| Age | Commit message (Collapse) | Author |
|---|
|
commit b3743c71b7c33a126d6d8942bb268775987400ec upstream.
For all the entries allocated from the ring cmd area, the memory is
something like the stack memory, which will always reserve the old
data, so the entry->req.iov_bidi_cnt maybe none zero.
On some environments, the crash could be reproduce very easy and some
not. The following is the crash core trace as reported by Damien:
[ 240.143969] CPU: 0 PID: 1285 Comm: iscsi_trx Not tainted 4.12.0-rc1+ #3
[ 240.150607] Hardware name: ASUS All Series/H87-PRO, BIOS 2104 10/28/2014
[ 240.157331] task: ffff8807de4f5800 task.stack: ffffc900047dc000
[ 240.163270] RIP: 0010:memcpy_erms+0x6/0x10
[ 240.167377] RSP: 0018:ffffc900047dfc68 EFLAGS: 00010202
[ 240.172621] RAX: ffffc9065db85540 RBX: ffff8807f7980000 RCX: 0000000000000010
[ 240.179771] RDX: 0000000000000010 RSI: ffff8807de574fe0 RDI: ffffc9065db85540
[ 240.186930] RBP: ffffc900047dfd30 R08: ffff8807de41b000 R09: 0000000000000000
[ 240.194088] R10: 0000000000000040 R11: ffff8807e9b726f0 R12: 00000006565726b0
[ 240.201246] R13: ffffc90007612ea0 R14: 000000065657d540 R15: 0000000000000000
[ 240.208397] FS: 0000000000000000(0000) GS:ffff88081fa00000(0000) knlGS:0000000000000000
[ 240.216510] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 240.222280] CR2: ffffc9065db85540 CR3: 0000000001c0f000 CR4: 00000000001406f0
[ 240.229430] Call Trace:
[ 240.231887] ? tcmu_queue_cmd+0x83c/0xa80
[ 240.235916] ? target_check_reservation+0xcd/0x6f0
[ 240.240725] __target_execute_cmd+0x27/0xa0
[ 240.244918] target_execute_cmd+0x232/0x2c0
[ 240.249124] ? __local_bh_enable_ip+0x64/0xa0
[ 240.253499] iscsit_execute_cmd+0x20d/0x270
[ 240.257693] iscsit_sequence_cmd+0x110/0x190
[ 240.261985] iscsit_get_rx_pdu+0x360/0xc80
[ 240.267565] ? iscsi_target_rx_thread+0x54/0xd0
[ 240.273571] iscsi_target_rx_thread+0x9a/0xd0
[ 240.279413] kthread+0x113/0x150
[ 240.284120] ? iscsi_target_tx_thread+0x1e0/0x1e0
[ 240.290297] ? kthread_create_on_node+0x40/0x40
[ 240.296297] ret_from_fork+0x2e/0x40
[ 240.301332] Code: 90 90 90 90 90 eb 1e 0f 1f 00 48 89 f8 48 89 d1 48
c1 e9 03 83 e2 07 f3 48 a5 89 d1 f3 a4 c3 66 0f 1f 44 00 00 48 89 f8 48
89 d1 <f3> a4 c3 0f 1f 80 00 00 00 00 48 89 f8 48 83 fa 20 72 7e 40 38
[ 240.321751] RIP: memcpy_erms+0x6/0x10 RSP: ffffc900047dfc68
[ 240.328838] CR2: ffffc9065db85540
[ 240.333667] ---[ end trace b7e5354cfb54d08b ]---
To fix this, just memset all the entry memory before using it, and
also to be more readable we adjust the bidi code.
Fixed: fe25cc34795(tcmu: Recalculate the tcmu_cmd size to save cmd area
memories)
Reported-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
Tested-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
Reported-by: Damien Le Moal <damien.lemoal@wdc.com>
Tested-by: Damien Le Moal <damien.lemoal@wdc.com>
Reviewed-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Cc: <stable@vger.kernel.org> # 4.12+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 9d62bc0e6d79b11e3298e831358155930fb8f5e3 upstream.
When feeding the tcmu's cmd ring, we need to flush the dcache page
for the cmd entry to make sure these kernel stores are visible to
user space mappings of that page.
For the none PAD cmd entry, this will be flushed at the end of the
tcmu_queue_cmd_ring().
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Reviewed-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 1d6ef276594a781686058802996e09c8550fd767 upstream.
This patch addresses a COMPARE_AND_WRITE se_device->caw_sem leak,
that would be triggered during normal se_cmd shutdown or abort
via __transport_wait_for_tasks().
This would occur because target_complete_cmd() would catch this
early and do complete_all(&cmd->t_transport_stop_comp), but since
target_complete_ok_work() or target_complete_failure_work() are
never called to invoke se_cmd->transport_complete_callback(),
the COMPARE_AND_WRITE specific callbacks never release caw_sem.
To address this special case, go ahead and release caw_sem
directly from target_complete_cmd().
(Remove '&& success' from check, to release caw_sem regardless
of scsi_status - nab)
Signed-off-by: Jiang Yi <jiangyilism@gmail.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 138d351eefb727ab9e41a3dc5f112ceb4f6e59f2 upstream.
This patch re-introduces part of a long standing login workaround that
was recently dropped by:
commit 1c99de981f30b3e7868b8d20ce5479fa1c0fea46
Author: Nicholas Bellinger <nab@linux-iscsi.org>
Date: Sun Apr 2 13:36:44 2017 -0700
iscsi-target: Drop work-around for legacy GlobalSAN initiator
Namely, the workaround for FirstBurstLength ended up being required by
Mellanox Flexboot PXE boot ROMs as reported by Robert.
So this patch re-adds the work-around for FirstBurstLength within
iscsi_check_proposer_for_optional_reply(), and makes the key optional
to respond when the initiator does not propose, nor respond to it.
Also as requested by Arun, this patch introduces a new TPG attribute
named 'login_keys_workaround' that controls the use of both the
FirstBurstLength workaround, as well as the two other existing
workarounds for gPXE iSCSI boot client.
By default, the workaround is enabled with login_keys_workaround=1,
since Mellanox FlexBoot requires it, and Arun has verified the Qlogic
MSFT initiator already proposes FirstBurstLength, so it's uneffected
by this re-adding this part of the original work-around.
Reported-by: Robert LeBlanc <robert@leblancnet.us>
Cc: Robert LeBlanc <robert@leblancnet.us>
Reviewed-by: Arun Easi <arun.easi@cavium.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
When iscsi WRITE underflow occurs there are two different scenarios
that can happen.
Normally in practice, when an EDTL vs. SCSI CDB TRANSFER LENGTH
underflow is detected, the iscsi immediate data payload is the
smaller SCSI CDB TRANSFER LENGTH.
That is, when a host fabric LLD is using a fixed size EDTL for
a specific control CDB, the SCSI CDB TRANSFER LENGTH and actual
SCSI payload ends up being smaller than EDTL. In iscsi, this
means the received iscsi immediate data payload matches the
smaller SCSI CDB TRANSFER LENGTH, because there is no more
SCSI payload to accept beyond SCSI CDB TRANSFER LENGTH.
However, it's possible for a malicous host to send a WRITE
underflow where EDTL is larger than SCSI CDB TRANSFER LENGTH,
but incoming iscsi immediate data actually matches EDTL.
In the wild, we've never had a iscsi host environment actually
try to do this.
For this special case, it's wrong to truncate part of the
control CDB payload and continue to process the command during
underflow when immediate data payload received was larger than
SCSI CDB TRANSFER LENGTH, so go ahead and reject and drop the
bogus payload as a defensive action.
Note this potential bug was originally relaxed by the following
for allowing WRITE underflow in MSFT FCP host environments:
commit c72c5250224d475614a00c1d7e54a67f77cd3410
Author: Roland Dreier <roland@purestorage.com>
Date: Wed Jul 22 15:08:18 2015 -0700
target: allow underflow/overflow for PR OUT etc. commands
Cc: Roland Dreier <roland@purestorage.com>
Cc: Mike Christie <mchristi@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: <stable@vger.kernel.org> # v4.3+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch fixes a BUG() in iscsit_close_session() that could be
triggered when iscsit_logout_post_handler() execution from within
tx thread context was not run for more than SECONDS_FOR_LOGOUT_COMP
(15 seconds), and the TCP connection didn't already close before
then forcing tx thread context to automatically exit.
This would manifest itself during explicit logout as:
[33206.974254] 1 connection(s) still exist for iSCSI session to iqn.1993-08.org.debian:01:3f5523242179
[33206.980184] INFO: NMI handler (kgdb_nmi_handler) took too long to run: 2100.772 msecs
[33209.078643] ------------[ cut here ]------------
[33209.078646] kernel BUG at drivers/target/iscsi/iscsi_target.c:4346!
Normally when explicit logout attempt fails, the tx thread context
exits and iscsit_close_connection() from rx thread context does the
extra cleanup once it detects conn->conn_logout_remove has not been
cleared by the logout type specific post handlers.
To address this special case, if the logout post handler in tx thread
context detects conn->tx_thread_active has already been cleared, simply
return and exit in order for existing iscsit_close_connection()
logic from rx thread context do failed logout cleanup.
Reported-by: Bart Van Assche <bart.vanassche@sandisk.com>
Tested-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Mike Christie <mchristi@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: stable@vger.kernel.org # 3.14+
Tested-by: Gary Guo <ghg@datera.io>
Tested-by: Chu Yuan Lin <cyl@datera.io>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch fixes a se_cmd->cmd_kref underflow during CMD_T_ABORTED
when a fabric driver drops it's second reference from below the
target_core_tmr.c based callers of transport_cmd_finish_abort().
Recently with the conversion of kref to refcount_t, this bug was
manifesting itself as:
[705519.601034] refcount_t: underflow; use-after-free.
[705519.604034] INFO: NMI handler (kgdb_nmi_handler) took too long to run: 20116.512 msecs
[705539.719111] ------------[ cut here ]------------
[705539.719117] WARNING: CPU: 3 PID: 26510 at lib/refcount.c:184 refcount_sub_and_test+0x33/0x51
Since the original kref atomic_t based kref_put() didn't check for
underflow and only invoked the final callback when zero was reached,
this bug did not manifest in practice since all se_cmd memory is
using preallocated tags.
To address this, go ahead and propigate the existing return from
transport_put_cmd() up via transport_cmd_finish_abort(), and
change transport_cmd_finish_abort() + core_tmr_handle_tas_abort()
callers to only do their local target_put_sess_cmd() if necessary.
Reported-by: Bart Van Assche <bart.vanassche@sandisk.com>
Tested-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Mike Christie <mchristi@redhat.com>
Cc: Hannes Reinecke <hare@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Himanshu Madhani <himanshu.madhani@qlogic.com>
Cc: Sagi Grimberg <sagig@mellanox.com>
Cc: stable@vger.kernel.org # 3.14+
Tested-by: Gary Guo <ghg@datera.io>
Tested-by: Chu Yuan Lin <cyl@datera.io>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
There are three timing problems in the kthread usages of iscsi_target_mod:
- np_thread of struct iscsi_np
- rx_thread and tx_thread of struct iscsi_conn
In iscsit_close_connection(), it calls
send_sig(SIGINT, conn->tx_thread, 1);
kthread_stop(conn->tx_thread);
In conn->tx_thread, which is iscsi_target_tx_thread(), when it receive
SIGINT the kthread will exit without checking the return value of
kthread_should_stop().
So if iscsi_target_tx_thread() exit right between send_sig(SIGINT...)
and kthread_stop(...), the kthread_stop() will try to stop an already
stopped kthread.
This is invalid according to the documentation of kthread_stop().
(Fix -ECONNRESET logout handling in iscsi_target_tx_thread and
early iscsi_target_rx_thread failure case - nab)
Signed-off-by: Jiang Yi <jiangyilism@gmail.com>
Cc: <stable@vger.kernel.org> # v3.12+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch fixes a OOPs originally introduced by:
commit bb048357dad6d604520c91586334c9c230366a14
Author: Nicholas Bellinger <nab@linux-iscsi.org>
Date: Thu Sep 5 14:54:04 2013 -0700
iscsi-target: Add sk->sk_state_change to cleanup after TCP failure
which would trigger a NULL pointer dereference when a TCP connection
was closed asynchronously via iscsi_target_sk_state_change(), but only
when the initial PDU processing in iscsi_target_do_login() from iscsi_np
process context was blocked waiting for backend I/O to complete.
To address this issue, this patch makes the following changes.
First, it introduces some common helper functions used for checking
socket closing state, checking login_flags, and atomically checking
socket closing state + setting login_flags.
Second, it introduces a LOGIN_FLAGS_INITIAL_PDU bit to know when a TCP
connection has dropped via iscsi_target_sk_state_change(), but the
initial PDU processing within iscsi_target_do_login() in iscsi_np
context is still running. For this case, it sets LOGIN_FLAGS_CLOSED,
but doesn't invoke schedule_delayed_work().
The original NULL pointer dereference case reported by MNC is now handled
by iscsi_target_do_login() doing a iscsi_target_sk_check_close() before
transitioning to FFP to determine when the socket has already closed,
or iscsi_target_start_negotiation() if the login needs to exchange
more PDUs (eg: iscsi_target_do_login returned 0) but the socket has
closed. For both of these cases, the cleanup up of remaining connection
resources will occur in iscsi_target_start_negotiation() from iscsi_np
process context once the failure is detected.
Finally, to handle to case where iscsi_target_sk_state_change() is
called after the initial PDU procesing is complete, it now invokes
conn->login_work -> iscsi_target_do_login_rx() to perform cleanup once
existing iscsi_target_sk_check_close() checks detect connection failure.
For this case, the cleanup of remaining connection resources will occur
in iscsi_target_do_login_rx() from delayed workqueue process context
once the failure is detected.
Reported-by: Mike Christie <mchristi@redhat.com>
Reviewed-by: Mike Christie <mchristi@redhat.com>
Tested-by: Mike Christie <mchristi@redhat.com>
Cc: Mike Christie <mchristi@redhat.com>
Reported-by: Hannes Reinecke <hare@suse.com>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Varun Prakash <varun@chelsio.com>
Cc: <stable@vger.kernel.org> # v3.12+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
We currently do
tcmu_free_device ->tcmu_netlink_event(TCMU_CMD_REMOVED_DEVICE) ->
uio_unregister_device -> kfree(tcmu_dev).
The problem is that the kernel does not wait for userspace to
do the close() on the uio device before freeing the tcmu_dev.
We can then hit a race where the kernel frees the tcmu_dev before
userspace does close() and so when close() -> release -> tcmu_release
is done, we try to access a freed tcmu_dev.
This patch made over the target-pending master branch moves the freeing
of the tcmu_dev to when the last reference has been dropped.
This also fixes a leak where if tcmu_configure_device was not called on a
device we did not free udev->name which was allocated at tcmu_alloc_device time.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
During v4.3 when the overflow/underflow check was relaxed by
commit c72c525022:
commit c72c5250224d475614a00c1d7e54a67f77cd3410
Author: Roland Dreier <roland@purestorage.com>
Date: Wed Jul 22 15:08:18 2015 -0700
target: allow underflow/overflow for PR OUT etc. commands
to allow underflow/overflow for Windows compliance + FCP, a
consequence was to allow control CDBs to process overflow
data for iscsi-target with immediate data as well.
As per Roland's original change, continue to allow underflow
cases for control CDBs to make Windows compliance + FCP happy,
but until overflow for control CDBs is supported tree-wide,
explicitly reject all control WRITEs with overflow following
pre v4.3.y logic.
Reported-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Roland Dreier <roland@purestorage.com>
Cc: <stable@vger.kernel.org> # v4.3+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending
Pull SCSI target updates from Nicholas Bellinger:
"Things were a lot more calm than previously expected. It's primarily
fixes in various areas, with most of the new functionality centering
around TCMU backend driver work that Xiubo Li has been driving.
Here's the summary on the feature side:
- Make T10-PI verify configurable for emulated (FILEIO + RD) backends
(Dmitry Monakhov)
- Allow target-core/TCMU pass-through to use in-kernel SPC-PR logic
(Bryant Ly + MNC)
- Add TCMU support for growing ring buffer size (Xiubo Li + MNC)
- Add TCMU support for global block data pool (Xiubo Li + MNC)
and on the bug-fix side:
- Fix COMPARE_AND_WRITE non GOOD status handling for READ phase
failures (Gary Guo + nab)
- Fix iscsi-target hang with explicitly changing per NodeACL
CmdSN number depth with concurrent login driven session
reinstatement. (Gary Guo + nab)
- Fix ibmvscsis fabric driver ABORT task handling (Bryant Ly)
- Fix target-core/FILEIO zero length handling (Bart Van Assche)
Also, there was an OOPs introduced with the WRITE_VERIFY changes that
I ended up reverting at the last minute, because as not unusual Bart
and I could not agree on the fix in time for -rc1. Since it's specific
to a conformance test, it's been reverted for now.
There is a separate patch in the queue to address the underlying
control CDB write overflow regression in >= v4.3 separate from the
WRITE_VERIFY revert here, that will be pushed post -rc1"
* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (30 commits)
Revert "target: Fix VERIFY and WRITE VERIFY command parsing"
IB/srpt: Avoid that aborting a command triggers a kernel warning
IB/srpt: Fix abort handling
target/fileio: Fix zero-length READ and WRITE handling
ibmvscsis: Do not send aborted task response
tcmu: fix module removal due to stuck thread
target: Don't force session reset if queue_depth does not change
iscsi-target: Set session_fall_back_to_erl0 when forcing reinstatement
target: Fix compare_and_write_callback handling for non GOOD status
tcmu: Recalculate the tcmu_cmd size to save cmd area memories
tcmu: Add global data block pool support
tcmu: Add dynamic growing data area feature support
target: fixup error message in target_tg_pt_gp_tg_pt_gp_id_store()
target: fixup error message in target_tg_pt_gp_alua_access_type_store()
target/user: PGR Support
target: Add WRITE_VERIFY_16
Documentation/target: add an example script to configure an iSCSI target
target: Use kmalloc_array() in transport_kmap_data_sg()
target: Use kmalloc_array() in compare_and_write_callback()
target: Improve size determinations in two functions
...
|
|
This reverts commit 0e2eb7d12eaa8e391bf5615d4271bb87a649caaa
Author: Bart Van Assche <bart.vanassche@sandisk.com>
Date: Thu Mar 30 10:12:39 2017 -0700
target: Fix VERIFY and WRITE VERIFY command parsing
This patch broke existing behaviour for WRITE_VERIFY because
it dropped the original SCF_SCSI_DATA_CDB assignment for
bytchk = 0 so target_cmd_size_check() no longer rejected
this case, allowing an overflow case to trigger an OOPs
in iscsi-target.
Since the short term and long term fixes are still being
discussed, revert it for now since it's late in the merge
window and try again in v4.13-rc1.
Conflicts:
drivers/target/target_core_sbc.c
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch fixes zero-length READ and WRITE handling in target/FILEIO,
which was broken a long time back by:
Since:
commit d81cb44726f050d7cf1be4afd9cb45d153b52066
Author: Paolo Bonzini <pbonzini@redhat.com>
Date: Mon Sep 17 16:36:11 2012 -0700
target: go through normal processing for all zero-length commands
which moved zero-length READ and WRITE completion out of target-core,
to doing submission into backend driver code.
To address this, go ahead and invoke target_complete_cmd() for any
non negative return value in fd_do_rw().
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Andy Grover <agrover@redhat.com>
Cc: David Disseldorp <ddiss@suse.de>
Cc: <stable@vger.kernel.org> # v3.7+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
We need to do a kthread_should_stop to check when kthread_stop has been
called.
This was a regression added in
b6df4b79a5514a9c6c53533436704129ef45bf76
tcmu: Add global data block pool support
so not sure if you wanted to merge it in with that patch or what.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Keeping in the idempotent nature of target_core_fabric_configfs.c,
if a queue_depth value is set and it's the same as the existing
value, don't attempt to force session reinstatement.
Reported-by: Raghu Krishnamurthy <rk@datera.io>
Cc: Raghu Krishnamurthy <rk@datera.io>
Tested-by: Gary Guo <ghg@datera.io>
Cc: Gary Guo <ghg@datera.io>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
While testing modification of per se_node_acl queue_depth forcing
session reinstatement via lio_target_nacl_cmdsn_depth_store() ->
core_tpg_set_initiator_node_queue_depth(), a hung task bug triggered
when changing cmdsn_depth invoked session reinstatement while an iscsi
login was already waiting for session reinstatement to complete.
This can happen when an outstanding se_cmd descriptor is taking a
long time to complete, and session reinstatement from iscsi login
or cmdsn_depth change occurs concurrently.
To address this bug, explicitly set session_fall_back_to_erl0 = 1
when forcing session reinstatement, so session reinstatement is
not attempted if an active session is already being shutdown.
This patch has been tested with two scenarios. The first when
iscsi login is blocked waiting for iscsi session reinstatement
to complete followed by queue_depth change via configfs, and
second when queue_depth change via configfs us blocked followed
by a iscsi login driven session reinstatement.
Note this patch depends on commit d36ad77f702 to handle multiple
sessions per se_node_acl when changing cmdsn_depth, and for
pre v4.5 kernels will need to be included for stable as well.
Reported-by: Gary Guo <ghg@datera.io>
Tested-by: Gary Guo <ghg@datera.io>
Cc: Gary Guo <ghg@datera.io>
Cc: <stable@vger.kernel.org> # v4.1+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Following the bugfix for handling non SAM_STAT_GOOD COMPARE_AND_WRITE
status during COMMIT phase in commit 9b2792c3da1, the same bug exists
for the READ phase as well.
This would manifest first as a lost SCSI response, and eventual
hung task during fabric driver logout or re-login, as existing
shutdown logic waited for the COMPARE_AND_WRITE se_cmd->cmd_kref
to reach zero.
To address this bug, compare_and_write_callback() has been changed
to set post_ret = 1 and return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE
as necessary to signal failure status.
Reported-by: Bill Borsari <wgb@datera.io>
Cc: Bill Borsari <wgb@datera.io>
Tested-by: Gary Guo <ghg@datera.io>
Cc: Gary Guo <ghg@datera.io>
Cc: <stable@vger.kernel.org> # v4.1+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
For the "struct tcmu_cmd_entry" in cmd area, the minimum size
will be sizeof(struct tcmu_cmd_entry) == 112 Bytes. And it could
fill about (sizeof(struct rsp) - sizeof(struct req)) /
sizeof(struct iovec) == 68 / 16 ~= 4 data regions(iov[4]) by
default.
For most tcmu_cmds, the data block indexes allocated from the
data area will be continuous. And for the continuous blocks they
will be merged into the same region using only one iovec. For
the current code, it will always allocates the same number of
iovecs with blocks for each tcmu_cmd, and it will wastes much
memories.
For example, when the block size is 4K and the DATA_OUT buffer
size is 64K, and the regions needed is less than 5(on my
environment is almost 99.7%). The current code will allocate
about 16 iovecs, and there will be (16 - 4) * sizeof(struct
iovec) = 192 Bytes cmd area memories wasted.
Here adds two helpers to calculate the base size and full size
of the tcmu_cmd. And will recalculate them again when it make sure
how many iovs is needed before insert it to cmd area.
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Acked-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
For each target there will be one ring, when the target number
grows larger and larger, it could eventually runs out of the
system memories.
In this patch for each target ring, currently for the cmd area
the size will be fixed to 8MB and for the data area the size
will grow from 0 to max 256K * PAGE_SIZE(1G for 4K page size).
For all the targets' data areas, they will get empty blocks
from the "global data block pool", which has limited to 512K *
PAGE_SIZE(2G for 4K page size) for now.
When the "global data block pool" has been used up, then any
target could wake up the unmap thread routine to shrink other
targets' data area memories. And the unmap thread routine will
always try to truncate the ring vma from the last using block
offset.
When user space has touched the data blocks out of tcmu_cmd
iov[], the tcmu_page_fault() will try to return one zeroed blocks.
Here we move the timeout's tcmu_handle_completions() into unmap
thread routine, that's to say when the timeout fired, it will
only do the tcmu_check_expired_cmd() and then wake up the unmap
thread to do the completions() and then try to shrink its idle
memories. Then the cmdr_lock could be a mutex and could simplify
this patch because the unmap_mapping_range() or zap_* may go to
sleep.
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Jianfei Hu <hujianfei@cmss.chinamobile.com>
Acked-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Currently for the TCMU, the ring buffer size is fixed to 64K cmd
area + 1M data area, and this will be bottlenecks for high iops.
The struct tcmu_cmd_entry {} size is fixed about 112 bytes with
iovec[N] & N <= 4, and the size of struct iovec is about 16 bytes.
If N == 0, the ratio will be sizeof(cmd entry) : sizeof(datas) ==
112Bytes : (N * 4096)Bytes = 28 : 0, no data area is need.
If 0 < N <=4, the ratio will be sizeof(cmd entry) : sizeof(datas)
== 112Bytes : (N * 4096)Bytes = 28 : (N * 1024), so the max will
be 28 : 1024.
If N > 4, the sizeof(cmd entry) will be [(N - 4) *16 + 112] bytes,
and its corresponding data size will be [N * 4096], so the ratio
of sizeof(cmd entry) : sizeof(datas) == [(N - 4) * 16 + 112)Bytes
: (N * 4096)Bytes == 4/1024 - 12/(N * 1024), so the max is about
4 : 1024.
When N is bigger, the ratio will be smaller.
As the initial patch, we will set the cmd area size to 2M, and
the cmd area size to 32M. The TCMU will dynamically grows the data
area from 0 to max 32M size as needed.
The cmd area memory will be allocated through vmalloc(), and the
data area's blocks will be allocated individually later when needed.
The allocated data area block memory will be managed via radix tree.
For now the bitmap still be the most efficient way to search and
manage the block index, this could be update later.
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Signed-off-by: Jianfei Hu <hujianfei@cmss.chinamobile.com>
Acked-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
When setting up an ALUA target port group with an invalid ID the
error message
kstrtoul() returned -22 for tg_pt_gp_id
is displayed, which is not really helpful.
Convert it to something sane.
And while we're at it, join the messages onto a single line.
Signed-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Bart van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
When setting up a target the error message:
Unable to do set ##_name ALUA state on non valid tg_pt_gp ID: 0
is displayed.
Apparently concatenation doesn't work in a string; one should be using
implicit string concatenation here.
Signed-off-by: Hannes Reinecke <hare@suse.com>
Reviewed-by: Bart van Assche <bart.vanassche@sandisk.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This adds initial PGR support for just TCMU, since tcmu doesn't
have the necessary IT_NEXUS info to process PGR in userspace,
so have those commands be processed in kernel.
HA support is not available yet, we will work on it if this patch
is acceptable.
Signed-off-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch addresses clients who needs write_verify_16 for
large volume groups such as AIX.
Signed-off-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
A multiplication for the size determination of a memory allocation
indicated that an array data structure should be processed.
Thus use the corresponding function "kmalloc_array".
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
* A multiplication for the size determination of a memory allocation
indicated that an array data structure should be processed.
Thus use the corresponding function "kmalloc_array".
This issue was detected by using the Coccinelle software.
* Replace the specification of a data structure by a pointer dereference
to make the corresponding size determination a bit safer according to
the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Replace the specification of two data structures by pointer dereferences
as the parameter for the operator "sizeof" to make the corresponding size
determinations a bit safer according to the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
The script "checkpatch.pl" pointed information out like the following.
WARNING: Possible unnecessary 'out of memory' message
Thus remove such statements here.
Link: http://events.linuxfoundation.org/sites/events/files/slides/LCJ16-Refactor_Strings-WSang_0.pdf
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
* Multiplications for the size determination of memory allocations
indicated that array data structures should be processed.
Thus use the corresponding function "kcalloc".
This issue was detected by using the Coccinelle software.
* Replace the specification of data structures by pointer dereferences
to make the corresponding size determination a bit safer according to
the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Replace the specification of four data structures by pointer dereferences
as the parameter for the operator "sizeof" to make the corresponding size
determinations a bit safer according to the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
The script "checkpatch.pl" pointed information out like the following.
WARNING: Possible unnecessary 'out of memory' message
Thus remove such statements here.
Link: http://events.linuxfoundation.org/sites/events/files/slides/LCJ16-Refactor_Strings-WSang_0.pdf
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
* A multiplication for the size determination of a memory allocation
indicated that an array data structure should be processed.
Thus use the corresponding function "kcalloc".
This issue was detected by using the Coccinelle software.
* Replace the specification of a data structure by a pointer dereference
to make the corresponding size determination a bit safer according to
the Linux coding style convention.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Currently ramdisk and fileio always perform PI verification
before and after backend IO. This approach is not very flexible.
Because some one may want to postpone this work to other layers in
IO stack. For example if we want to test blk_integrity_profile
testcase:
https://github.com/dmonakhov/xfstests/commit/dee408c868861d6b6871dbb3381facee7effdbe4
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
If we failed to read data from backing file (probably because some one
truncate file under us), we must zerofill cmd's data, otherwise it will
be returned as is. Most likely cmd's data are unitialized pages from
page cache. This result in information leak.
(Change BUG_ON into -EINVAL se_cmd failure - nab)
testcase: https://github.com/dmonakhov/xfstests/commit/e11a1b7b907ca67b1be51a1594025600767366d5
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Use the value of the BYTCHK field to determine the size of the
Data-Out buffer. For VERIFY, honor the VRPROTECT, DPO and FUA
fields. This patch avoids that LIO complains about a mismatch
between the expected transfer length and the SCSI CDB length
if the value of the BYTCHK field is 0.
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Max Lohrmann <post@wickenrode.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This commit updated persistent revervation out service action
code table in SPC-5 for development.
Signed-off-by: Zhu Lingshan <lszhu@suse.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
refcount_t type and corresponding API should be
used instead of atomic_t when the variable is used as
a reference counter. This allows to avoid accidental
refcounter overflows that might lead to use-after-free
situations.
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Hans Liljestrand <ishkamiel@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: David Windsor <dwindsor@gmail.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Pull block layer updates from Jens Axboe:
- Add BFQ IO scheduler under the new blk-mq scheduling framework. BFQ
was initially a fork of CFQ, but subsequently changed to implement
fairness based on B-WF2Q+, a modified variant of WF2Q. BFQ is meant
to be used on desktop type single drives, providing good fairness.
From Paolo.
- Add Kyber IO scheduler. This is a full multiqueue aware scheduler,
using a scalable token based algorithm that throttles IO based on
live completion IO stats, similary to blk-wbt. From Omar.
- A series from Jan, moving users to separately allocated backing
devices. This continues the work of separating backing device life
times, solving various problems with hot removal.
- A series of updates for lightnvm, mostly from Javier. Includes a
'pblk' target that exposes an open channel SSD as a physical block
device.
- A series of fixes and improvements for nbd from Josef.
- A series from Omar, removing queue sharing between devices on mostly
legacy drivers. This helps us clean up other bits, if we know that a
queue only has a single device backing. This has been overdue for
more than a decade.
- Fixes for the blk-stats, and improvements to unify the stats and user
windows. This both improves blk-wbt, and enables other users to
register a need to receive IO stats for a device. From Omar.
- blk-throttle improvements from Shaohua. This provides a scalable
framework for implementing scalable priotization - particularly for
blk-mq, but applicable to any type of block device. The interface is
marked experimental for now.
- Bucketized IO stats for IO polling from Stephen Bates. This improves
efficiency of polled workloads in the presence of mixed block size
IO.
- A few fixes for opal, from Scott.
- A few pulls for NVMe, including a lot of fixes for NVMe-over-fabrics.
From a variety of folks, mostly Sagi and James Smart.
- A series from Bart, improving our exposed info and capabilities from
the blk-mq debugfs support.
- A series from Christoph, cleaning up how handle WRITE_ZEROES.
- A series from Christoph, cleaning up the block layer handling of how
we track errors in a request. On top of being a nice cleanup, it also
shrinks the size of struct request a bit.
- Removal of mg_disk and hd (sorry Linus) by Christoph. The former was
never used by platforms, and the latter has outlived it's usefulness.
- Various little bug fixes and cleanups from a wide variety of folks.
* 'for-4.12/block' of git://git.kernel.dk/linux-block: (329 commits)
block: hide badblocks attribute by default
blk-mq: unify hctx delay_work and run_work
block: add kblock_mod_delayed_work_on()
blk-mq: unify hctx delayed_run_work and run_work
nbd: fix use after free on module unload
MAINTAINERS: bfq: Add Paolo as maintainer for the BFQ I/O scheduler
blk-mq-sched: alloate reserved tags out of normal pool
mtip32xx: use runtime tag to initialize command header
scsi: Implement blk_mq_ops.show_rq()
blk-mq: Add blk_mq_ops.show_rq()
blk-mq: Show operation, cmd_flags and rq_flags names
blk-mq: Make blk_flags_show() callers append a newline character
blk-mq: Move the "state" debugfs attribute one level down
blk-mq: Unregister debugfs attributes earlier
blk-mq: Only unregister hctxs for which registration succeeded
blk-mq-debugfs: Rename functions for registering and unregistering the mq directory
blk-mq: Let blk_mq_debugfs_register() look up the queue name
blk-mq: Register <dev>/queue/mq after having registered <dev>/queue
ide-pm: always pass 0 error to ide_complete_rq in ide_do_devset
ide-pm: always pass 0 error to __blk_end_request_all
..
|
|
This passes on the scsi_cmnd result field to users of passthrough
requests. Currently we abuse req->errors for this purpose, but that
field will go away in its current form.
Note that the old IDE code abuses the errors field in very creative
ways and stores all kinds of different values in it. I didn't dare
to touch this magic, so the abuses are brought forward 1:1.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Bart Van Assche <Bart.VanAssche@sandisk.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Now that we use the proper REQ_OP_WRITE_ZEROES operation everywhere we can
kill this hack.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Instead of bloating the generic struct request with it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
Instead of parsing address strings, use a generic
helper.
Acked-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@fb.com>
|
|
For the bidirectional case, the Data-Out buffer blocks will always at
the head of the tcmu_cmd's bitmap, and before gathering the Data-In
buffer, first of all it should skip the Data-Out ones, or the device
supporting BIDI commands won't work.
Fixed: 26418649eead ("target/user: Introduce data_bitmap, replace
data_length/data_head/data_tail")
Reported-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Tested-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Cc: stable@vger.kernel.org # 4.6+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Once upon a time back in 2009, a work-around was added to support
the GlobalSAN iSCSI initiator v3.3 for MacOSX, which during login
did not propose nor respond to MaxBurstLength, FirstBurstLength,
DefaultTime2Wait and DefaultTime2Retain keys.
The work-around in iscsi_check_proposer_for_optional_reply()
allowed the missing keys to be proposed, but did not require
waiting for a response before moving to full feature phase
operation. This allowed GlobalSAN v3.3 to work out-of-the
box, and for many years we didn't run into login interopt
issues with any other initiators..
Until recently, when Martin tried a QLogic 57840S iSCSI Offload
HBA on Windows 2016 which completed login, but subsequently
failed with:
Got unknown iSCSI OpCode: 0x43
The issue was QLogic MSFT side did not propose DefaultTime2Wait +
DefaultTime2Retain, so LIO proposes them itself, and immediately
transitions to full feature phase because of the GlobalSAN hack.
However, the QLogic MSFT side still attempts to respond to
DefaultTime2Retain + DefaultTime2Wait, even though LIO has set
ISCSI_FLAG_LOGIN_NEXT_STAGE3 + ISCSI_FLAG_LOGIN_TRANSIT
in last login response.
So while the QLogic MSFT side should have been proposing these
two keys to start, it was doing the correct thing per RFC-3720
attempting to respond to proposed keys before transitioning to
full feature phase.
All that said, recent versions of GlobalSAN iSCSI (v5.3.0.541)
does correctly propose the four keys during login, making the
original work-around moot.
So in order to allow QLogic MSFT to run unmodified as-is, go
ahead and drop this long standing work-around.
Reported-by: Martin Svec <martin.svec@zoner.cz>
Cc: Martin Svec <martin.svec@zoner.cz>
Cc: Himanshu Madhani <Himanshu.Madhani@cavium.com>
Cc: Arun Easi <arun.easi@cavium.com>
Cc: stable@vger.kernel.org # 3.1+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
Multiple threads could be writing to alua_access_state at
the same time, or there could be multiple STPGs in flight
(different initiators sending them or one initiator sending
them to different ports), or a combo of both and the
core_alua_do_transition_tg_pt calls will race with each other.
Because from the last patches we no longer delay running
core_alua_do_transition_tg_pt_work, there does not seem to be
any point in running that in a workqueue. And, we always
wait for it to complete one way or another, so we can sleep
in this code path. So, this patch made over target-pending just adds a
mutex and does the work core_alua_do_transition_tg_pt_work was doing in
core_alua_do_transition_tg_pt.
There is also no need to use an atomic for the
tg_pt_gp_alua_access_state. In core_alua_do_transition_tg_pt we will
test and set it under the transition mutex. And, it is a int/32 bits
so in the other places where it is read, we will never see it partially
updated.
Signed-off-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch changes iscsi-target to propagate iscsit_transport
->iscsit_queue_data_in() and ->iscsit_queue_status() callback
errors, back up into target-core.
This allows target-core to retry failed iscsit_transport
callbacks using internal queue-full logic.
Reported-by: Potnuri Bharat Teja <bharat@chelsio.com>
Reviewed-by: Potnuri Bharat Teja <bharat@chelsio.com>
Tested-by: Potnuri Bharat Teja <bharat@chelsio.com>
Cc: Potnuri Bharat Teja <bharat@chelsio.com>
Reported-by: Steve Wise <swise@opengridcomputing.com>
Cc: Steve Wise <swise@opengridcomputing.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
This patch fixes a set of queue-full response handling
bugs, where outgoing responses are leaked when a fabric
driver is propagating non -EAGAIN or -ENOMEM errors
to target-core.
It introduces TRANSPORT_COMPLETE_QF_ERR state used to
signal when CHECK_CONDITION status should be generated,
when fabric driver ->write_pending(), ->queue_data_in(),
or ->queue_status() callbacks fail with non -EAGAIN or
-ENOMEM errors, and data-transfer should not be retried.
Note all fabric driver -EAGAIN and -ENOMEM errors are
still retried indefinately with associated data-transfer
callbacks, following existing queue-full logic.
Also fix two missing ->queue_status() queue-full cases
related to CMD_T_ABORTED w/ TAS status handling.
Reported-by: Potnuri Bharat Teja <bharat@chelsio.com>
Reviewed-by: Potnuri Bharat Teja <bharat@chelsio.com>
Tested-by: Potnuri Bharat Teja <bharat@chelsio.com>
Cc: Potnuri Bharat Teja <bharat@chelsio.com>
Reported-by: Steve Wise <swise@opengridcomputing.com>
Cc: Steve Wise <swise@opengridcomputing.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
The t_data_nents and t_bidi_data_nents are the numbers of the
segments, but it couldn't be sure the block size equals to size
of the segment.
For the worst case, all the blocks are discontiguous and there
will need the same number of iovecs, that's to say: blocks == iovs.
So here just set the number of iovs to block count needed by tcmu
cmd.
Tested-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Reviewed-by: Mike Christie <mchristi@redhat.com>
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Cc: stable@vger.kernel.org # 3.18+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
|
If there has BIDI data, its first iov[] will overwrite the last
iov[] for se_cmd->t_data_sg.
To fix this, we can just increase the iov pointer, but this may
introuduce a new memory leakage bug: If the se_cmd->data_length
and se_cmd->t_bidi_data_sg->length are all not aligned up to the
DATA_BLOCK_SIZE, the actual length needed maybe larger than just
sum of them.
So, this could be avoided by rounding all the data lengthes up
to DATA_BLOCK_SIZE.
Reviewed-by: Mike Christie <mchristi@redhat.com>
Tested-by: Ilias Tsitsimpis <iliastsi@arrikto.com>
Reviewed-by: Bryant G. Ly <bryantly@linux.vnet.ibm.com>
Signed-off-by: Xiubo Li <lixiubo@cmss.chinamobile.com>
Cc: stable@vger.kernel.org # 3.18+
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
|
Xiubo Li
Jiang Yi
Nicholas Bellinger
Mike Christie
Linus Torvalds
Bart Van Assche
Hannes Reinecke
Bryant G. Ly
Markus Elfring
Dmitry Monakhov
Zhu Lingshan
Elena Reshetova
Christoph Hellwig
Sagi Grimberg