Hulk Robot reported a BUG_ON:
==================================================================
EXT4-fs error (device loop3): ext4_mb_generate_buddy:805: group 0,
block bitmap and bg descriptor inconsistent: 25 vs 31513 free clusters
kernel BUG at fs/ext4/ext4_jbd2.c:53!
invalid opcode: 0000 [#1] SMP KASAN PTI
CPU: 0 PID: 25371 Comm: syz-executor.3 Not tainted 5.10.0+ #1
RIP: 0010:ext4_put_nojournal fs/ext4/ext4_jbd2.c:53 [inline]
RIP: 0010:__ext4_journal_stop+0x10e/0x110 fs/ext4/ext4_jbd2.c:116
[...]
Call Trace:
ext4_write_inline_data_end+0x59a/0x730 fs/ext4/inline.c:795
generic_perform_write+0x279/0x3c0 mm/filemap.c:3344
ext4_buffered_write_iter+0x2e3/0x3d0 fs/ext4/file.c:270
ext4_file_write_iter+0x30a/0x11c0 fs/ext4/file.c:520
do_iter_readv_writev+0x339/0x3c0 fs/read_write.c:732
do_iter_write+0x107/0x430 fs/read_write.c:861
vfs_writev fs/read_write.c:934 [inline]
do_pwritev+0x1e5/0x380 fs/read_write.c:1031
[...]
==================================================================
Above issue may happen as follows:
cpu1 cpu2
__________________________|__________________________
do_pwritev
vfs_writev
do_iter_write
ext4_file_write_iter
ext4_buffered_write_iter
generic_perform_write
ext4_da_write_begin
vfs_fallocate
ext4_fallocate
ext4_convert_inline_data
ext4_convert_inline_data_nolock
ext4_destroy_inline_data_nolock
clear EXT4_STATE_MAY_INLINE_DATA
ext4_map_blocks
ext4_ext_map_blocks
ext4_mb_new_blocks
ext4_mb_regular_allocator
ext4_mb_good_group_nolock
ext4_mb_init_group
ext4_mb_init_cache
ext4_mb_generate_buddy --> error
ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)
ext4_restore_inline_data
set EXT4_STATE_MAY_INLINE_DATA
ext4_block_write_begin
ext4_da_write_end
ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)
ext4_write_inline_data_end
handle=NULL
ext4_journal_stop(handle)
__ext4_journal_stop
ext4_put_nojournal(handle)
ref_cnt = (unsigned long)handle
BUG_ON(ref_cnt == 0) ---> BUG_ON
The lock held by ext4_convert_inline_data is xattr_sem, but the lock
held by generic_perform_write is i_rwsem. Therefore, the two locks can
be concurrent.
To solve above issue, we add inode_lock() for ext4_convert_inline_data().
At the same time, move ext4_convert_inline_data() in front of
ext4_punch_hole(), remove similar handling from ext4_punch_hole().
Fixes: 0c8d414f16 ("ext4: let fallocate handle inline data correctly")
Cc: stable@vger.kernel.org
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220428134031.4153381-1-libaokun1@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Symlink's external data block is one kind of metadata block, and now
that almost all ext4 metadata block's page cache (e.g. directory blocks,
quota blocks...) belongs to bdev backing inode except the symlink. It
is essentially worked in data=journal mode like other regular file's
data block because probably in order to make it simple for generic VFS
code handling symlinks or some other historical reasons, but the logic
of creating external data block in ext4_symlink() is complicated. and it
also make things confused if user do not want to let the filesystem
worked in data=journal mode. This patch convert the final exceptional
case and make things clean, move the mapping of the symlink's external
data block to bdev like any other metadata block does.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20220424140936.1898920-3-yi.zhang@huawei.com
Current ext4_getblk() might sleep if some resources are not valid or
could be race with a concurrent extents modifing procedure. So we
cannot call ext4_getblk() and ext4_map_blocks() to get map blocks in
the atomic context in some fast path (e.g. the upcoming procedure of
getting symlink external block in the RCU context), even if the map
extents have already been check and cached.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Link: https://lore.kernel.org/r/20220424140936.1898920-2-yi.zhang@huawei.com
In __ext4_super() we always overwrote the user specified journal_ioprio
value with a default value, expecting parse_apply_sb_mount_options() to
later correctly set ctx->journal_ioprio to the user specified value.
However, if parse_apply_sb_mount_options() returned early because of
empty sbi->es_s->s_mount_opts, the correct journal_ioprio value was
never set.
This patch fixes __ext4_super() to only use the default value if the
user has not specified any value for journal_ioprio.
Similarly, the remount behavior was to either use journal_ioprio
value specified during initial mount, or use the default value
irrespective of the journal_ioprio value specified during remount.
This patch modifies this to first check if a new value for ioprio
has been passed during remount and apply it. If no new value is
passed, use the value specified during initial mount.
Signed-off-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Reviewed-by: Ritesh Harjani <riteshh@linux.ibm.com>
Tested-by: Ritesh Harjani <riteshh@linux.ibm.com>
Link: https://lore.kernel.org/r/20220418083545.45778-1-ojaswin@linux.ibm.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Otherwise nonaligned fstrim calls will works inconveniently for iterative
scanners, for example:
// trim [0,16MB] for group-1, but mark full group as trimmed
fstrim -o $((1024*1024*128)) -l $((1024*1024*16)) ./m
// handle [16MB,16MB] for group-1, do nothing because group already has the flag.
fstrim -o $((1024*1024*144)) -l $((1024*1024*16)) ./m
[ Update function documentation for ext4_trim_all_free -- TYT ]
Signed-off-by: Dmitry Monakhov <dmtrmonakhov@yandex-team.ru>
Link: https://lore.kernel.org/r/1650214995-860245-1-git-send-email-dmtrmonakhov@yandex-team.ru
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Now that we have kernel message at mount time, system administrator
could acquire the mount time, device and options easily. But we don't
have corresponding unmounting message at umount time, so we cannot know
if someone umount a filesystem easily. Some of the modern filesystems
(e.g. xfs) have the umounting kernel message, so add one for ext4
filesystem for convenience.
EXT4-fs (sdb): mounted filesystem with ordered data mode. Quota mode: none.
EXT4-fs (sdb): unmounting filesystem.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Link: https://lore.kernel.org/r/20220412145320.2669897-1-yi.zhang@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Pull kvm fixes from Paolo Bonzini:
"ARM:
- Take care of faults occuring between the PARange and IPA range by
injecting an exception
- Fix S2 faults taken from a host EL0 in protected mode
- Work around Oops caused by a PMU access from a 32bit guest when PMU
has been created. This is a temporary bodge until we fix it for
good.
x86:
- Fix potential races when walking host page table
- Fix shadow page table leak when KVM runs nested
- Work around bug in userspace when KVM synthesizes leaf 0x80000021
on older (pre-EPYC) or Intel processors
Generic (but affects only RISC-V):
- Fix bad user ABI for KVM_EXIT_SYSTEM_EVENT"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: work around QEMU issue with synthetic CPUID leaves
Revert "x86/mm: Introduce lookup_address_in_mm()"
KVM: x86/mmu: fix potential races when walking host page table
KVM: fix bad user ABI for KVM_EXIT_SYSTEM_EVENT
KVM: x86/mmu: Do not create SPTEs for GFNs that exceed host.MAXPHYADDR
KVM: arm64: Inject exception on out-of-IPA-range translation fault
KVM/arm64: Don't emulate a PMU for 32-bit guests if feature not set
KVM: arm64: Handle host stage-2 faults from 32-bit EL0
Pull x86 fixes from Borislav Petkov:
- A fix to disable PCI/MSI[-X] masking for XEN_HVM guests as that is
solely controlled by the hypervisor
- A build fix to make the function prototype (__warn()) as visible as
the definition itself
- A bunch of objtool annotation fixes which have accumulated over time
- An ORC unwinder fix to handle bad input gracefully
- Well, we thought the microcode gets loaded in time in order to
restore the microcode-emulated MSRs but we thought wrong. So there's
a fix for that to have the ordering done properly
- Add new Intel model numbers
- A spelling fix
* tag 'x86_urgent_for_v5.18_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/pci/xen: Disable PCI/MSI[-X] masking for XEN_HVM guests
bug: Have __warn() prototype defined unconditionally
x86/Kconfig: fix the spelling of 'becoming' in X86_KERNEL_IBT config
objtool: Use offstr() to print address of missing ENDBR
objtool: Print data address for "!ENDBR" data warnings
x86/xen: Add ANNOTATE_NOENDBR to startup_xen()
x86/uaccess: Add ENDBR to __put_user_nocheck*()
x86/retpoline: Add ANNOTATE_NOENDBR for retpolines
x86/static_call: Add ANNOTATE_NOENDBR to static call trampoline
objtool: Enable unreachable warnings for CLANG LTO
x86,objtool: Explicitly mark idtentry_body()s tail REACHABLE
x86,objtool: Mark cpu_startup_entry() __noreturn
x86,xen,objtool: Add UNWIND hint
lib/strn*,objtool: Enforce user_access_begin() rules
MAINTAINERS: Add x86 unwinding entry
x86/unwind/orc: Recheck address range after stack info was updated
x86/cpu: Load microcode during restore_processor_state()
x86/cpu: Add new Alderlake and Raptorlake CPU model numbers
Pull objtool fixes from Borislav Petkov:
"A bunch of objtool fixes to improve unwinding, sibling call detection,
fallthrough detection and relocation handling of weak symbols when the
toolchain strips section symbols"
* tag 'objtool_urgent_for_v5.18_rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Fix code relocs vs weak symbols
objtool: Fix type of reloc::addend
objtool: Fix function fallthrough detection for vmlinux
objtool: Fix sibling call detection in alternatives
objtool: Don't set 'jump_dest' for sibling calls
x86/uaccess: Don't jump between functions
Pull driver core fixes from Greg KH:
"Here are some small driver core and kernfs fixes for some reported
problems. They include:
- kernfs regression that is causing oopses in 5.17 and newer releases
- topology sysfs fixes for a few small reported problems.
All of these have been in linux-next for a while with no reported
issues"
* tag 'driver-core-5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
kernfs: fix NULL dereferencing in kernfs_remove
topology: Fix up build warning in topology_is_visible()
arch_topology: Do not set llc_sibling if llc_id is invalid
topology: make core_mask include at least cluster_siblings
topology/sysfs: Hide PPIN on systems that do not support it.
Pull char/misc driver fixes from Greg KH:
"Here are a small number of char/misc/other driver fixes for 5.18-rc5
Nothing major in here, this is mostly IIO driver fixes along with some
other small things:
- at25 driver fix for systems without a dma-able stack
- phy driver fixes for reported issues
- binder driver fixes for reported issues
All of these have been in linux-next without any reported problems"
* tag 'char-misc-5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (31 commits)
eeprom: at25: Use DMA safe buffers
binder: Gracefully handle BINDER_TYPE_FDA objects with num_fds=0
binder: Address corner cases in deferred copy and fixup
phy: amlogic: fix error path in phy_g12a_usb3_pcie_probe()
iio: imu: inv_icm42600: Fix I2C init possible nack
iio: dac: ltc2688: fix voltage scale read
interconnect: qcom: sdx55: Drop IP0 interconnects
interconnect: qcom: sc7180: Drop IP0 interconnects
phy: ti: Add missing pm_runtime_disable() in serdes_am654_probe
phy: mapphone-mdm6600: Fix PM error handling in phy_mdm6600_probe
phy: ti: omap-usb2: Fix error handling in omap_usb2_enable_clocks
bus: mhi: host: pci_generic: Flush recovery worker during freeze
bus: mhi: host: pci_generic: Add missing poweroff() PM callback
phy: ti: tusb1210: Fix an error handling path in tusb1210_probe()
phy: samsung: exynos5250-sata: fix missing device put in probe error paths
phy: samsung: Fix missing of_node_put() in exynos_sata_phy_probe
phy: ti: Fix missing of_node_put in ti_pipe3_get_sysctrl()
phy: ti: tusb1210: Make tusb1210_chg_det_states static
iio:dac:ad3552r: Fix an IS_ERR() vs NULL check
iio: sx9324: Fix default precharge internal resistance register
...
Pull tty/serial fixes from Greg KH:
"Here are some small serial driver fixes, and a larger number of GSM
line discipline fixes for 5.18-rc5.
These include:
- lots of tiny n_gsm fixes for issues to resolve a number of reported
problems. Seems that people are starting to actually use this code
again.
- 8250 driver fixes for some devices
- imx serial driver fix
- amba-pl011 driver fix
All of these have been in linux-next for a while with no reported
issues"
* tag 'tty-5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty: (27 commits)
tty: n_gsm: fix sometimes uninitialized warning in gsm_dlci_modem_output()
serial: 8250: Correct the clock for EndRun PTP/1588 PCIe device
serial: 8250: Also set sticky MCR bits in console restoration
tty: n_gsm: fix software flow control handling
tty: n_gsm: fix invalid use of MSC in advanced option
tty: n_gsm: fix broken virtual tty handling
Revert "serial: sc16is7xx: Clear RS485 bits in the shutdown"
tty: n_gsm: fix missing update of modem controls after DLCI open
serial: 8250: Fix runtime PM for start_tx() for empty buffer
serial: imx: fix overrun interrupts in DMA mode
serial: amba-pl011: do not time out prematurely when draining tx fifo
tty: n_gsm: fix incorrect UA handling
tty: n_gsm: fix reset fifo race condition
tty: n_gsm: fix missing tty wakeup in convergence layer type 2
tty: n_gsm: fix wrong signal octets encoding in MSC
tty: n_gsm: fix wrong command frame length field encoding
tty: n_gsm: fix wrong command retry handling
tty: n_gsm: fix missing explicit ldisc flush
tty: n_gsm: fix wrong DLCI release order
tty: n_gsm: fix insufficient txframe size
...
Pull USB fixes from Greg KH:
"Here are a number of small USB driver fixes for 5.18-rc5 for some
reported issues and new quirks. They include:
- dwc3 driver fixes
- xhci driver fixes
- typec driver fixes
- new usb-serial driver ids
- added new USB devices to existing quirk tables
- other tiny fixes
All of these have been in linux-next for a while with no reported
issues"
* tag 'usb-5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (31 commits)
usb: phy: generic: Get the vbus supply
usb: dwc3: gadget: Return proper request status
usb: dwc3: pci: add support for the Intel Meteor Lake-P
usb: dwc3: core: Only handle soft-reset in DCTL
usb: gadget: configfs: clear deactivation flag in configfs_composite_unbind()
usb: misc: eud: Fix an error handling path in eud_probe()
usb: core: Don't hold the device lock while sleeping in do_proc_control()
usb: dwc3: Try usb-role-switch first in dwc3_drd_init
usb: dwc3: core: Fix tx/rx threshold settings
usb: mtu3: fix USB 3.0 dual-role-switch from device to host
xhci: Enable runtime PM on second Alderlake controller
usb: dwc3: fix backwards compat with rockchip devices
dt-bindings: usb: samsung,exynos-usb2: add missing required reg
usb: misc: fix improper handling of refcount in uss720_probe()
USB: Fix ehci infinite suspend-resume loop issue in zhaoxin
usb: typec: tcpm: Fix undefined behavior due to shift overflowing the constant
usb: typec: rt1719: Fix build error without CONFIG_POWER_SUPPLY
usb: typec: ucsi: Fix role swapping
usb: typec: ucsi: Fix reuse of completion structure
usb: xhci: tegra:Fix PM usage reference leak of tegra_xusb_unpowergate_partitions
...
Pull SCSI fix from James Bottomley:
"One fix for an endless error loop with the target driver affecting
tapes"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: target: pscsi: Set SCF_TREAT_READ_AS_NORMAL flag only if there is valid data
Pull ARM SoC fixes from Arnd Bergmann:
- A fix for a regression caused by the previous set of bugfixes
changing tegra and at91 pinctrl properties.
More work is needed to figure out what this should actually be, but a
revert makes it work for the moment.
- Defconfig regression fixes for tegra after renamed symbols
- Build-time warning and static checker fixes for imx, op-tee, sunxi,
meson, at91, and omap
- More at91 DT fixes for audio, regulator and spi nodes
- A regression fix for Renesas Hyperflash memory probe
- A stability fix for amlogic boards, modifying the allowed cpufreq
states
- Multiple fixes for system suspend on omap2+
- DT fixes for various i.MX bugs
- A probe error fix for imx6ull-colibri MMC
- A MAINTAINERS file entry for samsung bug reports
* tag 'soc-fixes-5.18-3' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (42 commits)
Revert "arm: dts: at91: Fix boolean properties with values"
bus: sunxi-rsb: Fix the return value of sunxi_rsb_device_create()
Revert "arm64: dts: tegra: Fix boolean properties with values"
arm64: dts: imx8mn-ddr4-evk: Describe the 32.768 kHz PMIC clock
ARM: dts: imx6ull-colibri: fix vqmmc regulator
MAINTAINERS: add Bug entry for Samsung and memory controller drivers
memory: renesas-rpc-if: Fix HF/OSPI data transfer in Manual Mode
ARM: dts: logicpd-som-lv: Fix wrong pinmuxing on OMAP35
ARM: dts: am3517-evm: Fix misc pinmuxing
ARM: dts: am33xx-l4: Add missing touchscreen clock properties
ARM: dts: Fix mmc order for omap3-gta04
ARM: dts: at91: fix pinctrl phandles
ARM: dts: at91: sama5d4_xplained: fix pinctrl phandle name
ARM: dts: at91: Describe regulators on at91sam9g20ek
ARM: dts: at91: Map MCLK for wm8731 on at91sam9g20ek
ARM: dts: at91: Fix boolean properties with values
ARM: dts: at91: use generic node name for dataflash
ARM: dts: at91: align SPI NOR node name with dtschema
ARM: dts: at91: sama7g5ek: Align the impedance of the QSPI0's HSIO and PCB lines
ARM: dts: at91: sama7g5ek: enable pull-up on flexcom3 console lines
...
Pull clk fixes from Stephen Boyd:
"A semi-large pile of clk driver fixes this time around.
Nothing is touching the core so these fixes are fairly well contained
to specific devices that use these clk drivers.
- Some Allwinner SoC fixes to gracefully handle errors and mark an
RTC clk as critical so that the RTC keeps ticking.
- Fix AXI bus clks and RTC clk design for Microchip PolarFire SoC
driver introduced this cycle. This has some devicetree bits acked
by riscv maintainers. We're fixing it now so that the prior
bindings aren't released in a major kernel version.
- Remove a reset on Microchip PolarFire SoCs that broke when enabling
CONFIG_PM.
- Set a min/max for the Qualcomm graphics clk. This got broken by the
clk rate range patches introduced this cycle"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: sunxi: sun9i-mmc: check return value after calling platform_get_resource()
clk: sunxi-ng: sun6i-rtc: Mark rtc-32k as critical
riscv: dts: microchip: reparent mpfs clocks
clk: microchip: mpfs: add RTCREF clock control
clk: microchip: mpfs: re-parent the configurable clocks
dt-bindings: rtc: add refclk to mpfs-rtc
dt-bindings: clk: mpfs: add defines for two new clocks
dt-bindings: clk: mpfs document msspll dri registers
riscv: dts: microchip: fix usage of fic clocks on mpfs
clk: microchip: mpfs: mark CLK_ATHENA as critical
clk: microchip: mpfs: fix parents for FIC clocks
clk: qcom: clk-rcg2: fix gfx3d frequency calculation
clk: microchip: mpfs: don't reset disabled peripherals
clk: sunxi-ng: fix not NULL terminated coccicheck error
Pull block fixes from Jens Axboe:
- Revert of a patch that caused timestamp issues (Tejun)
- iocost warning fix (Tejun)
- bfq warning fix (Jan)
* tag 'block-5.18-2022-04-29' of git://git.kernel.dk/linux-block:
bfq: Fix warning in bfqq_request_over_limit()
Revert "block: inherit request start time from bio for BLK_CGROUP"
iocost: don't reset the inuse weight of under-weighted debtors
Pull io_uring fixes from Jens Axboe:
"Pretty boring:
- three patches just adding reserved field checks (me, Eugene)
- Fixing a potential regression with IOPOLL caused by a block change
(Joseph)"
Boring is good.
* tag 'io_uring-5.18-2022-04-29' of git://git.kernel.dk/linux-block:
io_uring: check that data field is 0 in ringfd unregister
io_uring: fix uninitialized field in rw io_kiocb
io_uring: check reserved fields for recv/recvmsg
io_uring: check reserved fields for send/sendmsg
Pull random number generator fixes from Jason Donenfeld:
- Eric noticed that the memmove() in crng_fast_key_erasure() was bogus,
so this has been changed to a memcpy() and the confusing situation
clarified with a detailed comment.
- [Half]SipHash documentation updates from Bagas and Eric, after Eric
pointed out that the use of HalfSipHash in random.c made a bit of the
text potentially misleading.
* tag 'random-5.18-rc5-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:
Documentation: siphash: disambiguate HalfSipHash algorithm from hsiphash functions
Documentation: siphash: enclose HalfSipHash usage example in the literal block
Documentation: siphash: convert danger note to warning for HalfSipHash
random: document crng_fast_key_erasure() destination possibility
Pull ceph client fixes from Ilya Dryomov:
"A fix for a NULL dereference that turns out to be easily triggerable
by fsync (marked for stable) and a false positive WARN and snap_rwsem
locking fixups"
* tag 'ceph-for-5.18-rc5' of https://github.com/ceph/ceph-client:
ceph: fix possible NULL pointer dereference for req->r_session
ceph: remove incorrect session state check
ceph: get snap_rwsem read lock in handle_cap_export for ceph_add_cap
libceph: disambiguate cluster/pool full log message
This reverts commit 0dc23d1a8e, which caused another regression
as the pinctrl code actually expects an integer value of 0 or 1
rather than a simple boolean property.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Synthesizing AMD leaves up to 0x80000021 caused problems with QEMU,
which assumes the *host* CPUID[0x80000000].EAX is higher or equal
to what KVM_GET_SUPPORTED_CPUID reports.
This causes QEMU to issue bogus host CPUIDs when preparing the input
to KVM_SET_CPUID2. It can even get into an infinite loop, which is
only terminated by an abort():
cpuid_data is full, no space for cpuid(eax:0x8000001d,ecx:0x3e)
To work around this, only synthesize those leaves if 0x8000001d exists
on the host. The synthetic 0x80000021 leaf is mostly useful on Zen2,
which satisfies the condition.
Fixes: f144c49e8c ("KVM: x86: synthesize CPUID leaf 0x80000021h if useful")
Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull perf tools fixes from Arnaldo Carvalho de Melo:
- Fix Intel PT (Processor Trace) timeless decoding with perf.data
directory.
- ARM SPE (Statistical Profiling Extensions) address fixes, for
synthesized events and for SPE events with physical addresses. Add a
simple 'perf test' entry to make sure this doesn't regress.
- Remove arch specific processing of kallsyms data to fixup symbol end
address, fixing excessive memory consumption in the annotation code.
* tag 'perf-tools-fixes-for-v5.18-2022-04-29' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
perf symbol: Remove arch__symbols__fixup_end()
perf symbol: Update symbols__fixup_end()
perf symbol: Pass is_kallsyms to symbols__fixup_end()
perf test: Add perf_event_attr test for Arm SPE
perf arm-spe: Fix SPE events with phys addresses
perf arm-spe: Fix addresses of synthesized SPE events
perf intel-pt: Fix timeless decoding with perf.data directory
Pull RISC-V fixes from Palmer Dabbelt:
- A fix to properly ensure a single CPU is running during patch_text().
- A defconfig update to include RPMSG_CTRL when RPMSG_CHAR was set,
necessary after a recent refactoring.
* tag 'riscv-for-linus-5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
RISC-V: configs: Configs that had RPMSG_CHAR now get RPMSG_CTRL
riscv: patch_text: Fixup last cpu should be master
Pull arm64 fix from Will Deacon:
"Rename and reallocate the PT_ARM_MEMTAG_MTE ELF segment type.
This is a fix to the MTE ELF ABI for a bug that was added during the
most recent merge window as part of the coredump support.
The issue is that the value assigned to the new PT_ARM_MEMTAG_MTE
segment type has already been allocated to PT_AARCH64_UNWIND by the
ELF ABI, so we've bumped the value and changed the name of the
identifier to be better aligned with the existing one"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
elf: Fix the arm64 MTE ELF segment name and value
Drop lookup_address_in_mm() now that KVM is providing it's own variant
of lookup_address_in_pgd() that is safe for use with user addresses, e.g.
guards against page tables being torn down. A variant that provides a
non-init mm is inherently dangerous and flawed, as the only reason to use
an mm other than init_mm is to walk a userspace mapping, and
lookup_address_in_pgd() does not play nice with userspace mappings, e.g.
doesn't disable IRQs to block TLB shootdowns and doesn't use READ_ONCE()
to ensure an upper level entry isn't converted to a huge page between
checking the PAGE_SIZE bit and grabbing the address of the next level
down.
This reverts commit 13c72c060f.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <YmwIi3bXr/1yhYV/@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fixes for (relatively) old bugs, to be merged in both the -rc and next
development trees:
* Fix potential races when walking host page table
* Fix bad user ABI for KVM_EXIT_SYSTEM_EVENT
* Fix shadow page table leak when KVM runs nested
KVM uses lookup_address_in_mm() to detect the hugepage size that the host
uses to map a pfn. The function suffers from several issues:
- no usage of READ_ONCE(*). This allows multiple dereference of the same
page table entry. The TOCTOU problem because of that may cause KVM to
incorrectly treat a newly generated leaf entry as a nonleaf one, and
dereference the content by using its pfn value.
- the information returned does not match what KVM needs; for non-present
entries it returns the level at which the walk was terminated, as long
as the entry is not 'none'. KVM needs level information of only 'present'
entries, otherwise it may regard a non-present PXE entry as a present
large page mapping.
- the function is not safe for mappings that can be torn down, because it
does not disable IRQs and because it returns a PTE pointer which is never
safe to dereference after the function returns.
So implement the logic for walking host page tables directly in KVM, and
stop using lookup_address_in_mm().
Cc: Sean Christopherson <seanjc@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Mingwei Zhang <mizhang@google.com>
Message-Id: <20220429031757.2042406-1-mizhang@google.com>
[Inline in host_pfn_mapping_level, ensure no semantic change for its
callers. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When KVM_EXIT_SYSTEM_EVENT was introduced, it included a flags
member that at the time was unused. Unfortunately this extensibility
mechanism has several issues:
- x86 is not writing the member, so it would not be possible to use it
on x86 except for new events
- the member is not aligned to 64 bits, so the definition of the
uAPI struct is incorrect for 32- on 64-bit userspace. This is a
problem for RISC-V, which supports CONFIG_KVM_COMPAT, but fortunately
usage of flags was only introduced in 5.18.
Since padding has to be introduced, place a new field in there
that tells if the flags field is valid. To allow further extensibility,
in fact, change flags to an array of 16 values, and store how many
of the values are valid. The availability of the new ndata field
is tied to a system capability; all architectures are changed to
fill in the field.
To avoid breaking compilation of userspace that was using the flags
field, provide a userspace-only union to overlap flags with data[0].
The new field is placed at the same offset for both 32- and 64-bit
userspace.
Cc: Will Deacon <will@kernel.org>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Peter Gonda <pgonda@google.com>
Cc: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reported-by: kernel test robot <lkp@intel.com>
Message-Id: <20220422103013.34832-1-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Disallow memslots and MMIO SPTEs whose gpa range would exceed the host's
MAXPHYADDR, i.e. don't create SPTEs for gfns that exceed host.MAXPHYADDR.
The TDP MMU bounds its zapping based on host.MAXPHYADDR, and so if the
guest, possibly with help from userspace, manages to coerce KVM into
creating a SPTE for an "impossible" gfn, KVM will leak the associated
shadow pages (page tables):
WARNING: CPU: 10 PID: 1122 at arch/x86/kvm/mmu/tdp_mmu.c:57
kvm_mmu_uninit_tdp_mmu+0x4b/0x60 [kvm]
Modules linked in: kvm_intel kvm irqbypass
CPU: 10 PID: 1122 Comm: set_memory_regi Tainted: G W 5.18.0-rc1+ #293
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
RIP: 0010:kvm_mmu_uninit_tdp_mmu+0x4b/0x60 [kvm]
Call Trace:
<TASK>
kvm_arch_destroy_vm+0x130/0x1b0 [kvm]
kvm_destroy_vm+0x162/0x2d0 [kvm]
kvm_vm_release+0x1d/0x30 [kvm]
__fput+0x82/0x240
task_work_run+0x5b/0x90
exit_to_user_mode_prepare+0xd2/0xe0
syscall_exit_to_user_mode+0x1d/0x40
entry_SYSCALL_64_after_hwframe+0x44/0xae
</TASK>
On bare metal, encountering an impossible gpa in the page fault path is
well and truly impossible, barring CPU bugs, as the CPU will signal #PF
during the gva=>gpa translation (or a similar failure when stuffing a
physical address into e.g. the VMCS/VMCB). But if KVM is running as a VM
itself, the MAXPHYADDR enumerated to KVM may not be the actual MAXPHYADDR
of the underlying hardware, in which case the hardware will not fault on
the illegal-from-KVM's-perspective gpa.
Alternatively, KVM could continue allowing the dodgy behavior and simply
zap the max possible range. But, for hosts with MAXPHYADDR < 52, that's
a (minor) waste of cycles, and more importantly, KVM can't reasonably
support impossible memslots when running on bare metal (or with an
accurate MAXPHYADDR as a VM). Note, limiting the overhead by checking if
KVM is running as a guest is not a safe option as the host isn't required
to announce itself to the guest in any way, e.g. doesn't need to set the
HYPERVISOR CPUID bit.
A second alternative to disallowing the memslot behavior would be to
disallow creating a VM with guest.MAXPHYADDR > host.MAXPHYADDR. That
restriction is undesirable as there are legitimate use cases for doing
so, e.g. using the highest host.MAXPHYADDR out of a pool of heterogeneous
systems so that VMs can be migrated between hosts with different
MAXPHYADDRs without running afoul of the allow_smaller_maxphyaddr mess.
Note that any guest.MAXPHYADDR is valid with shadow paging, and it is
even useful in order to test KVM with MAXPHYADDR=52 (i.e. without
any reserved physical address bits).
The now common kvm_mmu_max_gfn() is inclusive instead of exclusive.
The memslot and TDP MMU code want an exclusive value, but the name
implies the returned value is inclusive, and the MMIO path needs an
inclusive check.
Fixes: faaf05b00a ("kvm: x86/mmu: Support zapping SPTEs in the TDP MMU")
Fixes: 524a1e4e38 ("KVM: x86/mmu: Don't leak non-leaf SPTEs when zapping all SPTEs")
Cc: stable@vger.kernel.org
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: Ben Gardon <bgardon@google.com>
Cc: David Matlack <dmatlack@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220428233416.2446833-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM/arm64 fixes for 5.18, take #2
- Take care of faults occuring between the PARange and
IPA range by injecting an exception
- Fix S2 faults taken from a host EL0 in protected mode
- Work around Oops caused by a PMU access from a 32bit
guest when PMU has been created. This is a temporary
bodge until we fix it for good.
ARM: tegra: Default configuration fixes for v5.18
This contains two updates to the default configuration needed because of
a Kconfig symbol name change. This fixes a failure that was detected in
the NVIDIA automated test farm.
* tag 'tegra-for-5.18-arm-defconfig-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
ARM: config: multi v7: Enable NVIDIA Tegra video decoder driver
ARM: tegra_defconfig: Update CONFIG_TEGRA_VDE option
Link: https://lore.kernel.org/r/20220429080626.494150-1-thierry.reding@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fix return value in RSB bus driver
* tag 'sunxi-fixes-for-5.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
bus: sunxi-rsb: Fix the return value of sunxi_rsb_device_create()
Link: https://lore.kernel.org/r/Ymbkd+/dDmRJz66w@kista.localdomain
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
When a XEN_HVM guest uses the XEN PIRQ/Eventchannel mechanism, then
PCI/MSI[-X] masking is solely controlled by the hypervisor, but contrary to
XEN_PV guests this does not disable PCI/MSI[-X] masking in the PCI/MSI
layer.
This can lead to a situation where the PCI/MSI layer masks an MSI[-X]
interrupt and the hypervisor grants the write despite the fact that it
already requested the interrupt. As a consequence interrupt delivery on the
affected device is not happening ever.
Set pci_msi_ignore_mask to prevent that like it's done for XEN_PV guests
already.
Fixes: 809f9267bb ("xen: map MSIs into pirqs")
Reported-by: Jeremi Piotrowski <jpiotrowski@linux.microsoft.com>
Reported-by: Dusty Mabe <dustymabe@redhat.com>
Reported-by: Salvatore Bonaccorso <carnil@debian.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Noah Meyerhans <noahm@debian.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/87tuaduxj5.ffs@tglx
Pull drm fixes from Dave Airlie:
"Another relatively quiet week, amdgpu leads the way, some i915 display
fixes, and a single sunxi fix.
amdgpu:
- Runtime pm fix
- DCN memory leak fix in error path
- SI DPM deadlock fix
- S0ix fix
amdkfd:
- GWS fix
- GWS support for CRIU
i915:
- Fix#5284: Backlight control regression on XMG Core 15 e21
- Fix black display plane on Acer One AO532h
- Two smaller display fixes
sunxi:
- Single fix removing applying PHYS_OFFSET twice"
* tag 'drm-fixes-2022-04-29' of git://anongit.freedesktop.org/drm/drm:
drm/amdgpu: keep mmhub clock gating being enabled during s2idle suspend
drm/amd/pm: fix the deadlock issue observed on SI
drm/amd/display: Fix memory leak in dcn21_clock_source_create
drm/amdgpu: don't runtime suspend if there are displays attached (v3)
drm/amdkfd: CRIU add support for GWS queues
drm/amdkfd: Fix GWS queue count
drm/sun4i: Remove obsolete references to PHYS_OFFSET
drm/i915/fbc: Consult hw.crtc instead of uapi.crtc
drm/i915: Fix SEL_FETCH_PLANE_*(PIPE_B+) register addresses
drm/i915: Check EDID for HDR static metadata when choosing blc
drm/i915: Fix DISP_POS_Y and DISP_HEIGHT defines
