Skip to contentSkip to conversation

RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths

PATCH Zhu Yanjun 3 participants 5 messages Jul 26 1 message not in the archive InfiniBand

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

Zhu Yanjun

author
#1 lore raw
Quoted 104 lines · Weiming Shi

Commit 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages") dropped the access permission test from rxe_check_pagefault() and left only HMM_PFN_VALID. A page faulted in read-only, for example a page-cache folio behind a PROT_READ file mapping, then satisfies the check and ODP write operations (RDMA WRITE, RDMA READ response, SEND payload, atomics) modify it through kmap without ever breaking CoW.

An unprivileged user can register an ODP MR over such a mapping and have incoming RDMA traffic overwrite the page cache of a file it only holds O_RDONLY, including /etc/passwd or setuid binaries. This is the same primitive class as Dirty COW and CVE-2022-2590.

mlx5 has the missing invariant: its ODP path sets the device write bit only for pfns that carry HMM_PFN_WRITE. Restore it in rxe by requiring HMM_PFN_WRITE in rxe_check_pagefault() for every operation except RXE_PAGEFAULT_RDONLY. A write to a non-writable VMA now fails the one fault attempt with -EPERM from hmm_vma_fault() instead of re-faulting forever. For a writable VMA the fault breaks CoW and the write lands in the private page.

Keep pmem flushes on the read-only check. arch_wb_cache_pmem() never modifies memory, and the FLUSH access bits do not make the umem writable, so classifying flushes as writes would make every flush against a flush-only MR fail.

Fixes 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages")
---
  drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
  1 file changed, 11 insertions(+), 5 deletions(-)
drivers/infiniband/sw/rxe/rxe_odp.c +11 -5
index ff904d5e54a73..174bb2efc3c6a 100644
@@ -124,19 +124,23 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe, u64 start, u64 length,
  }
 
  static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, u64 iova,
-				       int length)
+				       int length, bool write)
  {
  	bool need_fault = false;
+	u64 access = HMM_PFN_VALID;
  	u64 addr;
  	int idx;
 
+	if (write)
+		access |= HMM_PFN_WRITE;
+
  	addr = iova & (~(BIT(umem_odp->page_shift) - 1));
 
  	/* Skim through all pages that are to be accessed. */
  	while (addr < iova + length) {
  		idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift;
 
-		if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) {
+		if ((umem_odp->map.pfn_list[idx] & access) != access) {
  			need_fault = true;
  			break;
  		}
@@ -159,6 +163,7 @@ static unsigned long rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u
  static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u32 flags)
  {
  	struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem);
+	bool write = !(flags & RXE_PAGEFAULT_RDONLY);
  	bool need_fault;
  	int err;
 
@@ -167,7 +172,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u
 
  	mutex_lock(&umem_odp->umem_mutex);
 
-	need_fault = rxe_check_pagefault(umem_odp, iova, length);
+	need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
  	if (need_fault) {
  		mutex_unlock(&umem_odp->umem_mutex);
 
@@ -177,7 +182,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u
  		if (err < 0)
  			return err;
 
-		need_fault = rxe_check_pagefault(umem_odp, iova, length);
+		need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
  		if (need_fault) {
  			mutex_unlock(&umem_odp->umem_mutex);
  			return -EFAULT;
@@ -339,8 +344,9 @@ int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, u64 iova,
  	int err;
  	u8 *va;
 
+	/* A flush never modifies memory; read-only access suffices. */
  	err = rxe_odp_map_range_and_lock(mr, iova, length,
-					 RXE_PAGEFAULT_DEFAULT);
+					 RXE_PAGEFAULT_RDONLY);
  	if (err)
  		return err;
 
Signature · 2 lines
Best Regards,
Yanjun.Zhu
author Reviewed-by
#2 lore raw

While exploiting this ODP flaw in soft-RoCE for privilege escalation in practice might require specific conditions, enforcing the HMM_PFN_WRITE check is undoubtedly the right choice to enforce strict access control and eliminate the security risk.

The special handling for pmem flush (keeping it read-only) is also a nice and correct touch.

Let us wait for the feedback from Leon and Jason.

Zhu Yanjun

Quoted 100 lines · Zhu Yanjun
Quoted 97 lines · Weiming Shi

Commit 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages") dropped the access permission test from rxe_check_pagefault() and left only HMM_PFN_VALID. A page faulted in read-only, for example a page-cache folio behind a PROT_READ file mapping, then satisfies the check and ODP write operations (RDMA WRITE, RDMA READ response, SEND payload, atomics) modify it through kmap without ever breaking CoW.

An unprivileged user can register an ODP MR over such a mapping and have incoming RDMA traffic overwrite the page cache of a file it only holds O_RDONLY, including /etc/passwd or setuid binaries. This is the same primitive class as Dirty COW and CVE-2022-2590.

mlx5 has the missing invariant: its ODP path sets the device write bit only for pfns that carry HMM_PFN_WRITE. Restore it in rxe by requiring HMM_PFN_WRITE in rxe_check_pagefault() for every operation except RXE_PAGEFAULT_RDONLY. A write to a non-writable VMA now fails the one fault attempt with -EPERM from hmm_vma_fault() instead of re-faulting forever. For a writable VMA the fault breaks CoW and the write lands in the private page.

Keep pmem flushes on the read-only check. arch_wb_cache_pmem() never modifies memory, and the FLUSH access bits do not make the umem writable, so classifying flushes as writes would make every flush against a flush-only MR fail.

Fixes 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages")

---
  drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----

  1 file changed, 11 insertions(+), 5 deletions(-)
drivers/infiniband/sw/rxe/rxe_odp.c +5 -1
index ff904d5e54a73..174bb2efc3c6a 100644
@@ -124,19 +124,23 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe, 
 u64 start, u64 length,
   }
     static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, u64 iova,
-                       int length)
+                       int length, bool write)
   {
       bool need_fault = false;
+    u64 access = HMM_PFN_VALID;
       u64 addr;
       int idx;
   +    if (write)
+        access |= HMM_PFN_WRITE;
+
       addr = iova & (~(BIT(umem_odp->page_shift) - 1));
         /* Skim through all pages that are to be accessed. */
       while (addr < iova + length) {
           idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift;
   -        if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) {
+        if ((umem_odp->map.pfn_list[idx] & access) != access) {
               need_fault = true;
               break;
           }
@@ -159,6 +163,7 @@ static unsigned long 

rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u
  static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u32 flags) {
      struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem);
+    bool write = !(flags & RXE_PAGEFAULT_RDONLY);
      bool need_fault;
      int err;
  @@ -167,7 +172,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u mutex_lock(&umem_odp->umem_mutex);
  -    need_fault = rxe_check_pagefault(umem_odp, iova, length);
+    need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
      if (need_fault) {
          mutex_unlock(&umem_odp->umem_mutex);
  @@ -177,7 +182,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u if (err < 0)
              return err;
  -        need_fault = rxe_check_pagefault(umem_odp, iova, length);
+        need_fault = rxe_check_pagefault(umem_odp, iova, length,
write);
          if (need_fault) {
              mutex_unlock(&umem_odp->umem_mutex);
              return -EFAULT;

drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
@@ -339,8 +344,9 @@ int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, 
 u64 iova,
       int err;
       u8 *va;

  +    /* A flush never modifies memory; read-only access suffices. */ err = rxe_odp_map_range_and_lock(mr, iova, length,
-                     RXE_PAGEFAULT_DEFAULT);
+                     RXE_PAGEFAULT_RDONLY);
      if (err)
          return err;

Signature · 2 lines
Best Regards,
Yanjun.Zhu
#3 lore raw
Quoted 7 lines · Zhu Yanjun

While exploiting this ODP flaw in soft-RoCE for privilege escalation in practice might require specific conditions, enforcing the HMM_PFN_WRITE check is undoubtedly the right choice to enforce strict access control and eliminate the security risk.

The special handling for pmem flush (keeping it read-only) is also a nice and correct touch.

Zhu, we already asked you, please reduce the amount of AI-generated emails.

Quoted 2 lines

Let us wait for the feedback from Leon and Jason.

There is nothing we can do here, as it is not available in Patchwork or lore. Please post the patch to the mailing list.

Thanks

Quoted 118 lines

Zhu Yanjun

Quoted 107 lines · Zhu Yanjun
Quoted 104 lines · Weiming Shi

Commit 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages") dropped the access permission test from rxe_check_pagefault() and left only HMM_PFN_VALID. A page faulted in read-only, for example a page-cache folio behind a PROT_READ file mapping, then satisfies the check and ODP write operations (RDMA WRITE, RDMA READ response, SEND payload, atomics) modify it through kmap without ever breaking CoW.

An unprivileged user can register an ODP MR over such a mapping and have incoming RDMA traffic overwrite the page cache of a file it only holds O_RDONLY, including /etc/passwd or setuid binaries. This is the same primitive class as Dirty COW and CVE-2022-2590.

mlx5 has the missing invariant: its ODP path sets the device write bit only for pfns that carry HMM_PFN_WRITE. Restore it in rxe by requiring HMM_PFN_WRITE in rxe_check_pagefault() for every operation except RXE_PAGEFAULT_RDONLY. A write to a non-writable VMA now fails the one fault attempt with -EPERM from hmm_vma_fault() instead of re-faulting forever. For a writable VMA the fault breaks CoW and the write lands in the private page.

Keep pmem flushes on the read-only check. arch_wb_cache_pmem() never modifies memory, and the FLUSH access bits do not make the umem writable, so classifying flushes as writes would make every flush against a flush-only MR fail.

Fixes 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO

pages")

---
  drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----

  1 file changed, 11 insertions(+), 5 deletions(-)
drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
index ff904d5e54a73..174bb2efc3c6a 100644
@@ -124,19 +124,23 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe,
 u64 start, u64 length,
   }

    static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, u64 iova,
-                       int length)
+                       int length, bool write)
  {
      bool need_fault = false;
+    u64 access = HMM_PFN_VALID;
      u64 addr;
      int idx;
  +    if (write)
+        access |= HMM_PFN_WRITE;
+
      addr = iova & (~(BIT(umem_odp->page_shift) - 1)); /* Skim through all pages that are to be accessed. */ while (addr < iova + length) {
          idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift;
  -        if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) { +        if ((umem_odp->map.pfn_list[idx] & access) != access) { need_fault = true;
              break;
          }

drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
@@ -159,6 +163,7 @@ static unsigned long

rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u32 flags)
  {
      struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); +    bool write = !(flags & RXE_PAGEFAULT_RDONLY);
      bool need_fault;
      int err;
  @@ -167,7 +172,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u mutex_lock(&umem_odp->umem_mutex);
  -    need_fault = rxe_check_pagefault(umem_odp, iova, length); +    need_fault = rxe_check_pagefault(umem_odp, iova, length, write); if (need_fault) { mutex_unlock(&umem_odp->umem_mutex);
  @@ -177,7 +182,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u
          if (err < 0)
              return err;
  -        need_fault = rxe_check_pagefault(umem_odp, iova, length); +        need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
          if (need_fault) { mutex_unlock(&umem_odp->umem_mutex);
              return -EFAULT;

drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
@@ -339,8 +344,9 @@ int rxe_odp_flush_pmem_iova(struct rxe_mr *mr,
 u64 iova,
       int err;
       u8 *va;

  +    /* A flush never modifies memory; read-only access suffices. */ err = rxe_odp_map_range_and_lock(mr, iova, length,
-                     RXE_PAGEFAULT_DEFAULT);
+                     RXE_PAGEFAULT_RDONLY);
      if (err)
          return err;

Signature · 2 lines
Best Regards,
Yanjun.Zhu
#4 lore raw
Quoted 14 lines · Leon Romanovsky <[email protected]> 于2026年7月26日周日 16:21写道:
Quoted 7 lines · Zhu Yanjun

While exploiting this ODP flaw in soft-RoCE for privilege escalation in practice might require specific conditions, enforcing the HMM_PFN_WRITE check is undoubtedly the right choice to enforce strict access control and eliminate the security risk.

The special handling for pmem flush (keeping it read-only) is also a nice and correct touch.

Zhu, we already asked you, please reduce the amount of AI-generated emails.

Quoted 1 line
Quoted 125 lines
Quoted 1 line

Let us wait for the feedback from Leon and Jason.

There is nothing we can do here, as it is not available in Patchwork or lore. Please post the patch to the mailing list.

Thanks

Quoted 118 lines

Zhu Yanjun

Quoted 98 lines · Zhu Yanjun
Quoted 96 lines · Weiming Shi

Commit 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages") dropped the access permission test from rxe_check_pagefault() and left only HMM_PFN_VALID. A page faulted in read-only, for example a page-cache folio behind a PROT_READ file mapping, then satisfies the check and ODP write operations (RDMA WRITE, RDMA READ response, SEND payload, atomics) modify it through kmap without ever breaking CoW.

An unprivileged user can register an ODP MR over such a mapping and have incoming RDMA traffic overwrite the page cache of a file it only holds O_RDONLY, including /etc/passwd or setuid binaries. This is the same primitive class as Dirty COW and CVE-2022-2590.

mlx5 has the missing invariant: its ODP path sets the device write bit only for pfns that carry HMM_PFN_WRITE. Restore it in rxe by requiring HMM_PFN_WRITE in rxe_check_pagefault() for every operation except RXE_PAGEFAULT_RDONLY. A write to a non-writable VMA now fails the one fault attempt with -EPERM from hmm_vma_fault() instead of re-faulting forever. For a writable VMA the fault breaks CoW and the write lands in the private page.

Keep pmem flushes on the read-only check. arch_wb_cache_pmem() never modifies memory, and the FLUSH access bits do not make the umem writable, so classifying flushes as writes would make every flush against a flush-only MR fail.

Fixes 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO

pages")

---
  drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
  1 file changed, 11 insertions(+), 5 deletions(-)
drivers/infiniband/sw/rxe/rxe_odp.c +5 -1
index ff904d5e54a73..174bb2efc3c6a 100644
@@ -124,19 +124,23 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe,
 u64 start, u64 length,
  }
    static inline bool rxe_check_pagefault(struct ib_umem_odp
 *umem_odp, u64 iova,
-                       int length)
+                       int length, bool write)
  {
      bool need_fault = false;
+    u64 access = HMM_PFN_VALID;
      u64 addr;
      int idx;
  +    if (write)
+        access |= HMM_PFN_WRITE;
+
      addr = iova & (~(BIT(umem_odp->page_shift) - 1));
        /* Skim through all pages that are to be accessed. */
      while (addr < iova + length) {
          idx = (addr - ib_umem_start(umem_odp)) >>
 umem_odp->page_shift;
  -        if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) {
+        if ((umem_odp->map.pfn_list[idx] & access) != access) {
              need_fault = true;
              break;
          }
@@ -159,6 +163,7 @@ static unsigned long

rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u
static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova,
int length, u32 flags)

  {
      struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem);
+    bool write = !(flags & RXE_PAGEFAULT_RDONLY);
      bool need_fault;
      int err;
  @@ -167,7 +172,7 @@ static int rxe_odp_map_range_and_lock(struct
rxe_mr *mr, u64 iova, int length, u
        mutex_lock(&umem_odp->umem_mutex);
  -    need_fault = rxe_check_pagefault(umem_odp, iova, length);
+    need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
      if (need_fault) {
          mutex_unlock(&umem_odp->umem_mutex);
  @@ -177,7 +182,7 @@ static int rxe_odp_map_range_and_lock(struct
rxe_mr *mr, u64 iova, int length, u
          if (err < 0)
              return err;
  -        need_fault = rxe_check_pagefault(umem_odp, iova, length);
+        need_fault = rxe_check_pagefault(umem_odp, iova, length,
write);
          if (need_fault) {
              mutex_unlock(&umem_odp->umem_mutex);
              return -EFAULT;
drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
@@ -339,8 +344,9 @@ int rxe_odp_flush_pmem_iova(struct rxe_mr *mr,
 u64 iova,
Quoted 9 lines

drivers/infiniband/sw/rxe/rxe_odp.c int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, · line 351

Quoted 8 lines
drivers/infiniband/sw/rxe/rxe_odp.c +1 -1
      int err;
      u8 *va;
  +    /* A flush never modifies memory; read-only access suffices. */
      err = rxe_odp_map_range_and_lock(mr, iova, length,
-                     RXE_PAGEFAULT_DEFAULT);
+                     RXE_PAGEFAULT_RDONLY);
      if (err)
          return err;

--
Best Regards,
Yanjun.Zhu

Hi Leon,

Thank you for the feedback. I will post the patch to mailing list as requested. Regarding the reproduction details, I will omit the PoC code in the public mailing list post for security reasons.

Best regards,
Weiming Shi

#5 lore raw
Quoted 141 lines · Weiming Shi
Quoted 14 lines · Leon Romanovsky <[email protected]> 于2026年7月26日周日 16:21写道:
Quoted 7 lines · Zhu Yanjun

While exploiting this ODP flaw in soft-RoCE for privilege escalation in practice might require specific conditions, enforcing the HMM_PFN_WRITE check is undoubtedly the right choice to enforce strict access control and eliminate the security risk.

The special handling for pmem flush (keeping it read-only) is also a nice and correct touch.

Zhu, we already asked you, please reduce the amount of AI-generated emails.

Quoted 1 line
Quoted 125 lines
Quoted 1 line

Let us wait for the feedback from Leon and Jason.

There is nothing we can do here, as it is not available in Patchwork or lore. Please post the patch to the mailing list.

Thanks

Quoted 118 lines

Zhu Yanjun

Quoted 98 lines · Zhu Yanjun
Quoted 96 lines · Weiming Shi

Commit 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO pages") dropped the access permission test from rxe_check_pagefault() and left only HMM_PFN_VALID. A page faulted in read-only, for example a page-cache folio behind a PROT_READ file mapping, then satisfies the check and ODP write operations (RDMA WRITE, RDMA READ response, SEND payload, atomics) modify it through kmap without ever breaking CoW.

An unprivileged user can register an ODP MR over such a mapping and have incoming RDMA traffic overwrite the page cache of a file it only holds O_RDONLY, including /etc/passwd or setuid binaries. This is the same primitive class as Dirty COW and CVE-2022-2590.

mlx5 has the missing invariant: its ODP path sets the device write bit only for pfns that carry HMM_PFN_WRITE. Restore it in rxe by requiring HMM_PFN_WRITE in rxe_check_pagefault() for every operation except RXE_PAGEFAULT_RDONLY. A write to a non-writable VMA now fails the one fault attempt with -EPERM from hmm_vma_fault() instead of re-faulting forever. For a writable VMA the fault breaks CoW and the write lands in the private page.

Keep pmem flushes on the read-only check. arch_wb_cache_pmem() never modifies memory, and the FLUSH access bits do not make the umem writable, so classifying flushes as writes would make every flush against a flush-only MR fail.

Fixes 0b261d7c1cd3 ("RDMA/rxe: Break endless pagefault loop for RO

pages")

---
  drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
  1 file changed, 11 insertions(+), 5 deletions(-)
drivers/infiniband/sw/rxe/rxe_odp.c +5 -1
index ff904d5e54a73..174bb2efc3c6a 100644
@@ -124,19 +124,23 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe,
 u64 start, u64 length,
  }
    static inline bool rxe_check_pagefault(struct ib_umem_odp
 *umem_odp, u64 iova,
-                       int length)
+                       int length, bool write)
  {
      bool need_fault = false;
+    u64 access = HMM_PFN_VALID;
      u64 addr;
      int idx;
  +    if (write)
+        access |= HMM_PFN_WRITE;
+
      addr = iova & (~(BIT(umem_odp->page_shift) - 1));
        /* Skim through all pages that are to be accessed. */
      while (addr < iova + length) {
          idx = (addr - ib_umem_start(umem_odp)) >>
 umem_odp->page_shift;
  -        if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) {
+        if ((umem_odp->map.pfn_list[idx] & access) != access) {
              need_fault = true;
              break;
          }
@@ -159,6 +163,7 @@ static unsigned long

rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u
static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova,
int length, u32 flags)

  {
      struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem);
+    bool write = !(flags & RXE_PAGEFAULT_RDONLY);
      bool need_fault;
      int err;
  @@ -167,7 +172,7 @@ static int rxe_odp_map_range_and_lock(struct
rxe_mr *mr, u64 iova, int length, u
        mutex_lock(&umem_odp->umem_mutex);
  -    need_fault = rxe_check_pagefault(umem_odp, iova, length);
+    need_fault = rxe_check_pagefault(umem_odp, iova, length, write);
      if (need_fault) {
          mutex_unlock(&umem_odp->umem_mutex);
  @@ -177,7 +182,7 @@ static int rxe_odp_map_range_and_lock(struct
rxe_mr *mr, u64 iova, int length, u
          if (err < 0)
              return err;
  -        need_fault = rxe_check_pagefault(umem_odp, iova, length);
+        need_fault = rxe_check_pagefault(umem_odp, iova, length,
write);
          if (need_fault) {
              mutex_unlock(&umem_odp->umem_mutex);
              return -EFAULT;
drivers/infiniband/sw/rxe/rxe_odp.c +0 -0
@@ -339,8 +344,9 @@ int rxe_odp_flush_pmem_iova(struct rxe_mr *mr,
 u64 iova,
Quoted 9 lines

drivers/infiniband/sw/rxe/rxe_odp.c int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, · line 351

Quoted 8 lines
drivers/infiniband/sw/rxe/rxe_odp.c +1 -1
      int err;
      u8 *va;
  +    /* A flush never modifies memory; read-only access suffices. */
      err = rxe_odp_map_range_and_lock(mr, iova, length,
-                     RXE_PAGEFAULT_DEFAULT);
+                     RXE_PAGEFAULT_RDONLY);
      if (err)
          return err;

--
Best Regards,
Yanjun.Zhu

Quoted 5 lines

Hi Leon,

Thank you for the feedback. I will post the patch to mailing list as requested. Regarding the reproduction details, I will omit the PoC code in the public mailing list post for security reasons.

There is no security concern here. RXE is a virtual device used for testing and prototyping.

Thanks

Quoted 3 lines

Best regards,
Weiming Shi