SG_CHUNK_SIZE is 128, making struct svc_rdma_rw_ctxt + the first
SGL array more than 4200 bytes in length, pushing the memory
allocation well into order 1.
Even so, the RDMA rw core doesn't seem to use more than max_send_sge
entries in that array (typically 32 or less), so that is all wasted
space.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
A send/recv_ctxt already records transport-related information
in the cq.id, thus there is no need to record the IP addresses of
the transport endpoints.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Update the DMA error flow tracepoints to report the completion ID of
the failing context. This ties the wait/failure to a particular
operation or request, which is more useful than knowing only the
failing transport.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Update the Send Queue's error flow tracepoints to report the
completion ID of the waiting or failing context. This ties the
wait/failure to a particular operation or request, which is a little
more useful than knowing only the transport that is about to close.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Two svcrdma-related transport locks can become quite contended.
Collate their use and make them easy to find in /proc/lock_stat for
better observability.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
DMA unmapping can take quite some time, so it should not be handled
in a single-threaded completion handler. Defer releasing write_info
structs to the recently-added workqueue.
With this patch, DMA unmapping can be handled in parallel, and it
does not cause head-of-queue blocking of Write completions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
DMA unmapping can take quite some time, so it should not be handled
in a single-threaded completion handler. Defer releasing send_ctxts
to the recently-added workqueue.
With this patch, DMA unmapping can be handled in parallel, and it
does not cause head-of-queue blocking of Send completions.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
To handle work in the background, set up an UNBOUND workqueue for
svcrdma. Subsequent patches will make use of it.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The original reason for allocating svc_rdma_recv_ctxt objects during
Receive completion was to ensure the objects were allocated on the
NUMA node closest to the underlying IB device.
Since commit c5d68d25bd ("svcrdma: Clean up allocation of
svc_rdma_recv_ctxt"), however, the device's favored node is
explicitly passed to the memory allocator.
To enable switching Receive completion to soft IRQ context, move
memory allocation out of completion handling, since it can be
costly, and it can sleep.
A limited number of objects is now allocated at "accept" time.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The svc_rdma_recv_ctxt free list uses a lockless list to avoid the
need for a spin lock in the fast path. llist_del_first(), which is
used by svc_rdma_recv_ctxt_get(), requires serialization, however,
when there are multiple list producers that are unserialized.
I mistakenly thought there was only one caller of
svc_rdma_recv_ctxt_get() (svc_rdma_refresh_recvs()), thus explicit
serialization would not be necessary. But there is another caller:
svc_rdma_bc_sendto(), and these two are not serialized against each
other. I haven't seen ill effects that I could directly ascribe to
a lack of serialization. It's just an observation based on code
audit.
When DMA-mapping before sending a Reply, the passed-in struct
svc_rdma_recv_ctxt is used only for its write and reply PCLs. These
are currently always empty in the backchannel case. So, instead of
passing a full svc_rdma_recv_ctxt object to
svc_rdma_map_reply_msg(), let's pass in just the Write and Reply
PCLs.
This change makes it unnecessary for the backchannel to acquire a
dummy svc_rdma_recv_ctxt object when sending an RPC Call. The need
for svc_rdma_recv_ctxt free list serialization is now completely
avoided.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Callback operations enum is defined in client and server, move it to
common header file.
Signed-off-by: ChenXiaoSong <chenxiaosong@kylinos.cn>
Acked-by: Anna Schumaker <Anna.Schumaker@netapp.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
We check "state" for NULL on the previous line so it can't be NULL here.
No need to check again.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/r/202312031425.LffZTarR-lkp@intel.com/
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Avoid the use of an atomic bitop, and prepare for adding a run-time
switch for using splice reads.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
RQ_SPLICE_OK is a bit of a layering violation. Also, a subsequent
patch is going to provide a mechanism for always disabling splice
reads.
Splicing is an issue only for NFS READs, so refactor nfsd_read() to
check the auth type directly instead of relying on an rq_flag
setting.
The new helper will be added into the NFSv4 read path in a
subsequent patch.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
NFSD will use this new API to determine whether nfsd_splice_read is
safe to use. This avoids the need to add a dependency to NFSD for
CONFIG_SUNRPC_GSS.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Al Viro notes that normal system calls hold f_pos_lock when calling
->iterate_shared and ->llseek; however nfsd_readdir() does not take
that mutex when calling these methods.
It should be safe however because the struct file acquired by
nfsd_readdir() is not visible to other threads.
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
This trace point was for debugging the DRC's garbage collection. In
the field it's just noise.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
workqueue: nfsd_file_delayed_close [nfsd] hogged CPU for >13333us 8
times, consider switching to WQ_UNBOUND
There's no harm in closing a cached file descriptor on another core.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The usage of read_seqbegin_or_lock() in nfsd_copy_write_verifier()
is wrong. "seq" is always even and thus "or_lock" has no effect,
this code can never take ->writeverf_lock for writing.
I guess this is fine, nfsd_copy_write_verifier() just copies 8 bytes
and nfsd_reset_write_verifier() is supposed to be very rare operation
so we do not need the adaptive locking in this case.
Yet the code looks wrong and sub-optimal, it can use read_seqbegin()
without changing the behaviour.
[ cel: Note also that it eliminates this Sparse warning:
fs/nfsd/nfssvc.c:360:6: warning: context imbalance in 'nfsd_copy_write_verifier' -
different lock contexts for basic block
]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: NeilBrown <neilb@suse.de>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
We've had a number of attempts at different NFSv4 client tracking
methods over the years, but now nfsdcld has emerged as the clear winner
since the others (recoverydir and the usermodehelper upcall) are
problematic.
As a case in point, the recoverydir backend uses MD5 hashes to encode
long form clientid strings, which means that nfsd repeatedly gets dinged
on FIPS audits, since MD5 isn't considered secure. Its use of MD5 is not
cryptographically significant, so there is no danger there, but allowing
us to compile that out allows us to sidestep the issue entirely.
As a prelude to eventually removing support for these client tracking
methods, add a new Kconfig option that enables them. Mark it deprecated
and make it default to N.
Acked-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
ida_alloc() and ida_free() should be preferred to the deprecated
ida_simple_get() and ida_simple_remove().
This is less verbose.
Note that the upper bound of ida_alloc_range() is inclusive while the one
of ida_simple_get() was exclusive. So calls have been updated accordingly.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
s/diffentiate/differentiate/
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Query dir responses don't provide enough information on reparse points
such as major/minor numbers and symlink targets other than reparse
tags, however we don't need to unconditionally revalidate them only
because they are reparse points. Instead, revalidate them only when
their ctime or reparse tag has changed.
For instance, Windows Server updates ctime of reparse points when
their data have changed.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Change SMB2_set_eof() to take eof as CPU order rather than __le64 and pass
it directly rather than by pointer. This moves the conversion down into
SMB_set_eof() rather than all of its callers and means we don't need to
undo it for the traceline.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Jeff Layton <jlayton@kernel.org>
cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
The kfree() function was called in one case by
the allocate_mr_list() function during error handling
even if the passed variable contained a null pointer.
This issue was detected by using the Coccinelle software.
Thus use another label.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Steve French <stfrench@microsoft.com>
Recently, cifs_chan_update_iface was modified to not
remove an iface if a suitable replacement was not found.
With that, there were two conditionals that were exactly
the same. This change removes that extra condition check.
Also, fixed a logging in the same function to indicate
the correct message.
Signed-off-by: Shyam Prasad N <sprasad@microsoft.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Parse reparse points in SMB3 posix query info as they will be
supported and required by the new specification.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Use smb2_compound_op() with SMB2_OP_GET_REPARSE to get reparse point.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Add support for creating symlinks via IO_REPARSE_TAG_SYMLINK reparse
points in SMB2+.
These are fully supported by most SMB servers and documented in
MS-FSCC. Also have the advantage of requiring fewer roundtrips as
their symlink targets can be parsed directly from CREATE responses on
STATUS_STOPPED_ON_SYMLINK errors.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202311260838.nx5mkj1j-lkp@intel.com/
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
The client was sending an SMB2_CREATE request without setting
OPEN_REPARSE_POINT flag thus failing the entire hardlink operation.
Fix this by setting OPEN_REPARSE_POINT in create options for
SMB2_CREATE request when the source inode is a repase point.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
The client was sending an SMB2_CREATE request without setting
OPEN_REPARSE_POINT flag thus failing the entire rename operation.
Fix this by setting OPEN_REPARSE_POINT in create options for
SMB2_CREATE request when the source inode is a repase point.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Reduce number of roundtrips to server when querying reparse points in
->query_path_info() by sending a single compound request of
create+get_reparse+get_info+close.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Add support for creating special files (e.g. char/block devices,
sockets, fifos) via NFS reparse points on SMB2+, which are fully
supported by most SMB servers and documented in MS-FSCC.
smb2_get_reparse_inode() creates the file with a corresponding reparse
point buffer set in @iov through a single roundtrip to the server.
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202311260746.HOJ039BV-lkp@intel.com/
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Make smb2_compound_op() accept up to MAX_COMPOUND(5) commands to be
sent over a single compounded request.
This will allow next commits to read and write reparse files through a
single roundtrip to the server.
Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Fixes no-op checkpatch errors and warnings.
Signed-off-by: Pierre Mariani <pierre.mariani@gmail.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Commit ba24ea1291 ("net/sched: Retire ipt action") removed the ipt action
but not the testcases. This patch removes the outstanding tdc tests.
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Swee Leong Ching says:
====================
net: stmmac: Enable Per DMA Channel interrupt
Add Per DMA Channel interrupt feature for DWXGMAC IP.
Patchset (link below) contains per DMA channel interrupt, But it was
achieved.
https://lore.kernel.org/lkml/20230821203328.GA2197059-
robh@kernel.org/t/#m849b529a642e1bff89c05a07efc25d6a94c8bfb4
Some of the changes in this patchset are based on reviewer comment on
patchset mentioned beforehand.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Enable per DMA channel interrupt that uses shared peripheral
interrupt (SPI), so only per channel TX and RX intr (TI/RI)
are handled by TX/RX ISR without calling common interrupt ISR.
Signed-off-by: Teoh Ji Sheng <ji.sheng.teoh@intel.com>
Signed-off-by: Swee Leong Ching <leong.ching.swee@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no support for per DMA channel interrupt for non-MSI platform,
where the MAC's per channel interrupt hooks up to interrupt controller(GIC)
through shared peripheral interrupt(SPI) to handle interrupt from TX/RX
transmit channel.
This patch generalize the existing MSI ISR to also support non-MSI
platform.
Signed-off-by: Teoh Ji Sheng <ji.sheng.teoh@intel.com>
Signed-off-by: Swee Leong Ching <leong.ching.swee@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add dt-bindings for per channel irq.
Signed-off-by: Rohan G Thomas <rohan.g.thomas@intel.com>
Signed-off-by: Swee Leong Ching <leong.ching.swee@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Marangi says:
====================
net: phy: at803x: even more generalization
This is part 3 of at803x required patches to split the PHY driver
in more specific PHY Family driver.
While adding support for a new PHY Family qca807x it was notice lots
of similarities with the qca808x cdt function. Hence this series
is done to make things easier in the future when qca807x PHY will be
submitted.
Changes v4:
- Fix Smatch warning
Changes v3:
- Rebase on top of net-next
Changes v2:
- Address request from Russell in a previous series on cdt get
status improvement
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Make read_status more generic in preparation on moving it to shared
library as other PHY Family Driver will have the exact same
implementation.
The only specific part was a check for AR8031/33 if 1000basex was used.
The check is moved to a dedicated function specific for those PHYs.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
QCA808x PHY Family supports Cable Diagnostic Test also for Cross Pair
Short.
Add all the define to make enable and support these additional tests.
Cross Short test was previously disabled by default, this is now changed
and enabled by default. In this mode, the mask changed a bit and length
is shifted based on the fault condition.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor qca808x cable test get status function to remove code
duplication and clean things up.
The same logic is applied to each pair hence it can be generalized and
moved to a common function.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>