| Age | Commit message (Collapse) | Author |
|---|
|
In trace.c, the function trace_create_maxlat_file() is defined behind the
#ifdef CONFIG_TRACER_MAX_TRACE block. The #else part defines it as:
#define trace_create_maxlat_file(tr, d_tracer) \
trace_create_file("tracing_max_latency", TRACE_MODE_WRITE, \
d_tracer, tr, &tracing_max_lat_fops)
But the one place that it it used has:
#ifdef CONFIG_TRACER_MAX_TRACE
trace_create_maxlat_file(tr, d_tracer);
#endif
Which is pointless and also wrong!
It only gets created when both CONFIG_TRACE_MAX_TRACE and CONFIG_FS_NOTIFY
is defined, but the file itself should not be dependent on
CONFIG_FS_NOTIFY. Always create that file when TRACE_MAX_TRACE is defined
regardless if FS_NOTIFY is or is not.
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Link: https://patch.msgid.link/20260207191101.0e014abd@robin
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
The function tracing_set_filter_buffering() is only used in
trace_events_hist.c. Move it to that file and make it static.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: https://patch.msgid.link/20260206195936.617080218@kernel.org
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
When the triggers were first created, they may not have had a file
parameter passed to them and things needed to be done generically.
But today, all triggers have a file parameter passed to them. Remove the
generic code and add a "if (WARN_ON_ONCE(!file))" to each trigger.
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Reviewed-by: Tom Zanussi <zanussi@kernel.org>
Link: https://patch.msgid.link/20260206101351.609d8906@gandalf.local.home
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Extend a KUnit test suite for the bitops API to cover more APIs from
include/asm-generic/bitops/instrumented-atomic.h.
- change_bit()
- test_and_set_bit()
- test_and_clear_bit()
- test_and_change_bit()
Verified on x86_64, i386, and arm64 architectures.
Sample KUnit output:
KTAP version 1
# Subtest: test_change_bit
ok 1 BITOPS_4
ok 2 BITOPS_7
ok 3 BITOPS_11
ok 4 BITOPS_31
ok 5 BITOPS_88
# test_change_bit: pass:5 fail:0 skip:0 total:5
ok 2 test_change_bit
KTAP version 1
# Subtest: test_test_and_set_bit_test_and_clear_bit
ok 1 BITOPS_4
ok 2 BITOPS_7
ok 3 BITOPS_11
ok 4 BITOPS_31
ok 5 BITOPS_88
# test_test_and_set_bit_test_and_clear_bit: pass:5 fail:0 skip:0 total:5
ok 3 test_test_and_set_bit_test_and_clear_bit
KTAP version 1
# Subtest: test_test_and_change_bit
ok 1 BITOPS_4
ok 2 BITOPS_7
ok 3 BITOPS_11
ok 4 BITOPS_31
ok 5 BITOPS_88
# test_test_and_change_bit: pass:5 fail:0 skip:0 total:5
ok 4 test_test_and_change_bit
Signed-off-by: Ryota Sakamoto <sakamo.ryota@gmail.com>
Signed-off-by: Yury Norov <ynorov@nvidia.com>
|
|
It seems we have at least one more file to be added to the MAINTAINERS
that corresponds to bit operations.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Yury Norov <ynorov@nvidia.com>
|
|
In the FIND_NTH_BIT macro, if the 'size' parameter is 0, both the
loop conditions and the modulo condition are not met. Consequently,
the 'tmp' variable remains uninitialized before being used in the
'found' label.
This results in the following smatch errors:
lib/find_bit.c:164 __find_nth_bit() error: uninitialized symbol 'tmp'.
lib/find_bit.c:171 __find_nth_and_bit() error: uninitialized symbol 'tmp'.
lib/find_bit.c:178 __find_nth_andnot_bit() error: uninitialized symbol 'tmp'.
lib/find_bit.c:187 __find_nth_and_andnot_bit() error: uninitialized symbol 'tmp'.
Initialize 'tmp' to 0 to ensure that fns() operates on a zeroed value
(no bits set) when size is 0, preventing the use of garbage values.
[Yury: size == 0 is generally a sign of error on client side, and in
this case, any returned value is OK because the returned value would be
greater than 'size'. Applying the patch to reduce the checker noise.]
Signed-off-by: Lee Yongjun <jun85566@gmail.com>
Signed-off-by: Yury Norov <ynorov@nvidia.com>
|
|
Add a KUnit test suite for the bitops API.
The existing 'lib/test_bitops.c' is preserved as-is because it contains
ad-hoc micro-benchmarks 'test_fns' and is intended to ensure no compiler
warnings from C=1 sparse checker or -Wextra compilations.
Introduce 'lib/tests/bitops_kunit.c' for functional regression testing. It
ports the test logic and data patterns from 'lib/test_bitops.c' to KUnit,
verifying correct behavior across various input patterns and
architecture-specific edge cases using isolated stack-allocated bitmaps.
The following test logic has been ported from test_bitops_startup() in
lib/test_bitops.c:
- set_bit() / clear_bit() / find_first_bit() validation ->
test_set_bit_clear_bit()
- get_count_order() validation -> test_get_count_order()
- get_count_order_long() validation -> test_get_count_order_long()
Also improve the find_first_bit() test to check the full bitmap length
(BITOPS_LENGTH) instead of omitting the last bit, ensuring the bitmap is
completely empty after cleanup.
Verified on x86_64, i386, and arm64 architectures.
Sample KUnit output:
KTAP version 1
# Subtest: bitops
# module: bitops_kunit
1..3
KTAP version 1
# Subtest: test_set_bit_clear_bit
ok 1 BITOPS_4
ok 2 BITOPS_7
ok 3 BITOPS_11
ok 4 BITOPS_31
ok 5 BITOPS_88
# test_set_bit_clear_bit: pass:5 fail:0 skip:0 total:5
ok 1 test_set_bit_clear_bit
KTAP version 1
# Subtest: test_get_count_order
ok 1 0x00000003
ok 2 0x00000004
ok 3 0x00001fff
ok 4 0x00002000
ok 5 0x50000000
ok 6 0x80000000
ok 7 0x80003000
# test_get_count_order: pass:7 fail:0 skip:0 total:7
ok 2 test_get_count_order
KTAP version 1
# Subtest: test_get_count_order_long
ok 1 0x0000000300000000
ok 2 0x0000000400000000
ok 3 0x00001fff00000000
ok 4 0x0000200000000000
ok 5 0x5000000000000000
ok 6 0x8000000000000000
ok 7 0x8000300000000000
# test_get_count_order_long: pass:7 fail:0 skip:0 total:7
ok 3 test_get_count_order_long
[Yury: trim Kconfig help message]
CC: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Wei Yang <richard.weiyang@gmail.com>
Signed-off-by: Ryota Sakamoto <sakamo.ryota@gmail.com>
Signed-off-by: Yury Norov <ynorov@nvidia.com>
|
|
For smbdirect it required to use different ports depending
on the RDMA protocol. E.g. for iWarp 5445 is needed
(as tcp port 445 already used by the raw tcp transport for SMB),
while InfiniBand, RoCEv1 and RoCEv2 use port 445, as they
use an independent port range (even for RoCEv2, which uses udp
port 4791 itself).
Currently ksmbd is not able to function correctly at
all if the system has iWarp (RDMA_NODE_RNIC) interface(s)
and any InfiniBand, RoCEv1 and/or RoCEv2 interface(s)
at the same time.
Now we do a wildcard listen on port 5445 only
for iWarp devices and another wildcard listen
on port 445 of any InfiniBand, RoCEv1 and/or RoCEv2
devices.
The wildcard listeners also work if there is
no device of the requested node_type, this
is the same logic as we had before, but before
we had to decide between port 5445 or 445
and now both are possible at the same time.
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Leon Romanovsky <leon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-rdma@vger.kernel.org
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
For smbdirect it required to use different ports depending
on the RDMA protocol. E.g. for iWarp 5445 is needed
(as tcp port 445 already used by the raw tcp transport for SMB),
while InfiniBand, RoCEv1 and RoCEv2 use port 445, as they
use an independent port range (even for RoCEv2, which uses udp
port 4791 itself).
And cifs.ko uses 5445 with a fallback to 445, which
means depending on the available interfaces, it tries
5445 in the RoCE range or may tries iWarp with 445
as a fallback. This leads to strange error messages
and strange network captures.
To avoid these problems they will be able to
use rdma_restrict_node_type(RDMA_NODE_RNIC) before
trying port 5445 and rdma_restrict_node_type(RDMA_NODE_IB_CA)
before trying port 445. It means we'll get early
-ENODEV early from rdma_resolve_addr() without any
network traffic and timeouts.
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Leon Romanovsky <leon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-rdma@vger.kernel.org
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
For smbdirect it required to use different ports depending
on the RDMA protocol. E.g. for iWarp 5445 is needed
(as tcp port 445 already used by the raw tcp transport for SMB),
while InfiniBand, RoCEv1 and RoCEv2 use port 445, as they
use an independent port range (even for RoCEv2, which uses udp
port 4791 itself).
Currently ksmbd is not able to function correctly at
all if the system has iWarp (RDMA_NODE_RNIC) interface(s)
and any InfiniBand, RoCEv1 and/or RoCEv2 interface(s)
at the same time.
And cifs.ko uses 5445 with a fallback to 445, which
means depending on the available interfaces, it tries
5445 in the RoCE range or may tries iWarp with 445
as a fallback. This leads to strange error messages
and strange network captures.
To avoid these problems they will be able to
use rdma_restrict_node_type(RDMA_NODE_RNIC) before
trying port 5445 and rdma_restrict_node_type(RDMA_NODE_IB_CA)
before trying port 445. It means we'll get early
-ENODEV early from rdma_resolve_addr() without any
network traffic and timeouts.
This is designed to be called before calling any
of rdma_bind_addr(), rdma_resolve_addr() or rdma_listen().
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: linux-rdma@vger.kernel.org
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Acked-by: Leon Romanovsky <leon@kernel.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
With smbdirect_send_batch processing we likely have requests without
IB_SEND_SIGNALED, which will be destroyed in the final request
that has IB_SEND_SIGNALED set.
If the connection is broken all requests are signaled
even without explicit IB_SEND_SIGNALED.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The server has similar logic and it makes sure that
request->wr is used instead of a stack struct ib_send_wr send_wr.
This makes sure send_done can see request->wr.send_flags
as the next commit will check for IB_SEND_SIGNALED
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When we are about to use the last send credit that was
granted to us by the peer, we need to wait until
we are ourself able to grant at least one credit
to the peer. Otherwise it might not be possible
for the peer to grant more credits.
The following sections in MS-SMBD are related to this:
3.1.5.1 Sending Upper Layer Messages
...
If Connection.SendCredits is 1 and the CreditsGranted field of the
message is 0, stop processing.
...
3.1.5.9 Managing Credits Prior to Sending
...
If Connection.ReceiveCredits is zero, or if Connection.SendCredits is
one and the Connection.SendQueue is not empty, the sender MUST allocate
and post at least one receive of size Connection.MaxReceiveSize and MUST
increment Connection.ReceiveCredits by the number allocated and posted.
If no receives are posted, the processing MUST return a value of zero to
indicate to the caller that no Send message can be currently performed.
...
This is a similar logic as we have in the server.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
It turns out that our code will corrupt the stream of
reassabled data transfer messages when we trigger an
immendiate (empty) send.
In order to fix this we'll have a single 'batch' credit per
connection. And code getting that credit is free to use
as much messages until remaining_length reaches 0, then
the batch credit it given back and the next logical send can
happen.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This will allow us to use similar logic as we have in
the server soon, so that we can share common code later.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This is basically a copy of smb_direct_{alloc,free}_sendmsg()
in the server, with just using ib_dma_unmap_page() in all
cases, which is the same as ib_dma_unmap_single().
We'll use this logic in common code in future.
(I basically backported it from my branch that
as already has everything in common).
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This is like smb_direct_post_send() in the server
and will simplify porting the smbdirect_send_batch
and credit related logic from the server.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
This simplifies the logic and prepares the use of
smbdirect_send_batch in order to make sure
all messages in a multi fragment send are grouped
together.
We'll add the smbdirect_send_batch processin
in a later patch.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
smbd_post_send_iter()
If we reach this the connection is already broken as
smbd_post_send() already called
smbd_disconnect_rdma_connection().
This will also simplify further changes.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We either reach this code path before we call
new_credits = manage_credits_prior_sending(sc),
which means new_credits is still 0
or the connection is already broken as
smbd_post_send() already called
smbd_disconnect_rdma_connection().
This will also simplify further changes.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
We don't need a stack variable in addition.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
In captures I saw that Windows was granting 191 credits in a batch
when its peer posted a lot of messages. We are asking for a
credit target of 255 and 191 is 252*3/4.
So we also use that logic in order to fill the
recv buffers available to the peer.
Fixes: 02548c477a90 ("smb: client: queue post_recv_credits_work also if the peer raises the credit target")
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The logic off managing recv credits by counting posted recv_io and
granted credits is racy.
That's because the peer might already consumed a credit,
but between receiving the incoming recv at the hardware
and processing the completion in the 'recv_done' functions
we likely have a window where we grant credits, which
don't really exist.
So we better have a decicated counter for the
available credits, which will be incremented
when we posted new recv buffers and drained when
we grant the credits to the peer.
Fixes: 5fb9b459b368 ("smb: client: count the number of posted recv_io messages in order to calculated credits")
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
With smbdirect_send_batch processing we likely have requests without
IB_SEND_SIGNALED, which will be destroyed in the final request
that has IB_SEND_SIGNALED set.
If the connection is broken all requests are signaled
even without explicit IB_SEND_SIGNALED.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
When we are about to use the last send credit that was
granted to us by the peer, we need to wait until
we are ourself able to grant at least one credit
to the peer. Otherwise it might not be possible
for the peer to grant more credits.
The following sections in MS-SMBD are related to this:
3.1.5.1 Sending Upper Layer Messages
...
If Connection.SendCredits is 1 and the CreditsGranted field of the
message is 0, stop processing.
...
3.1.5.9 Managing Credits Prior to Sending
...
If Connection.ReceiveCredits is zero, or if Connection.SendCredits is
one and the Connection.SendQueue is not empty, the sender MUST allocate
and post at least one receive of size Connection.MaxReceiveSize and MUST
increment Connection.ReceiveCredits by the number allocated and posted.
If no receives are posted, the processing MUST return a value of zero to
indicate to the caller that no Send message can be currently performed.
...
This problem was found by running this on Windows 2025
against ksmbd with required smb signing:
'frametest.exe -r 4k -t 20 -n 2000' after
'frametest.exe -w 4k -t 20 -n 2000'.
Link: https://lore.kernel.org/linux-cifs/b58fa352-2386-4145-b42e-9b4b1d484e17@samba.org/
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
It turns out that our code will corrupt the stream of
reassabled data transfer messages when we trigger an
immendiate (empty) send.
In order to fix this we'll have a single 'batch' credit per
connection. And code getting that credit is free to use
as much messages until remaining_length reaches 0, then
the batch credit it given back and the next logical send can
happen.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
In captures I saw that Windows was granting 191 credits in a batch
when its peer posted a lot of messages. We are asking for a
credit target of 255 and 191 is 252*3/4.
So we also use that logic in order to fill the
recv buffers available to the peer.
Fixes: a7eef6144c97 ("smb: server: queue post_recv_credits_work in put_recvmsg() and avoid count_avail_recvmsg")
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The logic off managing recv credits by counting posted recv_io and
granted credits is racy.
That's because the peer might already consumed a credit,
but between receiving the incoming recv at the hardware
and processing the completion in the 'recv_done' functions
we likely have a window where we grant credits, which
don't really exist.
So we better have a decicated counter for the
available credits, which will be incremented
when we posted new recv buffers and drained when
we grant the credits to the peer.
This fixes regression Namjae reported with
the 6.18 release.
Fixes: 89b021a72663 ("smb: server: manage recv credits by counting posted recv_io and granted credits")
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
It turns out that our code will corrupt the stream of
reassabled data transfer messages when we trigger an
immendiate (empty) send.
In order to fix this we'll have a single 'batch' credit per
connection. And code getting that credit is free to use
as much messages until remaining_length reaches 0, then
the batch credit it given back and the next logical send can
happen.
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
The logic off managing recv credits by counting posted recv_io and
granted credits is racy.
That's because the peer might already consumed a credit,
but between receiving the incoming recv at the hardware
and processing the completion in the 'recv_done' functions
we likely have a window where we grant credits, which
don't really exist.
So we better have a decicated counter for the
available credits, which will be incremented
when we posted new recv buffers and drained when
we grant the credits to the peer.
Fixes: 5fb9b459b368 ("smb: client: count the number of posted recv_io messages in order to calculated credits")
Fixes: 89b021a72663 ("smb: server: manage recv credits by counting posted recv_io and granted credits")
Cc: <stable@vger.kernel.org> # 6.18.x
Cc: Steve French <smfrench@gmail.com>
Cc: Tom Talpey <tom@talpey.com>
Cc: Long Li <longli@microsoft.com>
Cc: Namjae Jeon <linkinjeon@kernel.org>
Cc: linux-cifs@vger.kernel.org
Cc: samba-technical@lists.samba.org
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Label the SMB2 status entries with error codes as a prelude to
autogenerating the mapping table. This was done with the following script:
#!/usr/bin/perl -w
use strict;
my @error_mapping_table = ();
my %statuses = ();
#
# Read the status list
#
my $f = "fs/smb/common/smb2status.h";
open FILE, "<$f" || die $f;
my @contents = <FILE>;
close FILE || die;
my $l = 0;
foreach (@contents) {
$l++;
if (m!^#define\s*([A-Za-z0-9_]+)\s+cpu_to_le32[(]([0-9a-fA-Fx]+)[)]!) {
my $status = $1;
my $code = $2;
my $s;
next if ($status =~ /^STATUS_SEVERITY/);
if (exists($statuses{$status})) {
print("$f:$l: Duplicate declaration of ", $s->{status}, "\n");
} else {
my %_s = (
status => $status,
code => hex($code)
);
$statuses{$status} = \%_s;
}
}
}
#
# Read the SMB2 status => error mapping table
#
$f = "fs/smb/client/smb2maperror.c";
open(MFILE, "<$f") || die $f;
my @maperror = <MFILE>;
close MFILE || die;
my $parse = 0;
my $acc = "";
$l = 0;
foreach my $line (@maperror) {
chomp $line;
$l++;
if ($parse == 0 &&
$line =~ /^static const struct status_to_posix_error smb2_error_map_table/) {
$parse = 1;
next;
}
last if ($parse >= 1 && $line =~ /^[}];/);
if ($parse == 1) {
if ($line =~ /[\t][{].*[}],/) {
$acc = $line;
$parse = 3;
} elsif ($line =~ /[\t][{].*/) {
$acc = $line;
$parse = 2;
next;
} elsif ($line =~ m!^\s*/[*].*[*]/!) {
next;
} else {
die "smb2maperror.c:$l: unparseable mapping element\n";
}
}
if ($parse == 2) {
if ($line =~ /.*[}],/) {
$acc .= $line;
$parse = 3;
} else {
$acc .= $line;
}
}
if ($parse == 3) {
$acc =~ s/\s+//g;
if ($acc =~ m/[{](.*)[}]/) {
$acc = $1;
} else {
die "'$acc'";
}
my ($status, $error, $string) = split(/,/, $acc);
if (exists($statuses{$status})) {
my $s = $statuses{$status};
if (exists($s->{error})) {
print("$f:$l: Dup mapping entry $status\n");
}
$s->{error} = $error;
#print $s->{code}, " => ", $error, "\n";
} else {
print STDERR "$f:$l: Unknown status $status\n";
}
$acc = "";
$parse = 1;
}
}
#
# Sort the error table by STATUS_* value
#
my @order = ();
foreach my $status (keys(%statuses)) {
my $s = $statuses{$status};
push @order, $s;
unless (exists($s->{code})) {
print("Unknown code for ", $s->{status}, "\n")
} else {
#print("Code for ", $s->{status}, " is ", $s->{code}, "\n");
}
}
@order = sort( { $a->{code} <=> $b->{code} } @order);
foreach my $s (@order) {
my $status = $s->{status};
my $error = $s->{error};
my $code = $s->{code};
my $pad = " ";
if (length($status) < 32) {
my $n = 32 - length($status);
$pad = "\t" x ((($n-1) / 8) + 1);
}
#printf("#define %s%scpu_to_le32(0x%08x) // -%s\n", $status, $pad, $code, $error);
#printf("\t%s%s= 0x%08x, // -%s\n", $status, $pad, $code, $error);
}
#
# Decorate the definitions
#
my @output = ();
for (my $i = 0; $i <= $#contents; $i++) {
my $line = $contents[$i];
if ($line =~ m!^#define\s*([A-Za-z0-9_]+)\s+cpu_to_le32[(]([0-9a-fA-Fx]+)[)]!) {
my $status = $1;
my $code = $2;
if ($status =~ /^STATUS_SEVERITY/) {
push @output, $line;
next;
}
my $pad = " ";
if (length($status) < 40) {
my $n = 40 - length($status);
$pad = "\t" x ((($n-1) / 8) + 1);
}
my $s = $statuses{$1};
my $error = "EIO";
if (exists($s->{error})) {
$error = $s->{error};
} else {
if (!$s->{code}) {
print "ZERO ", $status, "\n";
$error = "0";
} else {
$error = "-EIO";
}
}
#my $rev = "#define " . $status . $pad . "_S(" . $2 . ", " . $error . ")\n";
my $rev = "#define " . $status . $pad . "cpu_to_le32(" . $2 . ") // " . $error . "\n";
push @output, $rev;
} else {
push @output, $line;
}
}
open FILE, ">fs/smb/common/smb2status.h" || die;
print FILE @output;
close FILE || die;
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <stfrench@microsoft.com>
cc: ChenXiaoSong <chenxiaosong@kylinos.cn>
cc: linux-cifs@vger.kernel.org
Signed-off-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Reviewed-by: David Howells <dhowells@redhat.com>
Link: https://lore.kernel.org/linux-cifs/20260106071507.1420900-2-chenxiaosong.chenxiaosong@linux.dev/
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Make the BCC accessor functions conditional.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1-specific connection management stuff to smb1ops.c and move
CIFSTCon() to cifssmb.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1-specific session setup stuff into smb1session.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1-specific message signing into smb1encrypt.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split a variety of bits from netmisc.c into other places, notably the error
table into smb1maperror.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1 bits from misc.c to smb1misc.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1 bits from cifs_debug.c to smb1debug.c.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Move some SMB1-specific inline funcs to smb1proto.h.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Return void* rather than struct smb_hdr* from from cifs_buf_get() and
cifs_small_buf_get() as SMB2/3 shouldn't be accessing smb_hdr.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Move the BCC access functions to smb1proto.h as they're only applicable to
SMB1.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Adjust the #include set after the removal of the SMB1 protocol defs from
cifspdu.h.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Split SMB1 protocol defs into smb1pdu.h. This should perhaps go in the
common/ directory.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Add some #includes to make sure things continue to compile as splitting
occurs.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Move some SMB1 received checking bits to smb1transport.c from misc.c
so that they're with the rest of the receive handling code.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Move some SMB1 receive bits to smb1transport.c from smb1ops.c where they're
mixed in with unrelated code to do with encoding, decoding and processing
PDUs.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Separate out SMB1 declarations scriptedly into smb1proto.h. Script below:
#!/usr/bin/perl -w
use strict;
unless (@ARGV) {
die "Usage: $0 <c_file1> [<c_file2> ...]\n";
}
# Data tracking
my %funcs = (); # Func name => { func prototype }
my %headers = (); # Header filename => { header content }
my %c_files = (); # C filename => { ordered func list, header pref }
my %cmarkers = (); # C filename marker => { header filename it's in }
# Parse state
my $pathname = "-";
my $lineno = 0;
sub error(@) {
print STDERR $pathname, ":", $lineno, ": ", @_, "\n";
exit(1);
}
sub pad($) {
# Reindent the function arguments to line the arguments up with the char
# after the opening bracket on the func argument list
my ($lines) = @_;
return $lines if ($#{$lines} <= 0);
my $has_empty = 0;
for (my $i = 0; $i <= $#{$lines}; $i++) {
$lines->[$i] =~ s/^[ \t]+//;
$has_empty = 1 if ($lines->[$i] eq "");
}
if ($has_empty) {
my @clean = grep /.+/, @{$lines};
$lines = \@clean;
}
my $indlen = index($lines->[0], "(");
return $lines if ($indlen < 0);
my $indent = "";
$indlen++;
$indent .= "\t" x ($indlen / 8);
$indent .= " " x ($indlen % 8);
my @padded = ();
my $acc = "";
my $len = -$indlen;
for (my $i = 0; $i <= $#{$lines}; $i++) {
my $argument = $lines->[$i];
my $arglen = length($argument);
my $last = ($i == $#{$lines} ? 1 : 0);
if ($i == 0 ||
$i == 1) {
$acc .= $argument;
$acc .= ";" if ($last);
$len += $arglen + $last;
next;
}
if (!$acc) {
$acc = $indent . $argument;
$acc .= ";" if ($last);
$len += $arglen + $last;
next;
}
if ($indlen + $len + 1 + $arglen + $last > 79) {
push @padded, $acc;
$acc = $indent . $argument;
$acc .= ";" if ($last);
$len = $arglen + $last;
next;
}
$acc .= " " . $argument;
$acc .= ";" if ($last);
$len += 1 + $arglen + $last;
}
push @padded, $acc if ($acc);
return \@padded;
}
sub earliest(@) {
my $ret = -1;
foreach (@_) {
$ret = $_ if ($ret < 0 || ($_ >= 0 && $_ < $ret));
}
return $ret;
}
foreach my $file (@ARGV) {
# Open the file for reading.
next if $file =~ /trace[.]h$/;
next if $file =~ /smbdirect[.][ch]$/;
open my $fh, "<$file"
or die "Could not open file '$file'";
$pathname = $file;
$lineno = 0;
my $filename;
my @file_content = ();
my @copy = ();
my $state = 0;
my $qual = "";
my $type = "";
my $funcname = "";
my @funcdef = ();
my $bracket = 0;
my $comment = 0;
my $smb1 = 0;
my $header = 0;
my $inline = 0;
my $file_marker = "";
my $config = "";
my $c_file = 0;
$filename = $pathname;
$filename =~ s!.*/!!;
if ($file =~ m!.h$!) {
my %new_h_file = (
path => $pathname,
fname => $filename,
content => [],
);
$header = \%new_h_file;
$headers{$filename} = \%new_h_file;
} elsif ($file =~ m!.c$!) {
my %new_c_file = (
path => $pathname,
fname => $filename,
funcs => [],
);
$c_file = \%new_c_file;
$c_files{$filename} = \%new_c_file;
} else {
warn("Ignoring unexpected file $file\n");
next;
}
$smb1 = 1 if ($file =~ m!/smb1ops.c|/cifssmb.c|/cifstransport.c!);
foreach my $line (<$fh>) {
$lineno++;
chomp($line);
push @copy, $line;
if (!$line) {
# Blank line
push @file_content, @copy;
@copy = ();
next;
}
# Handle continuation or end of block comment. Look for C file
# prototype insertion point markers.
if ($comment) {
if ($line =~ m![*]/!) {
if ($comment == 2 && $file_marker) {
$cmarkers{$file_marker} = $file_marker;
push @copy, "#C_MARKER " . $file_marker;
$file_marker = 0;
}
$comment = 0;
} else {
$comment++;
if ($comment == 2 && $line =~ m! [*] ([a-z][a-z_0-9]*[.][c])$!) {
$file_marker = $1;
print("Found file marker ", $file_marker, " in ", $filename, "\n");
}
}
push @file_content, @copy;
@copy = ();
next;
}
# Check cpp directives, particularly looking for SMB1 bits
if ($line =~ /^[#]/) {
if ($header) {
if ($line =~ /ifdef.*(CONFIG_[A-Z0-9_])/) {
error("multiconfig") if $config;
$config = $1;
$smb1++ if ($config eq "CONFIG_CIFS_ALLOW_INSECURE_LEGACY");
} elsif ($line =~ /endif/) {
$smb1-- if ($config eq "CONFIG_CIFS_ALLOW_INSECURE_LEGACY");
$config = "";
}
}
push @file_content, @copy;
@copy = ();
next;
}
# Exclude interference in finding func names and return types
if ($line =~ /^[{]/ ||
$line =~ /##/ ||
$line =~ /^[_a-z0-9A-Z]+:$/ || # goto label
$line =~ /^do [{]/ ||
$line =~ m!^//!) {
push @file_content, @copy;
@copy = ();
next;
}
# Start of a block comment
if ($line =~ m!^/[*]!) {
$comment = 1 unless ($line =~ m![*]/!);
push @file_content, @copy;
@copy = ();
next;
}
# End of a braced section, such as a function implementation
if ($line =~ /^[}]/) {
$type = "";
$qual = "";
$funcname = "";
@funcdef = ();
push @file_content, @copy;
@copy = ();
next;
}
if ($line =~ /^typedef/) {
$type = "";
$qual = "";
$funcname = "";
@funcdef = ();
push @file_content, @copy;
@copy = ();
next;
}
# Extract function qualifiers. There may be multiple of these in more
# or less any order. Some of them cause the func to be skipped (e.g. inline).
if ($line =~ /^(static|extern|inline|noinline|noinline_for_stack|__always_inline)\W/ ||
$line =~ /^(static|extern|inline|noinline|noinline_for_stack|__always_inline)$/) {
error("Unexpected qualifier '$1'") if ($state != 0);
while ($line =~ /^(static|extern|inline|noinline|noinline_for_stack|__always_inline)\W/ ||
$line =~ /^(static|extern|inline|noinline|noinline_for_stack|__always_inline)$/) {
$qual .= " " if ($qual);
$qual .= $1;
$inline = 1 if ($1 eq "inline");
$inline = 1 if ($1 eq "__always_inline");
$line = substr($line, length($1));
$line =~ s/^\s+//;
}
}
if ($state == 0) {
# Extract what we assume to be the return type
if ($line =~ /^\s/) {
push @file_content, @copy;
@copy = ();
next;
}
while ($line =~ /^(unsigned|signed|bool|char|short|int|long|void|const|volatile|(struct|union|enum)\s+[_a-zA-Z][_a-zA-Z0-9]*|[*]|__init|__exit|__le16|__le32|__le64|__be16|__be32|__be64)/) {
$type .= " " if $type;
$type .= $1;
$line = substr($line, length($1));
$line =~ s/^\s+//;
}
if ($line =~ /^struct [{]/) {
# Ignore structure definitions
$type = "";
$qual = "";
$funcname = "";
@funcdef = ();
push @file_content, @copy;
@copy = ();
next;
}
if (index($line, "=") >= 0) {
# Ignore assignments
$type = "";
$qual = "";
$funcname = "";
@funcdef = "";
push @file_content, @copy;
@copy = ();
next;
}
# Try and extract a function's type and name
while ($line =~ /(^[_a-zA-Z][_a-zA-Z0-9]*)/) {
my $name = $1;
$line = substr($line, length($name));
next if ($line =~ /^[{]/);
$line =~ s/^\s+//;
my $ch = substr($line, 0, 1);
last if ($ch eq "[" || $ch eq ";"); # Global variables
if ($ch eq "(") {
# Found the function name
$state = 1;
$line = substr($line, 1);
$funcname = $name;
my $tmp = $qual . $type . " " . $funcname . "(";
$tmp =~ s/[*] /*/;
push @funcdef, $tmp;
$bracket = 1;
last;
}
if ($type) {
last if (index($line, ";") >= 0 && index($line, "(") == -1);
error("Unexpected name '$name' after '$type'");
}
$type .= " " if $type;
$type .= $name;
if ($line =~ /^(\s*[*]+)/) {
my $ptr = $1;
$type .= $ptr;
$line = substr($line, length($ptr));
}
}
}
# Try and extract a function's argument list
my $from = 0;
if ($state == 1) {
while (1) {
my $o = index($line, "(", $from);
my $c = index($line, ")", $from);
my $m = index($line, ",", $from);
my $b = earliest($o, $c, $m);
if ($b < 0) {
push @funcdef, $line
unless ($line eq "");
last;
}
my $ch = substr($line, $b, 1);
# Push the arguments separately on to the list
if ($ch eq ",") {
push @funcdef, substr($line, 0, $b + 1);
$line = substr($line, $b + 1);
$from = 0;
} elsif ($ch eq "(") {
# Handle brackets in the argument list (e.g. function
# pointers)
$bracket++;
$from = $b + 1;
} elsif ($ch eq ")") {
$bracket--;
if ($bracket == 0) {
push @funcdef, substr($line, 0, $b + 1);
$line = substr($line, $b + 1);
$state = 2;
last;
}
$from = $b + 1;
}
}
}
if ($state == 2) {
$inline = 1 if ($qual =~ /inline/);
#print("QUAL $qual $type $funcname $inline ", $#funcdef, "\n");
if (!$header &&
$qual !~ /static/ &&
$funcname ne "__acquires" &&
$funcname ne "__releases" &&
$funcname ne "module_init" &&
$funcname ne "module_exit" &&
$funcname ne "module_param" &&
$funcname ne "module_param_call" &&
$funcname ne "PROC_FILE_DEFINE" &&
$funcname !~ /MODULE_/ &&
$funcname !~ /DEFINE_/) {
# Okay, we appear to have a function implementation
my $func;
my $dup = 0;
if (exists($funcs{$funcname})) {
$func = $funcs{$funcname};
if (exists $func->{body}) {
print("dup $funcname\n");
$dup = 1;
}
} else {
my %new_func = (
name => $funcname,
cond => "",
legacy => 0,
);
$func = \%new_func;
$funcs{$funcname} = $func;
$func->{body} = pad(\@funcdef);
}
$func->{body} = pad(\@funcdef);
$func->{legacy} = 1 if $smb1;
if ($funcname eq "cifs_inval_name_dfs_link_error") {
$func->{cond} = "#ifdef CONFIG_CIFS_DFS_UPCALL";
} elsif ($funcname eq "cifs_listxattr") {
$func->{cond} = "#ifdef CONFIG_CIFS_XATTR";
}
push @{$c_file->{funcs}}, $func
unless $dup;
} elsif (!$header || $inline) {
# Ignore inline function implementations and other weirdies
push @file_content, @copy;
} elsif ($header && !$inline) {
push @file_content, "#FUNCPROTO " . $funcname;
my $func;
if (exists($funcs{$funcname})) {
$func = $funcs{$funcname};
$func->{lineno} = $lineno;
$func->{pathname} = $pathname;
} else {
my %new_func = (
name => $funcname,
cond => "",
lineno => $lineno,
pathname => $pathname,
legacy => 0,
);
$func = \%new_func;
$funcs{$funcname} = $func;
}
$func->{legacy} = 1 if $smb1;
}
@funcdef = ();
$type = "";
$qual = "";
$funcname = "";
$inline = 0;
$state = 0;
@copy = ();
}
if ($line =~ /;/) {
$type = "";
$qual = "";
$funcname = "";
@funcdef = ();
$state = 0;
push @file_content, @copy;
@copy = ();
}
}
close($fh);
if ($header) {
$header->{content} = \@file_content;
}
}
sub write_header($)
{
my ($header) = @_;
my $path = $header->{path};
my $legacy = 0;
$legacy = 1 if ($path =~ m!smb1proto[.]h!);
my @output = ();
foreach my $line (@{$header->{content}}) {
if ($line =~ "^[#]C_MARKER (.*)") {
my $file_marker = $cmarkers{$1};
my $c_file = $c_files{$file_marker};
print("Found $line\n");
foreach my $func (@{$c_file->{funcs}}) {
print("func ", $func->{name}, "\n");
push @output, @{$func->{body}};
}
next;
} elsif ($line =~ "^[#]FUNCPROTO ([_a-zA-Z0-9]+)") {
my $funcname = $1;
my $func = $funcs{$funcname};
if (!$func->{body}) {
print($func->{pathname}, ":", $func->{lineno}, ": '", $funcname,
"' dead prototype\n");
next;
}
if ($func->{legacy} == $legacy) {
#push @output, $line;
push @output, @{$func->{body}};
}
} else {
push @output, $line;
}
}
open my $fh, ">$path"
or die "Could not open file '$path' for writing";
foreach my $f (@output) {
print($fh $f, "\n") or die $path;
}
close($fh) or die $path;
}
foreach my $h (keys(%headers)) {
write_header($headers{$h});
}
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
|
Rename cifstransport.c to smb1transport.c in order to give consistent names
SMB1-specific files.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: Enzo Matsumiya <ematsumiya@suse.de>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-kernel@vger.kernel.org
Acked-by: Enzo Matsumiya <ematsumiya@suse.de>
Signed-off-by: Steve French <stfrench@microsoft.com>
|
