* Re: [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
[not found] <20260725094805.916278-1-bestswngs@gmail.com>
@ 2026-07-26 0:08 ` Zhu Yanjun
2026-07-26 2:12 ` Zhu Yanjun
0 siblings, 1 reply; 6+ messages in thread
From: Zhu Yanjun @ 2026-07-26 0:08 UTC (permalink / raw)
To: Weiming Shi, security
Cc: Jason Gunthorpe, Zhu Yanjun, Qiang Hong, Xinyu Ma, Shaomin Chen,
Rui Ding, Miao Zhao, RDMA mailing list
在 2026/7/25 2:48, Weiming Shi 写道:
> 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")
> Cc: stable@vger.kernel.org
> Signed-off-by: Weiming Shi <bestswngs@gmail.com>
> Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
> Reported-by: Rui Ding <threonine42@gmail.com>
> Reported-by: Miao Zhao <muel@nova.gal>
> Tested-by: Qiang Hong <wanbafv@gmail.com>
> Tested-by: Xinyu Ma <mmmxny@gmail.com>
> ---
> drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
> 1 file changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c b/drivers/infiniband/sw/rxe/rxe_odp.c
> index ff904d5e54a73..174bb2efc3c6a 100644
> --- a/drivers/infiniband/sw/rxe/rxe_odp.c
> +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
> @@ -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;
>
--
Best Regards,
Yanjun.Zhu
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
2026-07-26 0:08 ` [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths Zhu Yanjun
@ 2026-07-26 2:12 ` Zhu Yanjun
2026-07-26 8:20 ` Leon Romanovsky
0 siblings, 1 reply; 6+ messages in thread
From: Zhu Yanjun @ 2026-07-26 2:12 UTC (permalink / raw)
To: Weiming Shi, security, yanjun.zhu@linux.dev
Cc: Jason Gunthorpe, Zhu Yanjun, Qiang Hong, Xinyu Ma, Shaomin Chen,
Rui Ding, Miao Zhao, RDMA mailing list, leon@kernel.org
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.
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Zhu Yanjun
在 2026/7/25 17:08, Zhu Yanjun 写道:
>
> 在 2026/7/25 2:48, Weiming Shi 写道:
>> 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")
>> Cc: stable@vger.kernel.org
>> Signed-off-by: Weiming Shi <bestswngs@gmail.com>
>> Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
>> Reported-by: Rui Ding <threonine42@gmail.com>
>> Reported-by: Miao Zhao <muel@nova.gal>
>> Tested-by: Qiang Hong <wanbafv@gmail.com>
>> Tested-by: Xinyu Ma <mmmxny@gmail.com>
>> ---
>> drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
>> 1 file changed, 11 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c
>> b/drivers/infiniband/sw/rxe/rxe_odp.c
>> index ff904d5e54a73..174bb2efc3c6a 100644
>> --- a/drivers/infiniband/sw/rxe/rxe_odp.c
>> +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
>> @@ -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;
>
--
Best Regards,
Yanjun.Zhu
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
2026-07-26 2:12 ` Zhu Yanjun
@ 2026-07-26 8:20 ` Leon Romanovsky
2026-07-26 10:42 ` Weiming Shi
0 siblings, 1 reply; 6+ messages in thread
From: Leon Romanovsky @ 2026-07-26 8:20 UTC (permalink / raw)
To: Zhu Yanjun
Cc: Weiming Shi, security, Jason Gunthorpe, Zhu Yanjun, Qiang Hong,
Xinyu Ma, Shaomin Chen, Rui Ding, Miao Zhao, RDMA mailing list
On Sat, Jul 25, 2026 at 07:12:42PM -0700, Zhu Yanjun wrote:
> 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.
>
> 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
>
> Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
>
> Zhu Yanjun
>
> 在 2026/7/25 17:08, Zhu Yanjun 写道:
> >
> > 在 2026/7/25 2:48, Weiming Shi 写道:
> > > 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")
> > > Cc: stable@vger.kernel.org
> > > Signed-off-by: Weiming Shi <bestswngs@gmail.com>
> > > Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
> > > Reported-by: Rui Ding <threonine42@gmail.com>
> > > Reported-by: Miao Zhao <muel@nova.gal>
> > > Tested-by: Qiang Hong <wanbafv@gmail.com>
> > > Tested-by: Xinyu Ma <mmmxny@gmail.com>
> > > ---
> > > drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
> > > 1 file changed, 11 insertions(+), 5 deletions(-)
> > >
> > > diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > index ff904d5e54a73..174bb2efc3c6a 100644
> > > --- a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > @@ -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;
> >
> --
> Best Regards,
> Yanjun.Zhu
>
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
2026-07-26 8:20 ` Leon Romanovsky
@ 2026-07-26 10:42 ` Weiming Shi
2026-07-26 12:11 ` Leon Romanovsky
0 siblings, 1 reply; 6+ messages in thread
From: Weiming Shi @ 2026-07-26 10:42 UTC (permalink / raw)
To: Leon Romanovsky
Cc: Zhu Yanjun, security, Jason Gunthorpe, Zhu Yanjun, Qiang Hong,
Xinyu Ma, Shaomin Chen, Rui Ding, Miao Zhao, RDMA mailing list
Leon Romanovsky <leon@kernel.org> 于2026年7月26日周日 16:21写道：
>
> On Sat, Jul 25, 2026 at 07:12:42PM -0700, Zhu Yanjun wrote:
> > 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.
>
> >
> > 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
>
> >
> > Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
> >
> > Zhu Yanjun
> >
> > 在 2026/7/25 17:08, Zhu Yanjun 写道:
> > >
> > > 在 2026/7/25 2:48, Weiming Shi 写道:
> > > > 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")
> > > > Cc: stable@vger.kernel.org
> > > > Signed-off-by: Weiming Shi <bestswngs@gmail.com>
> > > > Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
> > > > Reported-by: Rui Ding <threonine42@gmail.com>
> > > > Reported-by: Miao Zhao <muel@nova.gal>
> > > > Tested-by: Qiang Hong <wanbafv@gmail.com>
> > > > Tested-by: Xinyu Ma <mmmxny@gmail.com>
> > > > ---
> > > > drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
> > > > 1 file changed, 11 insertions(+), 5 deletions(-)
> > > >
> > > > diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > index ff904d5e54a73..174bb2efc3c6a 100644
> > > > --- a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > @@ -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;
> > >
> > --
> > 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
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
@ 2026-07-26 11:15 Weiming Shi
0 siblings, 0 replies; 6+ messages in thread
From: Weiming Shi @ 2026-07-26 11:15 UTC (permalink / raw)
To: linux-rdma
Cc: Jason Gunthorpe, Zhu Yanjun, Leon Romanovsky, Weiming Shi,
Hongqiang Luo, Xinyu Ma, Shaomin Chen, Rui Ding, Miao Zhao
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")
Cc: stable@vger.kernel.org
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
Reported-by: Weiming Shi <bestswngs@gmail.com>
Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
Reported-by: Rui Ding <threonine42@gmail.com>
Reported-by: Miao Zhao <muel@nova.gal>
Tested-by: Hongqiang Luo <wanbafv@gmail.com>
Tested-by: Xinyu Ma <mmmxny@gmail.com>
Tested-by: Zhanbo Ye <cainyzb@gmail.com>
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
---
drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c b/drivers/infiniband/sw/rxe/rxe_odp.c
index ff904d5e54a73..174bb2efc3c6a 100644
--- a/drivers/infiniband/sw/rxe/rxe_odp.c
+++ b/drivers/infiniband/sw/rxe/rxe_odp.c
@@ -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;
--
2.43.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths
2026-07-26 10:42 ` Weiming Shi
@ 2026-07-26 12:11 ` Leon Romanovsky
0 siblings, 0 replies; 6+ messages in thread
From: Leon Romanovsky @ 2026-07-26 12:11 UTC (permalink / raw)
To: Weiming Shi
Cc: Zhu Yanjun, security, Jason Gunthorpe, Zhu Yanjun, Qiang Hong,
Xinyu Ma, Shaomin Chen, Rui Ding, Miao Zhao, RDMA mailing list
On Sun, Jul 26, 2026 at 06:42:12PM +0800, Weiming Shi wrote:
> Leon Romanovsky <leon@kernel.org> 于2026年7月26日周日 16:21写道：
> >
> > On Sat, Jul 25, 2026 at 07:12:42PM -0700, Zhu Yanjun wrote:
> > > 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.
> >
> > >
> > > 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
> >
> > >
> > > Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
> > >
> > > Zhu Yanjun
> > >
> > > 在 2026/7/25 17:08, Zhu Yanjun 写道:
> > > >
> > > > 在 2026/7/25 2:48, Weiming Shi 写道:
> > > > > 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")
> > > > > Cc: stable@vger.kernel.org
> > > > > Signed-off-by: Weiming Shi <bestswngs@gmail.com>
> > > > > Reported-by: Shaomin Chen <eeesssooo020@gmail.com>
> > > > > Reported-by: Rui Ding <threonine42@gmail.com>
> > > > > Reported-by: Miao Zhao <muel@nova.gal>
> > > > > Tested-by: Qiang Hong <wanbafv@gmail.com>
> > > > > Tested-by: Xinyu Ma <mmmxny@gmail.com>
> > > > > ---
> > > > > drivers/infiniband/sw/rxe/rxe_odp.c | 16 +++++++++++-----
> > > > > 1 file changed, 11 insertions(+), 5 deletions(-)
> > > > >
> > > > > diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > > b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > > index ff904d5e54a73..174bb2efc3c6a 100644
> > > > > --- a/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > > +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
> > > > > @@ -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;
> > > >
> > > --
> > > 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.
There is no security concern here. RXE is a virtual device used for
testing and prototyping.
Thanks
>
> Best regards,
> Weiming Shi
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-07-26 12:11 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20260725094805.916278-1-bestswngs@gmail.com>
2026-07-26 0:08 ` [PATCH] RDMA/rxe: Restore HMM_PFN_WRITE check in ODP write paths Zhu Yanjun
2026-07-26 2:12 ` Zhu Yanjun
2026-07-26 8:20 ` Leon Romanovsky
2026-07-26 10:42 ` Weiming Shi
2026-07-26 12:11 ` Leon Romanovsky
2026-07-26 11:15 Weiming Shi
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox