Function hclge_set_vlan_filter_hw() is a bit too long, so add a new
function hclge_need_update_port_vlan() to simplify code and improve
code readability.
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
hclge_cfg_common_loopback() is a bit too long, so
encapsulate hclge_cfg_common_loopback_cmd_send() and
hclge_cfg_common_loopback_wait() two functions to
improve readability.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pass the FBC instance instead of the crtc to a bunch of places.
We also adjust intel_fbc_post_update() to do the
intel_fbc_get_reg_params() things instead of doing it from the lower
level function (which also gets called for front buffer tracking).
Nothing in there will change during front buffer updates.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211124113652.22090-8-ville.syrjala@linux.intel.com
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
There's no need to store all this stuff in intel_fbc_state_cache.
Just check it all against the plane/crtc states and store only
what we need. Probably more should get nuked still, but this
is a start.
So what we'll do is:
- each plane will check its own state and update its local
no_fbc_reason
- the per-plane no_fbc_reason (if any) then gets propagated
to the cache->no_fbc_reason while doing the actual update
- fbc->no_fbc_reason gets updated in the end with either
the value from the cache or directly from frontbuffer
tracking
It's still a bit messy, but should hopefuly get cleaned up
more in the future. At least now we can observe each plane's
reasons for rejecting FBC now more consistently, and we don't
have so mcuh redundant state store all over the place.
v2: store no_fbc_reason per-plane instead of per-pipe
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211124113652.22090-4-ville.syrjala@linux.intel.com
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
The ilk fbc watermark computation uses intel_fbc_is_active() which
is racy since we don't know whether FBC will be enabled or not at
some point. So let's just assume it will be if both HAS_FBC()
and the modparam agree.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211124113652.22090-2-ville.syrjala@linux.intel.com
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Let's just stick to 32bit mmio accesses so we can get rid
of the bare "uncore" reg access in display code. The register
are defined as 32bit in the spec anyway.
We could define a 64bit "de" variant I suppose, but doesn't
really make much sense just for this one case, and when we
start to use the DSB for this stuff we'd also need another
64bit variant for that. Just easier to do 32bit always.
While at it we can reorder stuff a bit so that we write the
registers in order of increasing offset (more or less).
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211201152552.7821-2-ville.syrjala@linux.intel.com
Reviewed-by: José Roberto de Souza <jose.souza@intel.com>
For Foxconn T99W175 device(sdx55 platform) in some host platform,
it would be unavailable once the host execute the err handler.
After checking, it's caused by the delay time too short to
get a successful reset.
Please see my test evidence as bewlow(BTW, I add some extra test logs
in function mhi_pci_reset_prepare and mhi_pci_reset_done):
When MHI_POST_RESET_DELAY_MS equals to 500ms:
Nov 4 14:30:03 jbd-ThinkEdge kernel: [ 146.222477] mhi mhi0: Device MHI is not in valid state
Nov 4 14:30:03 jbd-ThinkEdge kernel: [ 146.222628] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare reset
Nov 4 14:30:03 jbd-ThinkEdge kernel: [ 146.222631] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare mhi_soc_reset
Nov 4 14:30:03 jbd-ThinkEdge kernel: [ 146.222632] mhi mhi0: mhi_soc_reset write soc to reset
Nov 4 14:30:05 jbd-ThinkEdge kernel: [ 147.839993] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_done
Nov 4 14:30:05 jbd-ThinkEdge kernel: [ 147.902063] mhi-pci-generic 0000:2d:00.0: reset failed
When MHI_POST_RESET_DELAY_MS equals to 1000ms or 1500ms:
Nov 4 19:07:26 jbd-ThinkEdge kernel: [ 157.067857] mhi mhi0: Device MHI is not in valid state
Nov 4 19:07:26 jbd-ThinkEdge kernel: [ 157.068029] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare reset
Nov 4 19:07:26 jbd-ThinkEdge kernel: [ 157.068032] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare mhi_soc_reset
Nov 4 19:07:26 jbd-ThinkEdge kernel: [ 157.068034] mhi mhi0: mhi_soc_reset write soc to reset
Nov 4 19:07:29 jbd-ThinkEdge kernel: [ 159.607006] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_done
Nov 4 19:07:29 jbd-ThinkEdge kernel: [ 159.607152] mhi mhi0: Requested to power ON
Nov 4 19:07:51 jbd-ThinkEdge kernel: [ 181.302872] mhi mhi0: Failed to reset MHI due to syserr state
Nov 4 19:07:51 jbd-ThinkEdge kernel: [ 181.303011] mhi-pci-generic 0000:2d:00.0: failed to power up MHI controller
When MHI_POST_RESET_DELAY_MS equals to 2000ms:
Nov 4 17:51:08 jbd-ThinkEdge kernel: [ 147.180527] mhi mhi0: Failed to transition from PM state: Linkdown or Error Fatal Detect to: SYS ERROR Process
Nov 4 17:51:08 jbd-ThinkEdge kernel: [ 147.180535] mhi mhi0: Device MHI is not in valid state
Nov 4 17:51:08 jbd-ThinkEdge kernel: [ 147.180722] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare reset
Nov 4 17:51:08 jbd-ThinkEdge kernel: [ 147.180725] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_prepare mhi_soc_reset
Nov 4 17:51:08 jbd-ThinkEdge kernel: [ 147.180727] mhi mhi0: mhi_soc_reset write soc to reset
Nov 4 17:51:11 jbd-ThinkEdge kernel: [ 150.230787] mhi-pci-generic 0000:2d:00.0: mhi_pci_reset_done
Nov 4 17:51:11 jbd-ThinkEdge kernel: [ 150.230928] mhi mhi0: Requested to power ON
Nov 4 17:51:11 jbd-ThinkEdge kernel: [ 150.231173] mhi mhi0: Power on setup success
Nov 4 17:51:14 jbd-ThinkEdge kernel: [ 153.254747] mhi mhi0: Wait for device to enter SBL or Mission mode
I also tried big data like 3000, and it worked as well. 500ms may not be
enough for all support mhi device. We shall increase it to 2000ms
at least.
Link: https://lore.kernel.org/r/20211108113127.3938-1-slark_xiao@163.com
[mani: massaged commit message little bit, added Fixes tag and CCed stable]
Fixes: 8ccc3279fc ("mhi: pci_generic: Add support for reset")
Cc: stable@vger.kernel.org
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Slark Xiao <slark_xiao@163.com>
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Link: https://lore.kernel.org/r/20211126104951.35685-2-manivannan.sadhasivam@linaro.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tanix TX6 comes either with RTL8822BS or RTL8822CS wifi+bt combo module.
Wifi part is already enabled in tanix DTSI. Let's enable also bluetooth.
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Both, Tanix TX6 and Tanix TX6 mini, have SDIO wifi module, albeit
different. However, driver can be autoprobed via SDIO ID.
Add MMC1 node, so kernel can discover wifi module and load driver for
it.
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Tanix TX6 mini is less capable version of Tanix TX6 although it comes
with some features not present on Tanix TX6.
Basic specs:
- H6 SoC
- 2 GiB DDR3 RAM
- HDMI
- SPDIF
- 2x USB
- analogue audio
- CVBS
- SD card
- IR remote
- LED display
- fast ethernet
- XR819 wifi
- 16 GiB eMMC
Currently supported features doesn't differ that much from Tanix TX6,
but that will change soon.
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Tanix TX6 mini is, as its name suggest, very similar to Tanix TX6, but
with less features. It misses bluetooth support, it has less RAM, wifi
supports only 2.4G, it comes with different IR remote, etc.
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
There is another very similar device to Tanix TX6, namely Tanix TX6
mini. Because most of the board design is shared, it makes sense to have
common nodes in DTSI file.
Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Maxime Ripard <maxime@cerno.tech>
Fixes spurious wakeups from s0ix on Lenovo ThinkPad X1 Cargon Gen 9 on
lid close.
These wakeups are generated by interrupts from the ISH on changes to the
lid status.
By disabling the wake IRQ from the ISH we inhibit these spurious
wakeups while keeping the resume from LID open through the ACPI
interrupt.
Reports on the Lenovo forums indicate that Lenovo ThinkPad X1 Yoga Gen6
is also affected.
Fixes: ae02e5d40d ("HID: intel-ish-hid: ipc layer")
BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=214855
Signed-off-by: Thomas Weißschuh <linux@weissschuh.net>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Link: https://lore.kernel.org/r/20211130060117.3026-1-linux@weissschuh.net
The prodikeys HID driver only controls USB devices, yet did not have a
dependancy on USB_HID. This causes build errors on some configurations
like nios2 when building due to new changes to the prodikeys driver.
Reported-by: kernel test robot <lkp@intel.com>
Cc: stable@vger.kernel.org
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Link: https://lore.kernel.org/r/20211203081231.2856936-1-gregkh@linuxfoundation.org
The chicony HID driver only controls USB devices, yet did not have a
dependancy on USB_HID. This causes build errors on some configurations
like sparc when building due to new changes to the chicony driver.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: stable@vger.kernel.org
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Link: https://lore.kernel.org/r/20211203075927.2829218-1-gregkh@linuxfoundation.org
gt_pm selftest calculates engine ticks cycles and wall time
cycles by delta of respective engine elapsed TIMESTAMP and ktime
for period of 1000us.
It compares the engine ticks cycles with wall time cycles.
Disable local cpu interrupt so that interrupt handler does not
switch out the thread during measure_clocks() and prevent
miscalculation of engine tick cycles.
v2:
- nuke preempt_{disable,enable}, as disable_local_irq()
disable the preemption. (Chris)
Cc: Chris P Wilson <chris.p.wilson@intel.com>
Cc: Badal Nilawar <badal.nilawar@intel.com>
Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211130132005.6305-1-anshuman.gupta@intel.com
The trampoline_pgd only maps the 0xfffffff000000000-0xffffffffffffffff
range of kernel memory (with 4-level paging). This range contains the
kernel's text+data+bss mappings and the module mapping space but not the
direct mapping and the vmalloc area.
This is enough to get the application processors out of real-mode, but
for code that switches back to real-mode the trampoline_pgd is missing
important parts of the address space. For example, consider this code
from arch/x86/kernel/reboot.c, function machine_real_restart() for a
64-bit kernel:
#ifdef CONFIG_X86_32
load_cr3(initial_page_table);
#else
write_cr3(real_mode_header->trampoline_pgd);
/* Exiting long mode will fail if CR4.PCIDE is set. */
if (boot_cpu_has(X86_FEATURE_PCID))
cr4_clear_bits(X86_CR4_PCIDE);
#endif
/* Jump to the identity-mapped low memory code */
#ifdef CONFIG_X86_32
asm volatile("jmpl *%0" : :
"rm" (real_mode_header->machine_real_restart_asm),
"a" (type));
#else
asm volatile("ljmpl *%0" : :
"m" (real_mode_header->machine_real_restart_asm),
"D" (type));
#endif
The code switches to the trampoline_pgd, which unmaps the direct mapping
and also the kernel stack. The call to cr4_clear_bits() will find no
stack and crash the machine. The real_mode_header pointer below points
into the direct mapping, and dereferencing it also causes a crash.
The reason this does not crash always is only that kernel mappings are
global and the CR3 switch does not flush those mappings. But if theses
mappings are not in the TLB already, the above code will crash before it
can jump to the real-mode stub.
Extend the trampoline_pgd to contain all kernel mappings to prevent
these crashes and to make code which runs on this page-table more
robust.
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20211202153226.22946-5-joro@8bytes.org
Boris reported that in one of his randconfig builds, objtool got
infinitely stuck. Turns out there's trivial list corruption in the
pv_ops tracking when a function is both in a static table and in a code
assignment.
Avoid re-adding function to the pv_ops[] lists when they're already on
it.
Fixes: db2b0c5d7b ("objtool: Support pv_opsindirect calls for noinstr")
Reported-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Tested-by: Borislav Petkov <bp@alien8.de>
Link: https://lkml.kernel.org/r/20211202204534.GA16608@worktop.programming.kicks-ass.net
This workaround is documented a bit strangely in the bspec; it's listed
as an A0 workaround, but the description clarifies that the workaround
is implicitly handled by the hardware and what the driver really needs
to do is program a chicken bit to reenable some internal behavior.
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211116174818.2128062-3-matthew.d.roper@intel.com
Commit cd0fcf5af7 ("drm/i915: rename DISP_STEPPING->DISPLAY_STEP and
GT_STEPPING->GT_STEP") renamed all platforms' display stepping tests,
but the DG2 patches were still in-flight at that time and did not
incorporate the new naming scheme. Rename DG2's macro now for
consistency with other platforms.
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211116174818.2128062-2-matthew.d.roper@intel.com
Invalidate IC cache through pipe control command as part of the ctx
restore flow through indirect ctx pointer.
v2:
- Move pipe control from xcs indirect context to the rcs indirect
context. We'll eventually need this on the CCS engines too, but
support for those hasn't landed yet.
Cc: Chris Wilson <chris.p.wilson@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211116174818.2128062-5-matthew.d.roper@intel.com
If kstrtoint() fails then "lfs_num" is uninitialized and the warning
doesn't make any sense. Just delete it.
Fixes: 8ec8015a31 ("crypto: octeontx2 - add support to process the crypto request")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Use the FIELD_PREP() helper, instead of open-coding the same operation.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When caam queue is full (-ENOSPC), caam frees descriptor memory.
crypto-engine checks if retry support is true and h/w queue
is full(-ENOSPC), then requeue the crypto request.
During processing the requested descriptor again, caam gives below error.
(caam_jr 30902000.jr: 40000006: DECO: desc idx 0: Invalid KEY Command).
This patch adds a check to return when caam input ring is full
and retry support is true. so descriptor memory is not freed
and requeued request can be processed again.
Fixes: 2d653936eb ("crypto: caam - enable crypto-engine retry mechanism")
Signed-off-by: Gaurav Jain <gaurav.jain@nxp.com>
Reviewed-by: Horia Geantă <horia.geanta@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Use platform_get_irq_optional() to avoid a noisy error message when the
irq isn't specified. The irq is definitely optional given that we only
care about errors that are -EPROBE_DEFER here.
Cc: Thara Gopinath <thara.gopinath@linaro.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Clang warns:
drivers/fpga/stratix10-soc.c:431:9: warning: variable 'ret' is uninitialized when used here [-Wuninitialized]
return ret;
^~~
ret is only assigned in an error path now so just return 0 directly.
Fixes: 4ba0b2c294 ("fpga: mgr: Use standard dev_release for class driver")
Link: https://github.com/ClangBuiltLinux/linux/issues/1517
Reviewed-by: Russ Weight <russell.h.weight@intel.com>
Reviewed-by: Tom Rix <trix@redhat.com>
Acked-by: Xu Yilun <yilun.xu@intel.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Moritz Fischer <mdf@kernel.org>
There isn't any reason to not allow zero poll queues from user
viewpoint.
Also sometimes we need to compare io poll between poll mode and irq
mode, so not allowing poll queues is bad.
Fixes: 15dfc662ef ("null_blk: Fix handling of submit_queues and poll_queues attributes")
Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20211203023935.3424042-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Make sd_zbc_parse_report() use if/else when setting the write pointer,
instead of setting it unconditionally and then conditionally updating it.
Link: https://lore.kernel.org/r/20211201142821.64650-2-Niklas.Cassel@wdc.com
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
According to the ZBC (and ZAC) specification, a zone that has Zone Type set
to Conventional, must also have its Zone Condition set to "Not Write
Pointer".
Therefore, a conventional zone will never have Zone Condition set to
"Full", which means that we can omit the non-conventional prerequisite from
the zone full condition check.
Link: https://lore.kernel.org/r/20211201142821.64650-1-Niklas.Cassel@wdc.com
Suggested-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Niklas Cassel <niklas.cassel@wdc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Calling scsi_remove_host() before scsi_add_host() results in a crash:
BUG: kernel NULL pointer dereference, address: 0000000000000108
RIP: 0010:device_del+0x63/0x440
Call Trace:
device_unregister+0x17/0x60
scsi_remove_host+0xee/0x2a0
pm8001_pci_probe+0x6ef/0x1b90 [pm80xx]
local_pci_probe+0x3f/0x90
We cannot call scsi_remove_host() in pm8001_alloc() because scsi_add_host()
has not been called yet at that point in time.
Function call tree:
pm8001_pci_probe()
|
`- pm8001_pci_alloc()
| |
| `- pm8001_alloc()
| |
| `- scsi_remove_host()
|
`- scsi_add_host()
Link: https://lore.kernel.org/r/20211201041627.1592487-1-ipylypiv@google.com
Fixes: 05c6c029a4 ("scsi: pm80xx: Increase number of supported queues")
Reviewed-by: Vishakha Channapattan <vishakhavc@google.com>
Acked-by: Jack Wang <jinpu.wang@ionos.com>
Signed-off-by: Igor Pylypiv <ipylypiv@google.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
When we attempt to merge off the cached request path, we return NULL
if successful. This makes the caller believe that it's should allocate
a new request, and hence we end up with the bio both merged and associated
with a new request. This, predictably, leads to all sorts of crashes.
Pass in a pointer to the bio pointer, and clear it for the merge case.
Then the caller knows that the bio is already queued, and no new requests
need to get allocated.
Fixes: 5b13bc8a3f ("blk-mq: cleanup request allocation")
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Most SoC specific clock drivers are picked by respective SoC maintainer
and then sent to the clock maintainers on their way upstream.
This has however not been the case for the Qualcomm clock drivers -
which doesn't actually have a maintainer per MAINTAINERS and where the
framework maintainers have just carried the Qualcomm effort as well,
presumably as a result of Stephen's history.
Move the maintainership of the Qualcomm clock drivers to use the same
model as other SoC vendors and document the ownership by actually
introducing an entry in MAINTAINERS.
Cc: Stephen Boyd <sboyd@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: Stephen Boyd <sboyd@kernel.org>
Link: https://lore.kernel.org/r/20211203013901.3460496-1-bjorn.andersson@linaro.org
Currently rp_filter tests in fib_tests.sh:fib_rp_filter_test() are
failing. ping sockets are bound to dummy1 using the "-I" option
(SO_BINDTODEVICE), but socket lookup is failing when receiving ping
replies, since the routing table thinks they belong to dummy0.
For example, suppose ping is using a SOCK_RAW socket for ICMP messages.
When receiving ping replies, in __raw_v4_lookup(), sk->sk_bound_dev_if
is 3 (dummy1), but dif (skb_rtable(skb)->rt_iif) says 2 (dummy0), so the
raw_sk_bound_dev_eq() check fails. Similar things happen in
ping_lookup() for SOCK_DGRAM sockets.
These tests used to pass due to a bug [1] in iputils, where "ping -I"
actually did not bind ICMP message sockets to device. The bug has been
fixed by iputils commit f455fee41c07 ("ping: also bind the ICMP socket
to the specific device") in 2016, which is why our rp_filter tests
started to fail. See [2] .
Fixing the tests while keeping everything in one netns turns out to be
nontrivial. Rework the tests and build the following topology:
┌─────────────────────────────┐ ┌─────────────────────────────┐
│ network namespace 1 (ns1) │ │ network namespace 2 (ns2) │
│ │ │ │
│ ┌────┐ ┌─────┐ │ │ ┌─────┐ ┌────┐ │
│ │ lo │<───>│veth1│<────────┼────┼─>│veth2│<──────────>│ lo │ │
│ └────┘ ├─────┴──────┐ │ │ ├─────┴──────┐ └────┘ │
│ │192.0.2.1/24│ │ │ │192.0.2.1/24│ │
│ └────────────┘ │ │ └────────────┘ │
└─────────────────────────────┘ └─────────────────────────────┘
Consider sending an ICMP_ECHO packet A in ns2. Both source and
destination IP addresses are 192.0.2.1, and we use strict mode rp_filter
in both ns1 and ns2:
1. A is routed to lo since its destination IP address is one of ns2's
local addresses (veth2);
2. A is redirected from lo's egress to veth2's egress using mirred;
3. A arrives at veth1's ingress in ns1;
4. A is redirected from veth1's ingress to lo's ingress, again, using
mirred;
5. In __fib_validate_source(), fib_info_nh_uses_dev() returns false,
since A was received on lo, but reverse path lookup says veth1;
6. However A is not dropped since we have relaxed this check for lo in
commit 66f8209547 ("fib: relax source validation check for loopback
packets");
Making sure A is not dropped here in this corner case is the whole point
of having this test.
7. As A reaches the ICMP layer, an ICMP_ECHOREPLY packet, B, is
generated;
8. Similarly, B is redirected from lo's egress to veth1's egress (in
ns1), then redirected once again from veth2's ingress to lo's
ingress (in ns2), using mirred.
Also test "ping 127.0.0.1" from ns2. It does not trigger the relaxed
check in __fib_validate_source(), but just to make sure the topology
works with loopback addresses.
Tested with ping from iputils 20210722-41-gf9fb573:
$ ./fib_tests.sh -t rp_filter
IPv4 rp_filter tests
TEST: rp_filter passes local packets [ OK ]
TEST: rp_filter passes loopback packets [ OK ]
[1] https://github.com/iputils/iputils/issues/55
[2] f455fee41c
Reported-by: Hangbin Liu <liuhangbin@gmail.com>
Fixes: adb701d6cf ("selftests: add a test case for rp_filter")
Reviewed-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Acked-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20211201004720.6357-1-yepeilin.cs@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>