| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
On some small machines with little entropy, a quasi-unique hostname is
sometimes a relevant factor. I've seen, for example, 8 character
alpha-numeric serial numbers. In addition, the time at which the hostname
is set is usually a decent measurement of how long early boot took. So,
call add_device_randomness() on new hostnames, which feeds its arguments
to the RNG in addition to a fresh cycle counter.
Low cost hooks like this never hurt and can only ever help, and since
this costs basically nothing for an operation that is never a fast path,
this is an overall easy win.
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
Rather than going through the current-> indirection for utsname, at this
point in boot, init_utsname()==utsname(), so just use it directly that
way. Additionally, init_utsname() appears to be available nearly always,
so move it into random_init_early().
Suggested-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
As of the prior commit, the RNG will have incorporated both a cycle
counter value and RDRAND, in addition to various other environmental
noise. Therefore, using get_random_u32() will supply a stronger seed
than simply using random_get_entropy(). N.B.: random_get_entropy()
should be considered an internal API of random.c and not generally
consumed.
Cc: Alexander Potapenko <glider@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Marco Elver <elver@google.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
The full RNG initialization relies on some timestamps, made possible
with initialization functions like time_init() and timekeeping_init().
However, these are only available rather late in initialization.
Meanwhile, other things, such as memory allocator functions, make use of
the RNG much earlier.
So split RNG initialization into two phases. We can provide arch
randomness very early on, and then later, after timekeeping and such are
available, initialize the rest.
This ensures that, for example, slabs are properly randomized if RDRAND
is available. Without this, CONFIG_SLAB_FREELIST_RANDOM=y loses a degree
of its security, because its random seed is potentially deterministic,
since it hasn't yet incorporated RDRAND. It also makes it possible to
use a better seed in kfence, which currently relies on only the cycle
counter.
Another positive consequence is that on systems with RDRAND, running
with CONFIG_WARN_ALL_UNSEEDED_RANDOM=y results in no warnings at all.
One subtle side effect of this change is that on systems with no RDRAND,
RDTSC is now only queried by random_init() once, committing the moment
of the function call, instead of multiple times as before. This is
intentional, as the multiple RDTSCs in a loop before weren't
accomplishing very much, with jitter being better provided by
try_to_generate_entropy(). Plus, filling blocks with RDTSC is still
being done in extract_entropy(), which is necessarily called before
random bytes are served anyway.
Cc: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
Previously, the fast pool was dumped into the main pool periodically in
the fast pool's hard IRQ handler. This worked fine and there weren't
problems with it, until RT came around. Since RT converts spinlocks into
sleeping locks, problems cropped up. Rather than switching to raw
spinlocks, the RT developers preferred we make the transformation from
originally doing:
do_some_stuff()
spin_lock()
do_some_other_stuff()
spin_unlock()
to doing:
do_some_stuff()
queue_work_on(some_other_stuff_worker)
This is an ordinary pattern done all over the kernel. However, Sherry
noticed a 10% performance regression in qperf TCP over a 40gbps
InfiniBand card. Quoting her message:
> MT27500 Family [ConnectX-3] cards:
> Infiniband device 'mlx4_0' port 1 status:
> default gid: fe80:0000:0000:0000:0010:e000:0178:9eb1
> base lid: 0x6
> sm lid: 0x1
> state: 4: ACTIVE
> phys state: 5: LinkUp
> rate: 40 Gb/sec (4X QDR)
> link_layer: InfiniBand
>
> Cards are configured with IP addresses on private subnet for IPoIB
> performance testing.
> Regression identified in this bug is in TCP latency in this stack as reported
> by qperf tcp_lat metric:
>
> We have one system listen as a qperf server:
> [root@yourQperfServer ~]# qperf
>
> Have the other system connect to qperf server as a client (in this
> case, it’s X7 server with Mellanox card):
> [root@yourQperfClient ~]# numactl -m0 -N0 qperf 20.20.20.101 -v -uu -ub --time 60 --wait_server 20 -oo msg_size:4K:1024K:*2 tcp_lat
Rather than incur the scheduling latency from queue_work_on, we can
instead switch to running on the next timer tick, on the same core. This
also batches things a bit more -- once per jiffy -- which is okay now
that mix_interrupt_randomness() can credit multiple bits at once.
Reported-by: Sherry Yang <sherry.yang@oracle.com>
Tested-by: Paul Webb <paul.x.webb@oracle.com>
Cc: Sherry Yang <sherry.yang@oracle.com>
Cc: Phillip Goerl <phillip.goerl@oracle.com>
Cc: Jack Vogel <jack.vogel@oracle.com>
Cc: Nicky Veitch <nicky.veitch@oracle.com>
Cc: Colm Harrington <colm.harrington@oracle.com>
Cc: Ramanan Govindarajan <ramanan.govindarajan@oracle.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Sultan Alsawaf <sultan@kerneltoast.com>
Cc: stable@vger.kernel.org
Fixes: 58340f8e952b ("random: defer fast pool mixing to worker")
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
In order to avoid reading and dirtying two cache lines on every IRQ,
move the work_struct to the bottom of the fast_pool struct. add_
interrupt_randomness() always touches .pool and .count, which are
currently split, because .mix pushes everything down. Instead, move .mix
to the bottom, so that .pool and .count are always in the first cache
line, since .mix is only accessed when the pool is full.
Fixes: 58340f8e952b ("random: defer fast pool mixing to worker")
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
Since the most that's mixed into the pool is sizeof(long)*2, don't
credit more than that many bytes of entropy.
Fixes: e3e33fc2ea7f ("random: do not use input pool from hard IRQs")
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
If a hwrng source does not provide an entropy estimate, it currently
does not contribute at all to the CRNG. In order to help fix this, in
case add_hwgenerator_randomness() is called with the entropy parameter
set to zero, go to sleep until one reseed interval has passed.
While the hwrng thread currently only runs under conditions where this
is non-zero, this change is not harmful and prepares for future updates
to the hwrng core.
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Reviewed-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
Mix in randomness from hw-rng sources more frequently during early
boot, approximately once for every rng reseed.
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
Prior to 5.6, when /dev/random was opened with O_NONBLOCK, it would
return -EAGAIN if there was no entropy. When the pools were unified in
5.6, this was lost. The post 5.6 behavior of blocking until the pool is
initialized, and ignoring O_NONBLOCK in the process, went unnoticed,
with no reports about the regression received for two and a half years.
However, eventually this indeed did break somebody's userspace.
So we restore the old behavior, by returning -EAGAIN if the pool is not
initialized. Unlike the old /dev/random, this can only occur during
early boot, after which it never blocks again.
In order to make this O_NONBLOCK behavior consistent with other
expectations, also respect users reading with preadv2(RWF_NOWAIT) and
similar.
Fixes: 30c08efec888 ("random: make /dev/random be almost like /dev/urandom")
Reported-by: Guozihua <guozihua@huawei.com>
Reported-by: Zhongguohua <zhongguohua1@huawei.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: Andrew Lutomirski <luto@kernel.org>
Cc: stable@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
RHEL/Fedora RPM build checks are stricter, and complain when executable
files don't have a shebang line, e.g.
*** WARNING: ./kselftests/net/forwarding/sch_red.sh is executable but has no shebang, removing executable bit
Fix it by adding shebang line.
Fixes: 6cf0291f9517 ("selftests: forwarding: Add a RED test for SW datapath")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/20220922024453.437757-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
When reading the timestamp is required bnxt_tx_int() hands
over the ownership of the completed skb to the PTP worker.
The skb should not be used afterwards, as the worker may
run before the rest of our code and free the skb, leading
to a use-after-free.
Since dev_kfree_skb_any() accepts NULL make the loss of
ownership more obvious and set skb to NULL.
Fixes: 83bb623c968e ("bnxt_en: Transmit and retrieve packet timestamps")
Reviewed-by: Andy Gospodarek <gospo@broadcom.com>
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Link: https://lore.kernel.org/r/20220921201005.335390-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
In prestera_port_sfp_bind(), there are two refcounting bugs:
(1) we should call of_node_get() before of_find_node_by_name() as
it will automaitcally decrease the refcount of 'from' argument;
(2) we should call of_node_put() for the break of the iteration
for_each_child_of_node() as it will automatically increase and
decrease the 'child'.
Fixes: 52323ef75414 ("net: marvell: prestera: add phylink support")
Signed-off-by: Liang He <windhl@126.com>
Reviewed-by: Yevhen Orlov <yevhen.orlov@plvision.eu>
Link: https://lore.kernel.org/r/20220921133245.4111672-1-windhl@126.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
tfilter_put need to be called to put the refount got by tp->ops->get to
avoid possible refcount leak when chain->tmplt_ops != NULL and
chain->tmplt_ops != tp->ops.
Fixes: 7d5509fa0d3d ("net: sched: extend proto ops with 'put' callback")
Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Link: https://lore.kernel.org/r/20220921092734.31700-1-hbh25y@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There is a separate receive path for small packets (under 256 bytes).
Instead of allocating a new dma-capable skb to be used for the next packet,
this path allocates a skb and copies the data into it (reusing the existing
sbk for the next packet). There are two bytes of junk data at the beginning
of every packet. I believe these are inserted in order to allow aligned DMA
and IP headers. We skip over them using skb_reserve. Before copying over
the data, we must use a barrier to ensure we see the whole packet. The
current code only synchronizes len bytes, starting from the beginning of
the packet, including the junk bytes. However, this leaves off the final
two bytes in the packet. Synchronize the whole packet.
To reproduce this problem, ping a HME with a payload size between 17 and
214
$ ping -s 17 <hme_address>
which will complain rather loudly about the data mismatch. Small packets
(below 60 bytes on the wire) do not have this issue. I suspect this is
related to the padding added to increase the minimum packet size.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20220920235018.1675956-1-seanga2@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Prevent udp_read_skb() from flooding the syslog.
Suggested-by: Jakub Sitnicki <jakub@cloudflare.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Link: https://lore.kernel.org/r/20220921005915.2697-1-yepeilin.cs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
This bonding selftest used to cause a kernel oops on aarch64
and should be architectures agnostic.
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Fix a NULL dereference of the struct bonding.rr_tx_counter member because
if a bond is initially created with an initial mode != zero (Round Robin)
the memory required for the counter is never created and when the mode is
changed there is never any attempt to verify the memory is allocated upon
switching modes.
This causes the following Oops on an aarch64 machine:
[ 334.686773] Unable to handle kernel paging request at virtual address ffff2c91ac905000
[ 334.694703] Mem abort info:
[ 334.697486] ESR = 0x0000000096000004
[ 334.701234] EC = 0x25: DABT (current EL), IL = 32 bits
[ 334.706536] SET = 0, FnV = 0
[ 334.709579] EA = 0, S1PTW = 0
[ 334.712719] FSC = 0x04: level 0 translation fault
[ 334.717586] Data abort info:
[ 334.720454] ISV = 0, ISS = 0x00000004
[ 334.724288] CM = 0, WnR = 0
[ 334.727244] swapper pgtable: 4k pages, 48-bit VAs, pgdp=000008044d662000
[ 334.733944] [ffff2c91ac905000] pgd=0000000000000000, p4d=0000000000000000
[ 334.740734] Internal error: Oops: 96000004 [#1] SMP
[ 334.745602] Modules linked in: bonding tls veth rfkill sunrpc arm_spe_pmu vfat fat acpi_ipmi ipmi_ssif ixgbe igb i40e mdio ipmi_devintf ipmi_msghandler arm_cmn arm_dsu_pmu cppc_cpufreq acpi_tad fuse zram crct10dif_ce ast ghash_ce sbsa_gwdt nvme drm_vram_helper drm_ttm_helper nvme_core ttm xgene_hwmon
[ 334.772217] CPU: 7 PID: 2214 Comm: ping Not tainted 6.0.0-rc4-00133-g64ae13ed4784 #4
[ 334.779950] Hardware name: GIGABYTE R272-P31-00/MP32-AR1-00, BIOS F18v (SCP: 1.08.20211002) 12/01/2021
[ 334.789244] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 334.796196] pc : bond_rr_gen_slave_id+0x40/0x124 [bonding]
[ 334.801691] lr : bond_xmit_roundrobin_slave_get+0x38/0xdc [bonding]
[ 334.807962] sp : ffff8000221733e0
[ 334.811265] x29: ffff8000221733e0 x28: ffffdbac8572d198 x27: ffff80002217357c
[ 334.818392] x26: 000000000000002a x25: ffffdbacb33ee000 x24: ffff07ff980fa000
[ 334.825519] x23: ffffdbacb2e398ba x22: ffff07ff98102000 x21: ffff07ff981029c0
[ 334.832646] x20: 0000000000000001 x19: ffff07ff981029c0 x18: 0000000000000014
[ 334.839773] x17: 0000000000000000 x16: ffffdbacb1004364 x15: 0000aaaabe2f5a62
[ 334.846899] x14: ffff07ff8e55d968 x13: ffff07ff8e55db30 x12: 0000000000000000
[ 334.854026] x11: ffffdbacb21532e8 x10: 0000000000000001 x9 : ffffdbac857178ec
[ 334.861153] x8 : ffff07ff9f6e5a28 x7 : 0000000000000000 x6 : 000000007c2b3742
[ 334.868279] x5 : ffff2c91ac905000 x4 : ffff2c91ac905000 x3 : ffff07ff9f554400
[ 334.875406] x2 : ffff2c91ac905000 x1 : 0000000000000001 x0 : ffff07ff981029c0
[ 334.882532] Call trace:
[ 334.884967] bond_rr_gen_slave_id+0x40/0x124 [bonding]
[ 334.890109] bond_xmit_roundrobin_slave_get+0x38/0xdc [bonding]
[ 334.896033] __bond_start_xmit+0x128/0x3a0 [bonding]
[ 334.901001] bond_start_xmit+0x54/0xb0 [bonding]
[ 334.905622] dev_hard_start_xmit+0xb4/0x220
[ 334.909798] __dev_queue_xmit+0x1a0/0x720
[ 334.913799] arp_xmit+0x3c/0xbc
[ 334.916932] arp_send_dst+0x98/0xd0
[ 334.920410] arp_solicit+0xe8/0x230
[ 334.923888] neigh_probe+0x60/0xb0
[ 334.927279] __neigh_event_send+0x3b0/0x470
[ 334.931453] neigh_resolve_output+0x70/0x90
[ 334.935626] ip_finish_output2+0x158/0x514
[ 334.939714] __ip_finish_output+0xac/0x1a4
[ 334.943800] ip_finish_output+0x40/0xfc
[ 334.947626] ip_output+0xf8/0x1a4
[ 334.950931] ip_send_skb+0x5c/0x100
[ 334.954410] ip_push_pending_frames+0x3c/0x60
[ 334.958758] raw_sendmsg+0x458/0x6d0
[ 334.962325] inet_sendmsg+0x50/0x80
[ 334.965805] sock_sendmsg+0x60/0x6c
[ 334.969286] __sys_sendto+0xc8/0x134
[ 334.972853] __arm64_sys_sendto+0x34/0x4c
[ 334.976854] invoke_syscall+0x78/0x100
[ 334.980594] el0_svc_common.constprop.0+0x4c/0xf4
[ 334.985287] do_el0_svc+0x38/0x4c
[ 334.988591] el0_svc+0x34/0x10c
[ 334.991724] el0t_64_sync_handler+0x11c/0x150
[ 334.996072] el0t_64_sync+0x190/0x194
[ 334.999726] Code: b9001062 f9403c02 d53cd044 8b040042 (b8210040)
[ 335.005810] ---[ end trace 0000000000000000 ]---
[ 335.010416] Kernel panic - not syncing: Oops: Fatal exception in interrupt
[ 335.017279] SMP: stopping secondary CPUs
[ 335.021374] Kernel Offset: 0x5baca8eb0000 from 0xffff800008000000
[ 335.027456] PHYS_OFFSET: 0x80000000
[ 335.030932] CPU features: 0x0000,0085c029,19805c82
[ 335.035713] Memory Limit: none
[ 335.038756] Rebooting in 180 seconds..
The fix is to allocate the memory in bond_open() which is guaranteed
to be called before any packets are processed.
Fixes: 848ca9182a7d ("net: bonding: Use per-cpu rr_tx_counter")
CC: Jussi Maki <joamaki@gmail.com>
Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Since commit ece19502834d ("net: phy: micrel: 1588 support for LAN8814
phy") the handler always returns IRQ_HANDLED, except in an error case.
Before that commit, the interrupt status register was checked and if
it was empty, IRQ_NONE was returned. Restore that behavior to play nice
with the interrupt line being shared with others.
Fixes: ece19502834d ("net: phy: micrel: 1588 support for LAN8814 phy")
Signed-off-by: Michael Walle <michael@walle.cc>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Horatiu Vultur <horatiu.vultur@microchip.com>
Reviewed-by: Divya Koppera <Divya.Koppera@microchip.com>
Link: https://lore.kernel.org/r/20220920141619.808117-1-michael@walle.cc
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
There might be a potential race between SMC-R buffer map and
link group termination.
smc_smcr_terminate_all() | smc_connect_rdma()
--------------------------------------------------------------
| smc_conn_create()
for links in smcibdev |
schedule links down |
| smc_buf_create()
| \- smcr_buf_map_usable_links()
| \- no usable links found,
| (rmb->mr = NULL)
|
| smc_clc_send_confirm()
| \- access conn->rmb_desc->mr[]->rkey
| (panic)
During reboot and IB device module remove, all links will be set
down and no usable links remain in link groups. In such situation
smcr_buf_map_usable_links() should return an error and stop the
CLC flow accessing to uninitialized mr.
Fixes: b9247544c1bc ("net/smc: convert static link ID instances to support multiple links")
Signed-off-by: Wen Gu <guwen@linux.alibaba.com>
Link: https://lore.kernel.org/r/1663656189-32090-1-git-send-email-guwen@linux.alibaba.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
We currently check the MokSBState variable to decide whether we should
treat UEFI secure boot as being disabled, even if the firmware thinks
otherwise. This is used by shim to indicate that it is not checking
signatures on boot images. In the kernel, we use this to relax lockdown
policies.
However, in cases where shim is not even being used, we don't want this
variable to interfere with lockdown, given that the variable may be
non-volatile and therefore persist across a reboot. This means setting
it once will persistently disable lockdown checks on a given system.
So switch to the mirrored version of this variable, called MokSBStateRT,
which is supposed to be volatile, and this is something we can check.
Cc: <stable@vger.kernel.org> # v4.19+
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Peter Jones <pjones@redhat.com>
|
|
When booting the x86 kernel via EFI using the LoadImage/StartImage boot
services [as opposed to the deprecated EFI handover protocol], the setup
header is taken from the image directly, and given that EFI's LoadImage
has no Linux/x86 specific knowledge regarding struct bootparams or
struct setup_header, any absolute addresses in the setup header must
originate from the file and not from a prior loading stage.
Since we cannot generally predict where LoadImage() decides to load an
image (*), such absolute addresses must be treated as suspect: even if a
prior boot stage intended to make them point somewhere inside the
[signed] image, there is no way to validate that, and if they point at
an arbitrary location in memory, the setup_data nodes will not be
covered by any signatures or TPM measurements either, and could be made
to contain an arbitrary sequence of SETUP_xxx nodes, which could
interfere quite badly with the early x86 boot sequence.
(*) Note that, while LoadImage() does take a buffer/size tuple in
addition to a device path, which can be used to provide the image
contents directly, it will re-allocate such images, as the memory
footprint of an image is generally larger than the PE/COFF file
representation.
Cc: <stable@vger.kernel.org> # v5.10+
Link: https://lore.kernel.org/all/20220904165321.1140894-1-Jason@zx2c4.com/
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Jason A. Donenfeld <Jason@zx2c4.com>
|
|
The original patch added the static branch to handle the situation,
when assigning an XDP TX queue to every CPU is not possible,
so they have to be shared.
However, in the XDP transmit handler ice_xdp_xmit(), an error was
returned in such cases even before static condition was checked,
thus making queue sharing still impossible.
Fixes: 22bf877e528f ("ice: introduce XDP_TX fallback path")
Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com>
Reviewed-by: Alexander Lobakin <alexandr.lobakin@intel.com>
Link: https://lore.kernel.org/r/20220919134346.25030-1-larysa.zaremba@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
It needs to enter the namespace before reading a file.
Fixes: 4183a8d70a288627 ("perf tools: Allow synthesizing the build id for kernel/modules/tasks in PERF_RECORD_MMAP2")
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20220920222822.2171056-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
To pick the changes from:
7df548840c496b01 ("x86/bugs: Add "unknown" reporting for MMIO Stale Data")
This only causes these perf files to be rebuilt:
CC /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o
CC /tmp/build/perf/bench/mem-memset-x86-64-asm.o
And addresses this perf build warning:
Warning: Kernel ABI header at 'tools/arch/x86/include/asm/cpufeatures.h' differs from latest version at 'arch/x86/include/asm/cpufeatures.h'
diff -u tools/arch/x86/include/asm/cpufeatures.h arch/x86/include/asm/cpufeatures.h
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Link: https://lore.kernel.org/lkml/YysTRji90sNn2p5f@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
/proc/kallsyms and /proc/modules are compared before and after the copy
in order to ensure no changes during the copy.
However /proc/modules also might change due to reference counts changing
even though that does not make any difference.
Any modules loaded or unloaded should be visible in changes to kallsyms,
so it is not necessary to check /proc/modules also anyway.
Remove the comparison checking that /proc/modules is unchanged.
Fixes: fc1b691d7651d949 ("perf buildid-cache: Add ability to add kcore to the cache")
Reported-by: Daniel Dao <dqminh@cloudflare.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Daniel Dao <dqminh@cloudflare.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/r/20220914122429.8770-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Originally, (refer commit f90d194a867a5a1d ("perf evlist: Do not poll
events that use the system_wide flag") there wasn't much reason to poll
system-wide events because:
1. The mmaps get "merged" via set-output anyway (the per-cpu case)
2. perf reads all mmaps when any event is woken
3. system-wide mmaps do not fill up as fast as the mmaps for user
selected events
But there was 1 reason not to poll which was that it prevented correct
termination due to POLLHUP on all user selected events. That issue is
now easily resolved by using fdarray_flag__nonfilterable.
With the advent of commit ae4f8ae16a078964 ("libperf evlist: Allow
mixing per-thread and per-cpu mmaps"), system-wide mmaps can be used
also in the per-thread case where reason 1 does not apply.
Fix the omission of system-wide events from polling by using the
fdarray_flag__nonfilterable flag.
Example:
Before:
$ perf record --no-bpf-event -vvv -e intel_pt// --per-thread uname 2>err.txt
Linux
$ grep 'sys_perf_event_open.*=\|pollfd' err.txt
sys_perf_event_open: pid 155076 cpu -1 group_fd -1 flags 0x8 = 5
sys_perf_event_open: pid 155076 cpu -1 group_fd -1 flags 0x8 = 6
sys_perf_event_open: pid -1 cpu 0 group_fd -1 flags 0x8 = 7
sys_perf_event_open: pid -1 cpu 1 group_fd -1 flags 0x8 = 9
sys_perf_event_open: pid -1 cpu 2 group_fd -1 flags 0x8 = 10
sys_perf_event_open: pid -1 cpu 3 group_fd -1 flags 0x8 = 11
sys_perf_event_open: pid -1 cpu 4 group_fd -1 flags 0x8 = 12
sys_perf_event_open: pid -1 cpu 5 group_fd -1 flags 0x8 = 13
sys_perf_event_open: pid -1 cpu 6 group_fd -1 flags 0x8 = 14
sys_perf_event_open: pid -1 cpu 7 group_fd -1 flags 0x8 = 15
thread_data[0x55fb43c29e80]: pollfd[0] <- event_fd=5
thread_data[0x55fb43c29e80]: pollfd[1] <- event_fd=6
thread_data[0x55fb43c29e80]: pollfd[2] <- non_perf_event fd=4
After:
$ perf record --no-bpf-event -vvv -e intel_pt// --per-thread uname 2>err.txt
Linux
$ grep 'sys_perf_event_open.*=\|pollfd' err.txt
sys_perf_event_open: pid 156316 cpu -1 group_fd -1 flags 0x8 = 5
sys_perf_event_open: pid 156316 cpu -1 group_fd -1 flags 0x8 = 6
sys_perf_event_open: pid -1 cpu 0 group_fd -1 flags 0x8 = 7
sys_perf_event_open: pid -1 cpu 1 group_fd -1 flags 0x8 = 9
sys_perf_event_open: pid -1 cpu 2 group_fd -1 flags 0x8 = 10
sys_perf_event_open: pid -1 cpu 3 group_fd -1 flags 0x8 = 11
sys_perf_event_open: pid -1 cpu 4 group_fd -1 flags 0x8 = 12
sys_perf_event_open: pid -1 cpu 5 group_fd -1 flags 0x8 = 13
sys_perf_event_open: pid -1 cpu 6 group_fd -1 flags 0x8 = 14
sys_perf_event_open: pid -1 cpu 7 group_fd -1 flags 0x8 = 15
thread_data[0x55cc19e58e80]: pollfd[0] <- event_fd=5
thread_data[0x55cc19e58e80]: pollfd[1] <- event_fd=6
thread_data[0x55cc19e58e80]: pollfd[2] <- event_fd=7
thread_data[0x55cc19e58e80]: pollfd[3] <- event_fd=9
thread_data[0x55cc19e58e80]: pollfd[4] <- event_fd=10
thread_data[0x55cc19e58e80]: pollfd[5] <- event_fd=11
thread_data[0x55cc19e58e80]: pollfd[6] <- event_fd=12
thread_data[0x55cc19e58e80]: pollfd[7] <- event_fd=13
thread_data[0x55cc19e58e80]: pollfd[8] <- event_fd=14
thread_data[0x55cc19e58e80]: pollfd[9] <- event_fd=15
thread_data[0x55cc19e58e80]: pollfd[10] <- non_perf_event fd=4
Fixes: ae4f8ae16a078964 ("libperf evlist: Allow mixing per-thread and per-cpu mmaps")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220915122612.81738-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
With mixed per-thread and (system-wide) per-cpu maps, the "any cpu" value
-1 must be skipped when setting CPU mask bits.
Prior to commit cbd7bfc7fd99acdd ("tools/perf: Fix out of bound access
to cpu mask array") the invalid setting went unnoticed, but since then
it causes perf record to fail with an error.
Example:
Before:
$ perf record -e intel_pt// --per-thread uname
Failed to initialize parallel data streaming masks
After:
$ perf record -e intel_pt// --per-thread uname
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.068 MB perf.data ]
Fixes: ae4f8ae16a078964 ("libperf evlist: Allow mixing per-thread and per-cpu mmaps")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220915122612.81738-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
It uses PERF_EVENT_IOC_MODIFY_ATTRIBUTES ioctl. The kernel would return
ENOTTY if it's not supported. Update the skip reason in that case.
Committer notes:
On s/390 the args aren't used, so need to be marked __maybe_unused.
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20220914183338.546357-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
The missing header makes it hard for programs like elfutils to open
these files.
Fixes: 2d86612aacb7805f ("perf symbol: Correct address for bss symbols")
Reviewed-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Lieven Hey <lieven.hey@kdab.com>
Tested-by: Leo Yan <leo.yan@linaro.org>
Cc: Leo Yan <leo.yan@linaro.org>
Link: https://lore.kernel.org/r/20220915092910.711036-1-lieven.hey@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
Jason A. Donenfeld
Dominik Brodowski
Hangbin Liu
Jakub Kicinski
Liang He
Hangyu Hua
Sean Anderson
Peilin Ye
Jonathan Toppins
Michael Walle
Wen Gu
Ard Biesheuvel
Larysa Zaremba
Namhyung Kim
Arnaldo Carvalho de Melo
Adrian Hunter
Lieven Hey