FAILED: patch "[PATCH] IB/core: Make testing MR flags for writability a static" failed to apply to 4.4-stable tree
Subsystems (2)
[email protected]
maintainerThe patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
Quoted 1 line
From
08bb558ac11ab944e0539e78619d7b4c356278bdMon Sep 17 00:00:00 2001
From: Jack Morgenstein <[email protected]>
Date: Wed, 23 May 2018 15:30:30 +0300
Subject: [PATCH] IB/core: Make testing MR flags for writability a static
inline function
Make the MR writability flags check, which is performed in umem.c, a static inline function in file ib_verbs.h
This allows the function to be used by low-level infiniband drivers.
drivers/infiniband/core/umem.c +1 -10
@@ -119,16 +119,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, umem->length = size; umem->address = addr; umem->page_shift =PAGE_SHIFT ; -/* -* We ask for writable memory if any of the following -* access flags are set. "Local write" and "remote write" -* obviously require write access. "Remote atomic" can do -* things like fetch and add, which will modify memory, and -* "MW bind" can change permissions by binding a window. -*/ - umem->writable = !!(access & - (IB_ACCESS_LOCAL_WRITE |IB_ACCESS_REMOTE_WRITE | -IB_ACCESS_REMOTE_ATOMIC |IB_ACCESS_MW_BIND )); + umem->writable =ib_access_writable (access);if (access &IB_ACCESS_ON_DEMAND ) { ret =ib_umem_odp_get (context, umem, access);
include/rdma/ib_verbs.h +14 -0
@@ -3734,6 +3734,20 @@ static inline int ib_check_mr_access(int flags)return 0 ; } +static inline bool ib_access_writable (int access_flags) +{ +/* +* We have writable memory backing the MR if any of the following +* access flags are set. "Local write" and "remote write" obviously +* require write access. "Remote atomic" can do things like fetch and +* add, which will modify memory, and "MW bind" can change permissions +* by binding a window. +*/ +return access_flags & + (IB_ACCESS_LOCAL_WRITE |IB_ACCESS_REMOTE_WRITE | +IB_ACCESS_REMOTE_ATOMIC |IB_ACCESS_MW_BIND ); +} +/** * ib_check_mr_status: lightweight check of MR status. * This routine may provide status checks on a selected
Hi Greg,
Quoted 5 lines · On Sun, Jul 01, 2018 at 11:37:25AM +0200, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
The attached backported patch should apply to 4.4-stable and also to 4.9-stable.
Though I am confused about why this should be in stable, unless some later fix, required in stable, depends on this.
--
Regards
Sudip
Quoted 1 line
From
5a8146ce6d32fd2321a0935ea7169be2d785fd41Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <[email protected]>
Date: Wed, 23 May 2018 15:30:30 +0300
Subject: [PATCH] IB/core: Make testing MR flags for writability a static inline function
commit 08bb558ac11ab944e0539e78619d7b4c356278bd upstream
Make the MR writability flags check, which is performed in umem.c, a static inline function in file ib_verbs.h
This allows the function to be used by low-level infiniband drivers.
--- drivers/infiniband/core/umem.c | 11 +---------- include/rdma/ib_verbs.h | 14 ++++++++++++++ 2 files changed, 15 insertions(+), 10 deletions(-)
drivers/infiniband/core/umem.c +1 -10
@@ -122,16 +122,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, umem->address = addr; umem->page_size =PAGE_SIZE ; umem->pid =get_task_pid (current,PIDTYPE_PID ); -/* -* We ask for writable memory if any of the following -* access flags are set. "Local write" and "remote write" -* obviously require write access. "Remote atomic" can do -* things like fetch and add, which will modify memory, and -* "MW bind" can change permissions by binding a window. -*/ - umem->writable = !!(access & - (IB_ACCESS_LOCAL_WRITE |IB_ACCESS_REMOTE_WRITE | -IB_ACCESS_REMOTE_ATOMIC |IB_ACCESS_MW_BIND )); + umem->writable =ib_access_writable (access);if (access &IB_ACCESS_ON_DEMAND ) {put_pid (umem->pid);
include/rdma/ib_verbs.h +14 -0
@@ -3007,6 +3007,20 @@ static inline int ib_check_mr_access(int flags)return 0 ; } +static inline bool ib_access_writable (int access_flags) +{ +/* +* We have writable memory backing the MR if any of the following +* access flags are set. "Local write" and "remote write" obviously +* require write access. "Remote atomic" can do things like fetch and +* add, which will modify memory, and "MW bind" can change permissions +* by binding a window. +*/ +return access_flags & + (IB_ACCESS_LOCAL_WRITE |IB_ACCESS_REMOTE_WRITE | +IB_ACCESS_REMOTE_ATOMIC |IB_ACCESS_MW_BIND ); +} +/** * ib_check_mr_status: lightweight check of MR status. * This routine may provide status checks on a selected
Signature · 1 line
2.11.0
Greg KH
maintainerQuoted 9 lines · Sudip Mukherjee
Hi Greg,
Quoted 5 lines · On Sun, Jul 01, 2018 at 11:37:25AM +0200, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
Quoted 5 lines
The attached backported patch should apply to 4.4-stable and also to 4.9-stable.
Though I am confused about why this should be in stable, unless some later fix, required in stable, depends on this.
That's a good question. Can some IB developer please answer this before I queue this up?
thanks,
greg k-h
Quoted 10 lines · Greg KH
Quoted 9 lines · Sudip Mukherjee
Hi Greg,
Quoted 5 lines · On Sun, Jul 01, 2018 at 11:37:25AM +0200, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
Quoted 8 lines
Quoted 5 lines
The attached backported patch should apply to 4.4-stable and also to 4.9-stable.
Though I am confused about why this should be in stable, unless some later fix, required in stable, depends on this.
That's a good question. Can some IB developer please answer this before I queue this up?
It is required for this upstream patch:
commit d8f9cc328c8888369880e2527e9186d745f2bbf6
Author: Jack Morgenstein <[email protected]> Date: Wed May 23 15:30:31 2018 +0300
IB/mlx4: Mark user MR as writable if actual virtual memory is writable
To allow rereg_user_mr to modify the MR from read-only to writable without using get_user_pages again, we needed to define the initial MR as writable. However, this was originally done unconditionally, without taking into account the writability of the underlying virtual memory.
As a result, any attempt to register a read-only MR over read-only virtual memory failed.
To fix this, do not add the writable flag bit when the user virtual memory is not writable (e.g. const memory).
However, when the underlying memory is NOT writable (and we therefore do not define the initial MR as writable), the IB core adds a "force writable" flag to its user-pages request. If this succeeds, the reg_user_mr caller gets a writable copy of the original pages.
If the user-space caller then does a rereg_user_mr operation to enable writability, this will succeed. This should not be allowed, since the original virtual memory was not writable.
Quoted 17 lines · Jason Gunthorpe
Quoted 16 lines · Greg KH
Quoted 14 lines · Sudip Mukherjee
Hi Greg,
Quoted 5 lines · On Sun, Jul 01, 2018 at 11:37:25AM +0200, [email protected] wrote:
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.The attached backported patch should apply to 4.4-stable and also to 4.9-stable.
Though I am confused about why this should be in stable, unless some later fix, required in stable, depends on this.
Quoted 8 lines
Quoted 2 lines
That's a good question. Can some IB developer please answer this before I queue this up?
It is required for this upstream patch:
commit
d8f9cc328c8888369880e2527e9186d745f2bbf6Author: Jack Morgenstein <[email protected]> Date: Wed May 23 15:30:31 2018 +0300
Thanks Jason.
Looks like d8f9cc328c88 ("IB/mlx4: Mark user MR as writable if actual virtual memory is writable") will be needed in both 4.4-stable and 4.9-stable, but will need backporting.
--
Regards
Sudip