Select Git revision
Search by author
- Any Author
- authors
-
Angus Ainslie
angus.ainslie
-
Eric Kuzmenko
eric.kuzmenko
-
Evangelos Ribeiro Tzaras
evangelos.tzaras
-
Francois Techene
francois.techene
-
Guido Gunther
guido.gunther
-
JC Staudt
jcstaudt
-
Jonathon Hall
jonathon.hall
-
Julian Sparber
julian.sparber
-
Martin Kepplinger
martin.kepplinger
-
Mohammed Sadiq
mohammed.sadiq
-
Richard Kolla
richard.kolla
-
Sam Hewitt sam.hewitt
-
Sebastian Krzyszkowiak
sebastian.krzyszkowiak
- May 12, 2022
-
-
Mark Brown authored
commit 2e3a0d1b upstream. The AIU ACODEC has a custom put() operation which returns 0 when the value of the mux changes, meaning that events are not generated for userspace. Change to return 1 in this case, the function returns early in the case where there is no change. Signed-off-by:
-
Mark Brown authored
commit b4f5c6b2 upstream. The WM8958 DSP controls all return 0 on successful write, not a boolean value indicating if the write changed the value of the control. Fix this by returning 1 after a change, there is already a check at the start of each put() that skips the function in the case that there is no change. Signed-off-by:
-
ChiYuan Huang authored
commit 87c18514 upstream. Correct the reg 0x09 size to one byte. Signed-off-by:
-
Mark Brown authored
commit 08ef4840 upstream. The tone generator frequency control just returns 0 on successful write, not a boolean value indicating if there was a change or not. Compare what was written with the value that was there previously so that notifications are generated appropriately when the value changes. Signed-off-by:
-
Thomas Pfaff authored
commit 8707898e upstream. A kernel hang can be observed when running setserial in a loop on a kernel with force threaded interrupts. The sequence of events is: setserial open("/dev/ttyXXX") request_irq() do_stuff() -> serial interrupt -> wake(irq_thread) desc->threads_active++; close() free_irq() kthread_stop(irq_thread) synchronize_irq() <- hangs because desc->threads_active != 0 The thread is created in request_irq() and woken up, but does not get on a CPU to reach the actual thread function, which would handle the pending wake-up. kthread_stop() sets the should stop condition which makes the thread immediately exit, which in turn leaves the stale threads_active count around. This problem was introduced with commit 519cc865, which addressed a interrupt sharing issue in the PCIe code. Before that commit free_irq() invoked synchronize_irq(), which waits for the hard interrupt handler and also for associated threads to complete. To address the PCIe issue synchronize_irq() was replaced with __synchronize_hardirq(), which only waits for the hard interrupt handler to complete, but not for threaded handlers. This was done under the assumption, that the interrupt thread already reached the thread function and waits for a wake-up, which is guaranteed to be handled before acting on the stop condition. The problematic case, that the thread would not reach the thread function, was obviously overlooked. Make sure that the interrupt thread is really started and reaches thread_fn() before returning from __setup_irq(). This utilizes the existing wait queue in the interrupt descriptor. The wait queue is unused for non-shared interrupts. For shared interrupts the usage might cause a spurious wake-up of a waiter in synchronize_irq() or the completion of a threaded handler might cause a spurious wake-up of the waiter for the ready flag. Both are harmless and have no functional impact. [ tglx: Amended changelog ] Fixes: 519cc865 ("genirq: Synchronize only with single thread on free_irq()") Signed-off-by:
-
Filipe Manana authored
commit 4b73c55f upstream. The compression property only has effect on regular files and directories (so that it's propagated to files and subdirectories created inside a directory). For any other inode type (symlink, fifo, device, socket), it's pointless to set the compression property because it does nothing and ends up unnecessarily wasting leaf space due to the pointless xattr (75 or 76 bytes, depending on the compression value). Symlinks in particular are very common (for example, I have almost 10k symlinks under /etc, /usr and /var alone) and therefore it's worth to avoid wasting leaf space with the compression xattr. For example, the compression property can end up on a symlink or character device implicitly, through inheritance from a parent directory $ mkdir /mnt/testdir $ btrfs property set /mnt/testdir compression lzo $ ln -s yadayada /mnt/testdir/lnk $ mknod /mnt/testdir/dev c 0 0 Or explicitly like this: $ ln -s yadayda /mnt/lnk $ setfattr -h -n btrfs.compression -v lzo /mnt/lnk So skip the compression property on inodes that are neither a regular file nor a directory. CC: stable@vger.kernel.org # 5.4+ Signed-off-by:
-
Chung-Chiang Cheng authored
commit 0e852ab8 upstream. Compression and nodatacow are mutually exclusive. A similar issue was fixed by commit f37c563b ("btrfs: add missing check for nocow and compression inode flags"). Besides ioctl, there is another way to enable/disable/reset compression directly via xattr. The following steps will result in a invalid combination. $ touch bar $ chattr +C bar $ lsattr bar ---------------C-- bar $ setfattr -n btrfs.compression -v zstd bar $ lsattr bar --------c------C-- bar To align with the logic in check_fsflags, nocompress will also be unacceptable after this patch, to prevent mix any compression-related options with nodatacow. $ touch bar $ chattr +C bar $ lsattr bar ---------------C-- bar $ setfattr -n btrfs.compression -v zstd bar setfattr: bar: Invalid argument $ setfattr -n btrfs.compression -v no bar setfattr: bar: Invalid argument When both compression and nodatacow are enabled, then btrfs_run_delalloc_range prefers nodatacow and no compression happens. Reported-by:
-
Chung-Chiang Cheng authored
commit e6f9d696 upstream. inode_can_compress will be used outside of inode.c to check the availability of setting compression flag by xattr. This patch moves this function as an internal helper and renames it to btrfs_inode_can_compress. Reviewed-by:
-
Filipe Manana authored
commit 193b4e83 upstream. We are doing a BUG_ON() if we fail to update an inode after setting (or clearing) a xattr, but there's really no reason to not instead simply abort the transaction and return the error to the caller. This should be a rare error because we have previously reserved enough metadata space to update the inode and the delayed inode should have already been setup, so an -ENOSPC or -ENOMEM, which are the possible errors, are very unlikely to happen. So replace the BUG_ON()s with a transaction abort. CC: stable@vger.kernel.org # 4.9+ Reviewed-by:
-
Qu Wenruo authored
commit 9f73f1ae upstream. [BUG] For a 4K sector sized btrfs with v1 cache enabled and only mounted on systems with 4K page size, if it's mounted on subpage (64K page size) systems, it can cause the following warning on v1 space cache: BTRFS error (device dm-1): csum mismatch on free space cache BTRFS warning (device dm-1): failed to load free space cache for block group 84082688, rebuilding it now Although not a big deal, as kernel can rebuild it without problem, such warning will bother end users, especially if they want to switch the same btrfs seamlessly between different page sized systems. [CAUSE] V1 free space cache is still using fixed PAGE_SIZE for various bitmap, like BITS_PER_BITMAP. Such hard-coded PAGE_SIZE usage will cause various mismatch, from v1 cache size to checksum. Thus kernel will always reject v1 cache with a different PAGE_SIZE with csum mismatch. [FIX] Although we should fix v1 cache, it's already going to be marked deprecated soon. And we have v2 cache based on metadata (which is already fully subpage compatible), and it has almost everything superior than v1 cache. So just force subpage mount to use v2 cache on mount. Reported-by:
-
David Sterba authored
commit 3e1ad196 upstream. The new state allowing device addition with paused balance is not exported to user space so it can't recognize it and actually start the operation. Fixes: efc0e69c ("btrfs: introduce exclusive operation BALANCE_PAUSED state") CC: stable@vger.kernel.org # 5.17 Signed-off-by:
-
Tan Tee Min authored
commit 47f753c1 upstream. Based on DesignWare Ethernet QoS datasheet, we are seeing the limitation of Split Header (SPH) feature is not supported for Ipv4 fragmented packet. This SPH limitation will cause ping failure when the packets size exceed the MTU size. For example, the issue happens once the basic ping packet size is larger than the configured MTU size and the data is lost inside the fragmented packet, replaced by zeros/corrupted values, and leads to ping fail. So, disable the Split Header for Intel platforms. v2: Add fixes tag in commit message. Fixes: 67afd6d1("net: stmmac: Add Split Header support and enable it in XGMAC cores") Cc: <stable@vger.kernel.org> # 5.10.x Suggested-by:
-
Niels Dossche authored
commit a7ecbe92 upstream. card->local_node and card->bm_retries are both always accessed under card->lock. fw_core_handle_bus_reset has a check whose condition depends on card->local_node and whose body writes to card->bm_retries. Both of these accesses are not under card->lock. Move the lock acquiring of card->lock to before this check such that these accesses do happen when card->lock is held. fw_destroy_nodes is called inside the check. Since fw_destroy_nodes already acquires card->lock inside its function body, move this out to the callsites of fw_destroy_nodes. Also add a comment to indicate which locking is necessary when calling fw_destroy_nodes. Cc: <stable@vger.kernel.org> Signed-off-by:
-
Jakob Koschel authored
commit 94239738 upstream. When list_for_each_entry() completes the iteration over the whole list without breaking the loop, the iterator value will be a bogus pointer computed based on the head element. While it is safe to use the pointer to determine if it was computed based on the head element, either with list_entry_is_head() or &pos->member == head, using the iterator variable after the loop should be avoided. In preparation to limit the scope of a list iterator to the list traversal loop, use a dedicated pointer to point to the found element [1]. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1] Cc: <stable@vger.kernel.org> Signed-off-by:
-
Chengfeng Ye authored
commit b7c81f80 upstream. &e->event and e point to the same address, and &e->event could be freed in queue_event. So there is a potential uaf issue if we dereference e after calling queue_event(). Fix this by adding a temporary variable to maintain e->client in advance, this can avoid the potential uaf issue. Cc: <stable@vger.kernel.org> Signed-off-by:
-
Kurt Kanzenbach authored
commit 2c33d775 upstream. Mark the CLOCK_MONOTONIC fast time accessors as notrace. These functions are used in tracing to retrieve timestamps, so they should not recurse. Fixes: 4498e746 ("time: Parametrize all tk_fast_mono users") Fixes: f09cb9a1 ("time: Introduce tk_fast_raw") Reported-by:
-
Trond Myklebust authored
commit a3d0562d upstream. This reverts commit 7073ea87. We must not try to connect the socket while the transport is under construction, because the mechanisms to safely tear it down are not in place. As the code stands, we end up leaking the sockets on a connection error. Reported-by:
-
Zev Weiss authored
commit 08da09f0 upstream. CLEAR_FAULTS commands can apparently sometimes trigger catastrophic power output glitches on the ahe-50dc, so block them from being sent at all. Signed-off-by:
-
Nick Kossifidis authored
commit c6fe8119 upstream. In case the DTB provided by the bootloader/BootROM is before the kernel image or outside /memory, we won't be able to access it through the linear mapping, and get a segfault on setup_arch(). Currently OpenSBI relocates DTB but that's not always the case (e.g. if FW_JUMP_FDT_ADDR is not specified), and it's also not the most portable approach since the default FW_JUMP_FDT_ADDR of the generic platform relocates the DTB at a specific offset that may not be available. To avoid this situation copy DTB so that it's visible through the linear mapping. Signed-off-by:
-
Marek Marczykowski-Górecki authored
commit 19965d82 upstream. While technically Xen dom0 is a virtual machine too, it does have access to most of the hardware so it doesn't need to be considered a "passthrough". Commit b818a5d3 ("drm/amdgpu/gmc: use PCI BARs for APUs in passthrough") changed how FB is accessed based on passthrough mode. This breaks amdgpu in Xen dom0 with message like this: [drm:dc_dmub_srv_wait_idle [amdgpu]] *ERROR* Error waiting for DMUB idle: status=3 While the reason for this failure is unclear, the passthrough mode is not really necessary in Xen dom0 anyway. So, to unbreak booting affected kernels, disable passthrough mode in this case. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1985 Fixes: b818a5d3 ("drm/amdgpu/gmc: use PCI BARs for APUs in passthrough") Signed-off-by:
-
Harry Wentland authored
commit 3dfe85fa upstream. A faulty receiver might report an erroneous channel count. We should guard against reading beyond AUDIO_CHANNELS_COUNT as that would overflow the dpcd_pattern_period array. Signed-off-by:
-
Nicolin Chen authored
commit 95d4782c upstream. The arm_smmu_mm_invalidate_range function is designed to be called by mm core for Shared Virtual Addressing purpose between IOMMU and CPU MMU. However, the ways of two subsystems defining their "end" addresses are slightly different. IOMMU defines its "end" address using the last address of an address range, while mm core defines that using the following address of an address range: include/linux/mm_types.h: unsigned long vm_end; /* The first byte after our end address ... This mismatch resulted in an incorrect calculation for size so it failed to be page-size aligned. Further, it caused a dead loop at "while (iova < end)" check in __arm_smmu_tlb_inv_range function. This patch fixes the issue by doing the calculation correctly. Fixes: 2f7e8c55 ("iommu/arm-smmu-v3: Hook up ATC invalidation to mm ops") Cc: stable@vger.kernel.org Signed-off-by:
-
David Stevens authored
commit 59bf3557 upstream. Calculate the appropriate mask for non-size-aligned page selective invalidation. Since psi uses the mask value to mask out the lower order bits of the target address, properly flushing the iotlb requires using a mask value such that [pfn, pfn+pages) all lie within the flushed size-aligned region. This is not normally an issue because iova.c always allocates iovas that are aligned to their size. However, iovas which come from other sources (e.g. userspace via VFIO) may not be aligned. To properly flush the IOTLB, both the start and end pfns need to be equal after applying the mask. That means that the most efficient mask to use is the index of the lowest bit that is equal where all higher bits are also equal. For example, if pfn=0x17f and pages=3, then end_pfn=0x181, so the smallest mask we can use is 8. Any differences above the highest bit of pages are due to carrying, so by xnor'ing pfn and end_pfn and then masking out the lower order bits based on pages, we get 0xffffff00, where the first set bit is the mask we want to use. Fixes: 6fe1010d ("vfio/type1: DMA unmap chunking") Cc: stable@vger.kernel.org Signed-off-by:
-
Kyle Huey authored
commit 5eb84932 upstream. Zen renumbered some of the performance counters that correspond to the well known events in perf_hw_id. This code in KVM was never updated for that, so guest that attempt to use counters on Zen that correspond to the pre-Zen perf_hw_id values will silently receive the wrong values. This has been observed in the wild with rr[0] when running in Zen 3 guests. rr uses the retired conditional branch counter 00d1 which is incorrectly recognized by KVM as PERF_COUNT_HW_STALLED_CYCLES_BACKEND. [0] https://rr-project.org/ Signed-off-by:
-
Thomas Gleixner authored
commit 59f5ede3 upstream. The FPU usage related to task FPU management is either protected by disabling interrupts (switch_to, return to user) or via fpregs_lock() which is a wrapper around local_bh_disable(). When kernel code wants to use the FPU then it has to check whether it is possible by calling irq_fpu_usable(). But the condition in irq_fpu_usable() is wrong. It allows FPU to be used when: !in_interrupt() || interrupted_user_mode() || interrupted_kernel_fpu_idle() The latter is checking whether some other context already uses FPU in the kernel, but if that's not the case then it allows FPU to be used unconditionally even if the calling context interrupted a fpregs_lock() critical region. If that happens then the FPU state of the interrupted context becomes corrupted. Allow in kernel FPU usage only when no other context has in kernel FPU usage and either the calling context is not hard interrupt context or the hard interrupt did not interrupt a local bottomhalf disabled region. It's hard to find a proper Fixes tag as the condition was broken in one way or the other for a very long time and the eager/lazy FPU changes caused a lot of churn. Picked something remotely connected from the history. This survived undetected for quite some time as FPU usage in interrupt context is rare, but the recent changes to the random code unearthed it at least on a kernel which had FPU debugging enabled. There is probably a higher rate of silent corruption as not all issues can be detected by the FPU debugging code. This will be addressed in a subsequent change. Fixes: 5d2bd700 ("x86, fpu: decouple non-lazy/eager fpu restore from xsave") Reported-by:
-
Andrei Lalaev authored
commit e75f88ef upstream. Gpiolib interprets the elements of "gpio-reserved-ranges" as "start,size" because it clears "size" bits starting from the "start" bit in the according bitmap. So it has to use "greater" instead of "greater or equal" when performs bounds check to make sure that GPIOs are in the available range. Previous implementation skipped ranges that include the last GPIO in the range. I wrote the mail to the maintainers (https://lore.kernel.org/linux-gpio/20220412115554.159435-1-andrei.lalaev@emlid.com/T/#u ) of the questioned DTSes (because I couldn't understand how the maintainers interpreted this property), but I haven't received a response. Since the questioned DTSes use "gpio-reserved-ranges = <0 4>" (i.e., the beginning of the range), this patch doesn't affect these DTSes at all. TBH this patch doesn't break any existing DTSes because none of them reserve gpios at the end of range. Fixes: 726cb3ba ("gpiolib: Support 'gpio-reserved-ranges' property") Signed-off-by:
-
Brian Norris authored
commit 4bc31ede upstream. Way back in commit 4f25580f ("mmc: core: changes frequency to hs_max_dtr when selecting hs400es"), Rockchip engineers noticed that some eMMC don't respond to SEND_STATUS commands very reliably if they're still running at a low initial frequency. As mentioned in that commit, JESD84-B51 P49 suggests a sequence in which the host: 1. sets HS_TIMING 2. bumps the clock ("<= 52 MHz") 3. sends further commands It doesn't exactly require that we don't use a lower-than-52MHz frequency, but in practice, these eMMC don't like it. The aforementioned commit tried to get that right for HS400ES, although it's unclear whether this ever truly worked as committed into mainline, as other changes/refactoring adjusted the sequence in conflicting ways: 08573eaf ("mmc: mmc: do not use CMD13 to get status after speed mode switch") 53e60650 ("mmc: core: Allow CMD13 polling when switching to HS mode for mmc") In any case, today we do step 3 before step 2. Let's fix that, and also apply the same logic to HS200/400, where this eMMC has problems too. Resolves errors like this seen when booting some RK3399 Gru/Scarlet systems: [ 2.058881] mmc1: CQHCI version 5.10 [ 2.097545] mmc1: SDHCI controller on fe330000.mmc [fe330000.mmc] using ADMA [ 2.209804] mmc1: mmc_select_hs400es failed, error -84 [ 2.215597] mmc1: error -84 whilst initialising MMC card [ 2.417514] mmc1: mmc_select_hs400es failed, error -110 [ 2.423373] mmc1: error -110 whilst initialising MMC card [ 2.605052] mmc1: mmc_select_hs400es failed, error -110 [ 2.617944] mmc1: error -110 whilst initialising MMC card [ 2.835884] mmc1: mmc_select_hs400es failed, error -110 [ 2.841751] mmc1: error -110 whilst initialising MMC card Ealier versions of this patch bumped to 200MHz/HS200 speeds too early, which caused issues on, e.g., qcom-msm8974-fairphone-fp2. (Thanks for the report Luca!) After a second look, it appears that aligns with JESD84 / page 45 / table 28, so we need to keep to lower (HS / 52 MHz) rates first. Fixes: 08573eaf ("mmc: mmc: do not use CMD13 to get status after speed mode switch") Fixes: 53e60650 ("mmc: core: Allow CMD13 polling when switching to HS mode for mmc") Fixes: 4f25580f ("mmc: core: changes frequency to hs_max_dtr when selecting hs400es") Cc: Shawn Lin <shawn.lin@rock-chips.com> Link: https://lore.kernel.org/linux-mmc/11962455.O9o76ZdvQC@g550jk/ Reported-by:
-
Samuel Holland authored
commit e9f3fb52 upstream. Newer variants of the MMC controller support a 34-bit physical address space by using word addresses instead of byte addresses. However, the code truncates the DMA descriptor address to 32 bits before applying the shift. This breaks DMA for descriptors allocated above the 32-bit limit. Fixes: 3536b82e ("mmc: sunxi: add support for A100 mmc controller") Signed-off-by:
-
Shaik Sajida Bhanu authored
commit 3e5a8e84 upstream. Reset GCC_SDCC_BCR register before every fresh initilazation. This will reset whole SDHC-msm controller, clears the previous power control states and avoids, software reset timeout issues as below. [ 5.458061][ T262] mmc1: Reset 0x1 never completed. [ 5.462454][ T262] mmc1: sdhci: ============ SDHCI REGISTER DUMP =========== [ 5.469065][ T262] mmc1: sdhci: Sys addr: 0x00000000 | Version: 0x00007202 [ 5.475688][ T262] mmc1: sdhci: Blk size: 0x00000000 | Blk cnt: 0x00000000 [ 5.482315][ T262] mmc1: sdhci: Argument: 0x00000000 | Trn mode: 0x00000000 [ 5.488927][ T262] mmc1: sdhci: Present: 0x01f800f0 | Host ctl: 0x00000000 [ 5.495539][ T262] mmc1: sdhci: Power: 0x00000000 | Blk gap: 0x00000000 [ 5.502162][ T262] mmc1: sdhci: Wake-up: 0x00000000 | Clock: 0x00000003 [ 5.508768][ T262] mmc1: sdhci: Timeout: 0x00000000 | Int stat: 0x00000000 [ 5.515381][ T262] mmc1: sdhci: Int enab: 0x00000000 | Sig enab: 0x00000000 [ 5.521996][ T262] mmc1: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000 [ 5.528607][ T262] mmc1: sdhci: Caps: 0x362dc8b2 | Caps_1: 0x0000808f [ 5.535227][ T262] mmc1: sdhci: Cmd: 0x00000000 | Max curr: 0x00000000 [ 5.541841][ T262] mmc1: sdhci: Resp[0]: 0x00000000 | Resp[1]: 0x00000000 [ 5.548454][ T262] mmc1: sdhci: Resp[2]: 0x00000000 | Resp[3]: 0x00000000 [ 5.555079][ T262] mmc1: sdhci: Host ctl2: 0x00000000 [ 5.559651][ T262] mmc1: sdhci_msm: ----------- VENDOR REGISTER DUMP----------- [ 5.566621][ T262] mmc1: sdhci_msm: DLL sts: 0x00000000 | DLL cfg: 0x6000642c | DLL cfg2: 0x0020a000 [ 5.575465][ T262] mmc1: sdhci_msm: DLL cfg3: 0x00000000 | DLL usr ctl: 0x00010800 | DDR cfg: 0x80040873 [ 5.584658][ T262] mmc1: sdhci_msm: Vndr func: 0x00018a9c | Vndr func2 : 0xf88218a8 Vndr func3: 0x02626040 Fixes: 0eb0d9f4 ("mmc: sdhci-msm: Initial support for Qualcomm chipsets") Signed-off-by:
-
Takashi Sakamoto authored
commit eb9d84b0 upstream. ALSA fireworks driver has a bug in its initial state to return count shorter than expected by 4 bytes to userspace applications when handling response frame for Echo Audio Fireworks transaction. It's due to missing addition of the size for the type of event in ALSA firewire stack. Fixes: 555e8a8f ("ALSA: fireworks: Add command/response functionality into hwdep interface") Cc: <stable@vger.kernel.org> Signed-off-by:
-
Zihao Wang authored
commit 3b79954f upstream. Lenovo Yoga Duet 7 13ITL6 has Realtek ALC287 and built-in speakers do not work out of the box. The fix developed for Yoga 7i 14ITL5 also enables speaker output for this model. Signed-off-by:
-
Helge Deller authored
commit 5b89966b upstream. The Linux tool "lscpu" shows the double amount of CPUs if we have "model" and "model name" in two different lines in /proc/cpuinfo. This change combines the model and the model name into one line. Signed-off-by:
-
Helge Deller authored
commit 7962c089 upstream. This reverts commit d97180ad. It triggers RCU stalls at boot with a 32-bit kernel. Signed-off-by:
-
Helge Deller authored
commit 9dc4241b upstream. This reverts commit afdb4a5b. It triggers RCU stalls at boot with a 32-bit kernel. Signed-off-by:
-
Maciej W. Rozycki authored
commit f0a6c68f upstream. Fix the discrepancy between the two places we check for the CP0 counter erratum in along with the incorrect comparison of the R4400 revision number against 0x30 which matches none and consistently consider all R4000 and R4400 processors affected, as documented in processor errata publications[1][2][3], following the mapping between CP0 PRId register values and processor models: PRId | Processor Model ---------+-------------------- 00000422 | R4000 Revision 2.2 00000430 | R4000 Revision 3.0 00000440 | R4400 Revision 1.0 00000450 | R4400 Revision 2.0 00000460 | R4400 Revision 3.0 No other revision of either processor has ever been spotted. Contrary to what has been stated in commit ce202cbb ("[MIPS] Assume R4000/R4400 newer than 3.0 don't have the mfc0 count bug") marking the CP0 counter as buggy does not preclude it from being used as either a clock event or a clock source device. It just cannot be used as both at a time, because in that case clock event interrupts will be occasionally lost, and the use as a clock event device takes precedence. Compare against 0x4ff in `can_use_mips_counter' so that a single machine instruction is produced. References: [1] "MIPS R4000PC/SC Errata, Processor Revision 2.2 and 3.0", MIPS Technologies Inc., May 10, 1994, Erratum 53, p.13 [2] "MIPS R4400PC/SC Errata, Processor Revision 1.0", MIPS Technologies Inc., February 9, 1994, Erratum 21, p.4 [3] "MIPS R4400PC/SC Errata, Processor Revision 2.0 & 3.0", MIPS Technologies Inc., January 24, 1995, Erratum 14, p.3 Signed-off-by:
-
Corey Minyard authored
commit 9cc3aac4 upstream. KASAN report null-ptr-deref as follows: KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014 RIP: 0010:ipmi_unregister_smi+0x7d/0xd50 drivers/char/ipmi/ipmi_msghandler.c:3680 Call Trace: ipmi_ipmb_remove+0x138/0x1a0 drivers/char/ipmi/ipmi_ipmb.c:443 ipmi_ipmb_probe+0x409/0xda1 drivers/char/ipmi/ipmi_ipmb.c:548 i2c_device_probe+0x959/0xac0 drivers/i2c/i2c-core-base.c:563 really_probe+0x3f3/0xa70 drivers/base/dd.c:541 In ipmi_ipmb_probe(), 'iidev->intf' is not set before ipmi_register_smi() success. And in the error handling case, ipmi_ipmb_remove() is called to release resources, ipmi_unregister_smi() is called without check 'iidev->intf', this will cause KASAN null-ptr-deref issue. General kernel style is to allow NULL to be passed into unregister calls, so fix it that way. This allows a NULL check to be removed in other code. Fixes: 57c9e3c9 ("ipmi:ipmi_ipmb: Unregister the SMI on remove") Reported-by:
-
Corey Minyard authored
commit 3d092ef0 upstream. A chunk was dropped when the code handling send messages was rewritten. Those messages shouldn't be processed normally, they are just an indication that the message was successfully sent and the timers should be started for the real response that should be coming later. Add back in the missing chunk to just discard the message and go on. Fixes: 059747c2 ("ipmi: Add support for IPMB direct messages") Reported-by:
-
Thomas Gleixner authored
pci_irq_vector() can't be used in atomic context any longer. This conflicts with the usage of this function in nic_mbx_intr_handler(). age of this function in nic_mbx_intr_handler(). commit 6b292a04 upstream. Cache the Linux interrupt numbers in struct nicpf and use that cache in the interrupt handler to select the mailbox. Fixes: 495c66ac ("genirq/msi: Convert to new functions") Reported-by:
-
- May 09, 2022
-
-
Greg Kroah-Hartman authored
Link: https://lore.kernel.org/r/20220504153110.096069935@linuxfoundation.org Tested-by:
-
Alexey Kardashevskiy authored
commit d7997691 upstream. When ld detects unaligned relocations, it emits R_PPC64_UADDR64 relocations instead of R_PPC64_RELATIVE. Currently R_PPC64_UADDR64 are detected by arch/powerpc/tools/relocs_check.sh and expected not to work. Below is a simple chunk to trigger this behaviour (this disables optimization for the demonstration purposes only, this also happens with -O1/-O2 when CONFIG_PRINTK_INDEX=y, for example): \#pragma GCC push_options \#pragma GCC optimize ("O0") struct entry { const char *file; int line; } __attribute__((packed)); static const struct entry e1 = { .file = __FILE__, .line = __LINE__ }; static const struct entry e2 = { .file = __FILE__, .line = __LINE__ }; ... prom_printf("e1=%s %lx %lx\n", e1.file, (unsigned long) e1.file, mfmsr()); prom_printf("e2=%s %lx\n", e2.file, (unsigned long) e2.file); \#pragma GCC pop_options This adds support for UADDR64 for 64bit. This reuses __dynamic_symtab from the 32bit code which supports more relocation types already. Because RELACOUNT includes only R_PPC64_RELATIVE, this replaces it with RELASZ which is the size of all relocation records. Signed-off-by:
-