summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2025-09-10Merge tag 'iwlwifi-next-2025-09-09' of ↵Johannes Berg
https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next Miri Korenblit says: ==================== iwlwifi features, notably: - unused API cleanups - transport layer cleanups - new APIs support ==================== Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: cfg80211: Remove the redundant wiphy_devZheng tan
There is no need to call wiphy_dev again.Simplifying the code makes it more readable. Signed-off-by: Zheng tan <tanzheng@kylinos.cn> Reviewed-by: Arend van Spriel <arend.vanspriel@broadcom.com> Link: https://patch.msgid.link/20250910015556.219298-1-tanzheng@kylinos.cn Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: mac80211: fix incorrect commentMiri Korenblit
As opposed to what the comment says, we don't count in the skb size of the association request frame the length of the Per STA Profile of the association link. Fix the comment. Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250908122652.7022f33b1f33.Iac0d35744df883e8b96d71bbe8da518cc5d514bf@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: cfg80211: update the time stamps in hidden ssidMiri Korenblit
In hidden SSID we have separate BSS entries for the beacon and for the probe response(s). The BSS entry time stamps represent the age of the BSS; when was the last time we heard the BSS. When we receive a beacon of a hidden SSID it means that we heard that BSS, so it makes sense to indicate that in the probe response entries. Do that. Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250907115135.712745e498c0.I38186abf5d20dec6f6f2d42d2e1cdb50c6bfea25@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: mac80211: Fix HE capabilities element checkIlan Peer
The element data length check did not account for the extra octet used for the extension ID. Fix it. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com> Link: https://patch.msgid.link/20250907115109.8da0012e2286.I8c0c69a0011f7153c13b365b14dfef48cfe7c3e3@changeid Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: mac80211: add tx_handlers_drop statistics to ethtoolSarika Sharma
Currently tx_handlers_drop statistics are handled only for slow TX path and only at radio level. This also requires CONFIG_MAC80211_DEBUG_COUNTERS to be enabled to account the dropped packets. There is no way to check these stats for fast TX, at interface level and monitor without enabling the debug configuration. Hence, add a new counter at the sdata level to track packets dropped with reason as TX_DROP during transmission for fast path, slow path and other tx management packets. Expose this via ethtool statistics, to improve visibility into transmission failures at interface level and aid debugging and performance monitoring. Place the counter in ethtool with other available tx_* stats for better readability and accurate tracking. Sample output: root@buildroot:~# ethtool -S wlan0 NIC statistics: rx_packets: 5904 rx_bytes: 508122 rx_duplicates: 12 rx_fragments: 5900 rx_dropped: 12 tx_packets: 391487 tx_bytes: 600423383 tx_filtered: 0 tx_retry_failed: 10332 tx_retries: 1548 tx_handlers_drop: 4 .... Co-developed-by: Hari Chandrakanthan <quic_haric@quicinc.com> Signed-off-by: Hari Chandrakanthan <quic_haric@quicinc.com> Signed-off-by: Sarika Sharma <quic_sarishar@quicinc.com> Link: https://patch.msgid.link/20250822052110.513804-1-quic_sarishar@quicinc.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10wifi: virt_wifi: Fix page fault on connectJames Guan
This patch prevents page fault in __cfg80211_connect_result()[1] when connecting a virt_wifi device, while ensuring that virt_wifi can connect properly. [1] https://lore.kernel.org/linux-wireless/20250909063213.1055024-1-guan_yufei@163.com/ Closes: https://lore.kernel.org/linux-wireless/20250909063213.1055024-1-guan_yufei@163.com/ Signed-off-by: James Guan <guan_yufei@163.com> Link: https://patch.msgid.link/20250910111929.137049-1-guan_yufei@163.com [remove irrelevant network-manager instructions] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10riscv: dts: allwinner: rename devterm i2c-gpio node to comply with bindingConor Dooley
The i2c controller binding does not permit permit the node name to contain "gpio", resulting in two warnings: i2c-gpio-0 (i2c-gpio): $nodename:0: 'i2c-gpio-0' does not match '^i2c(@.+|-[a-z0-9]+)?$' i2c-gpio-0 (i2c-gpio): Unevaluated properties are not allowed ('#address-cells', '#size-cells', 'adc@54' were unexpected) Drop it to satisfy dtbs_check. Signed-off-by: Conor Dooley <conor.dooley@microchip.com> Link: https://patch.msgid.link/20250909-frown-wrinkle-f16df243a970@spud Signed-off-by: Chen-Yu Tsai <wens@csie.org>
2025-09-10drm/mediatek: clean up driver data initialisationJohan Hovold
The platform and drm devices are only used to look up the drm device and its driver data respectively when initialising the driver data during bind(). Drop the reference counts as soon as they have been used to make the code more readable. Note that the crtc count is never incremented on lookup failures. Signed-off-by: Johan Hovold <johan@kernel.org> Reviewed-by: CK Hu <ck.hu@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://patchwork.kernel.org/project/dri-devel/patch/20250829090345.21075-3-johan@kernel.org/ Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
2025-09-10soc: sunxi: sram: register regmap as sysconChen-Yu Tsai
If the system controller had a ethernet controller glue layer control register, a limited access regmap would be registered and tied to the system controller struct device for the ethernet driver to use. Until now, for the ethernet driver to acquire this regmap, it had to do a of_parse_phandle() + find device + dev_get_regmap() sequence. Since the syscon framework allows a provider to register a custom regmap for its device node, and the ethernet driver already uses syscon for one platform, this provides a much more easier way to pass the regmap. Use of_syscon_register_regmap() to register our regmap with the syscon framework so that consumers can retrieve it that way. Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com> Link: https://patch.msgid.link/20250908181059.1785605-5-wens@kernel.org Signed-off-by: Chen-Yu Tsai <wens@csie.org>
2025-09-10soc: sunxi: sram: add entry for a523Chen-Yu Tsai
The A523 has two Ethernet controllers. So in the system controller address space, there are two registers for Ethernet clock delays, one for each controller. Add a new entry for the A523 system controller that allows access to the second register. Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com> Link: https://patch.msgid.link/20250908181059.1785605-4-wens@kernel.org Signed-off-by: Chen-Yu Tsai <wens@csie.org>
2025-09-10i2c: core: Use fwnode_for_each_child_node_scoped()Jean-François Lessard
Replace the manual __free(fwnode_handle) iterator declaration with the new scoped iterator macro for cleaner, less error-prone code. This eliminates the need for explicit iterator variable declaration with the cleanup attribute, making the code more consistent with other scoped iterator usage patterns in the kernel. Signed-off-by: Jean-François Lessard <jefflessard3@gmail.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
2025-09-10drm/mediatek: fix potential OF node use-after-freeJohan Hovold
The for_each_child_of_node() helper drops the reference it takes to each node as it iterates over children and an explicit of_node_put() is only needed when exiting the loop early. Drop the recently introduced bogus additional reference count decrement at each iteration that could potentially lead to a use-after-free. Fixes: 1f403699c40f ("drm/mediatek: Fix device/node reference count leaks in mtk_drm_get_all_drm_priv") Cc: Ma Ke <make24@iscas.ac.cn> Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold <johan@kernel.org> Reviewed-by: CK Hu <ck.hu@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://patchwork.kernel.org/project/dri-devel/patch/20250829090345.21075-2-johan@kernel.org/ Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
2025-09-10device property: Add scoped fwnode child node iteratorsJean-François Lessard
Add scoped versions of fwnode child node iterators that automatically handle reference counting cleanup using the __free() attribute: - fwnode_for_each_child_node_scoped() - fwnode_for_each_available_child_node_scoped() These macros follow the same pattern as existing scoped iterators in the kernel, ensuring fwnode references are automatically released when the iterator variable goes out of scope. This prevents resource leaks and eliminates the need for manual cleanup in error paths. The implementation mirrors the non-scoped variants but uses __free(fwnode_handle) for automatic resource management, providing a safer and more convenient interface for drivers iterating over firmware node children. Signed-off-by: Jean-François Lessard <jefflessard3@gmail.com> Acked-by: Danilo Krummrich <dakr@kernel.org> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
2025-09-10um: Fix FD copy size in os_rcv_fd_msg()Tiwei Bie
When copying FDs, the copy size should not include the control message header (cmsghdr). Fix it. Fixes: 5cde6096a4dd ("um: generalize os_rcv_fd") Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: virtio_uml: Fix use-after-free after put_device in probeMiaoqian Lin
When register_virtio_device() fails in virtio_uml_probe(), the code sets vu_dev->registered = 1 even though the device was not successfully registered. This can lead to use-after-free or other issues. Fixes: 04e5b1fb0183 ("um: virtio: Remove device on disconnect") Signed-off-by: Miaoqian Lin <linmq006@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Don't mark stack executableRichard Weinberger
On one of my machines UML failed to start after enabling SELinux. UML failed to start because SELinux's execmod rule denies executable pages on a modified file mapping. Historically UML marks it's stack rwx. AFAICT, these days this is no longer needed, so let's remove PROT_EXEC. Signed-off-by: Richard Weinberger <richard@nod.at> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Remove unused ipi_pipe field from cpuinfo_umTiwei Bie
It's no longer used after the removal of the SMP implementation in TT mode by commit 28fa468f5316 ("um: Remove broken SMP support"). While at it, remove the outdated comment. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Remove unused cpu_data and current_cpu_data macrosTiwei Bie
These two macros have no users. Remove them. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Stop tracking virtual CPUs via mm_cpumask()Tiwei Bie
In UML, each user address space is represented as a separate stub process on the host. Therefore, user address spaces do not require TLB management on UML virtual CPUs, and it's unnecessary to track which virtual CPUs they have executed on. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Centralize stub size calculationsTiwei Bie
Currently, the stub size is calculated in multiple places. Define a macro that performs the calculation so that the code is easier to read and maintain. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Remove outdated comment about STUB_DATA_PAGESTiwei Bie
STUB_DATA_PAGES is no longer required to be a power of two since commit 91f0a0c5cc5b ("um: Calculate stub data address relative to stub code"). Remove the outdated comment. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Remove unused offset and child_err fields from stub_dataTiwei Bie
They are no longer used. Remove them. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Indent time-travel help messagesTiwei Bie
Indent the help messages for time-travel to make them consistent with the format of other help messages. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Fix help message for ssl-non-rawTiwei Bie
Add the missing option name in the help message. Additionally, switch to __uml_help(), because this is a global option rather than a per-channel option. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: vector: Fix indentation for help messageTiwei Bie
For consistency with other help messages, use four spaces for indentation instead of a tab plus a space. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Add missing trailing newline to help messagesTiwei Bie
Some help messages are missing a trailing newline. They should end with two newlines, but only one is present. Add the missing newline to make the --help output more readable. Signed-off-by: Tiwei Bie <tiwei.btw@antgroup.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: virtio-pci: implement .shutdown()Johannes Berg
We don't want queues stopped since the shutdown can still be doing things with PCI devices, so implement .shutdown() to avoid the generic virtio shutdown. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10um: Support SPARSE_IRQSinan Nalkaya
Motivation: IRQ KUnit tests are going to require CONFIG_SPARSE_IRQ [1] in order to: (a) reliably allocate additional (fake) IRQs and (b) ensure we can test managed affinity, which is only supported with SPARSE_IRQ. It seems that the only thing necessary for ARCH=um is to tell the genirq core to skip over our preallocated NR_IRQS. Tested with: $ ./tools/testing/kunit/kunit.py run [...] [13:55:58] Testing complete. Ran 676 tests: passed: 646, skipped: 30 [...] This compares with pre-patch results: Ran 672 tests: passed: 644, skipped: 28 i.e., we no longer skip tests that 'depend on SPARSE_IRQ', and existing tests all pass. [1] [PATCH v2 4/6] genirq/test: Depend on SPARSE_IRQ https://lore.kernel.org/all/CABVgOSngoD0fh1WEkUCEwSdk0Joypo3dA_Y_SjW+K=nVDnZs3Q@mail.gmail.com/ Signed-off-by: Sinan Nalkaya <sardok@gmail.com> [Brian: Adapted Sinan's patch; rewrote commit message] Signed-off-by: Brian Norris <briannorris@chromium.org> Tested-by: David Gow <davidgow@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2025-09-10Merge drm/drm-next into drm-intel-nextRodrigo Vivi
Catching up with some display dependencies. Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
2025-09-10blk-mq: fix stale nr_requests documentationYu Kuai
The nr_requests documentation is still the removed single queue, remove it and update to current blk-mq. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: remove blk_mq_tag_update_depth()Yu Kuai
This helper is not used now. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: fix potential deadlock while nr_requests grownYu Kuai
Allocate and free sched_tags while queue is freezed can deadlock[1], this is a long term problem, hence allocate memory before freezing queue and free memory after queue is unfreezed. [1] https://lore.kernel.org/all/0659ea8d-a463-47c8-9180-43c719e106eb@linux.ibm.com/ Fixes: e3a2b3f931f5 ("blk-mq: allow changing of queue depth through sysfs") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq-sched: add new parameter nr_requests in blk_mq_alloc_sched_tags()Yu Kuai
This helper only support to allocate the default number of requests, add a new parameter to support specific number of requests. Prepare to fix potential deadlock in the case nr_requests grow. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: split bitmap grow and resize case in blk_mq_update_nr_requests()Yu Kuai
No functional changes are intended, make code cleaner and prepare to fix the grow case in following patches. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: cleanup shared tags case in blk_mq_update_nr_requests()Yu Kuai
For shared tags case, all hctx->sched_tags/tags are the same, it doesn't make sense to call into blk_mq_tag_update_depth() multiple times for the same tags. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: convert to serialize updating nr_requests with update_nr_hwq_lockYu Kuai
request_queue->nr_requests can be changed by: a) switch elevator by updating nr_hw_queues b) switch elevator by elevator sysfs attribute c) configue queue sysfs attribute nr_requests Current lock order is: 1) update_nr_hwq_lock, case a,b 2) freeze_queue 3) elevator_lock, case a,b,c And update nr_requests is seriablized by elevator_lock() already, however, in the case c, we'll have to allocate new sched_tags if nr_requests grow, and do this with elevator_lock held and queue freezed has the risk of deadlock. Hence use update_nr_hwq_lock instead, make it possible to allocate memory if tags grow, meanwhile also prevent nr_requests to be changed concurrently. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: check invalid nr_requests in queue_requests_store()Yu Kuai
queue_requests_store() is the only caller of blk_mq_update_nr_requests(), and blk_mq_update_nr_requests() is the only caller of blk_mq_tag_update_depth(), however, they all have checkings for nr_requests input by user. Make code cleaner by moving all the checkings to the top function: 1) nr_requests > reserved tags; 2) if there is elevator, 4 <= nr_requests <= 2048; 3) if elevator is none, 4 <= nr_requests <= tag_set->queue_depth; Meanwhile, case 2 is the only case tags can grow and -ENOMEM might be returned. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: remove useless checkings in blk_mq_update_nr_requests()Yu Kuai
1) queue_requests_store() is the only caller of blk_mq_update_nr_requests(), where queue is already freezed, no need to check mq_freeze_depth; 2) q->tag_set must be set for request based device, and queue_is_mq() is already checked in blk_mq_queue_attr_visible(), no need to check q->tag_set. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-mq: remove useless checking in queue_requests_store()Yu Kuai
blk_mq_queue_attr_visible() already checked queue_is_mq(), no need to check this again in queue_requests_store(). Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10ublk: consolidate nr_io_ready and nr_queues_readyCaleb Sander Mateos
ublk_mark_io_ready() tracks whether all the ublk_device's I/Os have been fetched by incrementing ublk_queue's nr_io_ready count and incrementing ublk_device's nr_queues_ready count if the whole queue is ready. Simplify the logic by just tracking the total number of fetched I/Os on each ublk_device. When this count reaches nr_hw_queues * queue_depth, the ublk_device is ready to receive I/O. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/raid0: convert raid0_make_request() to use bio_submit_split_bioset()Yu Kuai
Currently, raid0_make_request() will remap the original bio to underlying disks to prevent reordered IO. Now that bio_submit_split_bioset() will put original bio to the head of current->bio_list, it's safe converting to use this helper and bio will still be ordered. CC: Jan Kara <jack@suse.cz> Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10block: fix ordering of recursive split IOYu Kuai
Currently, split bio will be chained to original bio, and original bio will be resubmitted to the tail of current->bio_list, waiting for split bio to be issued. However, if split bio get split again, the IO order will be messed up. This problem, on the one hand, will cause performance degradation, especially for mdraid with large IO size; on the other hand, will cause write errors for zoned block devices[1]. For example, in raid456 IO will first be split by max_sector from md_submit_bio(), and then later be split again by chunksize for internal handling: For example, assume max_sectors is 1M, and chunksize is 512k 1) issue a 2M IO: bio issuing: 0+2M current->bio_list: NULL 2) md_submit_bio() split by max_sector: bio issuing: 0+1M current->bio_list: 1M+1M 3) chunk_aligned_read() split by chunksize: bio issuing: 0+512k current->bio_list: 1M+1M -> 512k+512k 4) after first bio issued, __submit_bio_noacct() will contuine issuing next bio: bio issuing: 1M+1M current->bio_list: 512k+512k bio issued: 0+512k 5) chunk_aligned_read() split by chunksize: bio issuing: 1M+512k current->bio_list: 512k+512k -> 1536k+512k bio issued: 0+512k 6) no split afterwards, finally the issue order is: 0+512k -> 1M+512k -> 512k+512k -> 1536k+512k This behaviour will cause large IO read on raid456 endup to be small discontinuous IO in underlying disks. Fix this problem by placing split bio to the head of current->bio_list. Test script: test on 8 disk raid5 with 64k chunksize dd if=/dev/md0 of=/dev/null bs=4480k iflag=direct Test results: Before this patch 1) iostat results: Device r/s rMB/s rrqm/s %rrqm r_await rareq-sz aqu-sz %util md0 52430.00 3276.87 0.00 0.00 0.62 64.00 32.60 80.10 sd* 4487.00 409.00 2054.00 31.40 0.82 93.34 3.68 71.20 2) blktrace G stage: 8,0 0 486445 11.357392936 843 G R 14071424 + 128 [dd] 8,0 0 486451 11.357466360 843 G R 14071168 + 128 [dd] 8,0 0 486454 11.357515868 843 G R 14071296 + 128 [dd] 8,0 0 486468 11.357968099 843 G R 14072192 + 128 [dd] 8,0 0 486474 11.358031320 843 G R 14071936 + 128 [dd] 8,0 0 486480 11.358096298 843 G R 14071552 + 128 [dd] 8,0 0 486490 11.358303858 843 G R 14071808 + 128 [dd] 3) io seek for sdx: Noted io seek is the result from blktrace D stage, statistic of: ABS((offset of next IO) - (offset + len of previous IO)) Read|Write seek cnt 55175, zero cnt 25079 >=(KB) .. <(KB) : count ratio |distribution | 0 .. 1 : 25079 45.5% |########################################| 1 .. 2 : 0 0.0% | | 2 .. 4 : 0 0.0% | | 4 .. 8 : 0 0.0% | | 8 .. 16 : 0 0.0% | | 16 .. 32 : 0 0.0% | | 32 .. 64 : 12540 22.7% |##################### | 64 .. 128 : 2508 4.5% |##### | 128 .. 256 : 0 0.0% | | 256 .. 512 : 10032 18.2% |################# | 512 .. 1024 : 5016 9.1% |######### | After this patch: 1) iostat results: Device r/s rMB/s rrqm/s %rrqm r_await rareq-sz aqu-sz %util md0 87965.00 5271.88 0.00 0.00 0.16 61.37 14.03 90.60 sd* 6020.00 658.44 5117.00 45.95 0.44 112.00 2.68 86.50 2) blktrace G stage: 8,0 0 206296 5.354894072 664 G R 7156992 + 128 [dd] 8,0 0 206305 5.355018179 664 G R 7157248 + 128 [dd] 8,0 0 206316 5.355204438 664 G R 7157504 + 128 [dd] 8,0 0 206319 5.355241048 664 G R 7157760 + 128 [dd] 8,0 0 206333 5.355500923 664 G R 7158016 + 128 [dd] 8,0 0 206344 5.355837806 664 G R 7158272 + 128 [dd] 8,0 0 206353 5.355960395 664 G R 7158528 + 128 [dd] 8,0 0 206357 5.356020772 664 G R 7158784 + 128 [dd] 3) io seek for sdx Read|Write seek cnt 28644, zero cnt 21483 >=(KB) .. <(KB) : count ratio |distribution | 0 .. 1 : 21483 75.0% |########################################| 1 .. 2 : 0 0.0% | | 2 .. 4 : 0 0.0% | | 4 .. 8 : 0 0.0% | | 8 .. 16 : 0 0.0% | | 16 .. 32 : 0 0.0% | | 32 .. 64 : 7161 25.0% |############## | BTW, this looks like a long term problem from day one, and large sequential IO read is pretty common case like video playing. And even with this patch, in this test case IO is merged to at most 128k is due to block layer plug limit BLK_PLUG_FLUSH_SIZE, increase such limit can get even better performance. However, we'll figure out how to do this properly later. [1] https://lore.kernel.org/all/e40b076d-583d-406b-b223-005910a9f46f@acm.org/ Fixes: d89d87965dcb ("When stacked block devices are in-use (e.g. md or dm), the recursive calls") Reported-by: Tie Ren <tieren@fnnas.com> Closes: https://lore.kernel.org/all/7dro5o7u5t64d6bgiansesjavxcuvkq5p2pok7dtwkav7b7ape@3isfr44b6352/ Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10block: skip unnecessary checks for split bioYu Kuai
Lots of checks are already done while submitting this bio the first time, and there is no need to check them again when this bio is resubmitted after split. Hence open code should_fail_bio() and blk_throtl_bio() that are still necessary from submit_bio_split_bioset(). Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10blk-crypto: convert to use bio_submit_split_bioset()Yu Kuai
Unify bio split code, prepare to fix ordering of split IO. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/md-linear: convert to use bio_submit_split_bioset()Yu Kuai
Unify bio split code, prepare to fix reordered split IO. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/raid5: convert to use bio_submit_split_bioset()Yu Kuai
Unify bio split code, prepare to fix ordering of split IO. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/raid10: convert read/write to use bio_submit_split_bioset()Yu Kuai
Unify bio split code, prepare to fix ordering of split IO, the error path is modified a bit, however no functional changes are intended: - bio_submit_split_bioset() can fail the original bio directly by split error, set R10BIO_Uptodate in this case to notify raid_end_bio_io() that the original bio is returned already. - set R10BIO_Uptodate and set error value to -EIO is useless now, for r10_bio without R10BIO_Uptodate, -EIO will be returned for original bio. And discard is not handled, because discard is only split for unaligned head and tail, and this can be considered slow path, the reorder here does not matter much. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/raid10: add a new r10bio flag R10BIO_ReturnedYu Kuai
The new helper bio_submit_split_bioset() can failed the orginal bio on split errors, prepare to handle this case in raid_end_bio_io(). The flag name is refer to the r1bio flag name. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2025-09-10md/raid1: convert to use bio_submit_split_bioset()Yu Kuai
Unify bio split code, and prepare to fix ordering of split IO. Noted that bio_submit_split_bioset() can fail the original bio directly by split error, set R1BIO_Returned in this case to notify raid_end_bio_io() that the original bio is returned already. Signed-off-by: Yu Kuai <yukuai3@huawei.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
generated by cgit v1.3 (git 2.53.0) at 2026-11-14 18:15:07 +0000