Re: [PATCH 6.14 000/145] 6.14.8-rc1 review

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Wed, May 21, 2025 at 10:13:45PM +0800, Eric Naim wrote:
> On 5/21/25 13:35, Greg Kroah-Hartman wrote:
> > On Tue, May 20, 2025 at 09:42:13PM -0500, Mario Limonciello wrote:
> >> On 5/20/2025 4:34 PM, Eric Naim wrote:
> >>> Hi Greg,
> >>>
> >>> On 5/20/25 21:49, Greg Kroah-Hartman wrote:
> >>>> This is the start of the stable review cycle for the 6.14.8 release.
> >>>> There are 145 patches in this series, all will be posted as a response
> >>>> to this one.  If anyone has any issues with these being applied, please
> >>>> let me know.
> >>>>
> >>>> Responses should be made by Thu, 22 May 2025 12:57:37 +0000.
> >>>> Anything received after that time might be too late.
> >>>>
> >>>> The whole patch series can be found in one patch at:
> >>>> 	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.14.8-rc1.gz
> >>>> or in the git tree and branch at:
> >>>> 	git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.14.y
> >>>> and the diffstat can be found below.
> >>>>
> >>>> thanks,
> >>>>
> >>>> greg k-h
> >>>>
> >>>> -------------
> >>>> Pseudo-Shortlog of commits:
> >>>>
> >>>> Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> >>>>      Linux 6.14.8-rc1
> >>>>
> >>>> Dan Carpenter <dan.carpenter@linaro.org>
> >>>>      phy: tegra: xusb: remove a stray unlock
> >>>>
> >>>> Tiezhu Yang <yangtiezhu@loongson.cn>
> >>>>      perf tools: Fix build error for LoongArch
> >>>>
> >>>> Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> >>>>      mm/page_alloc: fix race condition in unaccepted memory handling
> >>>>
> >>>> Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> >>>>      drm/xe/gsc: do not flush the GSC worker from the reset path
> >>>>
> >>>> Maciej Falkowski <maciej.falkowski@linux.intel.com>
> >>>>      accel/ivpu: Flush pending jobs of device's workqueues
> >>>>
> >>>> Karol Wachowski <karol.wachowski@intel.com>
> >>>>      accel/ivpu: Fix missing MMU events if file_priv is unbound
> >>>>
> >>>> Karol Wachowski <karol.wachowski@intel.com>
> >>>>      accel/ivpu: Fix missing MMU events from reserved SSID
> >>>>
> >>>> Karol Wachowski <karol.wachowski@intel.com>
> >>>>      accel/ivpu: Move parts of MMU event IRQ handling to thread handler
> >>>>
> >>>> Karol Wachowski <karol.wachowski@intel.com>
> >>>>      accel/ivpu: Dump only first MMU fault from single context
> >>>>
> >>>> Maciej Falkowski <maciej.falkowski@linux.intel.com>
> >>>>      accel/ivpu: Use workqueue for IRQ handling
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: Refactor remove call with idxd_cleanup() helper
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: fix memory leak in error handling path of idxd_pci_probe
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: fix memory leak in error handling path of idxd_alloc
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: Add missing idxd cleanup to fix memory leak in remove call
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: Add missing cleanups in cleanup internals
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: Add missing cleanup for early error out in idxd_setup_internals
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: fix memory leak in error handling path of idxd_setup_groups
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: fix memory leak in error handling path of idxd_setup_engines
> >>>>
> >>>> Shuai Xue <xueshuai@linux.alibaba.com>
> >>>>      dmaengine: idxd: fix memory leak in error handling path of idxd_setup_wqs
> >>>>
> >>>> Yemike Abhilash Chandra <y-abhilashchandra@ti.com>
> >>>>      dmaengine: ti: k3-udma: Use cap_mask directly from dma_device structure instead of a local copy
> >>>>
> >>>> Ronald Wahl <ronald.wahl@legrand.com>
> >>>>      dmaengine: ti: k3-udma: Add missing locking
> >>>>
> >>>> Barry Song <baohua@kernel.org>
> >>>>      mm: userfaultfd: correct dirty flags set for both present and swap pte
> >>>>
> >>>> Wupeng Ma <mawupeng1@huawei.com>
> >>>>      mm: hugetlb: fix incorrect fallback for subpool
> >>>>
> >>>> hexue <xue01.he@samsung.com>
> >>>>      io_uring/uring_cmd: fix hybrid polling initialization issue
> >>>>
> >>>> Jens Axboe <axboe@kernel.dk>
> >>>>      io_uring/memmap: don't use page_address() on a highmem page
> >>>>
> >>>> Nathan Chancellor <nathan@kernel.org>
> >>>>      net: qede: Initialize qede_ll_ops with designated initializer
> >>>>
> >>>> Steven Rostedt <rostedt@goodmis.org>
> >>>>      ring-buffer: Fix persistent buffer when commit page is the reader page
> >>>>
> >>>> Ming Yen Hsieh <mingyen.hsieh@mediatek.com>
> >>>>      wifi: mt76: mt7925: fix missing hdr_trans_tlv command for broadcast wtbl
> >>>>
> >>>> Fedor Pchelkin <pchelkin@ispras.ru>
> >>>>      wifi: mt76: disable napi on driver removal
> >>>>
> >>>> Jarkko Sakkinen <jarkko@kernel.org>
> >>>>      tpm: Mask TPM RC in tpm2_start_auth_session()
> >>>>
> >>>> Aaron Kling <webgeek1234@gmail.com>
> >>>>      spi: tegra114: Use value to check for invalid delays
> >>>>
> >>>> Jethro Donaldson <devel@jro.nz>
> >>>>      smb: client: fix memory leak during error handling for POSIX mkdir
> >>>>
> >>>> Steve Siwinski <ssiwinski@atto.com>
> >>>>      scsi: sd_zbc: block: Respect bio vector limits for REPORT ZONES buffer
> >>>>
> >>>> Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
> >>>>      phy: renesas: rcar-gen3-usb2: Set timing registers only once
> >>>>
> >>>> Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>
> >>>>      phy: renesas: rcar-gen3-usb2: Fix role detection on unbind/bind
> >>>>
> >>>> Oleksij Rempel <o.rempel@pengutronix.de>
> >>>>      net: phy: micrel: remove KSZ9477 EEE quirks now handled by phylink
> >>>>
> >>>> Oleksij Rempel <o.rempel@pengutronix.de>
> >>>>      net: dsa: microchip: let phylink manage PHY EEE configuration on KSZ switches
> >>>>
> >>>> Ma Ke <make24@iscas.ac.cn>
> >>>>      phy: Fix error handling in tegra_xusb_port_init
> >>>>
> >>>> Wayne Chang <waynec@nvidia.com>
> >>>>      phy: tegra: xusb: Use a bitmask for UTMI pad power state tracking
> >>>>
> >>>> Steven Rostedt <rostedt@goodmis.org>
> >>>>      tracing: samples: Initialize trace_array_printk() with the correct function
> >>>>
> >>>> Ashish Kalra <ashish.kalra@amd.com>
> >>>>      x86/sev: Make sure pages are not skipped during kdump
> >>>>
> >>>> Ashish Kalra <ashish.kalra@amd.com>
> >>>>      x86/sev: Do not touch VMSA pages during SNP guest memory kdump
> >>>>
> >>>> pengdonglin <pengdonglin@xiaomi.com>
> >>>>      ftrace: Fix preemption accounting for stacktrace filter command
> >>>>
> >>>> pengdonglin <pengdonglin@xiaomi.com>
> >>>>      ftrace: Fix preemption accounting for stacktrace trigger command
> >>>>
> >>>> Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> >>>>      i2c: designware: Fix an error handling path in i2c_dw_pci_probe()
> >>>>
> >>>> Nathan Chancellor <nathan@kernel.org>
> >>>>      kbuild: Disable -Wdefault-const-init-unsafe
> >>>>
> >>>> Michael Kelley <mhklinux@outlook.com>
> >>>>      Drivers: hv: vmbus: Remove vmbus_sendpacket_pagebuffer()
> >>>>
> >>>> Michael Kelley <mhklinux@outlook.com>
> >>>>      Drivers: hv: Allow vmbus_sendpacket_mpb_desc() to create multiple ranges
> >>>>
> >>>> Michael Kelley <mhklinux@outlook.com>
> >>>>      hv_netvsc: Remove rmsg_pgcnt
> >>>>
> >>>> Michael Kelley <mhklinux@outlook.com>
> >>>>      hv_netvsc: Preserve contiguous PFN grouping in the page buffer array
> >>>>
> >>>> Michael Kelley <mhklinux@outlook.com>
> >>>>      hv_netvsc: Use vmbus_sendpacket_mpb_desc() to send VMBus messages
> >>>>
> >>>> Dragan Simic <dsimic@manjaro.org>
> >>>>      arm64: dts: rockchip: Remove overdrive-mode OPPs from RK3588J SoC dtsi
> >>>>
> >>>> Sam Edwards <cfsworks@gmail.com>
> >>>>      arm64: dts: rockchip: Allow Turing RK1 cooling fan to spin down
> >>>>
> >>>> Christian Hewitt <christianshewitt@gmail.com>
> >>>>      arm64: dts: amlogic: dreambox: fix missing clkc_audio node
> >>>>
> >>>> Hyejeong Choi <hjeong.choi@samsung.com>
> >>>>      dma-buf: insert memory barrier before updating num_fences
> >>>>
> >>>> Nicolas Chauvet <kwizart@gmail.com>
> >>>>      ALSA: usb-audio: Add sample rate quirk for Microdia JP001 USB Camera
> >>>>
> >>>> Christian Heusel <christian@heusel.eu>
> >>>>      ALSA: usb-audio: Add sample rate quirk for Audioengine D1
> >>>>
> >>>> Wentao Liang <vulab@iscas.ac.cn>
> >>>>      ALSA: es1968: Add error handling for snd_pcm_hw_constraint_pow2()
> >>>>
> >>>> Jeremy Linton <jeremy.linton@arm.com>
> >>>>      ACPI: PPTT: Fix processor subtable walk
> >>>>
> >>>> Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
> >>>>      gpio: pca953x: fix IRQ storm on system wake up
> >>>>
> >>>> Alexey Makhalov <alexey.makhalov@broadcom.com>
> >>>>      MAINTAINERS: Update Alexey Makhalov's email address
> >>>>
> >>>> Wayne Lin <Wayne.Lin@amd.com>
> >>>>      drm/amd/display: Avoid flooding unnecessary info messages
> >>>>
> >>>> Wayne Lin <Wayne.Lin@amd.com>
> >>>>      drm/amd/display: Correct the reply value when AUX write incomplete
> >>>>
> >>>> Philip Yang <Philip.Yang@amd.com>
> >>>>      drm/amdgpu: csa unmap use uninterruptible lock
> >>>>
> >>>> Tim Huang <tim.huang@amd.com>
> >>>>      drm/amdgpu: fix incorrect MALL size for GFX1151
> >>>>
> >>>> David (Ming Qiang) Wu <David.Wu3@amd.com>
> >>>>      drm/amdgpu: read back register after written for VCN v4.0.5
> >>>>
> >>>
> >>> This commit seems to breaking a couple of devices with the Phoenix APU, most notably the Ryzen AI chips. Note that this commit in mainline seems to work as intended, and after doing a little bit of digging, [1] landed in 6.15 and so this cherrypick may not be so trivial after all. Attached is a kernel trace highlighting the breakage caused by this commit, along with [2] for the full log.
> >>>
> >>> Also adding Alex, David and Mario to Ccs.
> >>>
> >>
> >> Just a minor correction - VCN 4.0.5 is on Strix.  So this report is not
> >> likely from a Phoenix APU.
> >>
> >> Nonetheless I agree; I suspect backporting
> >> ecc9ab4e924b7eb9e2c4a668162aaa1d9d60d08c will help the issue.
> > 
> > If it is required, someone is going to need to provide a working
> > version, as that does not apply cleanly as-is.
> > 
> > thanks,
> > 
> > greg k-h
> 
> Hi Greg,
> 
> I finally got someone to test the mentioned commit, and I (or they) can confirm that ecc9ab4e924b7eb9e2c4a668162aaa1d9d60d08c is indeed the missing commit needed for "drm/amdgpu: read back register after written for VCN v4.0.5".
> 
> I don't know if this will help, but I got everything to apply cleanly by first applying ecc9ab4e924b7eb9e2c4a668162aaa1d9d60d08c, then cherry-picking the mainline version of "drm/amdgpu: read back register after written for VCN v4.0.5" [1] since the one in stable-rc seems to have been adjusted for the stable tree and made it conflict.

I don't see the conflict that happened here, as I only apply patches
"cleanly".  But I trust you that things are odd here.

> -- 
> Regards,
>   Eric
> 
> [1] ee7360fc27d6045510f8fe459b5649b2af27811a

Ick.  Ok, let me go drop this commit from the tree now, and then can you
resubmit both of these as a tested series for them to be included
properly?

I'll do the same for both 6.12.y and 6.14.y now.

thanks,

greg k-h