
 
+    err = rangeset_add_range(hd->reserved_ranges, base_pfn, end_pfn);
+    if ( err )
+        return err;
+
     while ( base_pfn < end_pfn )
     {
-        int err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
-
+        err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
         if ( err )
             return err;
+
         base_pfn++;
     }
 
xen/drivers/passthrough/vtd/x86/vtd.c +20 -4
index dc37dce4b6..d65d3c60fb 100644
@@ -155,16 +155,32 @@ void __hwdom_init vtd_set_hwdom_mapping(struct domain *d)
         tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
         for ( j = 0; j < tmp; j++ )
         {
-            int ret = iommu_map_page(d, _bfn(pfn * tmp + j),
-                                     _mfn(pfn * tmp + j),
-                                     IOMMUF_readable|IOMMUF_writable);
+            bfn_t bfn = _bfn(pfn * tmp + j);
+            mfn_t mfn = _mfn(pfn * tmp + j);
 
+            int ret = iommu_map_page(d, bfn, mfn,
+                                     IOMMUF_readable | IOMMUF_writable);
             if ( !rc )
                rc = ret;
+
+            /*
+             * The only reason a reserved page would be mapped is that
+             * iommu_inclusive_mapping is set, in which case the BFN
+             * needs to be marked as reserved in the IOMMU.
+             */
+            if ( page_is_ram_type(pfn, RAM_TYPE_RESERVED) )
+            {
+                ASSERT(iommu_inclusive_mapping);
+
+                ret = rangeset_add_singleton(dom_iommu(d)->reserved_ranges,
+                                             bfn_x(bfn));
+                if ( !rc )
+                    rc = ret;
+            }
         }
 
         if ( rc )
-            printk(XENLOG_WARNING VTDPREFIX " d%d: IOMMU mapping failed: %d\n",
+            printk(XENLOG_WARNING VTDPREFIX " d%d: IOMMU mapping/reservation failed: %d\n",
                    d->domain_id, rc);
 
         if (!(i & (0xfffff >> (PAGE_SHIFT - PAGE_SHIFT_4K))))
xen/include/xen/iommu.h +6 -0
index 624784fec8..cc0be81b4e 100644
@@ -122,6 +122,12 @@ struct domain_iommu {
 
     /* Features supported by the IOMMU */
     DECLARE_BITMAP(features, IOMMU_FEAT_count);
+
+    /*
+     * BFN ranges that are reserved in the domain IOMMU mappings and
+     * must not be modified after initialization.
+     */
+    struct rangeset *reserved_ranges;
 };
 
 #define dom_iommu(d)              (&(d)->iommu)
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
Reviewed-by on v2 7/13
#29 lore raw

Re: [PATCH v2 07/13] iommu: track reserved ranges using a rangeset

Quoted 122 lines · Paul Durrant

Ranges that should be considered reserved in the IOMMU are not necessarily limited to RMRRs. If iommu_inclusive_mapping is set then any frame number falling within an E820 reserved region should also be considered as reserved in the IOMMU.
This patch adds a rangeset to the domain_iommu structure that is then used to track all reserved ranges. This will be needed by a subsequent patch to test whether it is safe to modify a particular IOMMU entry.

---
Cc: Jan Beulich <[email protected]>
Cc: Kevin Tian <[email protected]>

v2:
- New in v2.

---
 xen/drivers/passthrough/iommu.c       | 10 +++++++++-
 xen/drivers/passthrough/vtd/iommu.c   | 20 +++++++++++++-------
 xen/drivers/passthrough/vtd/x86/vtd.c | 24 ++++++++++++++++++++----
 xen/include/xen/iommu.h               |  6 ++++++
 4 files changed, 48 insertions(+), 12 deletions(-)
xen/drivers/passthrough/iommu.c +9 -1
index 7e59211a59..3fbd3ebaf6 100644
@@ -147,6 +147,10 @@ int iommu_domain_init(struct domain *d)
     if ( !iommu_enabled )
         return 0;
 
+    hd->reserved_ranges = rangeset_new(d, NULL, 0);
+    if ( !hd->reserved_ranges )
+        return -ENOMEM;
+
     hd->platform_ops = iommu_get_ops();
     return hd->platform_ops->init(d);
 }
@@ -249,12 +253,16 @@ int iommu_construct(struct domain *d)
 
 void iommu_domain_destroy(struct domain *d)
 {
-    if ( !iommu_enabled || !dom_iommu(d)->platform_ops )
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
         return;
 
     iommu_teardown(d);
 
     arch_iommu_domain_destroy(d);
+
+    rangeset_destroy(hd->reserved_ranges);
 }
 
 int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
xen/drivers/passthrough/vtd/iommu.c +13 -7
index 2aa97c9efe..7cd3813b9f 100644
@@ -1911,6 +1911,7 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
     unsigned long end_pfn = PAGE_ALIGN_4K(rmrr->end_address) >> PAGE_SHIFT_4K;
     struct mapped_rmrr *mrmrr;
     struct domain_iommu *hd = dom_iommu(d);
+    int err = 0;
 
     ASSERT(pcidevs_locked());
     ASSERT(rmrr->base_address < rmrr->end_address);
@@ -1924,8 +1925,6 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
         if ( mrmrr->base == rmrr->base_address &&
              mrmrr->end == rmrr->end_address )
         {
-            int ret = 0;
-
             if ( map )
             {
                 ++mrmrr->count;
@@ -1935,28 +1934,35 @@ static int rmrr_identity_mapping(struct domain *d, bool_t map,
             if ( --mrmrr->count )
                 return 0;
 
-            while ( base_pfn < end_pfn )
+            err = rangeset_remove_range(hd->reserved_ranges,
+                                        base_pfn, end_pfn);
+            while ( !err && base_pfn < end_pfn )
             {
                 if ( clear_identity_p2m_entry(d, base_pfn) )
-                    ret = -ENXIO;
+                    err = -ENXIO;
+
                 base_pfn++;
             }
 
             list_del(&mrmrr->list);
             xfree(mrmrr);
-            return ret;
+            return err;
         }
     }
 
     if ( !map )
         return -ENOENT;
 
+    err = rangeset_add_range(hd->reserved_ranges, base_pfn, end_pfn);
+    if ( err )
+        return err;
+
     while ( base_pfn < end_pfn )
     {
-        int err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
-
+        err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
         if ( err )
             return err;
+
         base_pfn++;
     }
 
xen/drivers/passthrough/vtd/x86/vtd.c +0 -0
index dc37dce4b6..d65d3c60fb 100644

xen/drivers/passthrough/vtd/x86/vtd.c void __hwdom_init vtd_set_hwdom_mapping(struct domain *d) · line 159

Quoted 8 lines
xen/drivers/passthrough/vtd/x86/vtd.c +2 -3
@@ -155,16 +155,32 @@ void __hwdom_init vtd_set_hwdom_mapping(struct domain *d)
         tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
         for ( j = 0; j < tmp; j++ )
         {
-            int ret = iommu_map_page(d, _bfn(pfn * tmp + j),
-                                     _mfn(pfn * tmp + j),
-                                     IOMMUF_readable|IOMMUF_writable);
+            bfn_t bfn = _bfn(pfn * tmp + j);
+            mfn_t mfn = _mfn(pfn * tmp + j);

Everything looks good to me except for this; but as this isn't my area of maintainership I'll just register my opinion and let the maintainers decide.

Everything else:

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#30 lore raw

Re: [PATCH v2 07/13] iommu: track reserved ranges using a rangeset

Quoted 154 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of George Dunlap
Sent: 11 July 2018 10:16
To: Paul Durrant <[email protected]>
Cc: xen-devel <[email protected]>; Kevin Tian <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH v2 07/13] iommu: track reserved ranges using a rangeset

Quoted 1 line · Paul Durrant

Ranges that should be considered reserved in the IOMMU are not

necessarily

Quoted 1 line

limited to RMRRs. If iommu_inclusive_mapping is set then any frame

number

Quoted 3 lines

falling within an E820 reserved region should also be considered as reserved in the IOMMU.
This patch adds a rangeset to the domain_iommu structure that is then

used

Quoted 10 lines

to track all reserved ranges. This will be needed by a subsequent patch to test whether it is safe to modify a particular IOMMU entry.

---
Cc: Jan Beulich <[email protected]>
Cc: Kevin Tian <[email protected]>

v2:
- New in v2.

Review

Quoted 8 lines
---
 xen/drivers/passthrough/iommu.c       | 10 +++++++++-
 xen/drivers/passthrough/vtd/iommu.c   | 20 +++++++++++++-------
 xen/drivers/passthrough/vtd/x86/vtd.c | 24 ++++++++++++++++++++----
 xen/include/xen/iommu.h               |  6 ++++++
 4 files changed, 48 insertions(+), 12 deletions(-)
(quoted diff) +0 -0

b/xen/drivers/passthrough/iommu.c

Review

Quoted 33 lines

index 7e59211a59..3fbd3ebaf6 100644

xen/drivers/passthrough/iommu.c +9 -1
@@ -147,6 +147,10 @@ int iommu_domain_init(struct domain *d)
     if ( !iommu_enabled )
         return 0;
 
+    hd->reserved_ranges = rangeset_new(d, NULL, 0);
+    if ( !hd->reserved_ranges )
+        return -ENOMEM;
+
     hd->platform_ops = iommu_get_ops();
     return hd->platform_ops->init(d);
 }
@@ -249,12 +253,16 @@ int iommu_construct(struct domain *d)
 
 void iommu_domain_destroy(struct domain *d)
 {
-    if ( !iommu_enabled || !dom_iommu(d)->platform_ops )
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
         return;
 
     iommu_teardown(d);
 
     arch_iommu_domain_destroy(d);
+
+    rangeset_destroy(hd->reserved_ranges);
 }
 
 int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
(quoted diff) +0 -0

b/xen/drivers/passthrough/vtd/iommu.c

xen/drivers/passthrough/vtd/iommu.c static int rmrr_identity_mapping(struct domain

Quoted 4 lines

index 2aa97c9efe..7cd3813b9f 100644

xen/drivers/passthrough/vtd/iommu.c +0 -0
@@ -1911,6 +1911,7 @@ static int rmrr_identity_mapping(struct domain

*d, bool_t map,

xen/drivers/passthrough/vtd/iommu.c static int rmrr_identity_mapping(struct domain · line 1911

Quoted 1 line
xen/drivers/passthrough/vtd/iommu.c +0 -0
     unsigned long end_pfn = PAGE_ALIGN_4K(rmrr->end_address) >>

PAGE_SHIFT_4K;

xen/drivers/passthrough/vtd/iommu.c static int rmrr_identity_mapping(struct domain

Quoted 7 lines
xen/drivers/passthrough/vtd/iommu.c +1 -0
     struct mapped_rmrr *mrmrr;
     struct domain_iommu *hd = dom_iommu(d);
+    int err = 0;
 
     ASSERT(pcidevs_locked());
     ASSERT(rmrr->base_address < rmrr->end_address);
@@ -1924,8 +1925,6 @@ static int rmrr_identity_mapping(struct domain

*d, bool_t map,

xen/drivers/passthrough/vtd/iommu.c static int rmrr_identity_mapping(struct domain

Quoted 9 lines
xen/drivers/passthrough/vtd/iommu.c +0 -2
         if ( mrmrr->base == rmrr->base_address &&
              mrmrr->end == rmrr->end_address )
         {
-            int ret = 0;
-
             if ( map )
             {
                 ++mrmrr->count;
@@ -1935,28 +1934,35 @@ static int rmrr_identity_mapping(struct domain

*d, bool_t map,

Review

Quoted 41 lines
xen/drivers/passthrough/vtd/iommu.c +12 -5
             if ( --mrmrr->count )
                 return 0;
 
-            while ( base_pfn < end_pfn )
+            err = rangeset_remove_range(hd->reserved_ranges,
+                                        base_pfn, end_pfn);
+            while ( !err && base_pfn < end_pfn )
             {
                 if ( clear_identity_p2m_entry(d, base_pfn) )
-                    ret = -ENXIO;
+                    err = -ENXIO;
+
                 base_pfn++;
             }
 
             list_del(&mrmrr->list);
             xfree(mrmrr);
-            return ret;
+            return err;
         }
     }
 
     if ( !map )
         return -ENOENT;
 
+    err = rangeset_add_range(hd->reserved_ranges, base_pfn, end_pfn);
+    if ( err )
+        return err;
+
     while ( base_pfn < end_pfn )
     {
-        int err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
-
+        err = set_identity_p2m_entry(d, base_pfn, p2m_access_rw, flag);
         if ( err )
             return err;
+
         base_pfn++;
     }
 
(quoted diff) +0 -0

b/xen/drivers/passthrough/vtd/x86/vtd.c

xen/drivers/passthrough/vtd/x86/vtd.c void __hwdom_init

Quoted 4 lines

index dc37dce4b6..d65d3c60fb 100644

xen/drivers/passthrough/vtd/x86/vtd.c +0 -0
@@ -155,16 +155,32 @@ void __hwdom_init

vtd_set_hwdom_mapping(struct domain *d)

xen/drivers/passthrough/vtd/x86/vtd.c void __hwdom_init vtd_set_hwdom_mapping(struct domain *d) · line 159

Quoted 8 lines
xen/drivers/passthrough/vtd/x86/vtd.c +2 -3
         tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
         for ( j = 0; j < tmp; j++ )
         {
-            int ret = iommu_map_page(d, _bfn(pfn * tmp + j),
-                                     _mfn(pfn * tmp + j),
-                                     IOMMUF_readable|IOMMUF_writable);
+            bfn_t bfn = _bfn(pfn * tmp + j);
+            mfn_t mfn = _mfn(pfn * tmp + j);
Quoted 4 lines

Everything looks good to me except for this; but as this isn't my area of maintainership I'll just register my opinion and let the maintainers decide.

Oh, that looks like it should have been in a previous patch.

Quoted 4 lines

Everything else:

Thanks,

Paul

Quoted 1 line

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#31 lore raw

PATCH v2 8/13 x86: add iommu_op to query reserved ranges

In reply to Paul Durrant

This patch adds an iommu_op to allow the domain IOMMU reserved ranges to be queried by the guest.

NOTE: The number of reserved ranges is determined by system firmware, in
conjunction with Xen command line options, and is expected to be small. Thus there is no pre-emption check within the operation.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- Re-implemented for v2 based on new rangeset.

---
 xen/arch/x86/iommu_op.c       | 167 ++++++++++++++++++++++++++++++++++++++++--
 xen/include/public/iommu_op.h |  39 ++++++++++
 xen/include/xlat.lst          |   2 +
 3 files changed, 202 insertions(+), 6 deletions(-)
xen/arch/x86/iommu_op.c +161 -6
index 280d0c967d..75c465a46a 100644
@@ -23,10 +23,68 @@
 #include <xen/guest_access.h>
 #include <xen/hypercall.h>
 
+struct get_reserved_ctxt {
+    unsigned int max_entries;
+    unsigned int nr_entries;
+    XEN_GUEST_HANDLE(xen_iommu_reserved_range_t) ranges;
+};
+
+static int get_reserved(unsigned long s, unsigned long e, void *arg)
+{
+    struct get_reserved_ctxt *ctxt = arg;
+
+    if ( ctxt->nr_entries < ctxt->max_entries )
+    {
+        xen_iommu_reserved_range_t range = {
+            .start_bfn = s,
+            .nr_frames = e - s,
+        };
+
+        if ( copy_to_guest_offset(ctxt->ranges, ctxt->nr_entries, &range,
+                                  1) )
+            return -EFAULT;
+    }
+
+    ctxt->nr_entries++;
+    return 0;
+}
+
+static int iommu_op_query_reserved(struct xen_iommu_op_query_reserved *op)
+{
+    struct domain *currd = current->domain;
+    struct domain_iommu *iommu = dom_iommu(currd);
+    struct get_reserved_ctxt ctxt = {
+        .max_entries = op->nr_entries,
+        .ranges = op->ranges,
+    };
+    int rc;
+
+    if ( op->pad )
+        return -EINVAL;
+
+    rc = rangeset_report_ranges(iommu->reserved_ranges, 0, ~0ul,
+                                get_reserved, &ctxt);
+    if ( rc )
+        return rc;
+
+    /* Pass back the actual number of reserved ranges */
+    op->nr_entries = ctxt.nr_entries;
+
+    if ( !guest_handle_is_null(ctxt.ranges) &&
+         ctxt.nr_entries > ctxt.max_entries )
+        return -ENOBUFS;
+
+    return 0;
+}
+
 static void iommu_op(xen_iommu_op_t *op)
 {
     switch ( op->op )
     {
+    case XEN_IOMMUOP_query_reserved:
+        op->status = iommu_op_query_reserved(&op->u.query_reserved);
+        break;
+
     default:
         op->status = -EOPNOTSUPP;
         break;
@@ -35,17 +93,33 @@ static void iommu_op(xen_iommu_op_t *op)
 
 int do_one_iommu_op(xen_iommu_op_buf_t *buf)
 {
-    xen_iommu_op_t op;
+    xen_iommu_op_t op = {};
+    size_t offset;
+    static const size_t op_size[] = {
+        [XEN_IOMMUOP_query_reserved] = sizeof(struct xen_iommu_op_query_reserved),
+    };
+
+    offset = offsetof(struct xen_iommu_op, u);
 
-    if ( buf->size < sizeof(op) )
+    if ( buf->size < offset )
         return -EFAULT;
 
-    if ( copy_from_guest((void *)&op, buf->h, sizeof(op)) )
+    if ( copy_from_guest((void *)&op, buf->h, offset) )
         return -EFAULT;
 
     if ( op.pad )
         return -EINVAL;
 
+    if ( op.op >= ARRAY_SIZE(op_size) )
+        return -EOPNOTSUPP;
+
+    if ( buf->size < offset + op_size[op.op] )
+        return -EFAULT;
+
+    if ( copy_from_guest_offset((void *)&op.u, buf->h, offset,
+                                op_size[op.op]) )
+        return -EFAULT;
+
     iommu_op(&op);
 
     if ( __copy_field_to_guest(guest_handle_cast(buf->h, xen_iommu_op_t),
@@ -101,24 +175,105 @@ long do_iommu_op(unsigned int nr_bufs,
 
 int compat_one_iommu_op(compat_iommu_op_buf_t *buf)
 {
-    compat_iommu_op_t cmp;
+    compat_iommu_op_t cmp = {};
+    size_t offset;
+    static const size_t op_size[] = {
+        [XEN_IOMMUOP_query_reserved] = sizeof(struct compat_iommu_op_query_reserved),
+    };
     xen_iommu_op_t nat;
+    unsigned int u;
+    int32_t status;
 
-    if ( buf->size < sizeof(cmp) )
+    offset = offsetof(struct compat_iommu_op, u);
+
+    if ( buf->size < offset )
         return -EFAULT;
 
-    if ( copy_from_compat((void *)&cmp, buf->h, sizeof(cmp)) )
+    if ( copy_from_compat((void *)&cmp, buf->h, offset) )
         return -EFAULT;
 
     if ( cmp.pad )
         return -EINVAL;
 
+    if ( cmp.op >= ARRAY_SIZE(op_size) )
+        return -EOPNOTSUPP;
+
+    if ( buf->size < offset + op_size[cmp.op] )
+        return -EFAULT;
+
+    if ( copy_from_compat_offset((void *)&cmp.u, buf->h, offset,
+                                 op_size[cmp.op]) )
+        return -EFAULT;
+
+    /*
+     * The xlat magic doesn't quite know how to handle the union so
+     * we need to fix things up here.
+     */
+#define XLAT_iommu_op_u_query_reserved XEN_IOMMUOP_query_reserved
+    u = cmp.op;
+
+#define XLAT_iommu_op_query_reserved_HNDL_ranges(_d_, _s_)            \
+    do                                                                \
+    {                                                                 \
+        if ( !compat_handle_is_null((_s_)->ranges) )                  \
+        {                                                             \
+            unsigned int *nr_entries = COMPAT_ARG_XLAT_VIRT_BASE;     \
+            xen_iommu_reserved_range_t *ranges =                      \
+                (void *)(nr_entries + 1);                             \
+                                                                      \
+            if ( sizeof(*nr_entries) +                                \
+                 (sizeof(*ranges) * (_s_)->nr_entries) >              \
+                 COMPAT_ARG_XLAT_SIZE )                               \
+                return -E2BIG;                                        \
+                                                                      \
+            *nr_entries = (_s_)->nr_entries;                          \
+            set_xen_guest_handle((_d_)->ranges, ranges);              \
+        }                                                             \
+        else                                                          \
+            set_xen_guest_handle((_d_)->ranges, NULL);                \
+    } while (false)
+
     XLAT_iommu_op(&nat, &cmp);
 
+#undef XLAT_iommu_op_query_reserved_HNDL_ranges
+
     iommu_op(&nat);
 
+    status = nat.status;
+
+#define XLAT_iommu_op_query_reserved_HNDL_ranges(_d_, _s_)            \
+    do                                                                \
+    {                                                                 \
+        if ( !compat_handle_is_null((_d_)->ranges) )                  \
+        {                                                             \
+            unsigned int *nr_entries = COMPAT_ARG_XLAT_VIRT_BASE;     \
+            xen_iommu_reserved_range_t *ranges =                      \
+                (void *)(nr_entries + 1);                             \
+            unsigned int j;                                           \
+                                                                      \
+            for ( j = 0;                                              \
+                  j < min_t(unsigned int, (_d_)->nr_entries,          \
+                            *nr_entries);                             \
+                  j++ )                                               \
+            {                                                         \
+                compat_iommu_reserved_range_t range;                  \
+                                                                      \
+                XLAT_iommu_reserved_range(&range, &ranges[j]);        \
+                                                                      \
+                if ( __copy_to_compat_offset((_d_)->ranges, j,        \
+                                             &range, 1) )             \
+                    status = -EFAULT;                                 \
+            }                                                         \
+        }                                                             \
+    } while (false)
+
     XLAT_iommu_op(&cmp, &nat);
 
+    /* status will have been modified if __copy_to_compat_offset() failed */
+    cmp.status = status;
+
+#undef XLAT_iommu_op_query_reserved_HNDL_ranges
+
     if ( __copy_field_to_compat(compat_handle_cast(buf->h,
                                                    compat_iommu_op_t),
                                 &cmp, status) )
xen/include/public/iommu_op.h +39 -0
index c3b68f665a..02213c12a4 100644
@@ -25,11 +25,50 @@
 
 #include "xen.h"
 
+typedef unsigned long xen_bfn_t;
+
+/* Structure describing a single range reserved in the IOMMU */
+struct xen_iommu_reserved_range {
+    xen_bfn_t start_bfn;
+    unsigned int nr_frames;
+    unsigned int pad;
+};
+typedef struct xen_iommu_reserved_range xen_iommu_reserved_range_t;
+DEFINE_XEN_GUEST_HANDLE(xen_iommu_reserved_range_t);
+
+/*
+ * XEN_IOMMUOP_query_reserved: Query ranges reserved in the IOMMU.
+ */
+#define XEN_IOMMUOP_query_reserved 1
+
+struct xen_iommu_op_query_reserved {
+    /*
+     * IN/OUT - On entry this is the number of entries available
+     *          in the ranges array below.
+     *          On exit this is the actual number of reserved ranges.
+     */
+    unsigned int nr_entries;
+    unsigned int pad;
+    /*
+     * OUT - This array is populated with reserved ranges. If it is
+     *       not sufficiently large then available entries are populated,
+     *       but the op status code will be set to -ENOBUFS.
+     *       It is permissable to set this to NULL if nr_entries is also
+     *       set to zero. In this case, on exit, nr_entries will still be
+     *       set to the actual number of reserved ranges but the status
+     *       code will be set to zero.
+     */
+    XEN_GUEST_HANDLE(xen_iommu_reserved_range_t) ranges;
+};
+
 struct xen_iommu_op {
     uint16_t op;    /* op type */
     uint16_t pad;
     int32_t status; /* op completion status: */
                     /* 0 for success otherwise, negative errno */
+    union {
+        struct xen_iommu_op_query_reserved query_reserved;
+    } u;
 };
 typedef struct xen_iommu_op xen_iommu_op_t;
 DEFINE_XEN_GUEST_HANDLE(xen_iommu_op_t);
xen/include/xlat.lst +2 -0
index 3b15c18c4e..93bcf4b4d0 100644
@@ -79,6 +79,8 @@
 ?	vcpu_hvm_x86_64			hvm/hvm_vcpu.h
 !	iommu_op			iommu_op.h
 !	iommu_op_buf			iommu_op.h
+!	iommu_op_query_reserved		iommu_op.h
+!	iommu_reserved_range		iommu_op.h
 ?	kexec_exec			kexec.h
 !	kexec_image			kexec.h
 !	kexec_range			kexec.h
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#32 lore raw

Re: [PATCH v2 08/13] x86: add iommu_op to query reserved ranges

Quoted 23 lines · Paul Durrant
xen/arch/x86/iommu_op.c +9 -3
@@ -35,17 +93,33 @@ static void iommu_op(xen_iommu_op_t *op)
 
 int do_one_iommu_op(xen_iommu_op_buf_t *buf)
 {
-    xen_iommu_op_t op;
+    xen_iommu_op_t op = {};
+    size_t offset;
+    static const size_t op_size[] = {
+        [XEN_IOMMUOP_query_reserved] = sizeof(struct xen_iommu_op_query_reserved),
+    };
+
+    offset = offsetof(struct xen_iommu_op, u);
 
-    if ( buf->size < sizeof(op) )
+    if ( buf->size < offset )
         return -EFAULT;
 
-    if ( copy_from_guest((void *)&op, buf->h, sizeof(op)) )
+    if ( copy_from_guest((void *)&op, buf->h, offset) )
         return -EFAULT;
 
     if ( op.pad )
         return -EINVAL;

xen/arch/x86/iommu_op.c static void iommu_op(xen_iommu_op_t *op) · line 121

Quoted 10 lines
xen/arch/x86/iommu_op.c +9 -0
 
+    if ( op.op >= ARRAY_SIZE(op_size) )
+        return -EOPNOTSUPP;
+
+    if ( buf->size < offset + op_size[op.op] )
+        return -EFAULT;
+
+    if ( copy_from_guest_offset((void *)&op.u, buf->h, offset,
+                                op_size[op.op]) )
+        return -EFAULT;

This looks like part of a potential SP1 gadget, so this needs to use array_index_nospec().

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#33 lore raw

Re: [PATCH v2 08/13] x86: add iommu_op to query reserved ranges

Quoted 38 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 11 July 2018 11:34
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH v2 08/13] x86: add iommu_op to query reserved ranges

xen/arch/x86/iommu_op.c static void iommu_op(xen_iommu_op_t *op) · line 99

Quoted 9 lines · Paul Durrant
xen/arch/x86/iommu_op.c +4 -1
@@ -35,17 +93,33 @@ static void iommu_op(xen_iommu_op_t *op)
 
 int do_one_iommu_op(xen_iommu_op_buf_t *buf)
 {
-    xen_iommu_op_t op;
+    xen_iommu_op_t op = {};
+    size_t offset;
+    static const size_t op_size[] = {
+        [XEN_IOMMUOP_query_reserved] = sizeof(struct

xen_iommu_op_query_reserved),

Quoted 15 lines
xen/arch/x86/iommu_op.c +5 -2
+    };
+
+    offset = offsetof(struct xen_iommu_op, u);
 
-    if ( buf->size < sizeof(op) )
+    if ( buf->size < offset )
         return -EFAULT;
 
-    if ( copy_from_guest((void *)&op, buf->h, sizeof(op)) )
+    if ( copy_from_guest((void *)&op, buf->h, offset) )
         return -EFAULT;
 
     if ( op.pad )
         return -EINVAL;
 
Quoted 13 lines

xen/arch/x86/iommu_op.c static void iommu_op(xen_iommu_op_t *op) · line 121

Quoted 9 lines
xen/arch/x86/iommu_op.c +9 -0
+    if ( op.op >= ARRAY_SIZE(op_size) )
+        return -EOPNOTSUPP;
+
+    if ( buf->size < offset + op_size[op.op] )
+        return -EFAULT;
+
+    if ( copy_from_guest_offset((void *)&op.u, buf->h, offset,
+                                op_size[op.op]) )
+        return -EFAULT;

This looks like part of a potential SP1 gadget, so this needs to use array_index_nospec().

Ok. There is similar code in dm ops too so I'll have a look while I'm at it.

Paul

Quoted 1 line

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#34 lore raw

PATCH v2 9/13 vtd: add lookup_page method to iommu_ops

In reply to Paul Durrant

This patch adds a new method to the VT-d IOMMU implementation to find the MFN currently mapped by the specified BFN along with a wrapper function in generic IOMMU code to call the implementation if it exists.

This functionality will be used by a subsequent patch.

---
Cc: Kevin Tian <[email protected]>
Cc: Jan Beulich <[email protected]>

v2:
- Addressed some comments from Jan.

---
 xen/drivers/passthrough/iommu.c     | 11 ++++++++++
 xen/drivers/passthrough/vtd/iommu.c | 40 +++++++++++++++++++++++++++++++++++++
 xen/drivers/passthrough/vtd/iommu.h |  1 +
 xen/include/xen/iommu.h             |  4 ++++
 4 files changed, 56 insertions(+)
xen/drivers/passthrough/iommu.c +11 -0
index 3fbd3ebaf6..f25aa3f1d6 100644
@@ -306,6 +306,17 @@ int iommu_unmap_page(struct domain *d, bfn_t bfn)
     return rc;
 }
 
+int iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                      unsigned int *flags)
+{
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
+        return -EOPNOTSUPP;
+
+    return hd->platform_ops->lookup_page(d, bfn, mfn, flags);
+}
+
 static void iommu_free_pagetables(unsigned long unused)
 {
     do {
xen/drivers/passthrough/vtd/iommu.c +40 -0
index 7cd3813b9f..438bef670d 100644
@@ -1831,6 +1831,45 @@ static int __must_check intel_iommu_unmap_page(struct domain *d,
     return dma_pte_clear_one(d, bfn_to_baddr(bfn));
 }
 
+static int intel_iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                                   unsigned int *flags)
+{
+    struct domain_iommu *hd = dom_iommu(d);
+    struct dma_pte *page = NULL, *pte = NULL, val;
+    u64 pg_maddr;
+
+    spin_lock(&hd->arch.mapping_lock);
+
+    pg_maddr = addr_to_dma_page_maddr(d, bfn_to_baddr(bfn), 0);
+    if ( pg_maddr == 0 )
+    {
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;
+    }
+
+    page = map_vtd_domain_page(pg_maddr);
+    pte = page + (bfn_x(bfn) & LEVEL_MASK);
+    val = *pte;
+    if ( !dma_pte_present(val) )
+    {
+        unmap_vtd_domain_page(page);
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;
+    }
+
+    unmap_vtd_domain_page(page);
+    spin_unlock(&hd->arch.mapping_lock);
+
+    *mfn = maddr_to_mfn(dma_pte_addr(val));
+    *flags = 0;
+    if ( dma_pte_prot(val) & DMA_PTE_READ )
+        *flags |= IOMMUF_readable;
+    if ( dma_pte_prot(val) & DMA_PTE_WRITE )
+        *flags |= IOMMUF_writable;
+
+    return 0;
+}
+
 int iommu_pte_flush(struct domain *d, uint64_t bfn, uint64_t *pte,
                     int order, int present)
 {
@@ -2662,6 +2701,7 @@ const struct iommu_ops intel_iommu_ops = {
     .teardown = iommu_domain_teardown,
     .map_page = intel_iommu_map_page,
     .unmap_page = intel_iommu_unmap_page,
+    .lookup_page = intel_iommu_lookup_page,
     .free_page_table = iommu_free_page_table,
     .reassign_device = reassign_device_ownership,
     .get_device_group_id = intel_iommu_group_id,
xen/drivers/passthrough/vtd/iommu.h +1 -0
index 72c1a2e3cd..9de994308f 100644
@@ -272,6 +272,7 @@ struct dma_pte {
 #define dma_set_pte_prot(p, prot) do { \
         (p).val = ((p).val & ~DMA_PTE_PROT) | ((prot) & DMA_PTE_PROT); \
     } while (0)
+#define dma_pte_prot(p) ((p).val & DMA_PTE_PROT)
 #define dma_pte_addr(p) ((p).val & PADDR_MASK & PAGE_MASK_4K)
 #define dma_set_pte_addr(p, addr) do {\
             (p).val |= ((addr) & PAGE_MASK_4K); } while (0)
xen/include/xen/iommu.h +4 -0
index cc0be81b4e..7c5d46df81 100644
@@ -100,6 +100,8 @@ void iommu_teardown(struct domain *d);
 int __must_check iommu_map_page(struct domain *d, bfn_t bfn,
                                 mfn_t mfn, unsigned int flags);
 int __must_check iommu_unmap_page(struct domain *d, bfn_t bfn);
+int __must_check iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                                   unsigned int *flags);
 
 enum iommu_feature
 {
@@ -198,6 +200,8 @@ struct iommu_ops {
     int __must_check (*map_page)(struct domain *d, bfn_t bfn, mfn_t mfn,
                                  unsigned int flags);
     int __must_check (*unmap_page)(struct domain *d, bfn_t bfn);
+    int __must_check (*lookup_page)(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                                    unsigned int *flags);
     void (*free_page_table)(struct page_info *);
 #ifdef CONFIG_X86
     void (*update_ire_from_apic)(unsigned int apic, unsigned int reg, unsigned int value);
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#35 lore raw

Re: [PATCH v2 09/13] vtd: add lookup_page method to iommu_ops

Quoted 64 lines · Paul Durrant

This patch adds a new method to the VT-d IOMMU implementation to find the MFN currently mapped by the specified BFN along with a wrapper function in generic IOMMU code to call the implementation if it exists.

This functionality will be used by a subsequent patch.

---
Cc: Kevin Tian <[email protected]>
Cc: Jan Beulich <[email protected]>

v2:
- Addressed some comments from Jan.

---
 xen/drivers/passthrough/iommu.c     | 11 ++++++++++
 xen/drivers/passthrough/vtd/iommu.c | 40 +++++++++++++++++++++++++++++++++++++
 xen/drivers/passthrough/vtd/iommu.h |  1 +
 xen/include/xen/iommu.h             |  4 ++++
 4 files changed, 56 insertions(+)
xen/drivers/passthrough/iommu.c +11 -0
index 3fbd3ebaf6..f25aa3f1d6 100644
@@ -306,6 +306,17 @@ int iommu_unmap_page(struct domain *d, bfn_t bfn)
     return rc;
 }
 
+int iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                      unsigned int *flags)
+{
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
+        return -EOPNOTSUPP;
+
+    return hd->platform_ops->lookup_page(d, bfn, mfn, flags);
+}
+
 static void iommu_free_pagetables(unsigned long unused)
 {
     do {
xen/drivers/passthrough/vtd/iommu.c +14 -0
index 7cd3813b9f..438bef670d 100644
@@ -1831,6 +1831,45 @@ static int __must_check intel_iommu_unmap_page(struct domain *d,
     return dma_pte_clear_one(d, bfn_to_baddr(bfn));
 }
 
+static int intel_iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                                   unsigned int *flags)
+{
+    struct domain_iommu *hd = dom_iommu(d);
+    struct dma_pte *page = NULL, *pte = NULL, val;
+    u64 pg_maddr;
+
+    spin_lock(&hd->arch.mapping_lock);
+
+    pg_maddr = addr_to_dma_page_maddr(d, bfn_to_baddr(bfn), 0);
+    if ( pg_maddr == 0 )
+    {
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;

xen/drivers/passthrough/vtd/iommu.c static int __must_check intel_iommu_unmap_page(struct domain *d, · line 1857

Quoted 10 lines
xen/drivers/passthrough/vtd/iommu.c +10 -0
+    }
+
+    page = map_vtd_domain_page(pg_maddr);
+    pte = page + (bfn_x(bfn) & LEVEL_MASK);
+    val = *pte;
+    if ( !dma_pte_present(val) )
+    {
+        unmap_vtd_domain_page(page);
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;

Should this be -EEXIST? Or maybe return MFN_INVALID?

Also, could you do the unmap / unlock first and then do the check, rather than duplicating things?

xen/drivers/passthrough/vtd/iommu.c static int __must_check intel_iommu_unmap_page(struct domain *d, · line 1868

Quoted 11 lines
xen/drivers/passthrough/vtd/iommu.c +11 -0
+    }
+
+    unmap_vtd_domain_page(page);
+    spin_unlock(&hd->arch.mapping_lock);
+
+    *mfn = maddr_to_mfn(dma_pte_addr(val));
+    *flags = 0;
+    if ( dma_pte_prot(val) & DMA_PTE_READ )
+        *flags |= IOMMUF_readable;
+    if ( dma_pte_prot(val) & DMA_PTE_WRITE )
+        *flags |= IOMMUF_writable;

This is a bit strange, since all dma_pte_prot() does is return val & DMA_PTE_READ|DMA_PTE_WRITE. Would it make sense to implement dma_pte_read() / dma_pte_write() instead (like dma_pte_present())?

Everything else looks good to me.

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#36 lore raw

Re: [PATCH v2 09/13] vtd: add lookup_page method to iommu_ops

Quoted 88 lines

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of George Dunlap
Sent: 11 July 2018 11:52
To: Paul Durrant <[email protected]>
Cc: xen-devel <[email protected]>; Kevin Tian <[email protected]>; Jan Beulich <[email protected]>
Subject: Re: [Xen-devel] [PATCH v2 09/13] vtd: add lookup_page method to iommu_ops

Quoted 1 line · Paul Durrant

This patch adds a new method to the VT-d IOMMU implementation to find

the

Quoted 1 line

MFN currently mapped by the specified BFN along with a wrapper function

in

Quoted 9 lines

generic IOMMU code to call the implementation if it exists.

This functionality will be used by a subsequent patch.

---
Cc: Kevin Tian <[email protected]>
Cc: Jan Beulich <[email protected]>

Quoted 5 lines

v2:
- Addressed some comments from Jan.

---
 xen/drivers/passthrough/iommu.c     | 11 ++++++++++
 xen/drivers/passthrough/vtd/iommu.c | 40

+++++++++++++++++++++++++++++++++++++

Review

Quoted 5 lines
 xen/drivers/passthrough/vtd/iommu.h |  1 +
 xen/include/xen/iommu.h             |  4 ++++
 4 files changed, 56 insertions(+)
(quoted diff) +0 -0

b/xen/drivers/passthrough/iommu.c

xen/drivers/passthrough/iommu.c int iommu_unmap_page(struct domain *d, bfn_t

Quoted 4 lines

index 3fbd3ebaf6..f25aa3f1d6 100644

xen/drivers/passthrough/iommu.c +0 -0
@@ -306,6 +306,17 @@ int iommu_unmap_page(struct domain *d, bfn_t

bfn)

Review

Quoted 18 lines
xen/drivers/passthrough/iommu.c +11 -0
     return rc;
 }
 
+int iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t *mfn,
+                      unsigned int *flags)
+{
+    const struct domain_iommu *hd = dom_iommu(d);
+
+    if ( !iommu_enabled || !hd->platform_ops )
+        return -EOPNOTSUPP;
+
+    return hd->platform_ops->lookup_page(d, bfn, mfn, flags);
+}
+
 static void iommu_free_pagetables(unsigned long unused)
 {
     do {
(quoted diff) +0 -0

b/xen/drivers/passthrough/vtd/iommu.c

xen/drivers/passthrough/vtd/iommu.c static int __must_check

Quoted 4 lines

index 7cd3813b9f..438bef670d 100644

xen/drivers/passthrough/vtd/iommu.c +0 -0
@@ -1831,6 +1831,45 @@ static int __must_check

intel_iommu_unmap_page(struct domain *d,

xen/drivers/passthrough/vtd/iommu.c static int __must_check intel_iommu_unmap_page(struct domain *d,

Quoted 4 lines
xen/drivers/passthrough/vtd/iommu.c +1 -0
     return dma_pte_clear_one(d, bfn_to_baddr(bfn));
 }
 
+static int intel_iommu_lookup_page(struct domain *d, bfn_t bfn, mfn_t

*mfn,

Quoted 17 lines
xen/drivers/passthrough/vtd/iommu.c +17 -0
+                                   unsigned int *flags)
+{
+    struct domain_iommu *hd = dom_iommu(d);
+    struct dma_pte *page = NULL, *pte = NULL, val;
+    u64 pg_maddr;
+
+    spin_lock(&hd->arch.mapping_lock);
+
+    pg_maddr = addr_to_dma_page_maddr(d, bfn_to_baddr(bfn), 0);
+    if ( pg_maddr == 0 )
+    {
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;
+    }
+
+    page = map_vtd_domain_page(pg_maddr);
+    pte = page + (bfn_x(bfn) & LEVEL_MASK);
Quoted 9 lines

xen/drivers/passthrough/vtd/iommu.c static int __must_check intel_iommu_unmap_page(struct domain *d, · line 1857

Quoted 6 lines
xen/drivers/passthrough/vtd/iommu.c +6 -0
+    val = *pte;
+    if ( !dma_pte_present(val) )
+    {
+        unmap_vtd_domain_page(page);
+        spin_unlock(&hd->arch.mapping_lock);
+        return -ENOMEM;

Should this be -EEXIST? Or maybe return MFN_INVALID?

Do you mean ENOENT? EEXIST implies it exists but it shouldn't, right?

Quoted 3 lines

Also, could you do the unmap / unlock first and then do the check, rather than duplicating things?

Sure.

Quoted 7 lines
Quoted 7 lines
xen/drivers/passthrough/vtd/iommu.c +7 -0
+    }
+
+    unmap_vtd_domain_page(page);
+    spin_unlock(&hd->arch.mapping_lock);
+
+    *mfn = maddr_to_mfn(dma_pte_addr(val));
+    *flags = 0;
Quoted 9 lines

xen/drivers/passthrough/vtd/iommu.c static int __must_check intel_iommu_unmap_page(struct domain *d, · line 1868

Quoted 4 lines
xen/drivers/passthrough/vtd/iommu.c +4 -0
+    if ( dma_pte_prot(val) & DMA_PTE_READ )
+        *flags |= IOMMUF_readable;
+    if ( dma_pte_prot(val) & DMA_PTE_WRITE )
+        *flags |= IOMMUF_writable;

This is a bit strange, since all dma_pte_prot() does is return val & DMA_PTE_READ|DMA_PTE_WRITE. Would it make sense to implement dma_pte_read() / dma_pte_write() instead (like dma_pte_present())?

Yes, I can do that.

Quoted 2 lines

Everything else looks good to me.

Thanks,

Paul

Quoted 1 line

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#37 lore raw

PATCH v2 10/13 x86: add iommu_op to enable modification of IOMMU mappings

In reply to Paul Durrant

This patch adds an iommu_op which checks whether it is possible or safe for a domain to modify its own IOMMU mappings and, if so, creates a rangeset to track modifications.

NOTE: The actual map and unmap operations are introduced by subsequent
patches.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- New in v2.

---
 xen/arch/x86/iommu_op.c         | 48 +++++++++++++++++++++++++++++++++++++++--
 xen/drivers/passthrough/iommu.c |  2 +-
 xen/include/public/iommu_op.h   |  6 ++++++
 xen/include/xen/iommu.h         |  3 +++
 4 files changed, 56 insertions(+), 3 deletions(-)
xen/arch/x86/iommu_op.c +46 -2
index 75c465a46a..e91d39b6c4 100644
@@ -77,6 +77,42 @@ static int iommu_op_query_reserved(struct xen_iommu_op_query_reserved *op)
     return 0;
 }
 
+static int iommu_op_enable_modification(void)
+{
+    struct domain *currd = current->domain;
+    struct domain_iommu *iommu = dom_iommu(currd);
+    const struct iommu_ops *ops = iommu->platform_ops;
+
+    /* Has modification already been enabled? */
+    if ( iommu->iommu_op_ranges )
+        return 0;
+
+    /*
+     * The IOMMU mappings cannot be modified if:
+     * - the IOMMU is not enabled or,
+     * - the current domain is dom0 and tranlsation is disabled or,
+     * - the IOMMU and EPT mappings are shared.
+     */
+    if ( !iommu_enabled ||
+         (is_hardware_domain(currd) && iommu_passthrough) ||
+         iommu_use_hap_pt(currd) )
+        return -EACCES;
+
+    /*
+     * The IOMMU implementation must provide the lookup method if
+     * modification of the mappings is to be supported.
+     */
+    if ( !ops->lookup_page )
+        return -EOPNOTSUPP;
+
+    iommu->iommu_op_ranges = rangeset_new(currd, NULL, 0);
+    if ( !iommu->iommu_op_ranges )
+        return -ENOMEM;
+
+    currd->need_iommu = 0; /* Disable identity GFN mapping */
+    return 0;
+}
+
 static void iommu_op(xen_iommu_op_t *op)
 {
     switch ( op->op )
@@ -85,6 +121,10 @@ static void iommu_op(xen_iommu_op_t *op)
         op->status = iommu_op_query_reserved(&op->u.query_reserved);
         break;
 
+    case XEN_IOMMUOP_enable_modification:
+        op->status = iommu_op_enable_modification();
+        break;
+
     default:
         op->status = -EOPNOTSUPP;
         break;
@@ -97,6 +137,7 @@ int do_one_iommu_op(xen_iommu_op_buf_t *buf)
     size_t offset;
     static const size_t op_size[] = {
         [XEN_IOMMUOP_query_reserved] = sizeof(struct xen_iommu_op_query_reserved),
+        [XEN_IOMMUOP_enable_modification] = 0,
     };
 
     offset = offsetof(struct xen_iommu_op, u);
@@ -116,7 +157,8 @@ int do_one_iommu_op(xen_iommu_op_buf_t *buf)
     if ( buf->size < offset + op_size[op.op] )
         return -EFAULT;
 
-    if ( copy_from_guest_offset((void *)&op.u, buf->h, offset,
+    if ( op_size[op.op] &&
+         copy_from_guest_offset((void *)&op.u, buf->h, offset,
                                 op_size[op.op]) )
         return -EFAULT;
 
@@ -179,6 +221,7 @@ int compat_one_iommu_op(compat_iommu_op_buf_t *buf)
     size_t offset;
     static const size_t op_size[] = {
         [XEN_IOMMUOP_query_reserved] = sizeof(struct compat_iommu_op_query_reserved),
+        [XEN_IOMMUOP_enable_modification] = 0,
     };
     xen_iommu_op_t nat;
     unsigned int u;
@@ -201,7 +244,8 @@ int compat_one_iommu_op(compat_iommu_op_buf_t *buf)
     if ( buf->size < offset + op_size[cmp.op] )
         return -EFAULT;
 
-    if ( copy_from_compat_offset((void *)&cmp.u, buf->h, offset,
+    if ( op_size[cmp.op] &&
+         copy_from_compat_offset((void *)&cmp.u, buf->h, offset,
                                  op_size[cmp.op]) )
         return -EFAULT;
 
xen/drivers/passthrough/iommu.c +1 -1
index f25aa3f1d6..f862d79a8c 100644
@@ -26,7 +26,6 @@ static void iommu_dump_p2m_table(unsigned char key);
 
 unsigned int __read_mostly iommu_dev_iotlb_timeout = 1000;
 integer_param("iommu_dev_iotlb_timeout", iommu_dev_iotlb_timeout);
-
 /*
  * The 'iommu' parameter enables the IOMMU.  Optional comma separated
  * value may contain:
@@ -263,6 +262,7 @@ void iommu_domain_destroy(struct domain *d)
     arch_iommu_domain_destroy(d);
 
     rangeset_destroy(hd->reserved_ranges);
+    rangeset_destroy(hd->iommu_op_ranges);
 }
 
 int iommu_map_page(struct domain *d, bfn_t bfn, mfn_t mfn,
xen/include/public/iommu_op.h +6 -0
index 02213c12a4..5a3148c247 100644
@@ -61,6 +61,12 @@ struct xen_iommu_op_query_reserved {
     XEN_GUEST_HANDLE(xen_iommu_reserved_range_t) ranges;
 };
 
+/*
+ * XEN_IOMMUOP_enable_modification: Enable operations that modify IOMMU
+ *                                  mappings.
+ */
+#define XEN_IOMMUOP_enable_modification 2
+
 struct xen_iommu_op {
     uint16_t op;    /* op type */
     uint16_t pad;
xen/include/xen/iommu.h +3 -0
index 7c5d46df81..08b163cbcb 100644
@@ -130,6 +130,9 @@ struct domain_iommu {
      * must not be modified after initialization.
      */
     struct rangeset *reserved_ranges;
+
+    /* Ranges under the control of iommu_op */
+    struct rangeset *iommu_op_ranges;
 };
 
 #define dom_iommu(d)              (&(d)->iommu)
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#38 lore raw

PATCH v2 11/13 memory: add get_paged_gfn() as a wrapper...

In reply to Paul Durrant

...for some uses of get_page_from_gfn().

There are many occurences of the following pattern in the code:

    q = <readonly look-up> ? P2M_ALLOC : P2M_UNSHARE;
    page = get_page_from_gfn(d, gfn, &p2mt, q);

    if ( p2m_is_paging(p2mt) )
    {
        if ( page )
            put_page(page);

        p2m_mem_paging_populate(d, gfn);
        return <-ENOENT or equivalent>;
    }

    if ( (q & P2M_UNSHARE) && p2m_is_shared(p2mt) )
    {
        if ( page )
            put_page(page);

        return <-ENOENT or equivalent>;
    }

    if ( !page )
        return <-EINVAL or equivalent>;

    if ( !p2m_is_ram(p2mt) ||
         (!<readonly look-up> && p2m_is_readonly(p2mt)) )
    {
        put_page(page);
        return <-EINVAL or equivalent>;
    }

There are some small differences between the exact way the occurrences are coded but the desired semantic is the same.

This patch introduces a new common implementation of this code in get_paged_gfn() and then converts the various open-coded patterns into calls to this new function.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- New in v2.

---
 xen/arch/x86/hvm/emulate.c | 32 +++++++-------------------------
 xen/arch/x86/hvm/hvm.c     | 16 ++--------------
 xen/common/grant_table.c   | 38 +++++++++++---------------------------
 xen/common/memory.c        | 42 ++++++++++++++++++++++++++++++++++--------
 xen/include/asm-arm/p2m.h  |  3 +++
 xen/include/asm-x86/p2m.h  |  2 ++
 6 files changed, 59 insertions(+), 74 deletions(-)
xen/arch/x86/hvm/emulate.c +7 -25
index 8385c62145..ad9bc4f251 100644
@@ -332,34 +332,16 @@ static int hvmemul_do_io_buffer(
 
 static int hvmemul_acquire_page(unsigned long gmfn, struct page_info **page)
 {
-    struct domain *curr_d = current->domain;
-    p2m_type_t p2mt;
-
-    *page = get_page_from_gfn(curr_d, gmfn, &p2mt, P2M_UNSHARE);
-
-    if ( *page == NULL )
-        return X86EMUL_UNHANDLEABLE;
-
-    if ( p2m_is_paging(p2mt) )
-    {
-        put_page(*page);
-        p2m_mem_paging_populate(curr_d, gmfn);
-        return X86EMUL_RETRY;
-    }
-
-    if ( p2m_is_shared(p2mt) )
+    switch ( get_paged_gfn(current->domain, _gfn(gmfn), false, NULL, page) )
     {
-        put_page(*page);
+    case -ENOENT:
         return X86EMUL_RETRY;
-    }
-
-    /* This code should not be reached if the gmfn is not RAM */
-    if ( p2m_is_mmio(p2mt) )
-    {
-        domain_crash(curr_d);
-
-        put_page(*page);
+    case -EINVAL:
         return X86EMUL_UNHANDLEABLE;
+    default:
+        ASSERT_UNREACHABLE();
+    case 0:
+        break;
     }
 
     return X86EMUL_OKAY;
xen/arch/x86/hvm/hvm.c +2 -14
index c23983cdff..906235932d 100644
@@ -2573,24 +2573,12 @@ static void *_hvm_map_guest_frame(unsigned long gfn, bool_t permanent,
                                   bool_t *writable)
 {
     void *map;
-    p2m_type_t p2mt;
     struct page_info *page;
     struct domain *d = current->domain;
+    p2m_type_t p2mt;
 
-    page = get_page_from_gfn(d, gfn, &p2mt,
-                             writable ? P2M_UNSHARE : P2M_ALLOC);
-    if ( (p2m_is_shared(p2mt) && writable) || !page )
-    {
-        if ( page )
-            put_page(page);
-        return NULL;
-    }
-    if ( p2m_is_paging(p2mt) )
-    {
-        put_page(page);
-        p2m_mem_paging_populate(d, gfn);
+    if ( get_paged_gfn(d, _gfn(gfn), !writable, &p2mt, &page) )
         return NULL;
-    }
 
     if ( writable )
     {
xen/common/grant_table.c +11 -27
index c6b99c4116..510f37f100 100644
@@ -375,39 +375,23 @@ static int get_paged_frame(unsigned long gfn, mfn_t *mfn,
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
-    int rc = GNTST_okay;
-    p2m_type_t p2mt;
-
-    *mfn = INVALID_MFN;
-    *page = get_page_from_gfn(rd, gfn, &p2mt,
-                              readonly ? P2M_ALLOC : P2M_UNSHARE);
-    if ( !*page )
-    {
-#ifdef P2M_SHARED_TYPES
-        if ( p2m_is_shared(p2mt) )
-            return GNTST_eagain;
-#endif
-#ifdef P2M_PAGES_TYPES
-        if ( p2m_is_paging(p2mt) )
-        {
-            p2m_mem_paging_populate(rd, gfn);
-            return GNTST_eagain;
-        }
-#endif
-        return GNTST_bad_page;
-    }
+    int rc;
 
-    if ( p2m_is_foreign(p2mt) )
+    rc = get_paged_gfn(rd, _gfn(gfn), readonly, NULL, page);
+    switch ( rc )
     {
-        put_page(*page);
-        *page = NULL;
-
+    case -ENOENT:
+        return GNTST_eagain;
+    case -EINVAL:
         return GNTST_bad_page;
+    default:
+        ASSERT_UNREACHABLE();
+    case 0:
+        break;
     }
 
     *mfn = page_to_mfn(*page);
-
-    return rc;
+    return GNTST_okay;
 }
 
 static inline void
xen/common/memory.c +34 -8

70 lines not on this page · raw message

xen/include/asm-arm/p2m.h +3 -0
index 8823707c17..a39a4faabd 100644
@@ -303,6 +303,9 @@ static inline struct page_info *get_page_from_gfn(
     return page;
 }
 
+int get_paged_gfn(struct domain *d, gfn_t gfn, bool readonly,
+                  p2m_type_t *p2mt_p, struct page_info **page_p);
+
 int get_page_type(struct page_info *page, unsigned long type);
 bool is_iomem_page(mfn_t mfn);
 static inline int get_page_and_type(struct page_info *page,
xen/include/asm-x86/p2m.h +2 -0
index d4b3cfcb6e..e890bcd3e1 100644
@@ -492,6 +492,8 @@ static inline struct page_info *get_page_from_gfn(
     return mfn_valid(_mfn(gfn)) && get_page(page, d) ? page : NULL;
 }
 
+int get_paged_gfn(struct domain *d, gfn_t gfn, bool readonly,
+                  p2m_type_t *p2mt_p, struct page_info **page_p);
 
 /* General conversion function from mfn to gfn */
 static inline unsigned long mfn_to_gfn(struct domain *d, mfn_t mfn)
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#39 lore raw

Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 34 lines · Paul Durrant

...for some uses of get_page_from_gfn().

There are many occurences of the following pattern in the code:

    q = <readonly look-up> ? P2M_ALLOC : P2M_UNSHARE;
    page = get_page_from_gfn(d, gfn, &p2mt, q);

    if ( p2m_is_paging(p2mt) )
    {
        if ( page )
            put_page(page);

        p2m_mem_paging_populate(d, gfn);
        return <-ENOENT or equivalent>;
    }

    if ( (q & P2M_UNSHARE) && p2m_is_shared(p2mt) )
    {
        if ( page )
            put_page(page);

        return <-ENOENT or equivalent>;
    }

    if ( !page )
        return <-EINVAL or equivalent>;

    if ( !p2m_is_ram(p2mt) ||
         (!<readonly look-up> && p2m_is_readonly(p2mt)) )
    {
        put_page(page);
        return <-EINVAL or equivalent>;
    }
Quoted 8 lines

There are some small differences between the exact way the occurrences are coded but the desired semantic is the same.

This patch introduces a new common implementation of this code in get_paged_gfn() and then converts the various open-coded patterns into calls to this new function.

This is a great idea.

It looks like this adds the restriction that the given gfn be ram (e.g., not MMIO) in all cases as well. It looks like that's what's wanted, but it would be good to point this out in the commit message (so people can verify that this change is warranted).

A few other comments...

Quoted 32 lines
xen/common/grant_table.c +1 -22
index c6b99c4116..510f37f100 100644
@@ -375,39 +375,23 @@ static int get_paged_frame(unsigned long gfn, mfn_t *mfn,
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
-    int rc = GNTST_okay;
-    p2m_type_t p2mt;
-
-    *mfn = INVALID_MFN;
-    *page = get_page_from_gfn(rd, gfn, &p2mt,
-                              readonly ? P2M_ALLOC : P2M_UNSHARE);
-    if ( !*page )
-    {
-#ifdef P2M_SHARED_TYPES
-        if ( p2m_is_shared(p2mt) )
-            return GNTST_eagain;
-#endif
-#ifdef P2M_PAGES_TYPES
-        if ( p2m_is_paging(p2mt) )
-        {
-            p2m_mem_paging_populate(rd, gfn);
-            return GNTST_eagain;
-        }
-#endif
-        return GNTST_bad_page;
-    }
+    int rc;
 
-    if ( p2m_is_foreign(p2mt) )

[snip]

Quoted 1 line
     {

[snip]

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn, mfn_t *mfn, · line 404

Quoted 3 lines
xen/common/grant_table.c +0 -3
-        put_page(*page);
-        *page = NULL;
-

Comparing before-and-after, this seems to remove this 'p2m_is_foreign()' check. Am I missing something?

xen/common/memory.c void destroy_ring_for_helper( · line 1578

Quoted 13 lines
xen/common/memory.c +2 -3
index 35da9ca80e..419b76ac38 100644
@@ -1574,30 +1574,31 @@ void destroy_ring_for_helper(
     }
 }
 
-int prepare_ring_for_helper(
-    struct domain *d, unsigned long gmfn, struct page_info **_page,
-    void **_va)
+int get_paged_gfn(struct domain *d, gfn_t gfn, bool readonly,
+                  p2m_type_t *p2mt_p, struct page_info **page_p)

This wants a comment somewhere describing exactly what the function does and what it will return -- probably here above the function definition itself would be the best.

Quoted 9 lines
xen/common/memory.c +3 -3
 {
-    struct page_info *page;
+    p2m_query_t q = readonly ? P2M_ALLOC : P2M_UNSHARE;
     p2m_type_t p2mt;
-    void *va;
+    struct page_info *page;
 
-    page = get_page_from_gfn(d, gmfn, &p2mt, P2M_UNSHARE);
+    page = get_page_from_gfn(d, gfn_x(gfn), &p2mt, q);

xen/common/memory.c void destroy_ring_for_helper( · line 1593

Quoted 10 lines
xen/common/memory.c +2 -1
 
 #ifdef CONFIG_HAS_MEM_PAGING
     if ( p2m_is_paging(p2mt) )
     {
         if ( page )
             put_page(page);
-        p2m_mem_paging_populate(d, gmfn);
+
+        p2m_mem_paging_populate(d, gfn_x(gfn));
         return -ENOENT;

I realize you're copying the return values of prepare_ring_for_helper(), but wouldn't -EAGAIN be more natural here?

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#40 lore raw

Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 55 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 11 July 2018 12:24
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Julien Grall <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 34 lines · Paul Durrant

...for some uses of get_page_from_gfn().

There are many occurences of the following pattern in the code:

    q = <readonly look-up> ? P2M_ALLOC : P2M_UNSHARE;
    page = get_page_from_gfn(d, gfn, &p2mt, q);

    if ( p2m_is_paging(p2mt) )
    {
        if ( page )
            put_page(page);

        p2m_mem_paging_populate(d, gfn);
        return <-ENOENT or equivalent>;
    }

    if ( (q & P2M_UNSHARE) && p2m_is_shared(p2mt) )
    {
        if ( page )
            put_page(page);

        return <-ENOENT or equivalent>;
    }

    if ( !page )
        return <-EINVAL or equivalent>;

    if ( !p2m_is_ram(p2mt) ||
         (!<readonly look-up> && p2m_is_readonly(p2mt)) )
    {
        put_page(page);
        return <-EINVAL or equivalent>;
    }
Quoted 1 line

There are some small differences between the exact way the occurrences

are

Quoted 6 lines

coded but the desired semantic is the same.

This patch introduces a new common implementation of this code in get_paged_gfn() and then converts the various open-coded patterns into calls to this new function.

Quoted 9 lines
Quoted 1 line

This is a great idea.

It looks like this adds the restriction that the given gfn be ram (e.g., not MMIO) in all cases as well. It looks like that's what's wanted, but it would be good to point this out in the commit message (so people can verify that this change is warranted).

Yes, that's what I meant by 'desired semantic' :-) I'll call out the restriction more explicitly.

Quoted 36 lines

A few other comments...

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn,

Quoted 5 lines
xen/common/grant_table.c +0 -0
index c6b99c4116..510f37f100 100644
@@ -375,39 +375,23 @@ static int get_paged_frame(unsigned long gfn,

mfn_t *mfn,

Quoted 27 lines
xen/common/grant_table.c +1 -22
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
-    int rc = GNTST_okay;
-    p2m_type_t p2mt;
-
-    *mfn = INVALID_MFN;
-    *page = get_page_from_gfn(rd, gfn, &p2mt,
-                              readonly ? P2M_ALLOC : P2M_UNSHARE);
-    if ( !*page )
-    {
-#ifdef P2M_SHARED_TYPES
-        if ( p2m_is_shared(p2mt) )
-            return GNTST_eagain;
-#endif
-#ifdef P2M_PAGES_TYPES
-        if ( p2m_is_paging(p2mt) )
-        {
-            p2m_mem_paging_populate(rd, gfn);
-            return GNTST_eagain;
-        }
-#endif
-        return GNTST_bad_page;
-    }
+    int rc;
 
-    if ( p2m_is_foreign(p2mt) )

[snip]

Quoted 9 lines
Quoted 1 line
     {

[snip]

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn, mfn_t *mfn, · line 404

Quoted 3 lines
xen/common/grant_table.c +0 -3
-        put_page(*page);
-        *page = NULL;
-

Comparing before-and-after, this seems to remove this 'p2m_is_foreign()' check. Am I missing something?

I may be. I thought p2m_is_ram() ruled out foreign pages (p2m_is_any_ram() being the way to include foreign maps if required). I'll check.

Quoted 8 lines
Quoted 8 lines
xen/common/memory.c +0 -0
index 35da9ca80e..419b76ac38 100644
@@ -1574,30 +1574,31 @@ void destroy_ring_for_helper(
     }
 }
 
Quoted 10 lines

xen/common/memory.c void destroy_ring_for_helper( · line 1578

Quoted 5 lines
xen/common/memory.c +2 -3
-int prepare_ring_for_helper(
-    struct domain *d, unsigned long gmfn, struct page_info **_page,
-    void **_va)
+int get_paged_gfn(struct domain *d, gfn_t gfn, bool readonly,
+                  p2m_type_t *p2mt_p, struct page_info **page_p)

This wants a comment somewhere describing exactly what the function does and what it will return -- probably here above the function definition itself would be the best.

Ok.

Quoted 20 lines
Quoted 9 lines
xen/common/memory.c +3 -3
 {
-    struct page_info *page;
+    p2m_query_t q = readonly ? P2M_ALLOC : P2M_UNSHARE;
     p2m_type_t p2mt;
-    void *va;
+    struct page_info *page;
 
-    page = get_page_from_gfn(d, gmfn, &p2mt, P2M_UNSHARE);
+    page = get_page_from_gfn(d, gfn_x(gfn), &p2mt, q);

xen/common/memory.c void destroy_ring_for_helper( · line 1593

Quoted 10 lines
xen/common/memory.c +2 -1
 
 #ifdef CONFIG_HAS_MEM_PAGING
     if ( p2m_is_paging(p2mt) )
     {
         if ( page )
             put_page(page);
-        p2m_mem_paging_populate(d, gmfn);
+
+        p2m_mem_paging_populate(d, gfn_x(gfn));
         return -ENOENT;
Quoted 3 lines

I realize you're copying the return values of prepare_ring_for_helper(), but wouldn't -EAGAIN be more natural here?

I may be able to swap for EAGAIN. I agree it seems more appropriate. Hopefully it won't complicate the callers too much.

Paul

Quoted 1 line

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#41 lore raw

Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 40 lines · Paul Durrant
Quoted 40 lines

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn,

Quoted 5 lines
xen/common/grant_table.c +0 -0
index c6b99c4116..510f37f100 100644
@@ -375,39 +375,23 @@ static int get_paged_frame(unsigned long gfn,

mfn_t *mfn,

Quoted 27 lines
xen/common/grant_table.c +1 -22
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
-    int rc = GNTST_okay;
-    p2m_type_t p2mt;
-
-    *mfn = INVALID_MFN;
-    *page = get_page_from_gfn(rd, gfn, &p2mt,
-                              readonly ? P2M_ALLOC : P2M_UNSHARE);
-    if ( !*page )
-    {
-#ifdef P2M_SHARED_TYPES
-        if ( p2m_is_shared(p2mt) )
-            return GNTST_eagain;
-#endif
-#ifdef P2M_PAGES_TYPES
-        if ( p2m_is_paging(p2mt) )
-        {
-            p2m_mem_paging_populate(rd, gfn);
-            return GNTST_eagain;
-        }
-#endif
-        return GNTST_bad_page;
-    }
+    int rc;
 
-    if ( p2m_is_foreign(p2mt) )

[snip]

Quoted 1 line
     {

[snip]

Quoted 3 lines
xen/common/grant_table.c +0 -3
-        put_page(*page);
-        *page = NULL;
-
Quoted 5 lines
Quoted 3 lines

Comparing before-and-after, this seems to remove this 'p2m_is_foreign()' check. Am I missing something?

I may be. I thought p2m_is_ram() ruled out foreign pages (p2m_is_any_ram() being the way to include foreign maps if required). I'll check.

Looks like you're right. But then, are you sure that's what we want for the other callers? Might we not need to do an emulation that ends up writing into a foreign mapping, for instance?

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#42 lore raw

Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 57 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 11 July 2018 14:05
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; Ian Jackson <[email protected]>; Julien Grall <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH v2 11/13] memory: add get_paged_gfn() as a wrapper...

Quoted 34 lines · Paul Durrant
Quoted 34 lines

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn,

Quoted 5 lines
xen/common/grant_table.c +0 -0
index c6b99c4116..510f37f100 100644
@@ -375,39 +375,23 @@ static int get_paged_frame(unsigned long gfn,

mfn_t *mfn,

Quoted 27 lines
xen/common/grant_table.c +1 -22
                            struct page_info **page, bool readonly,
                            struct domain *rd)
 {
-    int rc = GNTST_okay;
-    p2m_type_t p2mt;
-
-    *mfn = INVALID_MFN;
-    *page = get_page_from_gfn(rd, gfn, &p2mt,
-                              readonly ? P2M_ALLOC : P2M_UNSHARE);
-    if ( !*page )
-    {
-#ifdef P2M_SHARED_TYPES
-        if ( p2m_is_shared(p2mt) )
-            return GNTST_eagain;
-#endif
-#ifdef P2M_PAGES_TYPES
-        if ( p2m_is_paging(p2mt) )
-        {
-            p2m_mem_paging_populate(rd, gfn);
-            return GNTST_eagain;
-        }
-#endif
-        return GNTST_bad_page;
-    }
+    int rc;
 
-    if ( p2m_is_foreign(p2mt) )

[snip]

Quoted 7 lines
Quoted 7 lines
Quoted 1 line
     {

[snip]

xen/common/grant_table.c static int get_paged_frame(unsigned long gfn, mfn_t *mfn, · line 404

Quoted 3 lines
xen/common/grant_table.c +0 -3
-        put_page(*page);
-        *page = NULL;
-

Comparing before-and-after, this seems to remove this

'p2m_is_foreign()'

Quoted 3 lines
Quoted 2 lines

check. Am I missing something?

Quoted 7 lines
Quoted 1 line

I may be. I thought p2m_is_ram() ruled out foreign pages

(p2m_is_any_ram() being the way to include foreign maps if required). I'll check.

Looks like you're right. But then, are you sure that's what we want for the other callers? Might we not need to do an emulation that ends up writing into a foreign mapping, for instance?

If we do then I'd expect the emulation to know the domid that owns the page and thus the page would not be foreign to the specified domid. In fact, for x86 at least, get_page_from_gfn() will fail unless the domid of the page owner is specified so there's no prospect of the page being foreign in the p2mt check anyway.

Paul

Quoted 2 lines

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#43 lore raw

PATCH v2 12/13 x86: add iommu_ops to modify and flush IOMMU mappings

In reply to Paul Durrant

This patch adds iommu_ops to add (map) or remove (unmap) frames in the domain's IOMMU mappings, and an iommu_op to synchronize (flush) those manipulations with the hardware.

Mappings added by the map operation are tracked and only those mappings may be removed by a subsequent unmap operation. Frames are specified by the owning domain and GFN. It is, of course, permissable for a domain to map its own frames using DOMID_SELF.

NOTE: The owning domain and GFN must also be specified in the unmap
operation, as well as the BFN, so that they can be cross-checked with the existent mapping.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- Heavily re-worked in v2, including explicit tracking of mappings. This avoids the need to clear non-reserved mappings from IOMMU at start of day, which would be prohibitively slow on a large host.

---
 xen/arch/x86/iommu_op.c       | 134 ++++++++++++++++++++++++++++++++++++++++++
 xen/include/public/iommu_op.h |  43 ++++++++++++++
 xen/include/xlat.lst          |   2 +
 3 files changed, 179 insertions(+)
xen/arch/x86/iommu_op.c +134 -0
index e91d39b6c4..7174838a6c 100644
@@ -113,6 +113,116 @@ static int iommu_op_enable_modification(void)
     return 0;
 }
 
+static int iommuop_map(struct xen_iommu_op_map *op)
+{
+    struct domain *d, *currd = current->domain;
+    struct domain_iommu *iommu = dom_iommu(currd);
+    bool readonly = op->flags & XEN_IOMMUOP_map_readonly;
+    bfn_t bfn = _bfn(op->bfn);
+    struct page_info *page;
+    unsigned int prot;
+    int rc, ignore;
+
+    if ( op->pad || (op->flags & ~XEN_IOMMUOP_map_readonly) )
+        return -EINVAL;
+
+    if ( !iommu->iommu_op_ranges )
+        return -EOPNOTSUPP;
+
+    /* Check whether the specified BFN falls in a reserved region */
+    if ( rangeset_contains_singleton(iommu->reserved_ranges, bfn_x(bfn)) )
+        return -EINVAL;
+
+    d = rcu_lock_domain_by_any_id(op->domid);
+    if ( !d )
+        return -ESRCH;
+
+    rc = get_paged_gfn(d, _gfn(op->gfn), readonly, NULL, &page);
+    if ( rc )
+        goto unlock;
+
+    prot = IOMMUF_readable;
+    if ( !readonly )
+        prot |= IOMMUF_writable;
+
+    rc = -EIO;
+    if ( iommu_map_page(currd, bfn, page_to_mfn(page), prot) )
+        goto release;
+
+    rc = rangeset_add_singleton(iommu->iommu_op_ranges, bfn_x(bfn));
+    if ( rc )
+        goto unmap;
+
+    rc = 0;
+    goto unlock; /* retain mapping and reference */
+
+ unmap:
+    ignore = iommu_unmap_page(currd, bfn);
+
+ release:
+    put_page(page);
+
+ unlock:
+    rcu_unlock_domain(d);
+    return rc;
+}
+
+static int iommuop_unmap(struct xen_iommu_op_unmap *op)
+{
+    struct domain *d, *currd = current->domain;
+    struct domain_iommu *iommu = dom_iommu(currd);
+    bfn_t bfn = _bfn(op->bfn);
+    mfn_t mfn;
+    unsigned int prot;
+    struct page_info *page;
+    int rc;
+
+    if ( op->pad0 || op->pad1 )
+        return -EINVAL;
+
+    if ( !iommu->iommu_op_ranges )
+        return -EOPNOTSUPP;
+
+    if ( !rangeset_contains_singleton(iommu->iommu_op_ranges, bfn_x(bfn)) ||
+         iommu_lookup_page(currd, bfn, &mfn, &prot) ||
+         !mfn_valid(mfn) )
+        return -ENOENT;
+
+    d = rcu_lock_domain_by_any_id(op->domid);
+    if ( !d )
+        return -ESRCH;
+
+    rc = get_paged_gfn(d, _gfn(op->gfn), !(prot & IOMMUF_writable), NULL,
+                       &page);
+    if ( rc )
+        goto unlock;
+
+    put_page(page); /* release extra reference just taken */
+
+    rc = -EINVAL;
+    if ( !mfn_eq(page_to_mfn(page), mfn) )
+        goto unlock;
+
+    put_page(page); /* release reference taken in map */
+
+    rc = rangeset_remove_singleton(iommu->iommu_op_ranges, bfn_x(bfn));
+    if ( rc )
+        goto unlock;
+
+    if ( !iommu_unmap_page(currd, bfn) )
+        rc = -EIO;
+
+ unlock:
+    rcu_unlock_domain(d);
+
+    return rc;
+}
+
+static int iommuop_flush(void)
+{
+    return !iommu_iotlb_flush_all(current->domain) ? 0 : -EIO;
+}
+
 static void iommu_op(xen_iommu_op_t *op)
 {
     switch ( op->op )
@@ -125,6 +235,22 @@ static void iommu_op(xen_iommu_op_t *op)
         op->status = iommu_op_enable_modification();
         break;
 
+    case XEN_IOMMUOP_map:
+        this_cpu(iommu_dont_flush_iotlb) = 1;
+        op->status = iommuop_map(&op->u.map);
+        this_cpu(iommu_dont_flush_iotlb) = 0;
+        break;
+
+    case XEN_IOMMUOP_unmap:
+        this_cpu(iommu_dont_flush_iotlb) = 1;
+        op->status = iommuop_unmap(&op->u.unmap);
+        this_cpu(iommu_dont_flush_iotlb) = 0;
+        break;
+
+    case XEN_IOMMUOP_flush:
+        op->status = iommuop_flush();
+        break;
+
     default:
         op->status = -EOPNOTSUPP;
         break;
@@ -138,6 +264,9 @@ int do_one_iommu_op(xen_iommu_op_buf_t *buf)
     static const size_t op_size[] = {
         [XEN_IOMMUOP_query_reserved] = sizeof(struct xen_iommu_op_query_reserved),
         [XEN_IOMMUOP_enable_modification] = 0,
+        [XEN_IOMMUOP_map] = sizeof(struct xen_iommu_op_map),
+        [XEN_IOMMUOP_unmap] = sizeof(struct xen_iommu_op_unmap),
+        [XEN_IOMMUOP_flush] = 0,
     };
 
     offset = offsetof(struct xen_iommu_op, u);
@@ -222,6 +351,9 @@ int compat_one_iommu_op(compat_iommu_op_buf_t *buf)
     static const size_t op_size[] = {
         [XEN_IOMMUOP_query_reserved] = sizeof(struct compat_iommu_op_query_reserved),
         [XEN_IOMMUOP_enable_modification] = 0,
+        [XEN_IOMMUOP_map] = sizeof(struct compat_iommu_op_map),
+        [XEN_IOMMUOP_unmap] = sizeof(struct compat_iommu_op_unmap),
+        [XEN_IOMMUOP_flush] = 0,
     };
     xen_iommu_op_t nat;
     unsigned int u;
@@ -254,6 +386,8 @@ int compat_one_iommu_op(compat_iommu_op_buf_t *buf)
      * we need to fix things up here.
      */
 #define XLAT_iommu_op_u_query_reserved XEN_IOMMUOP_query_reserved
+#define XLAT_iommu_op_u_map XEN_IOMMUOP_map
+#define XLAT_iommu_op_u_unmap XEN_IOMMUOP_unmap
     u = cmp.op;
 
 #define XLAT_iommu_op_query_reserved_HNDL_ranges(_d_, _s_)            \
xen/include/public/iommu_op.h +43 -0

55 lines not on this page · raw message

xen/include/xlat.lst +2 -0
index 93bcf4b4d0..ed50216394 100644
@@ -79,7 +79,9 @@
 ?	vcpu_hvm_x86_64			hvm/hvm_vcpu.h
 !	iommu_op			iommu_op.h
 !	iommu_op_buf			iommu_op.h
+!	iommu_op_map			iommu_op.h
 !	iommu_op_query_reserved		iommu_op.h
+!	iommu_op_unmap			iommu_op.h
 !	iommu_reserved_range		iommu_op.h
 ?	kexec_exec			kexec.h
 !	kexec_image			kexec.h
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
#44 lore raw

Re: [PATCH v2 12/13] x86: add iommu_ops to modify and flush IOMMU mappings

Quoted 9 lines · Paul Durrant

This patch adds iommu_ops to add (map) or remove (unmap) frames in the domain's IOMMU mappings, and an iommu_op to synchronize (flush) those manipulations with the hardware.

Mappings added by the map operation are tracked and only those mappings may be removed by a subsequent unmap operation. Frames are specified by the owning domain and GFN. It is, of course, permissable for a domain to map its own frames using DOMID_SELF.

Quoted 5 lines

NOTE: The owning domain and GFN must also be specified in the unmap
operation, as well as the BFN, so that they can be cross-checked with the existent mapping.

The code on the whole looks correct, but I don't see any reference counting. The call to get_paged_gfn() in iommuop_unmap() kind of underlines the issue -- what's to stop the following sequence of events?

* iommuop_map() page A
* share(A)
* DMA write into A #

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#45 lore raw

Re: [PATCH v2 12/13] x86: add iommu_ops to modify and flush IOMMU mappings

Quoted 32 lines

-----Original Message-----
From: George Dunlap [mailto:[email protected]]
Sent: 11 July 2018 12:46
To: Paul Durrant <[email protected]>; [email protected]
Cc: Jan Beulich <[email protected]>; Andrew Cooper <[email protected]>; George Dunlap <[email protected]>; Ian Jackson <[email protected]>; Julien Grall <[email protected]>; Konrad Rzeszutek Wilk <[email protected]>; Stefano Stabellini <[email protected]>; Tim (Xen.org) <[email protected]>; Wei Liu <[email protected]>
Subject: Re: [PATCH v2 12/13] x86: add iommu_ops to modify and flush IOMMU mappings

Quoted 2 lines · Paul Durrant

This patch adds iommu_ops to add (map) or remove (unmap) frames in the domain's IOMMU mappings, and an iommu_op to synchronize (flush)

those

Quoted 3 lines

manipulations with the hardware.

Mappings added by the map operation are tracked and only those

mappings

Quoted 1 line

may be removed by a subsequent unmap operation. Frames are specified

by the

Quoted 8 lines

owning domain and GFN. It is, of course, permissable for a domain to map its own frames using DOMID_SELF.

NOTE: The owning domain and GFN must also be specified in the unmap
operation, as well as the BFN, so that they can be cross-checked with the existent mapping.

Quoted 8 lines

The code on the whole looks correct, but I don't see any reference counting. The call to get_paged_gfn() in iommuop_unmap() kind of underlines the issue -- what's to stop the following sequence of events?

* iommuop_map() page A
* share(A)
* DMA write into A #

I don't follow. In iommuop_map() we do a get_paged_gfn() and that reference is retained. In iommuop_unmap() there is another call to get_paged_gfn() but that is there to check that the specified gfn matches the mfn that's looked up in the IOMMU. Only if they match is the original page ref dropped.

Paul

Quoted 1 line

-George

_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel

#46 lore raw

PATCH v2 13/13 x86: extend the map and unmap iommu_ops to support grant references

In reply to Paul Durrant

This patch allows a domain to add or remove foreign frames from its IOMMU mappings by grant reference as well as GFN. This is necessary, for example, to support a PV network backend that needs to construct a packet buffer that can be directly accessed by a NIC.

---
Cc: Jan Beulich <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: George Dunlap <[email protected]>
Cc: Ian Jackson <[email protected]>
Cc: Julien Grall <[email protected]>
Cc: Konrad Rzeszutek Wilk <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Tim Deegan <[email protected]>
Cc: Wei Liu <[email protected]>

v2:
- New in v2.

---
 xen/arch/x86/iommu_op.c       |  68 ++++++++++++++------
 xen/common/grant_table.c      | 143 ++++++++++++++++++++++++++++++++++++++++++
 xen/include/public/iommu_op.h |  31 +++++++--
 xen/include/xen/grant_table.h |   7 +++
 4 files changed, 224 insertions(+), 25 deletions(-)
xen/arch/x86/iommu_op.c +49 -19
index 7174838a6c..cca1c59457 100644
@@ -22,6 +22,7 @@
 #include <xen/event.h>
 #include <xen/guest_access.h>
 #include <xen/hypercall.h>
+#include <xen/grant_table.h>
 
 struct get_reserved_ctxt {
     unsigned int max_entries;
@@ -120,10 +121,12 @@ static int iommuop_map(struct xen_iommu_op_map *op)
     bool readonly = op->flags & XEN_IOMMUOP_map_readonly;
     bfn_t bfn = _bfn(op->bfn);
     struct page_info *page;
+    mfn_t mfn;
     unsigned int prot;
     int rc, ignore;
 
-    if ( op->pad || (op->flags & ~XEN_IOMMUOP_map_readonly) )
+    if ( op->pad || (op->flags & ~(XEN_IOMMUOP_map_readonly |
+                                   XEN_IOMMUOP_map_gref)) )
         return -EINVAL;
 
     if ( !iommu->iommu_op_ranges )
@@ -137,16 +140,29 @@ static int iommuop_map(struct xen_iommu_op_map *op)
     if ( !d )
         return -ESRCH;
 
-    rc = get_paged_gfn(d, _gfn(op->gfn), readonly, NULL, &page);
-    if ( rc )
-        goto unlock;
+    if ( op->flags & XEN_IOMMUOP_map_gref )
+    {
+        rc = acquire_gref_for_iommu(d, op->u.gref, readonly, &mfn);
+        if ( rc )
+            goto unlock;
+
+        page = mfn_to_page(mfn);
+    }
+    else
+    {
+        rc = get_paged_gfn(d, _gfn(op->u.gfn), readonly, NULL, &page);
+        if ( rc )
+            goto unlock;
+
+        mfn = page_to_mfn(page);
+    }
 
     prot = IOMMUF_readable;
     if ( !readonly )
         prot |= IOMMUF_writable;
 
     rc = -EIO;
-    if ( iommu_map_page(currd, bfn, page_to_mfn(page), prot) )
+    if ( iommu_map_page(currd, bfn, mfn, prot) )
         goto release;
 
     rc = rangeset_add_singleton(iommu->iommu_op_ranges, bfn_x(bfn));
@@ -160,7 +176,10 @@ static int iommuop_map(struct xen_iommu_op_map *op)
     ignore = iommu_unmap_page(currd, bfn);
 
  release:
-    put_page(page);
+    if ( op->flags & XEN_IOMMUOP_map_gref )
+        release_gref_for_iommu(d, op->u.gref, readonly, mfn);
+    else
+        put_page(page);
 
  unlock:
     rcu_unlock_domain(d);
@@ -174,11 +193,10 @@ static int iommuop_unmap(struct xen_iommu_op_unmap *op)
     bfn_t bfn = _bfn(op->bfn);
     mfn_t mfn;
     unsigned int prot;
-    struct page_info *page;
+    bool readonly;
     int rc;
 
-    if ( op->pad0 || op->pad1 )
-        return -EINVAL;
+    if ( op->pad || (op->flags & ~XEN_IOMMUOP_unmap_gref) )
 
     if ( !iommu->iommu_op_ranges )
         return -EOPNOTSUPP;
@@ -188,28 +206,40 @@ static int iommuop_unmap(struct xen_iommu_op_unmap *op)
          !mfn_valid(mfn) )
         return -ENOENT;
 
+    readonly = !(prot & IOMMUF_writable);
+
     d = rcu_lock_domain_by_any_id(op->domid);
     if ( !d )
         return -ESRCH;
 
-    rc = get_paged_gfn(d, _gfn(op->gfn), !(prot & IOMMUF_writable), NULL,
-                       &page);
-    if ( rc )
-        goto unlock;
+    if ( op->flags & XEN_IOMMUOP_unmap_gref )
+    {
+        rc = release_gref_for_iommu(d, op->u.gref, readonly, mfn);
+        if ( rc )
+            goto unlock;
+    }
+    else
+    {
+        struct page_info *page;
+
+        rc = get_paged_gfn(d, _gfn(op->u.gfn), readonly, NULL, &page);
+        if ( rc )
+            goto unlock;
 
-    put_page(page); /* release extra reference just taken */
+        put_page(page); /* release extra reference just taken */
 
-    rc = -EINVAL;
-    if ( !mfn_eq(page_to_mfn(page), mfn) )
-        goto unlock;
+        rc = -EINVAL;
+        if ( !mfn_eq(page_to_mfn(page), mfn) )
+            goto unlock;
 
-    put_page(page); /* release reference taken in map */
+        put_page(page); /* release reference taken in map */
+    }
 
     rc = rangeset_remove_singleton(iommu->iommu_op_ranges, bfn_x(bfn));
     if ( rc )
         goto unlock;
 
-    if ( !iommu_unmap_page(currd, bfn) )
+    if ( iommu_unmap_page(currd, bfn) )
         rc = -EIO;
 
  unlock:
xen/common/grant_table.c +143 -0

149 lines not on this page · raw message

xen/include/public/iommu_op.h +25 -6
index 737e2c8cfe..485782a522 100644
@@ -24,6 +24,7 @@
 #define XEN_PUBLIC_IOMMU_OP_H
 
 #include "xen.h"
+#include "grant_table.h"
 
 typedef unsigned long xen_bfn_t;
 
@@ -79,12 +80,20 @@ struct xen_iommu_op_map {
 
 #define _XEN_IOMMUOP_map_readonly 0
 #define XEN_IOMMUOP_map_readonly (1 << (_XEN_IOMMUOP_map_readonly))
+#define _XEN_IOMMUOP_map_gref 1
+#define XEN_IOMMUOP_map_gref (1 << (_XEN_IOMMUOP_map_gref))
 
     uint32_t pad;
     /* IN - The IOMMU frame number which will hold the new mapping */
     xen_bfn_t bfn;
-    /* IN - The guest frame number of the page to be mapped */
-    xen_pfn_t gfn;
+    /*
+     * IN - The guest frame number or grant reference of the page to
+     * be mapped.
+     */
+    union {
+        xen_pfn_t gfn;
+        grant_ref_t gref;
+    } u;
 };
 
 /*
@@ -95,12 +104,22 @@ struct xen_iommu_op_map {
 struct xen_iommu_op_unmap {
     /* IN - The domid of the guest */
     domid_t domid;
-    uint16_t pad0;
-    uint32_t pad1;
+    uint16_t flags;
+
+#define _XEN_IOMMUOP_unmap_gref 0
+#define XEN_IOMMUOP_unmap_gref (1 << (_XEN_IOMMUOP_unmap_gref))
+
+    uint32_t pad;
     /* IN - The IOMMU frame number which holds the mapping to be removed */
     xen_bfn_t bfn;
-    /* IN - The guest frame number of the page that is mapped */
-    xen_pfn_t gfn;
+    /*
+     * IN - The guest frame number or grant reference of the page that
+     * is mapped.
+     */
+    union {
+        xen_pfn_t gfn;
+        grant_ref_t gref;
+    } u;
 };
 
 /*
xen/include/xen/grant_table.h +7 -0
index b3a95fda58..22c80c2238 100644
@@ -56,6 +56,13 @@ int mem_sharing_gref_to_gfn(struct grant_table *gt, grant_ref_t ref,
 int gnttab_map_frame(struct domain *d, unsigned long idx, gfn_t gfn,
                      mfn_t *mfn);
 
+int
+acquire_gref_for_iommu(struct domain *d, grant_ref_t gref,
+                       bool readonly, mfn_t *mfn);
+int
+release_gref_for_iommu(struct domain *d, grant_ref_t gref,
+                       bool readonly, mfn_t mfn);
+
 unsigned int gnttab_dom0_frames(void);
 
 #endif /* __XEN_GRANT_TABLE_H__ */
Signature · 7 lines
2.11.0


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel