Merge page ref overflow branch.
Jann Horn reported that he can overflow the page ref count with
sufficient memory (and a filesystem that is intentionally extremely
slow).
Admittedly it's not exactly easy. To have more than four billion
references to a page requires a minimum of 32GB of kernel memory just
for the pointers to the pages, much less any metadata to keep track of
those pointers. Jann needed a total of 140GB of memory and a specially
crafted filesystem that leaves all reads pending (in order to not ever
free the page references and just keep adding more).
Still, we have a fairly straightforward way to limit the two obvious
user-controllable sources of page references: direct-IO like page
references gotten through get_user_pages(), and the splice pipe page
duplication. So let's just do that.
* branch page-refs:
fs: prevent page refcount overflow in pipe_buf_get
mm: prevent get_user_pages() from overflowing page refcount
mm: add 'try_get_page()' helper function
mm: make page ref count overflow check tighter and more explicit
Change pipe_buf_get() to return a bool indicating whether it succeeded
in raising the refcount of the page (if the thing in the pipe is a page).
This removes another mechanism for overflowing the page refcount. All
callers converted to handle a failure.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull timer fix from Ingo Molnar:
"Fix the alarm_timer_remaining() return value"
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
alarmtimer: Return correct remaining time
Pull scheduler fix from Ingo Molnar:
"Fix a NULL pointer dereference crash in certain environments"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Do not re-read ->h_load_next during hierarchical load calculation
Pull perf fixes from Ingo Molnar:
"Six kernel side fixes: three related to NMI handling on AMD systems, a
race fix, a kexec initialization fix and a PEBS sampling fix"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/core: Fix perf_event_disable_inatomic() race
x86/perf/amd: Remove need to check "running" bit in NMI handler
x86/perf/amd: Resolve NMI latency issues for active PMCs
x86/perf/amd: Resolve race condition when disabling PMC
perf/x86/intel: Initialize TFA MSR
perf/x86/intel: Fix handling of wakeup_events for multi-entry PEBS
Pull locking fix from Ingo Molnar:
"Fixes a crash when accessing /proc/lockdep"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/lockdep: Zap lock classes even with lock debugging disabled
Thomas-Mich Richter reported he triggered a WARN()ing from event_function_local()
on his s390. The problem boils down to:
CPU-A CPU-B
perf_event_overflow()
perf_event_disable_inatomic()
@pending_disable = 1
irq_work_queue();
sched-out
event_sched_out()
@pending_disable = 0
sched-in
perf_event_overflow()
perf_event_disable_inatomic()
@pending_disable = 1;
irq_work_queue(); // FAILS
irq_work_run()
perf_pending_event()
if (@pending_disable)
perf_event_disable_local(); // WHOOPS
The problem exists in generic, but s390 is particularly sensitive
because it doesn't implement arch_irq_work_raise(), nor does it call
irq_work_run() from it's PMU interrupt handler (nor would that be
sufficient in this case, because s390 also generates
perf_event_overflow() from pmu::stop). Add to that the fact that s390
is a virtual architecture and (virtual) CPU-A can stall long enough
for the above race to happen, even if it would self-IPI.
Adding a irq_work_sync() to event_sched_in() would work for all hardare
PMUs that properly use irq_work_run() but fails for software PMUs.
Instead encode the CPU number in @pending_disable, such that we can
tell which CPU requested the disable. This then allows us to detect
the above scenario and even redirect the IPI to make up for the failed
queue.
Reported-by: Thomas-Mich Richter <tmricht@linux.ibm.com>
Tested-by: Thomas Richter <tmricht@linux.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Hendrik Brueckner <brueckner@linux.ibm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With skip set to 1, I get a traceback like this:
[ 106.867637] DMA-API: Mapped at:
[ 106.870784] afu_dma_map_region+0x2cd/0x4f0 [dfl_afu]
[ 106.875839] afu_ioctl+0x258/0x380 [dfl_afu]
[ 106.880108] do_vfs_ioctl+0xa9/0x720
[ 106.883688] ksys_ioctl+0x60/0x90
[ 106.887007] __x64_sys_ioctl+0x16/0x20
With the previous value of 2, afu_dma_map_region was being omitted. I
suspect that the code paths have simply changed since the value of 2 was
chosen a decade ago, but it's also possible that it varies based on which
mapping function was used, compiler inlining choices, etc. In any case,
it's best to err on the side of skipping less.
Signed-off-by: Scott Wood <swood@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
The following commit:
a0b0fd53e1 ("locking/lockdep: Free lock classes that are no longer in use")
changed the behavior of lockdep_free_key_range() from
unconditionally zapping lock classes into only zapping lock classes if
debug_lock == true. Not zapping lock classes if debug_lock == false leaves
dangling pointers in several lockdep datastructures, e.g. lock_class::name
in the all_lock_classes list.
The shell command "cat /proc/lockdep" causes the kernel to iterate the
all_lock_classes list. Hence the "unable to handle kernel paging request" cash
that Shenghui encountered by running cat /proc/lockdep.
Since the new behavior can cause cat /proc/lockdep to crash, restore the
pre-v5.1 behavior.
This patch avoids that cat /proc/lockdep triggers the following crash
with debug_lock == false:
BUG: unable to handle kernel paging request at fffffbfff40ca448
RIP: 0010:__asan_load1+0x28/0x50
Call Trace:
string+0xac/0x180
vsnprintf+0x23e/0x820
seq_vprintf+0x82/0xc0
seq_printf+0x92/0xb0
print_name+0x34/0xb0
l_show+0x184/0x200
seq_read+0x59e/0x6c0
proc_reg_read+0x11f/0x170
__vfs_read+0x4d/0x90
vfs_read+0xc5/0x1f0
ksys_read+0xab/0x130
__x64_sys_read+0x43/0x50
do_syscall_64+0x71/0x210
entry_SYSCALL_64_after_hwframe+0x49/0xbe
Reported-by: shenghui <shhuiw@foxmail.com>
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Fixes: a0b0fd53e1 ("locking/lockdep: Free lock classes that are no longer in use") # v5.1-rc1.
Link: https://lkml.kernel.org/r/20190403233552.124673-1-bvanassche@acm.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge misc fixes from Andrew Morton:
"14 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
kernel/sysctl.c: fix out-of-bounds access when setting file-max
mm/util.c: fix strndup_user() comment
sh: fix multiple function definition build errors
MAINTAINERS: add maintainer and replacing reviewer ARM/NUVOTON NPCM
MAINTAINERS: fix bad pattern in ARM/NUVOTON NPCM
mm: writeback: use exact memcg dirty counts
psi: clarify the units used in pressure files
mm/huge_memory.c: fix modifying of page protection by insert_pfn_pmd()
hugetlbfs: fix memory leak for resv_map
mm: fix vm_fault_t cast in VM_FAULT_GET_HINDEX()
lib/lzo: fix bugs for very short or empty input
include/linux/bitrev.h: fix constant bitrev
kmemleak: powerpc: skip scanning holes in the .bss section
lib/string.c: implement a basic bcmp
Commit 32a5ad9c22 ("sysctl: handle overflow for file-max") hooked up
min/max values for the file-max sysctl parameter via the .extra1 and
.extra2 fields in the corresponding struct ctl_table entry.
Unfortunately, the minimum value points at the global 'zero' variable,
which is an int. This results in a KASAN splat when accessed as a long
by proc_doulongvec_minmax on 64-bit architectures:
| BUG: KASAN: global-out-of-bounds in __do_proc_doulongvec_minmax+0x5d8/0x6a0
| Read of size 8 at addr ffff2000133d1c20 by task systemd/1
|
| CPU: 0 PID: 1 Comm: systemd Not tainted 5.1.0-rc3-00012-g40b114779944 #2
| Hardware name: linux,dummy-virt (DT)
| Call trace:
| dump_backtrace+0x0/0x228
| show_stack+0x14/0x20
| dump_stack+0xe8/0x124
| print_address_description+0x60/0x258
| kasan_report+0x140/0x1a0
| __asan_report_load8_noabort+0x18/0x20
| __do_proc_doulongvec_minmax+0x5d8/0x6a0
| proc_doulongvec_minmax+0x4c/0x78
| proc_sys_call_handler.isra.19+0x144/0x1d8
| proc_sys_write+0x34/0x58
| __vfs_write+0x54/0xe8
| vfs_write+0x124/0x3c0
| ksys_write+0xbc/0x168
| __arm64_sys_write+0x68/0x98
| el0_svc_common+0x100/0x258
| el0_svc_handler+0x48/0xc0
| el0_svc+0x8/0xc
|
| The buggy address belongs to the variable:
| zero+0x0/0x40
|
| Memory state around the buggy address:
| ffff2000133d1b00: 00 00 00 00 00 00 00 00 fa fa fa fa 04 fa fa fa
| ffff2000133d1b80: fa fa fa fa 04 fa fa fa fa fa fa fa 04 fa fa fa
| >ffff2000133d1c00: fa fa fa fa 04 fa fa fa fa fa fa fa 00 00 00 00
| ^
| ffff2000133d1c80: fa fa fa fa 00 fa fa fa fa fa fa fa 00 00 00 00
| ffff2000133d1d00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Fix the splat by introducing a unsigned long 'zero_ul' and using that
instead.
Link: http://lkml.kernel.org/r/20190403153409.17307-1-will.deacon@arm.com
Fixes: 32a5ad9c22 ("sysctl: handle overflow for file-max")
Signed-off-by: Will Deacon <will.deacon@arm.com>
Acked-by: Christian Brauner <christian@brauner.io>
Cc: Kees Cook <keescook@chromium.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Matteo Croce <mcroce@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull syscall-get-arguments cleanup and fixes from Steven Rostedt:
"Andy Lutomirski approached me to tell me that the
syscall_get_arguments() implementation in x86 was horrible and gcc
certainly gets it wrong.
He said that since the tracepoints only pass in 0 and 6 for i and n
repectively, it should be optimized for that case. Inspecting the
kernel, I discovered that all users pass in 0 for i and only one file
passing in something other than 6 for the number of arguments. That
code happens to be my own code used for the special syscall tracing.
That can easily be converted to just using 0 and 6 as well, and only
copying what is needed. Which is probably the faster path anyway for
that case.
Along the way, a couple of real fixes came from this as the
syscall_get_arguments() function was incorrect for csky and riscv.
x86 has been optimized to for the new interface that removes the
variable number of arguments, but the other architectures could still
use some loving and take more advantage of the simpler interface"
* tag 'trace-5.1-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
syscalls: Remove start and number from syscall_set_arguments() args
syscalls: Remove start and number from syscall_get_arguments() args
csky: Fix syscall_get_arguments() and syscall_set_arguments()
riscv: Fix syscall_get_arguments() and syscall_set_arguments()
tracing/syscalls: Pass in hardcoded 6 into syscall_get_arguments()
ptrace: Remove maxargs from task_current_syscall()
If a child irqchip calls irq_chip_set_wake_parent() but its parent irqchip
has the IRQCHIP_SKIP_SET_WAKE flag set an error is returned.
This is inconsistent behaviour vs. set_irq_wake_real() which returns 0 when
the irqchip has the IRQCHIP_SKIP_SET_WAKE flag set. It doesn't attempt to
walk the chain of parents and set irq wake on any chips that don't have the
flag set either. If the intent is to call the .irq_set_wake() callback of
the parent irqchip, then we expect irqchip implementations to omit the
IRQCHIP_SKIP_SET_WAKE flag and implement an .irq_set_wake() function that
calls irq_chip_set_wake_parent().
The problem has been observed on a Qualcomm sdm845 device where set wake
fails on any GPIO interrupts after applying work in progress wakeup irq
patches to the GPIO driver. The chain of chips looks like this:
QCOM GPIO -> QCOM PDC (SKIP) -> ARM GIC (SKIP)
The GPIO controllers parent is the QCOM PDC irqchip which in turn has ARM
GIC as parent. The QCOM PDC irqchip has the IRQCHIP_SKIP_SET_WAKE flag
set, and so does the grandparent ARM GIC.
The GPIO driver doesn't know if the parent needs to set wake or not, so it
unconditionally calls irq_chip_set_wake_parent() causing this function to
return a failure because the parent irqchip (PDC) doesn't have the
.irq_set_wake() callback set. Returning 0 instead makes everything work and
irqs from the GPIO controller can be configured for wakeup.
Make it consistent by returning 0 (success) from irq_chip_set_wake_parent()
when a parent chip has IRQCHIP_SKIP_SET_WAKE set.
[ tglx: Massaged changelog ]
Fixes: 08b55e2a92 ("genirq: Add irqchip_set_wake_parent")
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-gpio@vger.kernel.org
Cc: Lina Iyer <ilina@codeaurora.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20190325181026.247796-1-swboyd@chromium.org
Pull networking fixes from David Miller:
1) Several hash table refcount fixes in batman-adv, from Sven
Eckelmann.
2) Use after free in bpf_evict_inode(), from Daniel Borkmann.
3) Fix mdio bus registration in ixgbe, from Ivan Vecera.
4) Unbounded loop in __skb_try_recv_datagram(), from Paolo Abeni.
5) ila rhashtable corruption fix from Herbert Xu.
6) Don't allow upper-devices to be added to vrf devices, from Sabrina
Dubroca.
7) Add qmi_wwan device ID for Olicard 600, from Bjørn Mork.
8) Don't leave skb->next poisoned in __netif_receive_skb_list_ptype,
from Alexander Lobakin.
9) Missing IDR checks in mlx5 driver, from Aditya Pakki.
10) Fix false connection termination in ktls, from Jakub Kicinski.
11) Work around some ASPM issues with r8169 by disabling rx interrupt
coalescing on certain chips. From Heiner Kallweit.
12) Properly use per-cpu qstat values on NOLOCK qdiscs, from Paolo
Abeni.
13) Fully initialize sockaddr_in structures in SCTP, from Xin Long.
14) Various BPF flow dissector fixes from Stanislav Fomichev.
15) Divide by zero in act_sample, from Davide Caratti.
16) Fix bridging multicast regression introduced by rhashtable
conversion, from Nikolay Aleksandrov.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (106 commits)
ibmvnic: Fix completion structure initialization
ipv6: sit: reset ip header pointer in ipip6_rcv
net: bridge: always clear mcast matching struct on reports and leaves
libcxgb: fix incorrect ppmax calculation
vlan: conditional inclusion of FCoE hooks to match netdevice.h and bnx2x
sch_cake: Make sure we can write the IP header before changing DSCP bits
sch_cake: Use tc_skb_protocol() helper for getting packet protocol
tcp: Ensure DCTCP reacts to losses
net/sched: act_sample: fix divide by zero in the traffic path
net: thunderx: fix NULL pointer dereference in nicvf_open/nicvf_stop
net: hns: Fix sparse: some warnings in HNS drivers
net: hns: Fix WARNING when remove HNS driver with SMMU enabled
net: hns: fix ICMP6 neighbor solicitation messages discard problem
net: hns: Fix probabilistic memory overwrite when HNS driver initialized
net: hns: Use NAPI_POLL_WEIGHT for hns driver
net: hns: fix KASAN: use-after-free in hns_nic_net_xmit_hw()
flow_dissector: rst'ify documentation
ipv6: Fix dangling pointer when ipv6 fragment
net-gro: Fix GRO flush when receiving a GSO packet.
flow_dissector: document BPF flow dissector environment
...
The only users that calls syscall_get_arguments() with a variable and not a
hard coded '6' is ftrace_syscall_enter(). syscall_get_arguments() can be
optimized by removing a variable input, and always grabbing 6 arguments
regardless of what the system call actually uses.
Change ftrace_syscall_enter() to pass the 6 args into a local stack array
and copy the necessary arguments into the trace event as needed.
This is needed to remove two parameters from syscall_get_arguments().
Link: http://lkml.kernel.org/r/20161107213233.627583542@goodmis.org
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
A NULL pointer dereference bug was reported on a distribution kernel but
the same issue should be present on mainline kernel. It occured on s390
but should not be arch-specific. A partial oops looks like:
Unable to handle kernel pointer dereference in virtual kernel address space
...
Call Trace:
...
try_to_wake_up+0xfc/0x450
vhost_poll_wakeup+0x3a/0x50 [vhost]
__wake_up_common+0xbc/0x178
__wake_up_common_lock+0x9e/0x160
__wake_up_sync_key+0x4e/0x60
sock_def_readable+0x5e/0x98
The bug hits any time between 1 hour to 3 days. The dereference occurs
in update_cfs_rq_h_load when accumulating h_load. The problem is that
cfq_rq->h_load_next is not protected by any locking and can be updated
by parallel calls to task_h_load. Depending on the compiler, code may be
generated that re-reads cfq_rq->h_load_next after the check for NULL and
then oops when reading se->avg.load_avg. The dissassembly showed that it
was possible to reread h_load_next after the check for NULL.
While this does not appear to be an issue for later compilers, it's still
an accident if the correct code is generated. Full locking in this path
would have high overhead so this patch uses READ_ONCE to read h_load_next
only once and check for NULL before dereferencing. It was confirmed that
there were no further oops after 10 days of testing.
As Peter pointed out, it is also necessary to use WRITE_ONCE() to avoid any
potential problems with store tearing.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@vger.kernel.org>
Fixes: 685207963b ("sched: Move h_load calculation to task_h_load()")
Link: https://lkml.kernel.org/r/20190319123610.nsivgf3mjbjjesxb@techsingularity.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current sys_pidfd_send_signal() silently turns signals with explicit
SI_USER context that are sent to non-current tasks into signals with
kernel-generated siginfo.
This is unlike do_rt_sigqueueinfo(), which returns -EPERM in this case.
If a user actually wants to send a signal with kernel-provided siginfo,
they can do that with pidfd_send_signal(pidfd, sig, NULL, 0); so allowing
this case is unnecessary.
Instead of silently replacing the siginfo, just bail out with an error;
this is consistent with other interfaces and avoids special-casing behavior
based on security checks.
Fixes: 3eb39f4793 ("signal: add pidfd_send_signal() syscall")
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Christian Brauner <christian@brauner.io>
Pull CPU hotplug fixes from Thomas Gleixner:
"Two SMT/hotplug related fixes:
- Prevent crash when HOTPLUG_CPU is disabled and the CPU bringup
aborts. This is triggered with the 'nosmt' command line option, but
can happen by any abort condition. As the real unplug code is not
compiled in, prevent the fail by keeping the CPU in zombie state.
- Enforce HOTPLUG_CPU for SMP on x86 to avoid the above situation
completely. With 'nosmt' being a popular option it's required to
unplug the half brought up sibling CPUs (due to the MCE wreckage)
completely"
* 'smp-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/smp: Enforce CONFIG_HOTPLUG_CPU when SMP=y
cpu/hotplug: Prevent crash when CPU bringup fails on CONFIG_HOTPLUG_CPU=n
Pull core fixes from Thomas Gleixner:
"A small set of core updates:
- Make the watchdog respect the selected CPU mask again. That was
broken by the rework of the watchdog thread management and caused
inconsistent state and NMI watchdog being unstoppable.
- Ensure that the objtool build can find the libelf location.
- Remove dead kcore stub code"
* 'core-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
watchdog: Respect watchdog cpumask on CPU hotplug
objtool: Query pkg-config for libelf location
proc/kcore: Remove unused kclist_add_remap()
Daniel Borkmann says:
====================
pull-request: bpf 2019-03-29
The following pull-request contains BPF updates for your *net* tree.
The main changes are:
1) Bug fix in BTF deduplication that was mishandling an equivalence
comparison, from Andrii.
2) libbpf Makefile fixes to properly link against libelf for the shared
object and to actually export AF_XDP's xsk.h header, from Björn.
3) Fix use after free in bpf inode eviction, from Daniel.
4) Fix a bug in skb creation out of cpumap redirect, from Jesper.
5) Remove an unnecessary and triggerable WARN_ONCE() in max number
of call stack frames checking in verifier, from Paul.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
We want to avoid leaking pointer info from xdp_frame (that is placed in
top of frame) like commit 6dfb970d3d ("xdp: avoid leaking info stored in
frame data on page reuse"), and followup commit 97e19cce05 ("bpf:
reserve xdp_frame size in xdp headroom") that reserve this headroom.
These changes also affected how cpumap constructed SKBs, as xdpf->headroom
size changed, the skb data starting point were in-effect shifted with 32
bytes (sizeof xdp_frame). This was still okay, as the cpumap frame_size
calculation also included xdpf->headroom which were reduced by same amount.
A bug was introduced in commit 77ea5f4cbe ("bpf/cpumap: make sure
frame_size for build_skb is aligned if headroom isn't"), where the
xdpf->headroom became part of the SKB_DATA_ALIGN rounding up. This
round-up to find the frame_size is in principle still correct as it does
not exceed the 2048 bytes frame_size (which is max for ixgbe and i40e),
but the 32 bytes offset of pkt_data_start puts this over the 2048 bytes
limit. This cause skb_shared_info to spill into next frame. It is a little
hard to trigger, as the SKB need to use above 15 skb_shinfo->frags[] as
far as I calculate. This does happen in practise for TCP streams when
skb_try_coalesce() kicks in.
KASAN can be used to detect these wrong memory accesses, I've seen:
BUG: KASAN: use-after-free in skb_try_coalesce+0x3cb/0x760
BUG: KASAN: wild-memory-access in skb_release_data+0xe2/0x250
Driver veth also construct a SKB from xdp_frame in this way, but is not
affected, as it doesn't reserve/deduct the room (used by xdp_frame) from
the SKB headroom. Instead is clears the pointers via xdp_scrub_frame(),
and allows SKB to use this area.
The fix in this patch is to do like veth and instead allow SKB to (re)use
the area occupied by xdp_frame, by clearing via xdp_scrub_frame(). (This
does kill the idea of the SKB being able to access (mem) info from this
area, but I guess it was a bad idea anyhow, and it was already killed by
the veth changes.)
Fixes: 77ea5f4cbe ("bpf/cpumap: make sure frame_size for build_skb is aligned if headroom isn't")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
