2019-05-19 13:08:55 +01:00
|
|
|
|
2014-02-05 19:11:33 -05:00
|
|
|
|
2016-11-01 07:40:13 -06:00
|
|
|
|
2020-10-15 20:13:50 -07:00
|
|
|
|
2014-02-05 19:11:33 -05:00
|
|
|
|
|
|
|
|
|
2021-05-04 18:40:03 -07:00
|
|
|
|
2014-03-21 04:58:33 -04:00
|
|
|
|
|
|
|
|
|
2016-09-22 16:33:12 -04:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2018-12-03 17:52:09 -08:00
|
|
|
|
2020-01-21 17:05:11 +01:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2014-02-05 19:11:33 -05:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
2020-10-15 20:13:50 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2022-09-15 17:03:37 +02:00
|
|
|
|
2020-10-15 20:13:50 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
2025-11-17 17:43:42 +01:00
|
|
|
|
2025-11-17 17:43:41 +01:00
|
|
|
|
2025-11-17 17:43:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-17 17:43:41 +01:00
|
|
|
|
2025-11-17 17:43:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-11-17 17:43:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 17:03:37 +02:00
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
2021-05-02 11:57:37 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
|
|
|
|
|
2021-05-02 11:57:37 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-02 11:57:37 -04:00
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-02 14:54:16 +02:00
|
|
|
|
2015-03-11 10:43:31 -04:00
|
|
|
|
2021-07-05 17:26:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-07-05 17:26:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-07-05 17:26:28 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2014-03-05 19:28:09 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
2021-07-12 12:06:14 +02:00
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-03-05 19:28:09 -05:00
|
|
|
|
|
|
|
|
|
2014-03-15 04:05:57 -04:00
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
|
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
|
|
|
|
|
2021-09-07 19:58:54 -07:00
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-22 14:50:17 +01:00
|
|
|
|
|
|
|
|
|
2021-09-07 19:58:54 -07:00
|
|
|
|
|
|
|
|
|
2018-07-08 13:46:02 -07:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
|
|
|
|
|
2018-05-03 17:06:31 -07:00
|
|
|
|
2023-09-25 13:03:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: hwpoison: coredump: support recovery from dump_user_range()
dump_user_range() is used to copy the user page to a coredump file, but if
a hardware memory error occurred during copy, which called from
__kernel_write_iter() in dump_user_range(), it crashes,
CPU: 112 PID: 7014 Comm: mca-recover Not tainted 6.3.0-rc2 #425
pc : __memcpy+0x110/0x260
lr : _copy_from_iter+0x3bc/0x4c8
...
Call trace:
__memcpy+0x110/0x260
copy_page_from_iter+0xcc/0x130
pipe_write+0x164/0x6d8
__kernel_write_iter+0x9c/0x210
dump_user_range+0xc8/0x1d8
elf_core_dump+0x308/0x368
do_coredump+0x2e8/0xa40
get_signal+0x59c/0x788
do_signal+0x118/0x1f8
do_notify_resume+0xf0/0x280
el0_da+0x130/0x138
el0t_64_sync_handler+0x68/0xc0
el0t_64_sync+0x188/0x190
Generally, the '->write_iter' of file ops will use copy_page_from_iter()
and copy_page_from_iter_atomic(), change memcpy() to copy_mc_to_kernel()
in both of them to handle #MC during source read, which stop coredump
processing and kill the task instead of kernel panic, but the source
address may not always a user address, so introduce a new copy_mc flag in
struct iov_iter{} to indicate that the iter could do a safe memory copy,
also introduce the helpers to set/cleck the flag, for now, it's only used
in coredump's dump_user_range(), but it could expand to any other
scenarios to fix the similar issue.
Link: https://lkml.kernel.org/r/20230417045323.11054-1-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Tong Tiangen <tongtiangen@huawei.com>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2023-04-17 12:53:23 +08:00
|
|
|
|
|
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2016-09-22 16:33:12 -04:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2017-06-29 22:25:14 -04:00
|
|
|
|
2023-09-25 13:03:05 +01:00
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-03-30 13:11:07 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-11-27 20:27:08 -05:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2016-09-22 16:33:12 -04:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-27 20:27:08 -05:00
|
|
|
|
2017-06-29 21:45:10 -04:00
|
|
|
|
2014-11-27 20:27:08 -05:00
|
|
|
|
2017-05-29 12:22:50 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-03-30 14:52:45 -07:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-07-08 13:46:07 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-09-07 19:58:54 -07:00
|
|
|
|
2018-07-08 13:46:07 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-09-07 19:58:54 -07:00
|
|
|
|
|
|
|
|
|
2018-07-08 13:46:07 -07:00
|
|
|
|
Merge branch 'uaccess-work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull iov_iter hardening from Al Viro:
"This is the iov_iter/uaccess/hardening pile.
For one thing, it trims the inline part of copy_to_user/copy_from_user
to the minimum that *does* need to be inlined - object size checks,
basically. For another, it sanitizes the checks for iov_iter
primitives. There are 4 groups of checks: access_ok(), might_fault(),
object size and KASAN.
- access_ok() had been verified by whoever had set the iov_iter up.
However, that has happened in a function far away, so proving that
there's no path to actual copying bypassing those checks is hard
and proving that iov_iter has not been buggered in the meanwhile is
also not pleasant. So we want those redone in actual
copyin/copyout.
- might_fault() is better off consolidated - we know whether it needs
to be checked as soon as we enter iov_iter primitive and observe
the iov_iter flavour. No need to wait until the copyin/copyout. The
call chains are short enough to make sure we won't miss anything -
in fact, it's more robust that way, since there are cases where we
do e.g. forced fault-in before getting to copyin/copyout. It's not
quite what we need to check (in particular, combination of
iovec-backed and set_fs(KERNEL_DS) is almost certainly a bug, not a
cause to skip checks), but that's for later series. For now let's
keep might_fault().
- KASAN checks belong in copyin/copyout - at the same level where
other iov_iter flavours would've hit them in memcpy().
- object size checks should apply to *all* iov_iter flavours, not
just iovec-backed ones.
There are two groups of primitives - one gets the kernel object
described as pointer + size (copy_to_iter(), etc.) while another gets
it as page + offset + size (copy_page_to_iter(), etc.)
For the first group the checks are best done where we actually have a
chance to find the object size. In other words, those belong in inline
wrappers in uio.h, before calling into iov_iter.c. Same kind as we
have for inlined part of copy_to_user().
For the second group there is no object to look at - offset in page is
just a number, it bears no type information. So we do them in the
common helper called by iov_iter.c primitives of that kind. All it
currently does is checking that we are not trying to access outside of
the compound page; eventually we might want to add some sanity checks
on the page involved.
So the things we need in copyin/copyout part of iov_iter.c do not
quite match anything in uaccess.h (we want no zeroing, we *do* want
access_ok() and KASAN and we want no might_fault() or object size
checks done on that level). OTOH, these needs are simple enough to
provide a couple of helpers (static in iov_iter.c) doing just what we
need..."
* 'uaccess-work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
iov_iter: saner checks on copyin/copyout
iov_iter: sanity checks for copy to/from page primitives
iov_iter/hardening: move object size checks to inlined part
copy_{to,from}_user(): consolidate object size checks
copy_{from,to}_user(): move kasan checks and might_fault() out-of-line
2017-07-07 20:39:20 -07:00
|
|
|
|
2017-05-29 12:22:50 -07:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2017-05-29 12:22:50 -07:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-05-29 12:22:50 -07:00
|
|
|
|
Merge branch 'uaccess-work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull iov_iter hardening from Al Viro:
"This is the iov_iter/uaccess/hardening pile.
For one thing, it trims the inline part of copy_to_user/copy_from_user
to the minimum that *does* need to be inlined - object size checks,
basically. For another, it sanitizes the checks for iov_iter
primitives. There are 4 groups of checks: access_ok(), might_fault(),
object size and KASAN.
- access_ok() had been verified by whoever had set the iov_iter up.
However, that has happened in a function far away, so proving that
there's no path to actual copying bypassing those checks is hard
and proving that iov_iter has not been buggered in the meanwhile is
also not pleasant. So we want those redone in actual
copyin/copyout.
- might_fault() is better off consolidated - we know whether it needs
to be checked as soon as we enter iov_iter primitive and observe
the iov_iter flavour. No need to wait until the copyin/copyout. The
call chains are short enough to make sure we won't miss anything -
in fact, it's more robust that way, since there are cases where we
do e.g. forced fault-in before getting to copyin/copyout. It's not
quite what we need to check (in particular, combination of
iovec-backed and set_fs(KERNEL_DS) is almost certainly a bug, not a
cause to skip checks), but that's for later series. For now let's
keep might_fault().
- KASAN checks belong in copyin/copyout - at the same level where
other iov_iter flavours would've hit them in memcpy().
- object size checks should apply to *all* iov_iter flavours, not
just iovec-backed ones.
There are two groups of primitives - one gets the kernel object
described as pointer + size (copy_to_iter(), etc.) while another gets
it as page + offset + size (copy_page_to_iter(), etc.)
For the first group the checks are best done where we actually have a
chance to find the object size. In other words, those belong in inline
wrappers in uio.h, before calling into iov_iter.c. Same kind as we
have for inlined part of copy_to_user().
For the second group there is no object to look at - offset in page is
just a number, it bears no type information. So we do them in the
common helper called by iov_iter.c primitives of that kind. All it
currently does is checking that we are not trying to access outside of
the compound page; eventually we might want to add some sanity checks
on the page involved.
So the things we need in copyin/copyout part of iov_iter.c do not
quite match anything in uaccess.h (we want no zeroing, we *do* want
access_ok() and KASAN and we want no might_fault() or object size
checks done on that level). OTOH, these needs are simple enough to
provide a couple of helpers (static in iov_iter.c) doing just what we
need..."
* 'uaccess-work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
iov_iter: saner checks on copyin/copyout
iov_iter: sanity checks for copy to/from page primitives
iov_iter/hardening: move object size checks to inlined part
copy_{to,from}_user(): consolidate object size checks
copy_{from,to}_user(): move kasan checks and might_fault() out-of-line
2017-07-07 20:39:20 -07:00
|
|
|
|
2017-05-29 12:22:50 -07:00
|
|
|
|
|
|
|
|
|
2017-06-29 21:52:57 -04:00
|
|
|
|
|
|
|
|
|
2019-02-26 10:42:39 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-08-29 11:20:32 -07:00
|
|
|
|
2022-07-29 13:01:57 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-29 21:52:57 -04:00
|
|
|
|
2016-11-01 22:09:04 -04:00
|
|
|
|
2021-04-29 20:42:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-07-29 13:01:57 -04:00
|
|
|
|
2021-04-29 20:42:25 -04:00
|
|
|
|
2022-09-15 20:11:15 -04:00
|
|
|
|
|
|
|
|
|
2021-04-29 20:42:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-23 17:21:37 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-29 20:42:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-27 14:22:37 -05:00
|
|
|
|
|
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-11-17 13:38:46 -08:00
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
2023-03-22 18:57:03 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-27 14:22:37 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-07-29 12:54:53 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-27 12:33:24 -04:00
|
|
|
|
2022-07-29 12:54:53 -04:00
|
|
|
|
|
|
|
|
|
2021-04-27 12:33:24 -04:00
|
|
|
|
2022-07-29 12:54:53 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-02 17:25:59 -04:00
|
|
|
|
2022-07-29 12:54:53 -04:00
|
|
|
|
2014-11-27 14:22:37 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
|
|
|
|
|
2023-09-25 13:03:04 +01:00
|
|
|
|
|
|
|
|
|
2014-08-01 09:27:22 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
2023-07-15 23:03:52 -04:00
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
2017-06-29 21:52:57 -04:00
|
|
|
|
2023-07-15 23:03:52 -04:00
|
|
|
|
2016-09-22 16:33:12 -04:00
|
|
|
|
2023-07-15 23:03:52 -04:00
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
2023-07-15 23:03:52 -04:00
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
|
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
2023-06-06 16:43:32 -04:00
|
|
|
|
|
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2025-05-14 18:06:04 +01:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2021-01-09 16:03:01 +00:00
|
|
|
|
|
|
|
|
|
2022-06-06 23:44:33 -04:00
|
|
|
|
2021-01-09 16:03:01 +00:00
|
|
|
|
2022-06-06 23:44:33 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-09 16:03:01 +00:00
|
|
|
|
2022-06-06 23:44:33 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-09 16:03:01 +00:00
|
|
|
|
|
|
|
|
|
2021-04-23 12:58:53 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-23 12:58:53 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
|
|
|
|
|
2021-04-23 12:58:53 -04:00
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
|
|
|
|
|
2021-04-23 22:24:08 -04:00
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-23 12:58:53 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
2021-04-23 12:58:53 -04:00
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-05-08 13:54:47 -04:00
|
|
|
|
|
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
|
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2017-02-17 18:42:24 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-02 17:25:59 -04:00
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-06-02 17:25:59 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-02 17:25:59 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2015-01-23 01:08:07 -05:00
|
|
|
|
2014-11-24 14:46:11 -05:00
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-11-24 14:46:11 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2015-01-23 01:08:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2015-01-23 01:08:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-22 14:50:39 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
iov_iter.c: macros for iterating over iov_iter
iterate_all_kinds(iter, size, ident, step_iovec, step_bvec)
iterates through the ranges covered by iter (up to size bytes total),
repeating step_iovec or step_bvec for each of those. ident is
declared in expansion of that thing, either as struct iovec or
struct bvec, and it contains the range we are currently looking
at. step_bvec should be a void expression, step_iovec - a size_t
one, with non-zero meaning "stop here, that many bytes from this
range left". In the end, the amount actually handled is stored
in size.
iov_iter_copy_from_user_atomic() and iov_iter_alignment() converted
to it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2014-11-27 13:51:41 -05:00
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
iov_iter.c: macros for iterating over iov_iter
iterate_all_kinds(iter, size, ident, step_iovec, step_bvec)
iterates through the ranges covered by iter (up to size bytes total),
repeating step_iovec or step_bvec for each of those. ident is
declared in expansion of that thing, either as struct iovec or
struct bvec, and it contains the range we are currently looking
at. step_bvec should be a void expression, step_iovec - a size_t
one, with non-zero meaning "stop here, that many bytes from this
range left". In the end, the amount actually handled is stored
in size.
iov_iter_copy_from_user_atomic() and iov_iter_alignment() converted
to it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2014-11-27 13:51:41 -05:00
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
2024-01-23 15:24:46 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
2021-04-25 22:02:38 +01:00
|
|
|
|
2021-04-25 00:44:35 -04:00
|
|
|
|
|
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2016-04-08 19:05:19 -04:00
|
|
|
|
|
|
|
|
|
2016-12-21 21:55:02 -05:00
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
2016-04-08 19:05:19 -04:00
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
2016-04-08 19:05:19 -04:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
2016-09-22 16:33:12 -04:00
|
|
|
|
|
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
|
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-25 01:03:16 -04:00
|
|
|
|
|
|
|
|
|
2016-12-21 21:55:02 -05:00
|
|
|
|
2016-04-08 19:05:19 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
2022-06-17 13:35:35 -04:00
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 13:35:35 -04:00
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-23 11:07:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-09-30 11:55:00 -07:00
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:05 +01:00
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:05 +01:00
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:05 +01:00
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:05 +01:00
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 13:48:03 -04:00
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 13:48:03 -04:00
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2022-06-11 04:04:33 -04:00
|
|
|
|
|
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2020-02-10 10:00:21 +00:00
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2022-06-17 16:07:49 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-21 16:10:37 -04:00
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-21 16:10:37 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-21 16:10:37 -04:00
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
2022-06-21 15:55:19 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-10 20:38:20 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2023-06-14 16:03:41 +02:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2023-01-19 12:47:23 +00:00
|
|
|
|
2014-11-27 14:14:31 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
2022-06-11 16:44:21 -04:00
|
|
|
|
|
|
|
|
|
2014-11-27 14:14:31 -05:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
2018-10-20 00:57:56 +01:00
|
|
|
|
2021-07-12 12:06:14 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 16:07:49 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2022-06-10 20:38:20 -04:00
|
|
|
|
2014-11-27 14:14:31 -05:00
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
|
|
|
|
|
2022-06-21 16:10:37 -04:00
|
|
|
|
2022-06-17 14:45:41 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2025-03-10 14:35:24 +00:00
|
|
|
|
2025-04-01 22:47:12 +08:00
|
|
|
|
2025-03-10 14:35:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2022-06-11 04:04:33 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2021-04-25 09:14:44 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2022-06-10 20:38:20 -04:00
|
|
|
|
2023-06-14 16:03:41 +02:00
|
|
|
|
|
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-14 16:03:41 +02:00
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2022-06-17 13:54:15 -04:00
|
|
|
|
2023-06-14 16:03:41 +02:00
|
|
|
|
|
|
|
|
|
2022-06-10 20:38:20 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-14 16:03:41 +02:00
|
|
|
|
2022-06-10 20:38:20 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-10 13:05:12 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
|
|
|
|
|
2014-11-27 14:09:46 -05:00
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
|
|
|
|
|
2014-11-27 14:09:46 -05:00
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
2021-05-03 11:05:29 -04:00
|
|
|
|
|
|
|
|
|
2021-04-25 16:00:48 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2014-04-04 23:12:29 -04:00
|
|
|
|
2014-03-19 01:16:16 -04:00
|
|
|
|
2015-01-31 20:08:47 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-10-22 13:07:28 +01:00
|
|
|
|
iov_iter: use kmemdup_array for dup_iter to harden against overflow
While auditing the Linux 7.0-rc2 kernel, I identified a potential security
vulnerability in the iov_iter framework's memory allocation logic.
The dup_iter() function, which is exported via EXPORT_SYMBOL, currently
uses kmemdup() with a raw multiplication to allocate the duplicate iovec array:
new->iov = kmemdup(from->iov, nr_segs * sizeof(struct iovec), gfp);
The hazard here is that dup_iter() relies on a primitive multiplication without
any integrated overflow check. Since nr_segs is often derived from user-space
input, this line is vulnerable to integer overflow (on 32-bit systems or
via type narrowing), potentially leading to a small allocation followed by a
large out-of-bounds memory copy. Furthermore, it allows for unbounded memory
allocations, as the function lacks intrinsic knowledge of safe limits.
On the 7.0-rc2 branch, several high-impact callchains still rely on this
exported function:
drivers/usb/gadget/function/f_fs.c:
The ffs_epfile_read_iter() path demonstrates why relying on dup_iter() is
dangerous: it performs allocation based on user input before verifying driver
state. This confirms that dup_iter() must be hardened internally as it cannot
assume pre-validated input.
drivers/usb/gadget/legacy/inode.c:
The ep_read_iter() path illustrates how dup_iter()’s lack of boundary awareness
compounds resource risks. When combined with other allocations, it creates
a multiplier effect for kernel memory pressure.
This patch replaces kmemdup() with kmemdup_array(), which utilizes
check_mul_overflow() to ensure the allocation size is calculated safely,
hardening dup_iter() against malicious or malformed inputs from its callers
Signed-off-by: Wang Haoran <haoranwangsec@gmail.com>
Link: https://patch.msgid.link/20260413060655.1139141-1-haoranwangsec@gmail.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
2026-04-13 14:06:55 +08:00
|
|
|
|
|
|
|
|
|
2015-01-31 20:08:47 -05:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2015-01-31 20:08:47 -05:00
|
|
|
|
iov_iter: use kmemdup_array for dup_iter to harden against overflow
While auditing the Linux 7.0-rc2 kernel, I identified a potential security
vulnerability in the iov_iter framework's memory allocation logic.
The dup_iter() function, which is exported via EXPORT_SYMBOL, currently
uses kmemdup() with a raw multiplication to allocate the duplicate iovec array:
new->iov = kmemdup(from->iov, nr_segs * sizeof(struct iovec), gfp);
The hazard here is that dup_iter() relies on a primitive multiplication without
any integrated overflow check. Since nr_segs is often derived from user-space
input, this line is vulnerable to integer overflow (on 32-bit systems or
via type narrowing), potentially leading to a small allocation followed by a
large out-of-bounds memory copy. Furthermore, it allows for unbounded memory
allocations, as the function lacks intrinsic knowledge of safe limits.
On the 7.0-rc2 branch, several high-impact callchains still rely on this
exported function:
drivers/usb/gadget/function/f_fs.c:
The ffs_epfile_read_iter() path demonstrates why relying on dup_iter() is
dangerous: it performs allocation based on user input before verifying driver
state. This confirms that dup_iter() must be hardened internally as it cannot
assume pre-validated input.
drivers/usb/gadget/legacy/inode.c:
The ep_read_iter() path illustrates how dup_iter()’s lack of boundary awareness
compounds resource risks. When combined with other allocations, it creates
a multiplier effect for kernel memory pressure.
This patch replaces kmemdup() with kmemdup_array(), which utilizes
check_mul_overflow() to ensure the allocation size is calculated safely,
hardening dup_iter() against malicious or malformed inputs from its callers
Signed-off-by: Wang Haoran <haoranwangsec@gmail.com>
Link: https://patch.msgid.link/20260413060655.1139141-1-haoranwangsec@gmail.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <brauner@kernel.org>
2026-04-13 14:06:55 +08:00
|
|
|
|
|
|
|
|
|
2015-01-31 20:08:47 -05:00
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
2015-01-31 20:08:47 -05:00
|
|
|
|
|
|
|
|
|
saner iov_iter initialization primitives
iovec-backed iov_iter instances are assumed to satisfy several properties:
* no more than UIO_MAXIOV elements in iovec array
* total size of all ranges is no more than MAX_RW_COUNT
* all ranges pass access_ok().
The problem is, invariants of data structures should be established in the
primitives creating those data structures, not in the code using those
primitives. And iov_iter_init() violates that principle. For a while we
managed to get away with that, but once the use of iov_iter started to
spread, it didn't take long for shit to hit the fan - missed check in
sys_sendto() had introduced a roothole.
We _do_ have primitives for importing and validating iovecs (both native and
compat ones) and those primitives are almost always followed by shoving the
resulting iovec into iov_iter. Life would be considerably simpler (and safer)
if we combined those primitives with initializing iov_iter.
That gives us two new primitives - import_iovec() and compat_import_iovec().
Calling conventions:
iovec = iov_array;
err = import_iovec(direction, uvec, nr_segs,
ARRAY_SIZE(iov_array), &iovec,
&iter);
imports user vector into kernel space (into iov_array if it fits, allocated
if it doesn't fit or if iovec was NULL), validates it and sets iter up to
refer to it. On success 0 is returned and allocated kernel copy (or NULL
if the array had fit into caller-supplied one) is returned via iovec.
On failure all allocations are undone and -E... is returned. If the total
size of ranges exceeds MAX_RW_COUNT, the excess is silently truncated.
compat_import_iovec() expects uvec to be a pointer to user array of compat_iovec;
otherwise it's identical to import_iovec().
Finally, import_single_range() sets iov_iter backed by single-element iovec
covering a user-supplied range -
err = import_single_range(direction, address, size, iovec, &iter);
does validation and sets iter up. Again, size in excess of MAX_RW_COUNT gets
silently truncated.
Next commits will be switching the things up to use of those and reducing
the amount of iov_iter_init() instances.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-21 17:45:43 -04:00
|
|
|
|
2023-04-12 10:46:48 -06:00
|
|
|
|
2024-01-29 10:37:29 -08:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-01-29 10:37:29 -08:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2021-01-11 18:19:26 +01:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-16 14:43:55 +02:00
|
|
|
|
2023-03-30 14:53:51 -07:00
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
2023-03-30 14:53:51 -07:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2023-03-30 14:53:51 -07:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
2023-03-30 14:53:51 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
2026-02-21 16:37:42 -08:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2023-03-30 14:53:51 -07:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-24 14:37:19 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-01-31 14:13:15 +00:00
|
|
|
|
|
|
|
|
|
2023-03-24 14:37:19 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-24 14:37:19 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:39 +02:00
|
|
|
|
|
|
|
|
|
2016-10-08 11:18:07 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2016-10-08 11:18:07 +02:00
|
|
|
|
|
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
2016-10-08 11:18:07 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-05-14 16:02:22 -06:00
|
|
|
|
2016-10-08 11:18:07 +02:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
saner iov_iter initialization primitives
iovec-backed iov_iter instances are assumed to satisfy several properties:
* no more than UIO_MAXIOV elements in iovec array
* total size of all ranges is no more than MAX_RW_COUNT
* all ranges pass access_ok().
The problem is, invariants of data structures should be established in the
primitives creating those data structures, not in the code using those
primitives. And iov_iter_init() violates that principle. For a while we
managed to get away with that, but once the use of iov_iter started to
spread, it didn't take long for shit to hit the fan - missed check in
sys_sendto() had introduced a roothole.
We _do_ have primitives for importing and validating iovecs (both native and
compat ones) and those primitives are almost always followed by shoving the
resulting iovec into iov_iter. Life would be considerably simpler (and safer)
if we combined those primitives with initializing iov_iter.
That gives us two new primitives - import_iovec() and compat_import_iovec().
Calling conventions:
iovec = iov_array;
err = import_iovec(direction, uvec, nr_segs,
ARRAY_SIZE(iov_array), &iovec,
&iter);
imports user vector into kernel space (into iov_array if it fits, allocated
if it doesn't fit or if iovec was NULL), validates it and sets iter up to
refer to it. On success 0 is returned and allocated kernel copy (or NULL
if the array had fit into caller-supplied one) is returned via iovec.
On failure all allocations are undone and -E... is returned. If the total
size of ranges exceeds MAX_RW_COUNT, the excess is silently truncated.
compat_import_iovec() expects uvec to be a pointer to user array of compat_iovec;
otherwise it's identical to import_iovec().
Finally, import_single_range() sets iov_iter backed by single-element iovec
covering a user-supplied range -
err = import_single_range(direction, address, size, iovec, &iter);
does validation and sets iter up. Again, size in excess of MAX_RW_COUNT gets
silently truncated.
Next commits will be switching the things up to use of those and reducing
the amount of iov_iter_init() instances.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-21 17:45:43 -04:00
|
|
|
|
2020-09-25 06:51:40 +02:00
|
|
|
|
saner iov_iter initialization primitives
iovec-backed iov_iter instances are assumed to satisfy several properties:
* no more than UIO_MAXIOV elements in iovec array
* total size of all ranges is no more than MAX_RW_COUNT
* all ranges pass access_ok().
The problem is, invariants of data structures should be established in the
primitives creating those data structures, not in the code using those
primitives. And iov_iter_init() violates that principle. For a while we
managed to get away with that, but once the use of iov_iter started to
spread, it didn't take long for shit to hit the fan - missed check in
sys_sendto() had introduced a roothole.
We _do_ have primitives for importing and validating iovecs (both native and
compat ones) and those primitives are almost always followed by shoving the
resulting iovec into iov_iter. Life would be considerably simpler (and safer)
if we combined those primitives with initializing iov_iter.
That gives us two new primitives - import_iovec() and compat_import_iovec().
Calling conventions:
iovec = iov_array;
err = import_iovec(direction, uvec, nr_segs,
ARRAY_SIZE(iov_array), &iovec,
&iter);
imports user vector into kernel space (into iov_array if it fits, allocated
if it doesn't fit or if iovec was NULL), validates it and sets iter up to
refer to it. On success 0 is returned and allocated kernel copy (or NULL
if the array had fit into caller-supplied one) is returned via iovec.
On failure all allocations are undone and -E... is returned. If the total
size of ranges exceeds MAX_RW_COUNT, the excess is silently truncated.
compat_import_iovec() expects uvec to be a pointer to user array of compat_iovec;
otherwise it's identical to import_iovec().
Finally, import_single_range() sets iov_iter backed by single-element iovec
covering a user-supplied range -
err = import_single_range(direction, address, size, iovec, &iter);
does validation and sets iter up. Again, size in excess of MAX_RW_COUNT gets
silently truncated.
Next commits will be switching the things up to use of those and reducing
the amount of iov_iter_init() instances.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-21 17:45:43 -04:00
|
|
|
|
2020-09-25 06:51:41 +02:00
|
|
|
|
|
|
|
|
|
saner iov_iter initialization primitives
iovec-backed iov_iter instances are assumed to satisfy several properties:
* no more than UIO_MAXIOV elements in iovec array
* total size of all ranges is no more than MAX_RW_COUNT
* all ranges pass access_ok().
The problem is, invariants of data structures should be established in the
primitives creating those data structures, not in the code using those
primitives. And iov_iter_init() violates that principle. For a while we
managed to get away with that, but once the use of iov_iter started to
spread, it didn't take long for shit to hit the fan - missed check in
sys_sendto() had introduced a roothole.
We _do_ have primitives for importing and validating iovecs (both native and
compat ones) and those primitives are almost always followed by shoving the
resulting iovec into iov_iter. Life would be considerably simpler (and safer)
if we combined those primitives with initializing iov_iter.
That gives us two new primitives - import_iovec() and compat_import_iovec().
Calling conventions:
iovec = iov_array;
err = import_iovec(direction, uvec, nr_segs,
ARRAY_SIZE(iov_array), &iovec,
&iter);
imports user vector into kernel space (into iov_array if it fits, allocated
if it doesn't fit or if iovec was NULL), validates it and sets iter up to
refer to it. On success 0 is returned and allocated kernel copy (or NULL
if the array had fit into caller-supplied one) is returned via iovec.
On failure all allocations are undone and -E... is returned. If the total
size of ranges exceeds MAX_RW_COUNT, the excess is silently truncated.
compat_import_iovec() expects uvec to be a pointer to user array of compat_iovec;
otherwise it's identical to import_iovec().
Finally, import_single_range() sets iov_iter backed by single-element iovec
covering a user-supplied range -
err = import_single_range(direction, address, size, iovec, &iter);
does validation and sets iter up. Again, size in excess of MAX_RW_COUNT gets
silently truncated.
Next commits will be switching the things up to use of those and reducing
the amount of iov_iter_init() instances.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2015-03-21 17:45:43 -04:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-01-05 11:07:30 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-15 21:01:12 +02:00
|
|
|
|
2023-01-05 11:07:30 -08:00
|
|
|
|
2021-09-10 11:18:36 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-01-05 11:07:33 -08:00
|
|
|
|
|
|
|
|
|
2021-09-10 11:18:36 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-22 14:59:25 -04:00
|
|
|
|
|
|
|
|
|
2021-09-10 11:18:36 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-29 08:52:15 -06:00
|
|
|
|
2021-09-10 11:18:36 -06:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:06 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
2026-06-25 15:06:23 +01:00
|
|
|
|
2025-04-02 22:06:06 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:06 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:06 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2025-04-02 22:06:06 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-06-25 15:06:23 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2024-10-31 19:02:24 +08:00
|
|
|
|
2024-11-02 09:42:11 +08:00
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
2026-06-25 15:06:24 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
2024-10-24 07:00:15 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-09-08 17:03:20 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-06-25 15:06:25 +01:00
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-06-25 15:06:25 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
2026-06-25 15:06:25 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios
Define a data structure, struct folio_queue, to represent a sequence of
folios and a kernel-internal I/O iterator type, ITER_FOLIOQ, to allow a
list of folio_queue structures to be used to provide a buffer to
iov_iter-taking functions, such as sendmsg and recvmsg.
The folio_queue structure looks like:
struct folio_queue {
struct folio_batch vec;
u8 orders[PAGEVEC_SIZE];
struct folio_queue *next;
struct folio_queue *prev;
unsigned long marks;
unsigned long marks2;
};
It does not use a list_head so that next and/or prev can be set to NULL at
the ends of the list, allowing iov_iter-handling routines to determine that
they *are* the ends without needing to store a head pointer in the iov_iter
struct.
A folio_batch struct is used to hold the folio pointers which allows the
batch to be passed to batch handling functions. Two mark bits are
available per slot. The intention is to use at least one of them to mark
folios that need putting, but that might not be ultimately necessary.
Accessor functions are used to access the slots to do the masking and an
additional accessor function is used to indicate the size of the array.
The order of each folio is also stored in the structure to avoid the need
for iov_iter_advance() and iov_iter_revert() to have to query each folio to
find its size.
With careful barriering, this can be used as an extending buffer with new
folios inserted and new folio_queue structs added without the need for a
lock. Further, provided we always keep at least one struct in the buffer,
we can also remove consumed folios and consumed structs from the head end
as we without the need for locks.
[Questions/thoughts]
(1) To manage this, I need a head pointer, a tail pointer, a tail slot
number (assuming insertion happens at the tail end and the next
pointers point from head to tail). Should I put these into a struct
of their own, say "folio_queue_head" or "rolling_buffer"?
I will end up with two of these in netfs_io_request eventually, one
keeping track of the pagecache I'm dealing with for buffered I/O and
the other to hold a bounce buffer when we need one.
(2) Should I make the slots {folio,off,len} or bio_vec?
(3) This is intended to replace ITER_XARRAY eventually. Using an xarray
in I/O iteration requires the taking of the RCU read lock, doing
copying under the RCU read lock, walking the xarray (which may change
under us), handling retries and dealing with special values.
The advantage of ITER_XARRAY is that when we're dealing with the
pagecache directly, we don't need any allocation - but if we're doing
encrypted comms, there's a good chance we'd be using a bounce buffer
anyway.
This will require afs, erofs, cifs, orangefs and fscache to be
converted to not use this. afs still uses it for dirs and symlinks;
some of erofs usages should be easy to change, but there's one which
won't be so easy; ceph's use via fscache can be fixed by porting ceph
to netfslib; cifs is using xarray as a bounce buffer - that can be
moved to use sheaves instead; and orangefs has a similar problem to
erofs - maybe orangefs could use netfslib?
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jeff Layton <jlayton@kernel.org>
cc: Steve French <sfrench@samba.org>
cc: Ilya Dryomov <idryomov@gmail.com>
cc: Gao Xiang <xiang@kernel.org>
cc: Mike Marshall <hubcap@omnibond.com>
cc: netfs@lists.linux.dev
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
cc: linux-afs@lists.infradead.org
cc: linux-cifs@vger.kernel.org
cc: ceph-devel@vger.kernel.org
cc: linux-erofs@lists.ozlabs.org
cc: devel@lists.orangefs.org
Link: https://lore.kernel.org/r/20240814203850.2240469-13-dhowells@redhat.com/ # v2
Signed-off-by: Christian Brauner <brauner@kernel.org>
2024-06-19 00:20:42 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 21:50:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2026-01-26 06:53:35 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
