19 Dec, 2014
1 commit
-
Pull virtio fixes from Michael S Tsirkin:
"virtio 1.0 related fixesMost importantly, this fixes using virtio_pci as a module.
Further, the big virtio 1.0 conversion missed a couple of places.
This fixes them up.This isn't 100% sparse-clean yet because on many architectures
get_user triggers sparse warnings when used with __bitwise tag (when
same tag is on both pointer and value read).I posted a patchset to fix it up by adding __force on all arches that
don't already have it (many do), when that's merged these warnings
will go away"* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
virtio_pci: restore module attributes
mic/host: fix up virtio 1.0 APIs
vringh: update for virtio 1.0 APIs
vringh: 64 bit features
tools/virtio: add virtio 1.0 in vringh_test
tools/virtio: add virtio 1.0 in virtio_test
tools/virtio: enable -Werror
tools/virtio: 64 bit features
tools/virtio: fix vringh test
tools/virtio: more stubs
virtio: core support for config generation
virtio_pci: add VIRTIO_PCI_NO_LEGACY
virtio_pci: move probe to common file
virtio_pci_common.h: drop VIRTIO_PCI_NO_LEGACY
virtio_config: fix virtio_cread_bytes
virtio: set VIRTIO_CONFIG_S_FEATURES_OK on restore
16 Dec, 2014
2 commits
-
When switching everything over to virtio 1.0 memory access APIs,
I missed converting vringh.
Fortunately, it's straight-forward.Signed-off-by: Michael S. Tsirkin
-
Pass u64 everywhere.
Signed-off-by: Michael S. Tsirkin
12 Dec, 2014
1 commit
-
Pull networking updates from David Miller:
1) New offloading infrastructure and example 'rocker' driver for
offloading of switching and routing to hardware.This work was done by a large group of dedicated individuals, not
limited to: Scott Feldman, Jiri Pirko, Thomas Graf, John Fastabend,
Jamal Hadi Salim, Andy Gospodarek, Florian Fainelli, Roopa Prabhu2) Start making the networking operate on IOV iterators instead of
modifying iov objects in-situ during transfers. Thanks to Al Viro
and Herbert Xu.3) A set of new netlink interfaces for the TIPC stack, from Richard
Alpe.4) Remove unnecessary looping during ipv6 routing lookups, from Martin
KaFai Lau.5) Add PAUSE frame generation support to gianfar driver, from Matei
Pavaluca.6) Allow for larger reordering levels in TCP, which are easily
achievable in the real world right now, from Eric Dumazet.7) Add a variable of napi_schedule that doesn't need to disable cpu
interrupts, from Eric Dumazet.8) Use a doubly linked list to optimize neigh_parms_release(), from
Nicolas Dichtel.9) Various enhancements to the kernel BPF verifier, and allow eBPF
programs to actually be attached to sockets. From Alexei
Starovoitov.10) Support TSO/LSO in sunvnet driver, from David L Stevens.
11) Allow controlling ECN usage via routing metrics, from Florian
Westphal.12) Remote checksum offload, from Tom Herbert.
13) Add split-header receive, BQL, and xmit_more support to amd-xgbe
driver, from Thomas Lendacky.14) Add MPLS support to openvswitch, from Simon Horman.
15) Support wildcard tunnel endpoints in ipv6 tunnels, from Steffen
Klassert.16) Do gro flushes on a per-device basis using a timer, from Eric
Dumazet. This tries to resolve the conflicting goals between the
desired handling of bulk vs. RPC-like traffic.17) Allow userspace to ask for the CPU upon what a packet was
received/steered, via SO_INCOMING_CPU. From Eric Dumazet.18) Limit GSO packets to half the current congestion window, from Eric
Dumazet.19) Add a generic helper so that all drivers set their RSS keys in a
consistent way, from Eric Dumazet.20) Add xmit_more support to enic driver, from Govindarajulu
Varadarajan.21) Add VLAN packet scheduler action, from Jiri Pirko.
22) Support configurable RSS hash functions via ethtool, from Eyal
Perry.* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1820 commits)
Fix race condition between vxlan_sock_add and vxlan_sock_release
net/macb: fix compilation warning for print_hex_dump() called with skb->mac_header
net/mlx4: Add support for A0 steering
net/mlx4: Refactor QUERY_PORT
net/mlx4_core: Add explicit error message when rule doesn't meet configuration
net/mlx4: Add A0 hybrid steering
net/mlx4: Add mlx4_bitmap zone allocator
net/mlx4: Add a check if there are too many reserved QPs
net/mlx4: Change QP allocation scheme
net/mlx4_core: Use tasklet for user-space CQ completion events
net/mlx4_core: Mask out host side virtualization features for guests
net/mlx4_en: Set csum level for encapsulated packets
be2net: Export tunnel offloads only when a VxLAN tunnel is created
gianfar: Fix dma check map error when DMA_API_DEBUG is enabled
cxgb4/csiostor: Don't use MASTER_MUST for fw_hello call
net: fec: only enable mdio interrupt before phy device link up
net: fec: clear all interrupt events to support i.MX6SX
net: fec: reset fep link status in suspend function
net: sock: fix access via invalid file descriptor
net: introduce helper macro for_each_cmsghdr
...
10 Dec, 2014
1 commit
-
Note that the code _using_ ->msg_iter at that point will be very
unhappy with anything other than unshifted iovec-backed iov_iter.
We still need to convert users to proper primitives.Signed-off-by: Al Viro
09 Dec, 2014
10 commits
-
Signed-off-by: Jason Wang
Acked-by: Michael S. Tsirkin
Signed-off-by: Michael S. Tsirkin -
Include all endian conversions as required by virtio 1.0.
Don't set virtio 1.0 yet, since that requires ANY_LAYOUT
which we don't yet support.Signed-off-by: Michael S. Tsirkin
Acked-by: Paolo Bonzini -
Signed-off-by: Michael S. Tsirkin
-
Signed-off-by: Michael S. Tsirkin
Reviewed-by: Jason Wang
Reviewed-by: Cornelia Huck -
I had to add an explicit tag to suppress compiler warning:
gcc isn't smart enough to notice that
len is always initialized since function is called with size > 0.Signed-off-by: Michael S. Tsirkin
Reviewed-by: Cornelia Huck -
Signed-off-by: Michael S. Tsirkin
-
Most places in vhost can use __get/__put_user rather than
get/put_user since addresses are pre-validated.
This should be good for performance, but this also
will help make code sparse-clean: get/put_user macros
don't play well with __virtioXX bitwise tags.
Switch to get/put_user to __ variants everywhere in vhost.
There's one exception - for consistency switch that
as well, and add an explicit access_ok check.Signed-off-by: Michael S. Tsirkin
-
vhost/net keeps a copy of the used ring in host memory but (ab)uses
the length field for internal house-keeping. This works because the
length in the used ring for tx is always 0. In order to suppress sparse
warnings, we force native endianness here.
Note that these values are never exposed to guests.Signed-off-by: Michael S. Tsirkin
Reviewed-by: Jason Wang -
Add guest memory access wrappers to handle virtio endianness
conversions.Signed-off-by: Michael S. Tsirkin
Reviewed-by: Jason Wang
Reviewed-by: Cornelia Huck -
We need to use bit 32 for virtio 1.0.
Make vhost_has_feature bool to avoid discarding high bits.Cc: Sergei Shtylyov
Cc: Ben Hutchings
Signed-off-by: Michael S. Tsirkin
Reviewed-by: Jason Wang
29 Oct, 2014
1 commit
-
This patch addresses a bug where individual vhost-scsi configfs endpoint
groups can be removed from below while active exports to QEMU userspace
still exist, resulting in an OOPs.It adds a configfs_depend_item() in vhost_scsi_set_endpoint() to obtain
an explicit dependency on se_tpg->tpg_group in order to prevent individual
vhost-scsi WWPN endpoints from being released via normal configfs methods
while an QEMU ioctl reference still exists.Also, add matching configfs_undepend_item() in vhost_scsi_clear_endpoint()
to release the dependency, once QEMU's reference to the individual group
at /sys/kernel/config/target/vhost/$WWPN/$TPGT is released.(Fix up vhost_scsi_clear_endpoint() error path - DanC)
Cc: Michael S. Tsirkin
Cc: Paolo Bonzini
Cc: Stefan Hajnoczi
Cc: # 3.6+
Signed-off-by: Nicholas Bellinger
23 Jun, 2014
2 commits
-
Now that we have kvfree, use it in vhost-scsi instead of
the open-coded version.Cc: Nicholas Bellinger
Signed-off-by: Michael S. Tsirkin -
Commit 23cc5a991c ("vhost-net: extend device allocation to vmalloc")
added another open-coded version of kvfree (which is available since
v3.15-rc5), nuke it.Signed-off-by: Romain Francoise
Signed-off-by: Michael S. Tsirkin
13 Jun, 2014
1 commit
-
Pull SCSI target updates from Nicholas Bellinger:
"The highlights this round include:- Add support for T10 PI pass-through between vhost-scsi +
virtio-scsi (MST + Paolo + MKP + nab)
- Add support for T10 PI in qla2xxx target mode (Quinn + MKP + hch +
nab, merged through scsi.git)
- Add support for percpu-ida pre-allocation in qla2xxx target code
(Quinn + nab)
- A number of iser-target fixes related to hardening the network
portal shutdown path (Sagi + Slava)
- Fix response length residual handling for a number of control CDBs
(Roland + Christophe V.)
- Various iscsi RFC conformance fixes in the CHAP authentication path
(Tejas and Calsoft folks + nab)
- Return TASK_SET_FULL status for tcm_fc(FCoE) DataIn + Response
failures (Vasu + Jun + nab)
- Fix long-standing ABORT_TASK + session reset hang (nab)
- Convert iser-initiator + iser-target to include T10 bytes into EDTL
(Sagi + Or + MKP + Mike Christie)
- Fix NULL pointer dereference regression related to XCOPY introduced
in v3.15 + CC'ed to v3.12.y (nab)"* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (34 commits)
target: Fix NULL pointer dereference for XCOPY in target_put_sess_cmd
vhost-scsi: Include prot_bytes into expected data transfer length
TARGET/sbc,loopback: Adjust command data length in case pi exists on the wire
libiscsi, iser: Adjust data_length to include protection information
scsi_cmnd: Introduce scsi_transfer_length helper
target: Report correct response length for some commands
target/sbc: Check that the LBA and number of blocks are correct in VERIFY
target/sbc: Remove sbc_check_valid_sectors()
Target/iscsi: Fix sendtargets response pdu for iser transport
Target/iser: Fix a wrong dereference in case discovery session is over iser
iscsi-target: Fix ABORT_TASK + connection reset iscsi_queue_req memory leak
target: Use complete_all for se_cmd->t_transport_stop_comp
target: Set CMD_T_ACTIVE bit for Task Management Requests
target: cleanup some boolean tests
target/spc: Simplify INQUIRY EVPD=0x80
tcm_fc: Generate TASK_SET_FULL status for response failures
tcm_fc: Generate TASK_SET_FULL status for DataIN failures
iscsi-target: Reject mutual authentication with reflected CHAP_C
iscsi-target: Remove no-op from iscsit_tpg_del_portal_group
iscsi-target: Fix CHAP_A parameter list handling
...
12 Jun, 2014
2 commits
-
Pull vhost infrastructure updates from Michael S. Tsirkin:
"This reworks vhost core dropping unnecessary RCU uses in favor of VQ
mutexes which are used on fast path anyway. This fixes worst-case
latency for users which change the memory mappings a lot. Memory
allocation for vhost-net now supports fallback on vmalloc (same as for
vhost-scsi) this makes it possible to create the device on systems
where memory is very fragmented, with slightly lower performance"* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
vhost: move memory pointer to VQs
vhost: move acked_features to VQs
vhost: replace rcu with mutex
vhost-net: extend device allocation to vmalloc -
This patch updates vhost_scsi_get_tag() to accept the combined
expected data transfer length + T10 PI bytes as the value passed
into target_submit_cmd().This is required now that target-core logic in commit 14ef9200
expects to subtract se_cmd->prot_length from se_cmd->data_length.Cc: Paolo Bonzini
Cc: Michael S. Tsirkin
Cc: Martin K. Petersen
Cc: Sagi Grimberg
Signed-off-by: Nicholas Bellinger
09 Jun, 2014
4 commits
-
commit 2ae76693b8bcabf370b981cd00c36cd41d33fabc
vhost: replace rcu with mutex
replaced rcu sync for memory accesses with VQ mutex locl/unlock.
This is correct since all accesses are under VQ mutex, but incomplete:
we still do useless rcu lock/unlock operations, someone might copy this
code into some other context where this won't be right.
This use of RCU is also non standard and hard to understand.
Let's copy the pointer to each VQ structure, this way
the access rules become straight-forward, and there's
no need for RCU anymore.Reported-by: Eric Dumazet
Signed-off-by: Michael S. Tsirkin -
Refactor code to make sure features are only accessed
under VQ mutex. This makes everything simpler, no need
for RCU here anymore.Signed-off-by: Michael S. Tsirkin
-
All memory accesses are done under some VQ mutex.
So lock/unlock all VQs is a faster equivalent of synchronize_rcu()
for memory access changes.
Some guests cause a lot of these changes, so it's helpful
to make them faster.Reported-by: "Gonglei (Arei)"
Signed-off-by: Michael S. Tsirkin -
Michael Mueller provided a patch to reduce the size of
vhost-net structure as some allocations could fail under
memory pressure/fragmentation. We are still left with
high order allocations though.This patch is handling the problem at the core level, allowing
vhost structures to use vmalloc() if kmalloc() failed.As vmalloc() adds overhead on a critical network path, add __GFP_REPEAT
to kzalloc() flags to do this fallback only when really needed.People are still looking at cleaner ways to handle the problem
at the API level, probably passing in multiple iovecs.
This hack seems consistent with approaches
taken since then by drivers/vhost/scsi.c and net/core/dev.cBased on patch by Romain Francoise.
Cc: Michael Mueller
Signed-off-by: Romain Francoise
Acked-by: Michael S. Tsirkin
03 Jun, 2014
4 commits
-
This patch updates vhost_scsi_handle_vq() to check for the existance
of virtio_scsi_cmd_req_pi comparing vq->iov[0].iov_len in order to
calculate seperate data + protection SGLs from data_num.Also update tcm_vhost_submission_work() to pass the pre-allocated
cmd->tvc_prot_sgl[] memory into target_submit_cmd_map_sgls(), and
update vhost_scsi_get_tag() parameters to accept scsi_tag, lun, and
task_attr.Cc: Michael S. Tsirkin
Cc: Paolo Bonzini
Cc: Martin K. Petersen
Cc: Christoph Hellwig
Cc: Hannes Reinecke
Cc: Sagi Grimberg
Cc: H. Peter Anvin
Acked-by: Michael S. Tsirkin
Signed-off-by: Nicholas Bellinger -
This patch adds vhost_scsi_map_iov_to_prot() to perform the mapping of
T10 data integrity memory between virtio iov + struct scatterlist using
get_user_pages_fast() following existing code.As with vhost_scsi_map_iov_to_sgl(), this does sanity checks against the
total prot_sgl_count vs. pre-allocated SGLs, and loops across protection
iovs using vhost_scsi_map_to_sgl() to perform the actual memory mapping.Also update tcm_vhost_release_cmd() to release associated tvc_prot_sgl[]
struct page.Cc: Michael S. Tsirkin
Cc: Paolo Bonzini
Cc: Martin K. Petersen
Cc: Christoph Hellwig
Cc: Hannes Reinecke
Cc: Sagi Grimberg
Cc: H. Peter Anvin
Acked-by: Michael S. Tsirkin
Signed-off-by: Nicholas Bellinger -
This patch updates tcm_vhost_make_nexus() to pre-allocate per descriptor
tcm_vhost_cmd->tvc_prot_sgl[] used to expose protection SGLs from within
virtio-scsi guest memory to vhost-scsi.Cc: Michael S. Tsirkin
Cc: Paolo Bonzini
Cc: Martin K. Petersen
Cc: Christoph Hellwig
Cc: Hannes Reinecke
Acked-by: Michael S. Tsirkin
Signed-off-by: Nicholas Bellinger -
Move the overflow check for sgl_count > TCM_VHOST_PREALLOC_SGLS into
vhost_scsi_map_iov_to_sgl() so that it's based on the total number
of SGLs for all IOVs, instead of single IOVs.Also, rename TCM_VHOST_PREALLOC_PAGES -> TCM_VHOST_PREALLOC_UPAGES
to better describe pointers to user-space pages.Cc: Michael S. Tsirkin
Cc: Paolo Bonzini
Cc: Martin K. Petersen
Cc: Christoph Hellwig
Cc: Hannes Reinecke
Acked-by: Michael S. Tsirkin
Signed-off-by: Nicholas Bellinger
18 Apr, 2014
1 commit
-
Mostly scripted conversion of the smp_mb__* barriers.
Signed-off-by: Peter Zijlstra
Acked-by: Paul E. McKenney
Link: http://lkml.kernel.org/n/tip-55dhyhocezdw1dg7u19hmh1u@git.kernel.org
Cc: Linus Torvalds
Cc: linux-arch@vger.kernel.org
Signed-off-by: Ingo Molnar
13 Apr, 2014
1 commit
-
Pull SCSI target updates from Nicholas Bellinger:
"Here are the target pending updates for v3.15-rc1. Apologies in
advance for waiting until the second to last day of the merge window
to send these out.The highlights this round include:
- iser-target support for T10 PI (DIF) offloads (Sagi + Or)
- Fix Task Aborted Status (TAS) handling in target-core (Alex Leung)
- Pass in transport supported PI at session initialization (Sagi + MKP + nab)
- Add WRITE_INSERT + READ_STRIP T10 PI support in target-core (nab + Sagi)
- Fix iscsi-target ERL=2 ASYNC_EVENT connection pointer bug (nab)
- Fix tcm_fc use-after-free of ft_tpg (Andy Grover)
- Use correct ib_sg_dma primitives in ib_isert (Mike Marciniszyn)Also, note the virtio-scsi + vhost-scsi changes to expose T10 PI
metadata into KVM guest have been left-out for now, as there where a
few comments from MST + Paolo that where not able to be addressed in
time for v3.15. Please expect this feature for v3.16-rc1"* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (43 commits)
ib_srpt: Use correct ib_sg_dma primitives
target/tcm_fc: Rename ft_tport_create to ft_tport_get
target/tcm_fc: Rename ft_{add,del}_lport to {add,del}_wwn
target/tcm_fc: Rename structs and list members for clarity
target/tcm_fc: Limit to 1 TPG per wwn
target/tcm_fc: Don't export ft_lport_list
target/tcm_fc: Fix use-after-free of ft_tpg
target: Add check to prevent Abort Task from aborting itself
target: Enable READ_STRIP emulation in target_complete_ok_work
target/sbc: Add sbc_dif_read_strip software emulation
target: Enable WRITE_INSERT emulation in target_execute_cmd
target/sbc: Add sbc_dif_generate software emulation
target/sbc: Only expose PI read_cap16 bits when supported by fabric
target/spc: Only expose PI mode page bits when supported by fabric
target/spc: Only expose PI inquiry bits when supported by fabric
target: Pass in transport supported PI at session initialization
target/iblock: Fix double bioset_integrity_free bug
Target/sbc: Initialize COMPARE_AND_WRITE write_sg scatterlist
target/rd: T10-Dif: RAM disk is allocating more space than required.
iscsi-target: Fix ERL=2 ASYNC_EVENT connection pointer bug
...
07 Apr, 2014
2 commits
-
In order to support local WRITE_INSERT + READ_STRIP operations for
non PI enabled fabrics, the fabric driver needs to be able signal
what protection offload operations are supported.This is done at session initialization time so the modes can be
signaled by individual se_wwn + se_portal_group endpoints, as well
as optionally across different transports on the same endpoint.For iser-target, set TARGET_PROT_ALL if the underlying ib_device
has already signaled PI offload support, and allow this to be
exposed via a new iscsit_transport->iscsit_get_sup_prot_ops()
callback.For loopback, set TARGET_PROT_ALL to signal SCSI initiator mode
operation.For all other drivers, set TARGET_PROT_NORMAL to disable fabric
level PI.Cc: Martin K. Petersen
Cc: Sagi Grimberg
Cc: Or Gerlitz
Cc: Quinn Tran
Cc: Giridhar Malavali
Signed-off-by: Nicholas Bellinger -
Now that TASK_ABORTED status is not generated for all cases by
TMR ABORT_TASK + LUN_RESET, a new TFO->abort_task() caller is
necessary in order to give fabric drivers a chance to unmap
hardware / software resources before the se_cmd descriptor is
released via the normal TFO->release_cmd() codepath.This patch adds TFO->aborted_task() in core_tmr_abort_task()
in place of the original transport_send_task_abort(), and
also updates all fabric drivers to implement this caller.The fabric drivers that include changes to perform cleanup
via ->aborted_task() are:- iscsi-target
- iser-target
- srpt
- tcm_qla2xxxThe fabric drivers that currently set ->aborted_task() to
NOPs are:- loopback
- tcm_fc
- usb-gadget
- sbp-target
- vhost-scsiFor the latter five, there appears to be no additional cleanup
required before invoking TFO->release_cmd() to release the
se_cmd descriptor.v2 changes:
- Move ->aborted_task() call into transport_cmd_finish_abort (Alex)Cc: Alex Leung
Cc: Mark Rustad
Cc: Roland Dreier
Cc: Vu Pham
Cc: Chris Boot
Cc: Sebastian Andrzej Siewior
Cc: Michael S. Tsirkin
Cc: Giridhar Malavali
Cc: Saurav Kashyap
Cc: Quinn Tran
Cc: Sagi Grimberg
Cc: Or Gerlitz
Signed-off-by: Nicholas Bellinger
02 Apr, 2014
1 commit
-
Signed-off-by: Al Viro
29 Mar, 2014
2 commits
-
vhost fails to validate negative error code
from vhost_get_vq_desc causing
a crash: we are using -EFAULT which is 0xfffffff2
as vector size, which exceeds the allocated size.The code in question was introduced in commit
8dd014adfea6f173c1ef6378f7e5e7924866c923
vhost-net: mergeable buffers supportCVE-2014-0055
Signed-off-by: Michael S. Tsirkin
Signed-off-by: David S. Miller -
When mergeable buffers are disabled, and the
incoming packet is too large for the rx buffer,
get_rx_bufs returns success.This was intentional in order for make recvmsg
truncate the packet and then handle_rx would
detect err != sock_len and drop it.Unfortunately we pass the original sock_len to
recvmsg - which means we use parts of iov not fully
validated.Fix this up by detecting this overrun and doing packet drop
immediately.CVE-2014-0077
Signed-off-by: Michael S. Tsirkin
Signed-off-by: David S. Miller
02 Mar, 2014
1 commit
-
Pull SCSI target fixes from Nicholas Bellinger:
"The bulk of the series are bugfixes for qla2xxx target NPIV support
that went in for v3.14-rc1. Also included are a few DIF related
fixes, a qla2xxx fix (Cc'ed to stable) from Greg W., and vhost/scsi
protocol version related fix from Venkatesh.Also just a heads up that a series to address a number of issues with
iser-target active I/O reset/shutdown is still being tested, and will
be included in a separate -rc6 PULL request"* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
vhost/scsi: Check LUN structure byte 0 is set to 1, per spec
qla2xxx: Fix kernel panic on selective retransmission request
Target/sbc: Don't use sg as iterator in sbc_verify_read
target: Add DIF sense codes in transport_generic_request_failure
target/sbc: Fix sbc_dif_copy_prot addr offset bug
tcm_qla2xxx: Fix NAA formatted name for NPIV WWPNs
tcm_qla2xxx: Perform configfs depend/undepend for base_tpg
tcm_qla2xxx: Add NPIV specific enable/disable attribute logic
qla2xxx: Check + fail when npiv_vports_inuse exists in shutdown
qla2xxx: Fix qlt_lport_register base_vha callback race
25 Feb, 2014
1 commit
-
The virtio spec requires byte 0 of the virtio-scsi LUN structure
to be '1'.Signed-off-by: Venkatesh Srinivas
Reviewed-by: Paolo Bonzini
Signed-off-by: Nicholas Bellinger
14 Feb, 2014
2 commits
-
vhost_zerocopy_callback accesses VQ right after it drops a ubuf
reference. In theory, this could race with device removal which waits
on the ubuf kref, and crash on use after free.Do all accesses within rcu read side critical section, and synchronize
on release.Since callbacks are always invoked from bh, synchronize_rcu_bh seems
enough and will help release complete a bit faster.Signed-off-by: Michael S. Tsirkin
Acked-by: Jason Wang
Signed-off-by: David S. Miller -
vhost checked the counter within the refcnt before decrementing. It
really wanted to know that it is the one that has the last reference, as
a way to batch freeing resources a bit more efficiently.Note: we only let refcount go to 0 on device release.
This works well but we now access the ref counter twice so there's a
race: all users might see a high count and decide to defer freeing
resources.
In the end no one initiates freeing resources until the last reference
is gone (which is on VM shotdown so might happen after a looooong time).Let's do what we probably should have done straight away:
switch from kref to plain atomic, documenting the
semantics, return the refcount value atomically after decrement,
then use that to avoid the deadlock.Reported-by: Qin Chuanyu
Signed-off-by: Michael S. Tsirkin
Acked-by: Jason Wang
Signed-off-by: David S. Miller