As of 07940c369a ("PCI: dwc: Fix MSI page leakage in suspend/resume"),
the PCIe designware host driver has been using the driver data allocation
for the msi_msg DMA mapping which can result in a DMA_MAPPING_ERROR due to
the DMA overflow check in dma_direct_map_page() when the address is greater
than 32 bits (reported in [1]). The commit was trying to address a memory
leak on suspend/resume by moving the MSI mapping to dw_pcie_host_init(),
but subsequently dropped the page allocation thinking it wasn't needed.
To fix the DMA mapping issue as well as make msi_msg DMA'able, switch back
to allocating a 32-bit page for the msi_msg. To avoid the suspend/resume
leak, allocate the page in dw_pcie_host_init() since that shouldn't be
called during suspend/resume.
[1] https://lore.kernel.org/all/Yo0soniFborDl7+C@google.com/
Signed-off-by: Will McVicker <willmcvicker@google.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rob Herring <robh@kernel.org>
The DWC PCIe RC/EP/DM IP core configuration parameters determine the number
of inbound and outbound iATU windows, alignment requirements (which is also
the minimum window size), minimum and maximum sizes. If internal ATU is
enabled, the former settings are determined by CX_ATU_MIN_REGION_SIZE; the
latter are determined by CX_ATU_MAX_REGION_SIZE.
Determine the required alignment and maximum size supported by the
controller and log it to help verify whether the requested inbound or
outbound memory mappings can be fully created.
Note 1. The extended iATU regions have been supported since DWC PCIe
v4.60a. There is no need in testing the upper limit register availability
for the older cores.
Note 2. The regions alignment is determined with using the fls() method
since the lower four bits of the ATU Limit register can be occupied with
the Circular Buffer Increment setting, which can be initialized with zeros.
Link: https://lore.kernel.org/r/20220624143947.8991-13-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Previously __dw_pcie_prog_outbound_atu() duplicated a lot of code between
the iatu_unroll_enabled version and the PCIE_ATU_VIEWPORT version:
__dw_pcie_prog_outbound_atu
if (iatu_unroll_enabled)
dw_pcie_prog_outbound_atu_unroll
dw_pcie_writel_ob_unroll(PCIE_ATU_UNR_LOWER_BASE, ...)
dw_pcie_writel_ob_unroll(PCIE_ATU_UNR_UPPER_BASE, ...)
...
return
dw_pcie_writel_dbi(PCIE_ATU_VIEWPORT, ...)
dw_pcie_writel_dbi(PCIE_ATU_LOWER_BASE, ...)
dw_pcie_writel_dbi(PCIE_ATU_UPPER_BASE, ...)
...
Unify those by pushing the unroll address computation and viewport
selection down into dw_pcie_writel_atu() so we can use the same
dw_pcie_writel_atu_ob() accessor for both paths:
__dw_pcie_prog_outbound_atu
dw_pcie_writel_atu_ob(PCIE_ATU_LOWER_BASE, ...)
dw_pcie_writel_atu
dw_pcie_select_atu # new
if (iatu_unroll_enabled)
return pci->atu_base + PCIE_ATU_UNROLL_BASE(...)
dw_pcie_writel_dbi(PCIE_ATU_VIEWPORT, ...)
return pci->atu_base
dw_pcie_write(base + reg)
dw_pcie_writel_atu_ob(PCIE_ATU_UPPER_BASE, ...)
...
In the non-unroll case, this does involve more MMIO writes to
PCIE_ATU_VIEWPORT, but it's mainly in initialization paths and the code
simplification is significant.
[bhelgaas: commit log, simplify dw_pcie_select_atu()]
Link: https://lore.kernel.org/r/20220624143947.8991-12-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Previously callers of dw_pcie_disable_atu() supplied enum
dw_pcie_region_type (DW_PCIE_REGION_INBOUND, DW_PCIE_REGION_OUTBOUND),
which dw_pcie_disable_atu() converted to the PCIE_ATU_REGION_DIR_IB or
PCIE_ATU_REGION_DIR_OB values needed to program the ATU registers.
Simplify the code by dropping the dw_pcie_region_type enum and passing
PCIE_ATU_REGION_DIR_IB or PCIE_ATU_REGION_DIR_OB directly.
Reorder dw_pcie_disable_atu() arguments to (dir, index) since "index"
indicates an ATU window in the regions of the corresponding direction.
[bhelgaas: commit log]
Link: https://lore.kernel.org/r/20220624143947.8991-11-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Previously dw_pcie_ep_set_bar() converted the BAR PCI_BASE_ADDRESS_SPACE
bit to the internal dw_pcie_as_type enum (DW_PCIE_AS_MEM, DW_PCIE_AS_IO)
and passed it down to dw_pcie_prog_inbound_atu(), which converted the enum
to the PCIE_ATU_TYPE_MEM/PCIE_ATU_TYPE_IO values needed to program the ATU
registers.
Simplify the code by dropping the dw_pcie_as_type enum and passing
PCIE_ATU_TYPE_MEM or PCIE_ATU_TYPE_IO directly.
Reorder inbound ATU function arguments to match the outbound functions,
with address-related parameters at the end.
[bhelgaas: commit log]
Link: https://lore.kernel.org/r/20220624143947.8991-10-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Since DWC PCIe v4.70a, the controller version and version type can be read
from the PORT_LOGIC.PCIE_VERSION_OFF and PORT_LOGIC.PCIE_VERSION_TYPE_OFF
registers respectively.
Read the version from those registers and warn if if's different from the
version we got from the device tree.
We can only read the version after platform-specific drivers have done any
DBI-related initialization, such as reference clock activation.
[bhelgaas: commit log]
Link: https://lore.kernel.org/r/20220624143947.8991-5-Sergey.Semin@baikalelectronics.ru
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Pull irq updates from Thomas Gleixner:
"Updates for interrupt core and drivers:
Core:
- Fix a few inconsistencies between UP and SMP vs interrupt
affinities
- Small updates and cleanups all over the place
New drivers:
- LoongArch interrupt controller
- Renesas RZ/G2L interrupt controller
Updates:
- Hotpath optimization for SiFive PLIC
- Workaround for broken PLIC edge triggered interrupts
- Simall cleanups and improvements as usual"
* tag 'irq-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
irqchip/mmp: Declare init functions in common header file
irqchip/mips-gic: Check the return value of ioremap() in gic_of_init()
genirq: Use for_each_action_of_desc in actions_show()
irqchip / ACPI: Introduce ACPI_IRQ_MODEL_LPIC for LoongArch
irqchip: Add LoongArch CPU interrupt controller support
irqchip: Add Loongson Extended I/O interrupt controller support
irqchip/loongson-liointc: Add ACPI init support
irqchip/loongson-pch-msi: Add ACPI init support
irqchip/loongson-pch-pic: Add ACPI init support
irqchip: Add Loongson PCH LPC controller support
LoongArch: Prepare to support multiple pch-pic and pch-msi irqdomain
LoongArch: Use ACPI_GENERIC_GSI for gsi handling
genirq/generic_chip: Export irq_unmap_generic_chip
ACPI: irq: Allow acpi_gsi_to_irq() to have an arch-specific fallback
APCI: irq: Add support for multiple GSI domains
LoongArch: Provisionally add ACPICA data structures
irqdomain: Use hwirq_max instead of revmap_size for NOMAP domains
irqdomain: Report irq number for NOMAP domains
irqchip/gic-v3: Fix comment typo
dt-bindings: interrupt-controller: renesas,rzg2l-irqc: Document RZ/V2L SoC
...
Pull timer updates from Thomas Gleixner:
"Timers, timekeeping and related drivers update:
Core:
- Make wait_event_hrtimeout() aware of RT/DL tasks
New drivers:
- R-Car Gen4 timer
- Tegra186 timer
- Mediatek MT6795 CPUXGPT timer
Updates:
- Rework suspend/resume handling in timer drivers so it
takes inactive clocks into account.
- The usual device tree compatible add ons
- Small fixed and cleanups all over the place"
* tag 'timers-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (24 commits)
wait: Fix __wait_event_hrtimeout for RT/DL tasks
clocksource/drivers/sun5i: Remove unnecessary (void*) conversions
dt-bindings: timer: allwinner,sun4i-a10-timer: Add D1 compatible
dt-bindings: timer: ingenic,tcu: use absolute path to other schema
clocksource/drivers/sun4i: Remove unnecessary (void*) conversions
dt-bindings: timer: renesas,cmt: Fix R-Car Gen4 fall-out
clocksource/drivers/tegra186: Put Kconfig option 'tristate' to 'bool'
clocksource/drivers/timer-ti-dm: Make driver selection bool for TI K3
clocksource/drivers/timer-ti-dm: Add compatible for am6 SoCs
clocksource/drivers/timer-ti-dm: Make timer selectable for ARCH_K3
clocksource/drivers/timer-ti-dm: Move inline functions to driver for am6
clocksource/drivers/sh_cmt: Add R-Car Gen4 support
dt-bindings: timer: renesas,cmt: R-Car V3U is R-Car Gen4
dt-bindings: timer: renesas,cmt: Add r8a779f0 and generic Gen4 CMT support
clocksource/drivers/timer-microchip-pit64b: Fix compilation warnings
clocksource/drivers/timer-microchip-pit64b: Use mchp_pit64b_{suspend, resume}
clocksource/drivers/timer-microchip-pit64b: Remove suspend/resume ops for ce
thermal/drivers/rcar_gen3_thermal: Add r8a779f0 support
clocksource/drivers/timer-mediatek: Implement CPUXGPT timers
dt-bindings: timer: mediatek: Add CPUX System Timer and MT6795 compatible
...
If a subchannel is busy when a close is performed, the subchannel
needs to be quiesced and left nice and tidy, so nothing unexpected
(like a solicited interrupt) shows up while in the closed state.
Unfortunately, the return code from this call isn't checked,
so any busy subchannel is treated as a failing one.
Fix that, so that the close on a busy subchannel happens normally.
Signed-off-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20220728204914.2420989-4-farman@linux.ibm.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
SPI NOR core changes:
- move SECT_4K_PMC flag out of the core as it's a vendor specific flag
- s/addr_width/addr_nbytes: address width means the number of IO lines
used for the address, whereas in the code it is used as the number of
address bytes.
- do not change nor->addr_nbytes at SFDP parsing time. At the SFDP parsing
time we should not change members of struct spi_nor, but instead fill
members of struct spi_nor_flash_parameters which could later on be used
by the callers.
- track flash's internal address mode so that we can use 4B opcodes
together with opcodes that don't have a 4B opcode correspondent.
SPI NOR manufacturer drivers changes:
- esmt: Rename "f25l32qa" flash name to "f25l32qa-2s".
- micron-st: Skip FSR reading if SPI controller does not support it to
allow flashes that support FSR to work even when attached to such SPI
controllers.
- spansion: Add s25hl-t/s25hs-t IDs and fixups.
Pull perf events updates from Ingo Molnar:
- Fix Intel Alder Lake PEBS memory access latency & data source
profiling info bugs.
- Use Intel large-PEBS hardware feature in more circumstances, to
reduce PMI overhead & reduce sampling data.
- Extend the lost-sample profiling output with the PERF_FORMAT_LOST ABI
variant, which tells tooling the exact number of samples lost.
- Add new IBS register bits definitions.
- AMD uncore events: Add PerfMonV2 DF (Data Fabric) enhancements.
* tag 'perf-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/ibs: Add new IBS register bits into header
perf/x86/intel: Fix PEBS data source encoding for ADL
perf/x86/intel: Fix PEBS memory access info encoding for ADL
perf/core: Add a new read format to get a number of lost samples
perf/x86/amd/uncore: Add PerfMonV2 RDPMC assignments
perf/x86/amd/uncore: Add PerfMonV2 DF event format
perf/x86/amd/uncore: Detect available DF counters
perf/x86/amd/uncore: Use attr_update for format attributes
perf/x86/amd/uncore: Use dynamic events array
x86/events/intel/ds: Enable large PEBS for PERF_SAMPLE_WEIGHT_TYPE
Pull locking updates from Ingo Molnar:
"This was a fairly quiet cycle for the locking subsystem:
- lockdep: Fix a handful of the more complex lockdep_init_map_*()
primitives that can lose the lock_type & cause false reports. No
such mishap was observed in the wild.
- jump_label improvements: simplify the cross-arch support of initial
NOP patching by making it arch-specific code (used on MIPS only),
and remove the s390 initial NOP patching that was superfluous"
* tag 'locking-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/lockdep: Fix lockdep_init_map_*() confusion
jump_label: make initial NOP patching the special case
jump_label: mips: move module NOP patching into arch code
jump_label: s390: avoid pointless initial NOP patching
The "ways" variable comes from the user. The ways_to_cxl() function
has an upper bound but it doesn't check for negatives. Make
the "ways" variable an unsigned int to fix this bug.
Fixes: 80d10a6cee ("cxl/region: Add interleave geometry attributes")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/Yueo3NV2hFCXx1iV@kili
[djbw: fixup interleave_ways_store() to only accept unsigned input]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In the case of sk->dccps_qpolicy == DCCPQ_POLICY_PRIO, dccp_qpolicy_full
will drop a skb when qpolicy is full. And the lock in dccp_sendmsg is
released before sock_alloc_send_skb and then relocked after
sock_alloc_send_skb. The following conditions may lead dccp_qpolicy_push
to add skb to an already full sk_write_queue:
thread1--->lock
thread1--->dccp_qpolicy_full: queue is full. drop a skb
thread1--->unlock
thread2--->lock
thread2--->dccp_qpolicy_full: queue is not full. no need to drop.
thread2--->unlock
thread1--->lock
thread1--->dccp_qpolicy_push: add a skb. queue is full.
thread1--->unlock
thread2--->lock
thread2--->dccp_qpolicy_push: add a skb!
thread2--->unlock
Fix this by moving dccp_qpolicy_full.
Fixes: b1308dc015 ("[DCCP]: Set TX Queue Length Bounds via Sysctl")
Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Link: https://lore.kernel.org/r/20220729110027.40569-1-hbh25y@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The prototype of input features of ionic_set_nic_features() is
netdev_features_t, but the vlan_flags is using the private
definition of ionic drivers. It should use the variable
ctx.cmd.lif_setattr.features, rather than features to check
the vlan flags. So fixes it.
Fixes: beead698b1 ("ionic: Add the basic NDO callbacks for netdev support")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Acked-by: Shannon Nelson <snelson@pensando.io>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
vsi->current_netdev_flags is used store the current net device
flags, not the active netdevice features. So it should use
vsi->netdev->featurs, rather than vsi->current_netdev_flags
to check NETIF_F_HW_VLAN_CTAG_FILTER.
Fixes: 1babaf77f4 ("ice: Advertise 802.1ad VLAN filtering and offloads for PF netdev")
Signed-off-by: Jian Shen <shenjian15@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Acked-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eric Dumazet says:
====================
net: rose: fix module unload issues
Bernard Pidoux reported that unloading rose module could lead
to infamous "unregistered_netdevice:" issues.
First patch is the fix, stable candidate.
Second patch is adding netdev ref tracker to af_rose.
I chose net-next to not inflict merge conflicts, because
Jakub changed dev_put_track() to netdev_put_track() in net-next.
====================
Link: https://lore.kernel.org/r/20220729091233.1030680-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This will help debugging netdevice refcount problems with
CONFIG_NET_DEV_REFCNT_TRACKER=y
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tested-by: Bernard Pidoux <f6bvp@free.fr>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Bernard reported that trying to unload rose module would lead
to infamous messages:
unregistered_netdevice: waiting for rose0 to become free. Usage count = xx
This patch solves the issue, by making sure each socket referring to
a netdevice holds a reference count on it, and properly releases it
in rose_release().
rose_dev_first() is also fixed to take a device reference
before leaving the rcu_read_locked section.
Following patch will add ref_tracker annotations to ease
future bug hunting.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Reported-by: Bernard Pidoux <f6bvp@free.fr>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Tested-by: Bernard Pidoux <f6bvp@free.fr>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pull scheduler updates from Ingo Molnar:
"Load-balancing improvements:
- Improve NUMA balancing on AMD Zen systems for affine workloads.
- Improve the handling of reduced-capacity CPUs in load-balancing.
- Energy Model improvements: fix & refine all the energy fairness
metrics (PELT), and remove the conservative threshold requiring 6%
energy savings to migrate a task. Doing this improves power
efficiency for most workloads, and also increases the reliability
of energy-efficiency scheduling.
- Optimize/tweak select_idle_cpu() to spend (much) less time
searching for an idle CPU on overloaded systems. There's reports of
several milliseconds spent there on large systems with large
workloads ...
[ Since the search logic changed, there might be behavioral side
effects. ]
- Improve NUMA imbalance behavior. On certain systems with spare
capacity, initial placement of tasks is non-deterministic, and such
an artificial placement imbalance can persist for a long time,
hurting (and sometimes helping) performance.
The fix is to make fork-time task placement consistent with runtime
NUMA balancing placement.
Note that some performance regressions were reported against this,
caused by workloads that are not memory bandwith limited, which
benefit from the artificial locality of the placement bug(s). Mel
Gorman's conclusion, with which we concur, was that consistency is
better than random workload benefits from non-deterministic bugs:
"Given there is no crystal ball and it's a tradeoff, I think
it's better to be consistent and use similar logic at both fork
time and runtime even if it doesn't have universal benefit."
- Improve core scheduling by fixing a bug in
sched_core_update_cookie() that caused unnecessary forced idling.
- Improve wakeup-balancing by allowing same-LLC wakeup of idle CPUs
for newly woken tasks.
- Fix a newidle balancing bug that introduced unnecessary wakeup
latencies.
ABI improvements/fixes:
- Do not check capabilities and do not issue capability check denial
messages when a scheduler syscall doesn't require privileges. (Such
as increasing niceness.)
- Add forced-idle accounting to cgroups too.
- Fix/improve the RSEQ ABI to not just silently accept unknown flags.
(No existing tooling is known to have learned to rely on the
previous behavior.)
- Depreciate the (unused) RSEQ_CS_FLAG_NO_RESTART_ON_* flags.
Optimizations:
- Optimize & simplify leaf_cfs_rq_list()
- Micro-optimize set_nr_{and_not,if}_polling() via try_cmpxchg().
Misc fixes & cleanups:
- Fix the RSEQ self-tests on RISC-V and Glibc 2.35 systems.
- Fix a full-NOHZ bug that can in some cases result in the tick not
being re-enabled when the last SCHED_RT task is gone from a
runqueue but there's still SCHED_OTHER tasks around.
- Various PREEMPT_RT related fixes.
- Misc cleanups & smaller fixes"
* tag 'sched-core-2022-08-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)
rseq: Kill process when unknown flags are encountered in ABI structures
rseq: Deprecate RSEQ_CS_FLAG_NO_RESTART_ON_* flags
sched/core: Fix the bug that task won't enqueue into core tree when update cookie
nohz/full, sched/rt: Fix missed tick-reenabling bug in dequeue_task_rt()
sched/core: Always flush pending blk_plug
sched/fair: fix case with reduced capacity CPU
sched/core: Use try_cmpxchg in set_nr_{and_not,if}_polling
sched/core: add forced idle accounting for cgroups
sched/fair: Remove the energy margin in feec()
sched/fair: Remove task_util from effective utilization in feec()
sched/fair: Use the same cpumask per-PD throughout find_energy_efficient_cpu()
sched/fair: Rename select_idle_mask to select_rq_mask
sched, drivers: Remove max param from effective_cpu_util()/sched_cpu_util()
sched/fair: Decay task PELT values during wakeup migration
sched/fair: Provide u64 read for 32-bits arch helper
sched/fair: Introduce SIS_UTIL to search idle CPU based on sum of util_avg
sched: only perform capability check on privileged operation
sched: Remove unused function group_first_cpu()
sched/fair: Remove redundant word " *"
selftests/rseq: check if libc rseq support is registered
...
Pull slab updates from Vlastimil Babka:
- An addition of 'accounted' flag to slab allocation tracepoints to
indicate memcg_kmem accounting, by Vasily
- An optimization of memcg handling in freeing paths, by Muchun
- Various smaller fixes and cleanups
* tag 'slab-for-5.20_or_6.0' of git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab:
mm/slab_common: move generic bulk alloc/free functions to SLOB
mm/sl[au]b: use own bulk free function when bulk alloc failed
mm: slab: optimize memcg_slab_free_hook()
mm/tracing: add 'accounted' entry into output of allocation tracepoints
tools/vm/slabinfo: Handle files in debugfs
mm/slub: Simplify __kmem_cache_alias()
mm, slab: fix bad alignments
