If we're doing uncached readdir, allocate multiple pages in order to
try to avoid duplicate RPC calls for the same getdents() call.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
If the server is handing out monotonically increasing readdir cookie values,
then we can optimise away searches through pages that contain cookies that
lie outside our search range.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
If the server insists on using the readdir verifiers in order to allow
cookies to expire, then we should ensure that we cache the verifier
with the cookie, so that we can return an error if the application
tries to use the expired cookie.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
If the server returns NFS4ERR_NOT_SAME or tells us that the cookie is
bad in response to a READDIR call, then we should empty the page cache
so that we can fill it from scratch again.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
If we're ever going to allow support for servers that use the readdir
verifier, then that use needs to be managed by the middle layers as
those need to be able to reject cookies from other verifiers.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
The descriptor and the struct nfs_entry are both large structures,
so don't allocate them from the stack.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Clean up nfs_do_filldir().
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Remove the redundant caching of the credential in struct
nfs_open_dir_context.
Pass the buffer size as an argument to nfs_readdir_xdr_filler().
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Support readdir buffers of up to 1MB in size so that we can read
large directories using few RPC calls.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
We don't need to store a hash, so replace struct qstr with a simple
const char pointer and length.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
The kmapped pointer is only used once per loop to check if we need to
exit.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
If a readdir call returns more data than we can fit into one page
cache page, then allocate a new one for that data rather than
discarding the data.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Refactor to use pagecache_get_page() so that we can fill the page
in multiple stages.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Clean up handling of the case where there are no entries in the readdir
reply.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Since the 'eof_index' is only ever used as a flag, make it so.
Also add a flag to detect if the page has been completely filled.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Ensure that the contents of struct nfs_open_dir_context are consistent
by setting them under the file->f_lock from a private copy (that is
known to be consistent).
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Reviewed-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Tested-by: Dave Wysochanski <dwysocha@redhat.com>
Currently, the client will always ask for security_labels if the server
returns that it supports that feature regardless of any LSM modules
(such as Selinux) enforcing security policy. This adds performance
penalty to the READDIR operation.
Client adjusts superblock's support of the security_label based on
the server's support but also current client's configuration of the
LSM modules. Thus, prior to using the default bitmask in READDIR,
this patch checks the server's capabilities and then instructs
READDIR to remove FATTR4_WORD2_SECURITY_LABEL from the bitmask.
v5: fixing silly mistakes of the rushed v4
v4: simplifying logic
v3: changing label's initialization per Ondrej's comment
v2: dropping selinux hook and using the sb cap.
Suggested-by: Ondrej Mosnacek <omosnace@redhat.com>
Suggested-by: Scott Mayhew <smayhew@redhat.com>
Signed-off-by: Olga Kornievskaia <kolga@netapp.com>
Fixes: 2b0143b5c9 ("VFS: normal filesystems (and lustre): d_inode() annotations")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Currently, we wake up the tasks by priority queue ordering, which means
that we ignore the batching that is supposed to help with QoS issues.
Fixes: c049f8ea9a ("SUNRPC: Remove the bh-safe lock requirement on the rpc_wait_queue->lock")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
We need to respect the NFS_MOUNT_SOFTREVAL flag in _nfs4_proc_lookupp,
by timing out if the server is unavailable.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
In order to use the open_by_filehandle() operations on NFSv3, we need
to be able to emulate lookupp() so that nfs_get_parent() can be used
to convert disconnected dentries into connected ones.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
We want to reuse the lookup code in NFSv3 in order to emulate the
NFSv4 lookupp operation.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
The NO_160 flag specifies if the device doesn't have 160 MHz support,
but we errorneously assumed the opposite. If the flag was set, we
were considering that 160 MHz was supported, but it's actually the
opposite. Fix it by inverting the bits, i.e. NO_160 is 0x1 and 160
is 0x0.
Fixes: d6f2134a38 ("iwlwifi: add mac/rf types and 160MHz to the device tables")
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/iwlwifi.20201202143859.375bec857ccb.I83884286b688965293e9810381808039bd7eedae@changeid
Since commit b4868b44c5 ("NFSv4: Wait for stateid updates after
CLOSE/OPEN_DOWNGRADE"), every inter server copy operation suffers 5
seconds delay regardless of the size of the copy. The delay is from
nfs_set_open_stateid_locked when the check by nfs_stateid_is_sequential
fails because the seqid in both nfs4_state and nfs4_stateid are 0.
Fix __nfs42_ssc_open to delay setting of NFS_OPEN_STATE in nfs4_state,
until after the call to update_open_stateid, to indicate this is the 1st
open. This fix is part of a 2 patches, the other patch is the fix in the
source server to return the stateid for COPY_NOTIFY request with seqid 1
instead of 0.
Fixes: ce0887ac96 ("NFSD add nfs4 inter ssc to nfsd4_copy")
Signed-off-by: Dai Ngo <dai.ngo@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
This commit updates the BMPS exit path to be consistent with downstream in
terms of exiting BMPS mode. Downstream sets the flag to send a NULL data
frame to the host on exiting BMPS.
This will tell the AP to send any queued frames to the STA immediately.
Verified the relevant bit toggle in wireshark.
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201120021403.2646574-2-bryan.odonoghue@linaro.org
qmi_msg_handler[] and ath10k_qmi_ops are only used as input arguments
to qmi_handle_init() which accepts const pointers to both qmi_ops and
qmi_msg_handler. Make them const to allow the compiler to put them in
read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122234031.33432-3-rikard.falkeborn@gmail.com
Should an error occur after calling 'ath10k_usb_create()', it should be
undone by a corresponding 'ath10k_usb_destroy()' call
Fixes: 4db66499df ("ath10k: add initial USB support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122170358.1346065-1-christophe.jaillet@wanadoo.fr
If 'ath10k_usb_create()' fails, we should release some resources and report
an error instead of silently continuing.
Fixes: 4db66499df ("ath10k: add initial USB support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122170342.1346011-1-christophe.jaillet@wanadoo.fr
The wmi service available event has been
extended to contain extra 128 bit for new services
to be indicated by firmware.
Currently the presence of any optional TLVs in
the wmi service available event leads to a parsing
error with the below error message:
ath10k_snoc 18800000.wifi: failed to parse svc_avail tlv: -71
The wmi service available event parsing should
not return error for the newly added optional TLV.
Fix this parsing for service available event message.
Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.3.2.2-00720-QCAHLSWMTPL-1
Fixes: cea19a6ce8 ("ath10k: add WMI_SERVICE_AVAILABLE_EVENT support")
Signed-off-by: Rakesh Pillai <pillair@codeaurora.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/1605501291-23040-1-git-send-email-pillair@codeaurora.org
If 'kzalloc' fails, we must return an error code.
While at it, remove a useless initialization of 'err' which could hide the
issue.
Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201122173943.1366167-1-christophe.jaillet@wanadoo.fr
By switching to an XFS-backed export, I am able to reproduce the
ibcomp worker crash on my client with xfstests generic/013.
For the failing LISTXATTRS operation, xdr_inline_pages() is called
with page_len=12 and buflen=128.
- When ->send_request() is called, rpcrdma_marshal_req() does not
set up a Reply chunk because buflen is smaller than the inline
threshold. Thus rpcrdma_convert_iovs() does not get invoked at
all and the transport's XDRBUF_SPARSE_PAGES logic is not invoked
on the receive buffer.
- During reply processing, rpcrdma_inline_fixup() tries to copy
received data into rq_rcv_buf->pages because page_len is positive.
But there are no receive pages because rpcrdma_marshal_req() never
allocated them.
The result is that the ibcomp worker faults and dies. Sometimes that
causes a visible crash, and sometimes it results in a transport hang
without other symptoms.
RPC/RDMA's XDRBUF_SPARSE_PAGES support is not entirely correct, and
should eventually be fixed or replaced. However, my preference is
that upper-layer operations should explicitly allocate their receive
buffers (using GFP_KERNEL) when possible, rather than relying on
XDRBUF_SPARSE_PAGES.
Reported-by: Olga kornievskaia <kolga@netapp.com>
Suggested-by: Olga kornievskaia <kolga@netapp.com>
Fixes: c10a75145f ("NFSv4.2: add the extended attribute proc functions.")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Olga kornievskaia <kolga@netapp.com>
Reviewed-by: Frank van der Linden <fllinden@amazon.com>
Tested-by: Olga kornievskaia <kolga@netapp.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
It is rather easy to add more entries to ath11k_skb_cb while forgetting the
size limit of ieee80211_tx_info->driver_data. So just check this during the
build to reduce the change of accidental buffer overflow in the skbuff->cb.
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201119154235.263250-3-sven@narfation.org
It was observed that the codepath for the ATH11K_SKB_HW_80211_ENCAP was
used even when the IEEE80211_TX_CTRL_HW_80211_ENCAP was not enabled for a
an skbuff. This became even more prominent when the QCAs wlan-open patchset
for ath11k [1] was applied and a sane looking fix just caused crashes when
injecting frames via a monitor interface (for example with ratechecker):
[ 86.963152] Unable to handle kernel NULL pointer dereference at virtual address 00000338
[ 86.963192] pgd = ffffffc0008f0000
[ 86.971034] [00000338] *pgd=0000000051706003, *pud=0000000051706003, *pmd=0000000051707003, *pte=00e800000b000707
[ 86.984292] Internal error: Oops: 96000006 [#1] PREEMPT SMP
[...]
[ 87.713339] [<ffffffbffc802480>] ieee80211_tx_status_8023+0xf8/0x220 [mac80211]
[ 87.715654] [<ffffffbffc98bad4>] ath11k_dp_tx_completion_handler+0x42c/0xa10 [ath11k]
[ 87.722924] [<ffffffbffc989190>] ath11k_dp_service_srng+0x70/0x3c8 [ath11k]
[ 87.730831] [<ffffffbffca03460>] 0xffffffbffca03460
[ 87.737599] [<ffffffc00046ef58>] net_rx_action+0xf8/0x288
[ 87.742462] [<ffffffc000097554>] __do_softirq+0xfc/0x220
[ 87.748014] [<ffffffc000097900>] irq_exit+0x98/0xe8
[ 87.753396] [<ffffffc0000cf188>] __handle_domain_irq+0x90/0xb8
[ 87.757999] [<ffffffc000081ca4>] gic_handle_irq+0x6c/0xc8
[ 87.763899] Exception stack(0xffffffc00081bdc0 to 0xffffffc00081bef0)
Problem is that the state of ath11k_skb_cb->flags must be considered
unknown and could contain anything when it is not manually initialized. So
it could also contain ATH11K_SKB_HW_80211_ENCAP. And this can result in the
code to assume that the ath11k_skb_cb->vif is set - even when this is not
always the case for non ATH11K_SKB_HW_80211_ENCAP transmissions.
Tested-on: IPQ8074 hw2.0 WLAN.HK.2.4.0.1.r1-00026-QCAHKSWPL_SILICONZ-2
[1] https://source.codeaurora.org/quic/qsdk/oss/system/feeds/wlan-open/tree/mac80211/patches?h=NHSS.QSDK.11.4.r3
(162 patches at the moment which are often not upstreamed but essential
to get ath11k working)
Fixes: e7f33e0c52 ("ath11k: add tx hw 802.11 encapsulation offloading support")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201119154235.263250-2-sven@narfation.org
The driver_data area of ieee80211_tx_info is used in ath11k for
ath11k_skb_cb. The first function in the TX patch which rewrites it to
ath11k_skb_cb is already ath11k_mac_op_tx. No one else in the code path
must use it for something else before it reinitializes it. Otherwise the
data has to be considered uninitialized or corrupt.
But the ieee80211_tx_info.control shares exactly the same area as
ieee80211_tx_info.driver_data and ath11k is still using it. This results in
best case in a
ath11k c000000.wifi1: no vif found for mgmt frame, flags 0x0
or (slightly worse) in a kernel oops.
Instead, the interesting data must be moved first into the ath11k_skb_cb
and ieee80211_tx_info.control must then not be used anymore.
Tested-on: IPQ8074 hw2.0 WLAN.HK.2.4.0.1.r1-00026-QCAHKSWPL_SILICONZ-2
Fixes: d5c65159f2 ("ath11k: driver for Qualcomm IEEE 802.11ax devices")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Kalle Valo <kvalo@codeaurora.org>
Link: https://lore.kernel.org/r/20201119154235.263250-1-sven@narfation.org
In preparation to enable -Wimplicit-fallthrough for Clang, fix multiple
warnings by explicitly adding a couple break statements and replacing /*
fall through */ comments with the new pseudo-keyword macro fallthrough;
instead of just letting the code fall through to the next case.
Notice that Clang doesn't recognize /* fall through */ comments as implicit
fall-through markings.
Link: https://github.com/KSPP/linux/issues/115
Link: https://lore.kernel.org/r/2ae1cafd858238b85fc5e7fe5cc183843e21ec9f.1605896059.git.gustavoars@kernel.org
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In preparation to enable -Wimplicit-fallthrough for Clang, fix multiple
warnings by explicitly adding multiple break statements instead of just
letting the code fall through to the next case, and by adding fallthrough
statements in places where the code is intended to fall through, and
finally by replacing /* FALLTHROUGH */ comments with the new pseudo-keyword
macro fallthrough.
Link: https://github.com/KSPP/linux/issues/115
Link: https://lore.kernel.org/r/1a7cd2f77623e6ab46bbec0b6103b18491419206.1605896059.git.gustavoars@kernel.org
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
