linux-stable.git/drivers/xen/gntdev.c, branch v6.0

xen/gntdev: Ignore failure to unmap INVALID_GRANT_HANDLE

2022-07-11T09:39:45+00:00

The error paths of gntdev_mmap() can call unmap_grant_pages() even
though not all of the pages have been successfully mapped.  This will
trigger the WARN_ON()s in __unmap_grant_pages_done().  The number of
warnings can be very large; I have observed thousands of lines of
warnings in the systemd journal.

Avoid this problem by only warning on unmapping failure if the handle
being unmapped is not INVALID_GRANT_HANDLE.  The handle field of any
page that was not successfully mapped will be INVALID_GRANT_HANDLE, so
this catches all cases where unmapping can legitimately fail.

Fixes: dbe97cff7dd9 ("xen/gntdev: Avoid blocking in unmap_grant_pages()")
Cc: stable@vger.kernel.org
Suggested-by: Juergen Gross 
Signed-off-by: Demi Marie Obenour 
Reviewed-by: Oleksandr Tyshchenko 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/20220710230522.1563-1-demi@invisiblethingslab.com
Signed-off-by: Juergen Gross

xen/gntdev: Avoid blocking in unmap_grant_pages()

2022-06-23T13:29:18+00:00

unmap_grant_pages() currently waits for the pages to no longer be used.
In https://github.com/QubesOS/qubes-issues/issues/7481, this lead to a
deadlock against i915: i915 was waiting for gntdev's MMU notifier to
finish, while gntdev was waiting for i915 to free its pages.  I also
believe this is responsible for various deadlocks I have experienced in
the past.

Avoid these problems by making unmap_grant_pages async.  This requires
making it return void, as any errors will not be available when the
function returns.  Fortunately, the only use of the return value is a
WARN_ON(), which can be replaced by a WARN_ON when the error is
detected.  Additionally, a failed call will not prevent further calls
from being made, but this is harmless.

Because unmap_grant_pages is now async, the grant handle will be sent to
INVALID_GRANT_HANDLE too late to prevent multiple unmaps of the same
handle.  Instead, a separate bool array is allocated for this purpose.
This wastes memory, but stuffing this information in padding bytes is
too fragile.  Furthermore, it is necessary to grab a reference to the
map before making the asynchronous call, and release the reference when
the call returns.

It is also necessary to guard against reentrancy in gntdev_map_put(),
and to handle the case where userspace tries to map a mapping whose
contents have not all been freed yet.

Fixes: 745282256c75 ("xen/gntdev: safely unmap grants in case they are still in use")
Cc: stable@vger.kernel.org
Signed-off-by: Demi Marie Obenour 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/20220622022726.2538-1-demi@invisiblethingslab.com
Signed-off-by: Juergen Gross

xen/gntdev: fix unmap notification order

2022-01-06T07:52:22+00:00

While working with Xen's libxenvchan library I have faced an issue with
unmap notifications sent in wrong order if both UNMAP_NOTIFY_SEND_EVENT
and UNMAP_NOTIFY_CLEAR_BYTE were requested: first we send an event channel
notification and then clear the notification byte which renders in the below
inconsistency (cli_live is the byte which was requested to be cleared on unmap):

[  444.514243] gntdev_put_map UNMAP_NOTIFY_SEND_EVENT map->notify.event 6
libxenvchan_is_open cli_live 1
[  444.515239] __unmap_grant_pages UNMAP_NOTIFY_CLEAR_BYTE at 14

Thus it is not possible to reliably implement the checks like
- wait for the notification (UNMAP_NOTIFY_SEND_EVENT)
- check the variable (UNMAP_NOTIFY_CLEAR_BYTE)
because it is possible that the variable gets checked before it is cleared
by the kernel.

To fix that we need to re-order the notifications, so the variable is first
gets cleared and then the event channel notification is sent.
With this fix I can see the correct order of execution:

[   54.522611] __unmap_grant_pages UNMAP_NOTIFY_CLEAR_BYTE at 14
[   54.537966] gntdev_put_map UNMAP_NOTIFY_SEND_EVENT map->notify.event 6
libxenvchan_is_open cli_live 0

Cc: stable@vger.kernel.org
Signed-off-by: Oleksandr Andrushchenko 
Reviewed-by: Boris Ostrovsky 
Link: https://lore.kernel.org/r/20211210092817.580718-1-andr2000@gmail.com
Signed-off-by: Juergen Gross

Xen/gntdev: don't ignore kernel unmapping error

2021-09-20T14:57:08+00:00

While working on XSA-361 and its follow-ups, I failed to spot another
place where the kernel mapping part of an operation was not treated the
same as the user space part. Detect and propagate errors and add a 2nd
pr_debug().

Signed-off-by: Jan Beulich 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/c2513395-74dc-aea3-9192-fd265aa44e35@suse.com
Signed-off-by: Juergen Gross

xen: assume XENFEAT_gnttab_map_avail_bits being set for pv guests

2021-08-30T09:57:50+00:00

XENFEAT_gnttab_map_avail_bits is always set in Xen 4.0 and newer.
Remove coding assuming it might be zero.

Signed-off-by: Juergen Gross 
Acked-by: Peter Zijlstra (Intel) 
Reviewed-by: Boris Ostrovsky 
Link: https://lore.kernel.org/r/20210730071804.4302-4-jgross@suse.com
Signed-off-by: Juergen Gross

xen/gntdev: fix gntdev_mmap() error exit path

2021-05-10T07:32:00+00:00

Commit d3eeb1d77c5d0af ("xen/gntdev: use mmu_interval_notifier_insert")
introduced an error in gntdev_mmap(): in case the call of
mmu_interval_notifier_insert_locked() fails the exit path should not
call mmu_interval_notifier_remove(), as this might result in NULL
dereferences.

One reason for failure is e.g. a signal pending for the running
process.

Fixes: d3eeb1d77c5d0af ("xen/gntdev: use mmu_interval_notifier_insert")
Cc: stable@vger.kernel.org
Reported-by: Marek Marczykowski-Górecki 
Tested-by: Marek Marczykowski-Górecki 
Signed-off-by: Juergen Gross 
Reviewed-by: Luca Fancellu 
Link: https://lore.kernel.org/r/20210423054038.26696-1-jgross@suse.com
Signed-off-by: Juergen Gross

Xen/gntdev: don't needlessly use kvcalloc()

2021-03-10T22:39:29+00:00

Requesting zeroed memory when all of it will be overwritten subsequently
by all ones is a waste of processing bandwidth. In fact, rather than
recording zeroed ->grants[], fill that array too with more appropriate
"invalid" indicators.

Signed-off-by: Jan Beulich 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/9a726be2-4893-8ffe-0ef1-b70dd1c229b1@suse.com
Signed-off-by: Boris Ostrovsky

Xen/gnttab: introduce common INVALID_GRANT_{HANDLE,REF}

2021-03-10T22:39:29+00:00

It's not helpful if every driver has to cook its own. Generalize
xenbus'es INVALID_GRANT_HANDLE and pcifront's INVALID_GRANT_REF (which
shouldn't have expanded to zero to begin with). Use the constants in
p2m.c and gntdev.c right away, and update field types where necessary so
they would match with the constants' types (albeit without touching
struct ioctl_gntdev_grant_ref's ref field, as that's part of the public
interface of the kernel and would require introducing a dependency on
Xen's grant_table.h public header).

Signed-off-by: Jan Beulich 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/db7c38a5-0d75-d5d1-19de-e5fe9f0b9c48@suse.com
Signed-off-by: Boris Ostrovsky

Xen/gntdev: don't needlessly allocate k{,un}map_ops[]

2021-03-10T22:39:29+00:00

They're needed only in the not-auto-translate (i.e. PV) case; there's no
point in allocating memory that's never going to get accessed.

Signed-off-by: Jan Beulich 
Reviewed-by: Juergen Gross 
Link: https://lore.kernel.org/r/180d50cb-5531-8952-4bf0-d65c554638ed@suse.com
Signed-off-by: Boris Ostrovsky

Xen/gntdev: correct error checking in gntdev_map_grant_pages()

2021-02-15T07:52:27+00:00

Failure of the kernel part of the mapping operation should also be
indicated as an error to the caller, or else it may assume the
respective kernel VA is okay to access.

Furthermore gnttab_map_refs() failing still requires recording
successfully mapped handles, so they can be unmapped subsequently. This
in turn requires there to be a way to tell full hypercall failure from
partial success - preset map_op status fields such that they won't
"happen" to look as if the operation succeeded.

Also again use GNTST_okay instead of implying its value (zero).

This is part of XSA-361.

Signed-off-by: Jan Beulich 
Cc: stable@vger.kernel.org
Reviewed-by: Juergen Gross 
Signed-off-by: Juergen Gross