Merge series from Xichao Zhao <zhao.xichao@vivo.com>:
The dev_err_probe() doesn't do anything when error is '-ENOMEM'. Therefore,
remove the useless call to dev_err_probe(), and just return the value instead.
Merge series from srinivas.kandagatla@oss.qualcomm.com:
Sorry to resend this series once again, as some of the patches seems
to be dropped/rejected by email client from previous send.
This patchset:
- cleans up some of the audioreach tokens which are unused
- adds missing documentation
- add support for static calibration support which is required for ECNS
an speaker protection support.
Tested this with Single Mic ECNS on SM8450 platform.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQQqUNBr3gm4hGXdBJlZ7Krx/gZQ6wUCaKShewAKCRBZ7Krx/gZQ
6+p0AQD0vMecDWcRALFSInfGofuozEZ+98i3G0stlzU8KrvvpQD/a2jEfxOZOBph
0RL73MLt+eSl03tnsbqGRoHrcH9epQM=
=3DGl
-----END PGP SIGNATURE-----
Merge tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull mount fixes from Al Viro:
"Fixes for several recent mount-related regressions"
* tag 'pull-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
change_mnt_propagation(): calculate propagation source only if we'll need it
use uniform permission checks for all mount propagation changes
propagate_umount(): only surviving overmounts should be reparented
fix the softlockups in attach_recursive_mnt()
-----BEGIN PGP SIGNATURE-----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=x8IS
-----END PGP SIGNATURE-----
Merge tag 'ovl-fixes-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/overlayfs/vfs
Pull overlayfs fixes from Amir Goldstein:
"Fixes for two fallouts from Neil's directory locking changes"
* tag 'ovl-fixes-6.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/overlayfs/vfs:
ovl: fix possible double unlink
ovl: use I_MUTEX_PARENT when locking parent in ovl_create_temp()
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCaKRttQAKCRCRxhvAZXjc
onwmAP98oaMku7CttHEVwJj8KD7luXvZWbvB23TPGmF6BNWg9wEAraks5EzZZJy3
+4xWn10b6R+gXUqvwqr+bf0ufk3c+gc=
=Nbg0
-----END PGP SIGNATURE-----
Merge tag 'vfs-6.17-rc3.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs
Pull vfs fixes from Christian Brauner:
- Fix two memory leaks in pidfs
- Prevent changing the idmapping of an already idmapped mount without
OPEN_TREE_CLONE through open_tree_attr()
- Don't fail listing extended attributes in kernfs when no extended
attributes are set
- Fix the return value in coredump_parse()
- Fix the error handling for unbuffered writes in netfs
- Fix broken data integrity guarantees for O_SYNC writes via iomap
- Fix UAF in __mark_inode_dirty()
- Keep inode->i_blkbits constant in fuse
- Fix coredump selftests
- Fix get_unused_fd_flags() usage in do_handle_open()
- Rename EXPORT_SYMBOL_GPL_FOR_MODULES to EXPORT_SYMBOL_FOR_MODULES
- Fix use-after-free in bh_read()
- Fix incorrect lflags value in the move_mount() syscall
* tag 'vfs-6.17-rc3.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:
signal: Fix memory leak for PIDFD_SELF* sentinels
kernfs: don't fail listing extended attributes
coredump: Fix return value in coredump_parse()
fs/buffer: fix use-after-free when call bh_read() helper
pidfs: Fix memory leak in pidfd_info()
netfs: Fix unbuffered write error handling
fhandle: do_handle_open() should get FD with user flags
module: Rename EXPORT_SYMBOL_GPL_FOR_MODULES to EXPORT_SYMBOL_FOR_MODULES
fs: fix incorrect lflags value in the move_mount syscall
selftests/coredump: Remove the read() that fails the test
fuse: keep inode->i_blkbits constant
iomap: Fix broken data integrity guarantees for O_SYNC writes
selftests/mount_setattr: add smoke tests for open_tree_attr(2) bug
open_tree_attr: do not allow id-mapping changes without OPEN_TREE_CLONE
fs: writeback: fix use-after-free in __mark_inode_dirty()
Fix smb3_init_transform_rq() to initialise buffer to NULL before calling
netfs_alloc_folioq_buffer() as netfs assumes it can append to the buffer it
is given. Setting it to NULL means it should start a fresh buffer, but the
value is currently undefined.
Fixes: a2906d3316 ("cifs: Switch crypto buffer to use a folio_queue rather than an xarray")
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <sfrench@samba.org>
cc: Paulo Alcantara <pc@manguebit.org>
cc: linux-cifs@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
We only need it when mount in question was sending events downstream (then
recepients need to switch to new master) or the mount is being turned into
slave (then we need a new master for it).
That wouldn't be a big deal, except that it causes quite a bit of work
when umount_tree() is taking a large peer group out. Adding a trivial
"don't bother calling propagation_source() unless we are going to use
its results" logics improves the things quite a bit.
We are still doing unnecessary work on bulk removals from propagation graph,
but the full solution for that will have to wait for the next merge window.
Fixes: 955336e204 "do_make_slave(): choose new master sanely"
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
do_change_type() and do_set_group() are operating on different
aspects of the same thing - propagation graph. The latter
asks for mounts involved to be mounted in namespace(s) the caller
has CAP_SYS_ADMIN for. The former is a mess - originally it
didn't even check that mount *is* mounted. That got fixed,
but the resulting check turns out to be too strict for userland -
in effect, we check that mount is in our namespace, having already
checked that we have CAP_SYS_ADMIN there.
What we really need (in both cases) is
* only touch mounts that are mounted. That's a must-have
constraint - data corruption happens if it get violated.
* don't allow to mess with a namespace unless you already
have enough permissions to do so (i.e. CAP_SYS_ADMIN in its userns).
That's an equivalent of what do_set_group() does; let's extract that
into a helper (may_change_propagation()) and use it in both
do_set_group() and do_change_type().
Fixes: 12f147ddd6 "do_change_type(): refuse to operate on unmounted/not ours mounts"
Acked-by: Andrei Vagin <avagin@gmail.com>
Reviewed-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Tested-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
... as the comments in reparent() clearly say. As it is, we reparent
*all* overmounts of the mounts being taken out, including those that
are taken out themselves. It's not only a potentially massive slowdown
(on a pathological setup we might end up with O(N^2) time for N mounts
being kicked out), it can end up with incorrect ->overmount in the
surviving mounts.
Fixes: f0d0ba1998 "Rewrite of propagate_umount()"
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In case when we mounting something on top of a large stack of overmounts,
all of them being peers of each other, we get quadratic time by the
depth of overmount stack. Easily fixed by doing commit_tree() before
reparenting the overmount; simplifies commit_tree() as well - it doesn't
need to skip the already mounted stuff that had been reparented on top
of the new mounts.
Since we are holding mount_lock through both reparenting and call of
commit_tree(), the order does not matter from the mount hash point
of view.
Reported-by: "Lai, Yi" <yi1.lai@linux.intel.com>
Tested-by: "Lai, Yi" <yi1.lai@linux.intel.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Fixes: 663206854f "copy_tree(): don't link the mounts via mnt_list"
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>:
Set of patches that brings no new functionality but makes it easier to
maintain and read the avs-driver code. There is one 'fix' among them -
the third patch, described below.
First, debug-related code - mainly the data-probing feature - is mixed
with non-debug PCM code. Separate it into debug.h and update the pcm.c
file.
Next, as the probing-board is not tied to any topology file, it shall
not using struct avs_soc_component descriptor but the basic struct
snd_soc_component one. While on the first sight this is just bump in
LOCs, runtime is simplified as no topology checks and related code will
be engaged.
With probing-board addressed and debug code relocated:
- address the component-teardown issues when the
componet->name points to a string which is shared by multiple components
and/or machine boards. devm_kstrdup() makes this go away.
- streamline the naming scheme for the functions engaged in the
components registration. This is purely a readability improvement.
Merge series from Shimrra Shai <shimrrashai@gmail.com>:
This is a revision of a previous patch submission of mine wherein one
part was accepted and one part was not. It was originally phrased for
a particular device using the ES8323 CODEC, and I was advised to make
the mixer controls user-selectable so it was not tied to the needs of
any specific device. This submission brings the DAC and mixer controls
for basic audio playback on this CODEC into line with the DAPM
infrastructure and enables their operation through the standard mixer
interface.
The current consumer documentation does not include bulk operations,
providing an example of how to acquire multiple regulators by calling
regulator_get() multiple times. That solution is valid and slightly
simpler for a small amount of regulators, but it does not scale well.
Document the bulk operations to get, enable and disable regulators.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Link: https://patch.msgid.link/20250819-reg_consumer_doc-v1-1-b631fc0d35a3@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The "priv->cdata" pointer isn't an error pointer; this should be a NULL
check instead. Otherwise it leads to a NULL pointer dereference in the
caller, rt5133_probe().
Fixes: 714165e1c4 ("regulator: rt5133: Add RT5133 PMIC regulator Support")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://patch.msgid.link/aKMc1oK-7yY4cD3K@stanley.mountain
Signed-off-by: Mark Brown <broonie@kernel.org>
There is a copy and paste error and we accidentally use "PTR_ERR(rdev)"
instead of "error". The "rdev" pointer is valid at this point.
Also there is no need to print the error code in the error message
because dev_err_probe() already prints that. So clean up the error
message a bit.
Fixes: 38c9f98db2 ("regulator: tps65219: Add support for TPS65215 Regulator IRQs")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://patch.msgid.link/aKRGmVdbvT1HBvm8@stanley.mountain
Signed-off-by: Mark Brown <broonie@kernel.org>
The DTS code coding style expects exactly one space around '='
character.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Merge series from Mohammad Rafi Shaik <mohammad.rafi.shaik@oss.qualcomm.com>:
On some Qualcomm platforms such as QCS6490-RB3Gen2, the multiple
WSA8830/WSA8835 speakers share a common reset (shutdown) GPIO.
To handle such cases, use the reset controller framework along with the
"reset-gpio" driver.
Tested on:
- QCS6490-RB3Gen2
Move the x86 specific helpers for getting kvm_{amd,intel} module params to
x86 where they belong. Expose the module-agnostic helpers globally, there
is nothing secret about the logic.
Link: https://lore.kernel.org/r/20250806225159.1687326-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Check that the return value of KVM_GET_VCPU_MMAP_SIZE is non-negative
before comparing with sizeof(kvm_run). If KVM_GET_VCPU_MMAP_SIZE fails,
it will return -1, and `-1 > sizeof(kvm_run)` is true, so the ASSERT
passes.
There are no other locations in tools/testing/selftests/kvm that make
the same mistake.
Signed-off-by: James Houghton <jthoughton@google.com>
Link: https://lore.kernel.org/r/20250711001742.1965347-1-jthoughton@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use MMU read lock to recover TDP MMU NX huge pages. To prevent
concurrent modification of the list of potential huge pages, iterate over
the list under tdp_mmu_pages_lock protection and unaccount the page
before dropping the lock.
Zapping under MMU read lock unblocks vCPUs which are waiting for MMU
read lock, which solves a guest jitter issue on Windows VMs which were
observing an increase in network latency.
Do not zap an SPTE if:
- The SPTE is a root page.
- The SPTE does not point at the SP's page table.
If the SPTE does not point at the SP's page table, then something else
has change the SPTE, so KVM cannot safely zap it.
Warn if zapping SPTE fails and current SPTE is still pointing to same
page table, as it should be impossible for the CMPXCHG to fail due to all
other write scenarios being mutually exclusive.
There is always a race between dirty logging, vCPU faults, and NX huge
page recovery for backing a gfn by an NX huge page or an executable
small page. Unaccounting sooner during the list traversal increases the
window of that race, but functionally, it is okay. Accounting doesn't
protect against iTLB multi-hit bug, it is there purely to prevent KVM
from bouncing a gfn between two page sizes. The only downside is that a
vCPU will end up doing more work in tearing down all the child SPTEs.
This should be a very rare race.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Co-developed-by: James Houghton <jthoughton@google.com>
Signed-off-by: James Houghton <jthoughton@google.com>
Link: https://lore.kernel.org/r/20250707224720.4016504-4-jthoughton@google.com
[sean: clean up kvm_mmu_sp_dirty_logging_enabled() and the changelog]
Signed-off-by: Sean Christopherson <seanjc@google.com>
kvm_tdp_mmu_zap_sp() is only used for NX huge page recovery, so rename
it to kvm_tdp_mmu_zap_possible_nx_huge_page(). In a future commit, this
function will be changed to include logic specific to NX huge page
recovery.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Signed-off-by: James Houghton <jthoughton@google.com>
https://lore.kernel.org/r/20250707224720.4016504-3-jthoughton@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Track possible NX huge pages for the TDP MMU separately from Shadow MMUs
in anticipation of doing recovery for the TDP MMU while holding mmu_lock
for read instead of write.
Use a small structure to hold the list of pages along with the number of
pages/entries in the list, as relying on kvm->stat.nx_lpage_splits to
calculating the number of pages to recover would result in over-zapping
when both TDP and Shadow MMUs are active.
Suggested-by: Sean Christopherson <seanjc@google.com>
Suggested-by: David Matlack <dmatlack@google.com>
Signed-off-by: Vipin Sharma <vipinsh@google.com>
Co-developed-by: James Houghton <jthoughton@google.com>
Signed-off-by: James Houghton <jthoughton@google.com>
Link: https://lore.kernel.org/r/20250707224720.4016504-2-jthoughton@google.com
[sean: rewrite changelog, use #ifdef instead of dummy KVM_TDP_MMU #define]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Introduce support for the PCI-Express Gen3 controller found in the
MT8196 (and MT6991) SoC by adding a compatible string and platform
specific data.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Link: https://patch.msgid.link/20250703120847.121826-4-angelogioacchino.delregno@collabora.com
Add compatible strings for MT8196 and MT6991 (which are fully
compatible between each other) and clock definitions.
These new SoCs don't have tl_96m and tl_32k clocks, but need
an AHB to APB bus clock and a low power clock.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
[mani: changed 'maxItems' to 'minItems' for reset]
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://patch.msgid.link/20250703120847.121826-3-angelogioacchino.delregno@collabora.com
In preparation to add support for the PCI-Express Gen3 controller
found in newer MediaTek SoCs, such as the Dimensity 9400 MT6991
and the MT8196 Chromebook SoC, add the definition for the PCIE
Resource Control register and a new sys_clk_rdy_time_us variable
in platform data.
If sys_clk_rdy_time_us is found (> 0), set the new value in the
aforementioned register only after configuring the controller to
RC mode, as this may otherwise be reset.
Overriding the register defaults for SYS_CLK_RDY_TIME allows to
work around sys_clk_rdy signal glitching in MT6991 and MT8196.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
[mani: used FIELD_MODIFY() to simplify mask and update]
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Link: https://patch.msgid.link/20250703120847.121826-2-angelogioacchino.delregno@collabora.com
In several code paths, such as xe_pt_create(), the vm->xef field is used
to determine whether a VM originates from userspace or the kernel.
Previously, this handler was only assigned in xe_vm_create_ioctl(),
after the VM was created by xe_vm_create(). However, xe_vm_create()
triggers page table creation, and that function assumes vm->xef should
be already set. This could lead to incorrect origin detection.
To fix this problem and ensure consistency in the initialization of
the VM object, let's move the assignment of this handler to
xe_vm_create.
v2:
- take reference to the xe file object only when xef is not NULL
- release the reference to the xe file object on the error path (Matthew)
Fixes: 7f387e6012 ("drm/xe: add XE_BO_FLAG_PINNED_LATE_RESTORE")
Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250811104358.2064150-2-piotr.piorkowski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
(cherry picked from commit 9337166fa1)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Partially revert commit e1db856bd2 ("usb: xhci: remove '0' write to
write-1-to-clear register") because the patch cleared the Interrupt Pending
bit during interrupt enabling and disabling. The Interrupt Pending bit
should only be cleared when the driver has handled the interrupt.
Ideally, all interrupts should be handled before disabling the interrupt;
consequently, no interrupt should be pending when enabling the interrupt.
For this reason, keep the debug message informing if an interrupt is still
pending when an interrupt is disabled.
Because the Interrupt Pending bit is write-1-to-clear, writing '0' to it
ensures that the state does not change.
Link: https://lore.kernel.org/linux-usb/20250818231103.672ec7ed@foxbook
Fixes: e1db856bd2 ("usb: xhci: remove '0' write to write-1-to-clear register")
Closes: https://bbs.archlinux.org/viewtopic.php?id=307641
cc: stable@vger.kernel.org # 6.16+
Signed-off-by: Niklas Neronin <niklas.neronin@linux.intel.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20250819125844.2042452-3-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xHC controller may immediately reuse a slot_id after it's disabled,
giving it to a new enumerating device before the xhci driver freed
all resources related to the disabled device.
In such a scenario, device-A with slot_id equal to 1 is disconnecting
while device-B is enumerating, device-B will fail to enumerate in the
follow sequence.
1.[device-A] send disable slot command
2.[device-B] send enable slot command
3.[device-A] disable slot command completed and wakeup waiting thread
4.[device-B] enable slot command completed with slot_id equal to 1 and
wakeup waiting thread
5.[device-B] driver checks that slot_id is still in use (by device-A) in
xhci_alloc_virt_device, and fail to enumerate due to this
conflict
6.[device-A] xhci->devs[slot_id] set to NULL in xhci_free_virt_device
To fix driver's slot_id resources conflict, clear xhci->devs[slot_id] and
xhci->dcbba->dev_context_ptrs[slot_id] pointers in the interrupt context
when disable slot command completes successfully. Simultaneously, adjust
function xhci_free_virt_device to accurately handle device release.
[minor smatch warning and commit message fix -Mathias]
Cc: stable@vger.kernel.org
Fixes: 7faac1953e ("xhci: avoid race between disable slot command and host runtime suspend")
Signed-off-by: Weitao Wang <WeitaoWang-oc@zhaoxin.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20250819125844.2042452-2-mathias.nyman@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For external phy, clk_phy should be optional, and some external phy
need the clock input from clk_phy. This patch adds support for setting
clk_phy for external phy.
Signed-off-by: David Wu <david.wu@rock-chips.com>
Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
Link: https://patch.msgid.link/20250815023515.114-1-kernel@airkyi.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Test that threaded state (in the persistent NAPI config) gets updated
even when NAPI with given ID is not allocated at the time.
This test is validating commit ccba9f6baa ("net: update NAPI threaded
config even for disabled NAPIs").
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Joe Damato <joe@dama.to>
Link: https://patch.msgid.link/20250815013314.2237512-1-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This patch enhances RX buffer handling in the mana driver by allocating
pages from a page pool and slicing them into MTU-sized fragments, rather
than dedicating a full page per packet. This approach is especially
beneficial on systems with large base page sizes like 64KB.
Key improvements:
- Proper integration of page pool for RX buffer allocations.
- MTU-sized buffer slicing to improve memory utilization.
- Reduce overall per Rx queue memory footprint.
- Automatic fallback to full-page buffers when:
* Jumbo frames are enabled (MTU > PAGE_SIZE / 2).
* The XDP path is active, to avoid complexities with fragment reuse.
Testing on VMs with 64KB pages shows around 200% throughput improvement.
Memory efficiency is significantly improved due to reduced wastage in page
allocations. Example: We are now able to fit 35 rx buffers in a single 64kb
page for MTU size of 1500, instead of 1 rx buffer per page previously.
Tested:
- iperf3, iperf2, and nttcp benchmarks.
- Jumbo frames with MTU 9000.
- Native XDP programs (XDP_PASS, XDP_DROP, XDP_TX, XDP_REDIRECT) for
testing the XDP path in driver.
- Memory leak detection (kmemleak).
- Driver load/unload, reboot, and stress scenarios.
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Reviewed-by: Saurabh Sengar <ssengar@linux.microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Dipayaan Roy <dipayanroy@linux.microsoft.com>
Link: https://patch.msgid.link/20250814140410.GA22089@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
No point in going down into the iomap mapping loop when we know it
will be rejected.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hans Holmberg <hans.holmberg@wdc.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
XFS processes truncating unlinked inodes asynchronously and thus the free
space pool only sees them with a delay. The non-zoned write path thus
calls into inodegc to accelerate this processing before failing an
allocation due the lack of free blocks. Do the same for the zoned space
reservation.
Fixes: 0bb2193056 ("xfs: add support for zoned space reservations")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hans Holmberg <hans.holmberg@wdc.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
This was my first attempt at caching the last used zone. But it turns out
for O_DIRECT or RWF_DONTCACHE that operate concurrently or in very short
sequence, the bmap btree does not record a written extent yet, so it fails.
Because it then still finds the last written zone it can lead to a weird
ping-pong around a few zones with writers seeing different values.
Remove it entirely as the later added xfs_cached_zone actually does a
much better job enforcing the locality as the zone is associated with the
inode in the MRU cache as soon as the zone is selected.
Fixes: 4e4d520755 ("xfs: add the zoned space allocator")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hans Holmberg <hans.holmberg@wdc.com>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
Merge the mmc fixes for v6.17-rc[n] into the next branch, to allow them to
get tested together with the new mmc changes that are targeted for v6.18.
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
The controller suspends the clock between blocks when reading from the
MMC as part of its flow-control, called read block gap. At higher clock
speed and with IO delay between the controller and the MMC, this clock
pause can happen too late, during the read of the next block and
trigger a read error.
To prevent this, the delay can be programmed for each mode via the pair
of registers HRS37/38. This delay is obtained during tuning, by trying
a multi-block read and increasing the delay until the read succeeds.
For now, the tuning is only done in HS200, as the read error has only
been observed at that speed.
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Benoît Monin <benoit.monin@bootlin.com>
Link: https://lore.kernel.org/r/20250818-mobileye-emmc-for-upstream-4-v4-6-34ecb3995e96@bootlin.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Provide a function to the MMC hosts to read some blocks of data as part
of their tuning.
This function only returns the status of the read operation, not the
data read.
Signed-off-by: Benoît Monin <benoit.monin@bootlin.com>
Link: https://lore.kernel.org/r/20250818-mobileye-emmc-for-upstream-4-v4-5-34ecb3995e96@bootlin.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Add a helper to check for the missing CMD23 quirk, similar to other
quirk helpers. Also reorder the helpers to match the order of the quirk
bits defined in include/linux/mmc/card.h.
Signed-off-by: Benoît Monin <benoit.monin@bootlin.com>
Link: https://lore.kernel.org/r/20250818-mobileye-emmc-for-upstream-4-v4-2-34ecb3995e96@bootlin.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Use the standard error pointer macro to shorten the code and simplify.
Signed-off-by: Xichao Zhao <zhao.xichao@vivo.com>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20250812092908.101867-1-zhao.xichao@vivo.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
This patch adds support for Over Current Protection (OCP) to the Realtek
USB card reader driver.
The OCP mechanism protects the hardware by detecting and handling current
overload conditions.
This implementation includes:
- Register configurations to enable OCP monitoring.
- Handling of OCP interrupt events and associated error reporting.
- Card power management changes in response to OCP triggers.
This enhancement improves the robustness of the driver when operating in
environments where electrical anomalies may occur, particularly with SD
and MS card interfaces.
Signed-off-by: Ricky Wu <ricky_wu@realtek.com>
Link: https://lore.kernel.org/r/20250812030811.2426112-1-ricky_wu@realtek.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Add timeout handling to wait_for_completion calls in memstick_set_rw_addr()
and memstick_alloc_card() to prevent indefinite blocking in case of
hardware or communication failures.
Signed-off-by: Jiayi Li <lijiayi@kylinos.cn>
Link: https://lore.kernel.org/r/20250804024825.1565078-1-lijiayi@kylinos.cn
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
As per the RZ/{G2L,G3E} HW manual SD_BUF0 can be accessed by 16/32/64
bits. Most of the data transfer in SD/SDIO/eMMC mode is more than 8 bytes.
During testing it is found that, if the DMA buffer is not aligned to 128
bit it fallback to PIO mode. In such cases, 64-bit access is much more
efficient than the current 16-bit.
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20250730164618.233117-2-biju.das.jz@bp.renesas.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Wrong actual clock reported, if the SD clock division ratio is other
than 1:1(bits DIV[7:0] in SD_CLK_CTRL are set to 11111111).
On high speed mode, cat /sys/kernel/debug/mmc1/ios
Without the patch:
clock: 50000000 Hz
actual clock: 200000000 Hz
After the fix:
clock: 50000000 Hz
actual clock: 50000000 Hz
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/20250629203859.170850-1-biju.das.jz@bp.renesas.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
