I didn't really want to do this, but as part of all the other changes to
the user copy loops, I've been looking at this horror.
I tried to clean it up multiple times, but every time I just found more
problems, and the way it's written, it's just too hard to fix them.
For example, the code is written to do quad-word alignment, and will use
regular byte accesses to get to that point. That's fairly simple, but
it means that any initial 8-byte alignment will be done with cached
copies.
However, the code then is very careful to do any 4-byte _tail_ accesses
using an uncached 4-byte write, and that was claimed to be relevant in
commit a82eee7424 ("x86/uaccess/64: Handle the caching of 4-byte
nocache copies properly in __copy_user_nocache()").
So if you do a 4-byte copy using that function, it carefully uses a
4-byte 'movnti' for the destination. But if you were to do a 12-byte
copy that is 4-byte aligned, it would _not_ do a 4-byte 'movnti'
followed by a 8-byte 'movnti' to keep it all uncached.
Instead, it would align the destination to 8 bytes using a
byte-at-a-time loop, and then do a 8-byte 'movnti' for the final 8
bytes.
The main caller that cares is __copy_user_flushcache(), which knows
about this insanity, and has odd cases for it all. But I just can't
deal with looking at this kind of "it does one case right, and another
related case entirely wrong".
And the code really wasn't fixable without hard drugs, which I try to
avoid.
So instead, rewrite it in a form that hopefully not only gets this
right, but is a bit more maintainable. Knock wood.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This reverts commit 14624d7247.
The termination table usage is requires for DMFS steering mode as firmware
doesn't support mixed table destinations list which causes following
syndrome with hairpin rules:
[81922.283225] mlx5_core 0000:08:00.0: mlx5_cmd_out_err:803:(pid 25977): SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0xaca205), err(-22)
Fixes: 14624d7247 ("net/mlx5e: Don't use termination table when redundant")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
On failing to create promisc flow steering table, the pointer is
returned with an error. Nullify it so unloading the driver won't try to
destroy a non existing table.
Failing to create promisc table may happen over BF devices when the ARM
side is going through a firmware tear down. The host side start a
reload flow. While the driver unloads, it tries to remove the promisc
table. Remove WARN in this state as it is a valid error flow.
Fixes: 1c46d7409f ("net/mlx5e: Optimize promiscuous mode")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Use the same timeout for sync reset flow and health recovery flow, since
the former involves driver's recovery from firmware reset, which is
similar to health recovery. Otherwise, in some cases, such as a firmware
upgrade on the DPU, the firmware pre-init bit may not be ready within
current timeout and the driver will abort loading back after reset.
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Fixes: 37ca95e62e ("net/mlx5: Increase FW pre-init timeout for health recovery")
Reviewed-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
This reverts commit 5977ac3910.
Revert this patch as we need the "recovery" arg back in mlx5_load_one()
function. This arg will be used in the next patch for using recovery
timeout during sync reset flow.
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
On representor init rx error flow the flow steering pointer is being
released so mlx5e_attach_netdev() doesn't have a valid fs pointer
in its error flow. Make sure the pointer is nullified when released
and add a check in mlx5e_fs_cleanup() to verify fs is not null
as representor cleanup callback would be called anyway.
Fixes: af8bbf7300 ("net/mlx5e: Convert mlx5e_flow_steering member of mlx5e_priv to pointer")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
The cited commit causes a regression. Tunnel device is not released
after tc update skb if skb needs to be freed. The following error
message will be printed:
unregister_netdevice: waiting for vxlan1 to become free. Usage count = 11
Fix it by releasing tunnel device if skb needs to be freed.
Fixes: 93a1ab2c54 ("net/mlx5: Refactor tc miss handling to a single function")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Source port rewrite (forward to ovs internal port or statck device) isn't
supported in the rule of split action. So there is no indirect table in
split rule. The cited commit destroyes indirect table in split rule. The
indirect table for other rules will be destroyed wrongly. It will cause
traffic loss.
Fix it by removing the destroy function in split rule. And also remove
the destroy function in error flow.
Fixes: 10742efc20 ("net/mlx5e: VF tunnel TX traffic offloading")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Currently when creating per vport table, create flags are hardcoded.
Devlink encap mode is set based on user input and HW capability.
Create per vport table based on devlink encap mode.
Fixes: c796bb7cd2 ("net/mlx5: E-switch, Generalize per vport table API")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
The code already clones post action attributes in
mlx5e_clone_flow_attr_for_post_act(). Creating another copy in
mlx5e_tc_post_act_add() is a erroneous leftover from original
implementation. Instead, assign handle->attribute to post_attr provided by
the caller. Note that cloning the attribute second time is not just
wasteful but also causes issues like second copy not being properly updated
in neigh update code which leads to following use-after-free:
Feb 21 09:02:00 c-237-177-40-045 kernel: BUG: KASAN: use-after-free in mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_report+0xbb/0x1a0
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel: __kasan_kmalloc+0x7a/0x90
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_free_info+0x2a/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: ____kasan_slab_free+0x11a/0x1b0
Feb 21 09:02:00 c-237-177-40-045 kernel: page dumped because: kasan: bad access detected
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0: mlx5_cmd_out_err:803:(pid 8833): SET_FLOW_TABLE_ENTRY(0x936) op_mod(0x0) failed, status bad resource state(0x9), syndrome (0xf2ff71), err(-22)
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0 enp8s0f0: Failed to add post action rule
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_core 0000:08:00.0: mlx5e_tc_encap_flows_add:190:(pid 8833): Failed to update flow post acts, -22
Feb 21 09:02:00 c-237-177-40-045 kernel: Call Trace:
Feb 21 09:02:00 c-237-177-40-045 kernel: <TASK>
Feb 21 09:02:00 c-237-177-40-045 kernel: dump_stack_lvl+0x57/0x7d
Feb 21 09:02:00 c-237-177-40-045 kernel: print_report+0x170/0x471
Feb 21 09:02:00 c-237-177-40-045 kernel: ? mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_report+0xbb/0x1a0
Feb 21 09:02:00 c-237-177-40-045 kernel: ? mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_cmd_set_fte+0x200d/0x24c0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: ? __module_address.part.0+0x62/0x200
Feb 21 09:02:00 c-237-177-40-045 kernel: ? mlx5_cmd_stub_create_flow_table+0xd0/0xd0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: ? __raw_spin_lock_init+0x3b/0x110
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_cmd_create_fte+0x80/0xb0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: add_rule_fg+0xe80/0x19c0 [mlx5_core]
--
Feb 21 09:02:00 c-237-177-40-045 kernel: Allocated by task 13476:
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel: __kasan_kmalloc+0x7a/0x90
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_packet_reformat_alloc+0x7b/0x230 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_tc_tun_create_header_ipv4+0x977/0xf10 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_attach_encap+0x15b4/0x1e10 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: post_process_attr+0x305/0xa30 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_tc_add_fdb_flow+0x4c0/0xcf0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: __mlx5e_add_fdb_flow+0x7cf/0xe90 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_configure_flower+0xcaa/0x4b90 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_rep_setup_tc_cls_flower+0x99/0x1b0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_rep_setup_tc_cb+0x133/0x1e0 [mlx5_core]
--
Feb 21 09:02:00 c-237-177-40-045 kernel: Freed by task 8833:
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_stack+0x1e/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_set_track+0x21/0x30
Feb 21 09:02:00 c-237-177-40-045 kernel: kasan_save_free_info+0x2a/0x40
Feb 21 09:02:00 c-237-177-40-045 kernel: ____kasan_slab_free+0x11a/0x1b0
Feb 21 09:02:00 c-237-177-40-045 kernel: __kmem_cache_free+0x1de/0x400
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5_packet_reformat_dealloc+0xad/0x100 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_tc_encap_flows_del+0x3c0/0x500 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_rep_update_flows+0x40c/0xa80 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: mlx5e_rep_neigh_update+0x473/0x7a0 [mlx5_core]
Feb 21 09:02:00 c-237-177-40-045 kernel: process_one_work+0x7c2/0x1310
Feb 21 09:02:00 c-237-177-40-045 kernel: worker_thread+0x59d/0xec0
Feb 21 09:02:00 c-237-177-40-045 kernel: kthread+0x28f/0x330
Fixes: 8300f22526 ("net/mlx5e: Create new flow attr for multi table actions")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
To be consistent with the other enum keys use OP_MOD
instead of OP_MODE.
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Remove unused function which also seems a duplicate
of esw_port_metadata_set().
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
The passded esw->dev is redundant as esw being passed and esw->dev being
used inside.
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Linking the NAPI to the rq page_pool to improve page_pool cache
usage during skb recycling.
Here are the observed improvements for a iperf single stream
test case:
- For 1500 MTU and legacy rq, seeing a 20% improvement of cache usage.
- For 9K MTU, seeing 33-40 % page_pool cache usage improvements for
both striding and legacy rq (depending if the application is running on
the same core as the rq or not).
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
When the XDP handler marks the data for transmission (XDP_TX),
it is incorrect to release the page fragment. Instead, the
fragments should be marked as MLX5E_WQE_FRAG_SKIP_RELEASE
because XDP will release the page directly to the page_pool
(page_pool_put_defragged_page) after TX completion.
The linear case already does this. This patch fixes the
nonlinear part as well.
Also, the looping over the fragments was incorrect: When handling
pages after XDP_TX in the legacy rq nonlinear case, the loop was
skipping the first wqe fragment.
Fixes: 3f93f82988 ("net/mlx5e: RX, Defer page release in legacy rq for better recycling")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
mlx5e_free_rx_descs is responsible for calling the dealloc_wqe op which
returns pages to the page_pool. This can happen during flush or close.
For XSK, the regular RQ is flushed (when replaced by the XSK RQ) and
also closed later. This is normally not a problem as the wqe list is
empty on a second call to mlx5e_free_rx_descs. However, for striding RQ,
the previously released wqes from the list will appear as missing
and will be released a second time by mlx5e_free_rx_missing_descs.
This patch sets the no release bits on the striding RQ wqes in the
dealloc_wqe op to prevent releasing the pages a second time.
Please note that the bits are set only in the control path during
close and not in the data path.
Fixes: 4c2a132368 ("net/mlx5e: RX, Defer page release in striding rq for better recycling")
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Create a new devlink health reporter for representor interface, which
reports the values of representor vnic diagnostic counters when diagnosed.
This patch will allow admins to monitor VF diagnostic counters through
the representor-interface vnic reporter.
Example of usage:
$ devlink health diagnose pci/0000:08:00.0/65537 reporter vnic
vNIC env counters:
total_error_queues: 0 send_queue_priority_update_flow: 0
comp_eq_overrun: 0 async_eq_overrun: 0 cq_overrun: 0
invalid_command: 0 quota_exceeded_command: 0
nic_receive_steering_discard: 0
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Create a vnic devlink health reporter for PFs/VFs interfaces.
The reporter's diagnose callback displays the values of vNIC/vport
transport debug counters of PFs/VFs, as follows:
$ devlink health diagnose pci/0000:08:00.0 reporter vnic
vNIC env counters:
total_error_queues: 0 send_queue_priority_update_flow: 0
comp_eq_overrun: 0 async_eq_overrun: 0 cq_overrun: 0
invalid_command: 0 quota_exceeded_command: 0
nic_receive_steering_discard: 0
Moreover, add documentation on the reporter functionality and the
counters description.
While at it, expose the vNIC counters diagnose function to be used by
the downstream patch, which will reveal the counters for representor
interfaces.
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
This reverts commit 606e6a72e2 which exposes
the vnic diagnostic counters via debugfs. Instead, The upcoming series will
expose the same counters through devlink health reporter.
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
This reverts commit 4fe1b3a5f8, which
exposes the steering dropped packets counter via debugfs. The upcoming
series will expose the counter via devlink health reporter instead
of debugfs.
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Add counters for number of buddies that are currently in use per domain
per buddy type (STE, MODIFY-HEADER, MODIFY-PATTERN).
Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Add additinal items to domain info dump: Linux version and device name.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
When certain ICM chunk is no longer needed, it needs to be freed.
Fully freeing ICM memory involves issuing FW SYNC_STEERING command.
This is very time consuming, and it is impractical to do it for every
freed chunk.
Instead, we manage these 'freed' chunks in hot list (list of chunks
that are not required by SW any more, but HW might still access them).
When size of the hot list reaches certain threshold, we purge it and
issue SYNC_STEERING FW command.
There is one threshold for all the different ICM types, which is not
optimal, as different ICM types require different approach: STEs pool
is very large, and it is very 'dynamic' in its nature, so letting hot
list to become too large will result in a significant perf hiccup when
purging the hot list. Modify action is much smaller and less dynamic,
so we can let the hot list to grow to almost the size of the whole pool.
This patch fixes this problem: instead of having same hot memory
threshold for all the pools, sync operation will be triggered in
accordance with the ICM type.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
The steering dump parser expects to see 0 as rewrite num of actions
in case pattern/args aren't supported - parsing of legacy modify header
is based on this assumption.
Fix this to align to parser's expectation.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
The existing clk documentation has a section that talks about the
clk_ignore_unused kernel parameter. Add additional documentation that
describes how to log which clocks the kernel disables on bootup. This
will log messages like the following to the console on bootup:
[ 1.268115] clk: Disabling unused clocks
[ 1.272167] clk_disable: gcc_usb_clkref_en
[ 1.276389] clk_disable: gcc_usb30_sec_sleep_clk
[ 1.281131] clk_disable: gcc_usb30_prim_sleep_clk
...
Signed-off-by: Brian Masney <bmasney@redhat.com>
Link: https://lore.kernel.org/r/20230411192153.289688-1-bmasney@redhat.com
[jc: turned parameters into a literal block]
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
There is a typo in the sentence "A kernel developer must be
conscience ...". The word conscience should be conscious.
This patch fixes it.
Signed-off-by: Lin Yu Chen <starpt.official@gmail.com>
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Link: https://lore.kernel.org/r/20230412183739.89894-1-starpt.official@gmail.com
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
The "Select the recipients for your patch" part about CC-ing mailing
lists is a bit vague and might be understood that only some lists should
be Cc-ed. That's not what most of the maintainers expect. For given
code, associated mailing list must always be CC-ed, because the list is
used for reviewing and testing patches. Example are the Devicetree
bindings patches, which are tested iff Devicetree mailing list is CC-ed.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20230413165501.47442-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Turns out rules about PT_INTERP, PT_GNU_STACK and PT_GNU_PROPERTY
program headers are slightly different.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Link: https://lore.kernel.org/r/88d3f1bb-f4e0-4c40-9304-3843513a1262@p183
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
ST's STiH415 and STiH416 platforms support have been removed since
a long time already. This commit updates the sti related documentation
overview to remove related entries and update the sti part to add
STiH407/STiH410 and STiH418 platforms which are still actively
supported.
Signed-off-by: Alain Volmat <avolmat@me.com>
Link: https://lore.kernel.org/r/20230416185349.18156-1-avolmat@me.com
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
We have long disabled the "html_use_smartypants" option to prevent Sphinx
from mangling "--" sequences (among others). Unfortunately, Sphinx changed
that option to "smartquotes" in the 1.6.6 release, and seemingly didn't see
fit to warn about the use of the obsolete option, resulting in the
aforementioned mangling returning. Disable this behavior again and hope
that the option name stays stable for a while.
Reported-by: Zipeng Zhang <zhangzipeng0@foxmail.com>
Link: https://lore.kernel.org/lkml/tencent_CB1A298D31FD221496FF657CD7EF406E6605@qq.com
Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Eduard Zingerman says:
====================
Florian Westphal found a bug in test_loader.c processing of __retval tag.
Because of this bug the function test_loader.c:do_prog_test_run()
never executed and all __retval test tags were ignored. See [1].
Fix for this bug uncovers two additional bugs:
- During test_verifier tests migration to inline assembly (see [2])
I missed the fact that some tests require maps to contain mock values;
- Some issue with a new refcounted_kptr test, which causes kernel to
produce dead lock and refcount saturation warnings when subject to
libbpf's bpf_test_run_opts().
This series fixes the bug in __retval() processing, and address the
issue with test maps not being populated. The issue in refcounted_kptr
is not addressed, __retval tags in those tests are commented out.
I found that the following tests depend on test maps being populated:
- progs/verifier_array_access.c
- verifier/value_ptr_arith.c (planned for migration to inline assembly)
Given the small amount of these tests I decided to opt for simple
non-generic solution (see patch #4).
[1] https://lore.kernel.org/bpf/f4c4aee644425842ee6aa8edf1da68f0a8260e7c.camel@gmail.com/T/
[2] https://lore.kernel.org/bpf/20230325025524.144043-1-eddyz87@gmail.com/
====================
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Two test cases:
- "valid read map access into a read-only array 1" and
- "valid read map access into a read-only array 2"
Expect that map_array_ro map is filled with mock data. This logic was
not taken into acount during initial test conversion.
This commit modifies prog_tests/verifier.c entry point for this test
to fill the map.
Fixes: a3c830ae02 ("selftests/bpf: verifier/array_access.c converted to inline assembly")
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-5-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
When a test case is annotated with __retval tag the test_loader engine
would use libbpf's bpf_prog_test_run_opts() to do a test run of the
program and compare retvals.
This commit allows to perform arbitrary actions on bpf object right
before test loader invokes bpf_prog_test_run_opts(). This could be
used to setup some state for program execution, e.g. fill some maps.
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-4-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Florian Westphal found a bug in and suggested a fix for test_loader.c
processing of __retval tag. Because of this bug the function
test_loader.c:do_prog_test_run() never executed and all __retval test
tags were ignored.
If this bug is fixed a number of test cases from
progs/verifier_array_access.c fail with retval not matching the
expected value. This test was recently converted to use test_loader.c
and inline assembly in [1]. When doing the conversion I missed the
important detail of test_verifier.c operation: when it creates
fixup_map_array_ro, fixup_map_array_wo and fixup_map_array_small it
populates these maps with a dummy record.
Disabling the __retval checks for the affected verifier_array_access
in this commit to avoid false-postivies in any potential bisects.
The issue is addressed in the next patch.
I verified that the __retval tags are now respected by changing
expected return values for all tests annotated with __retval, and
checking that these tests started to fail.
[1] https://lore.kernel.org/bpf/20230325025524.144043-1-eddyz87@gmail.com/
Fixes: 19a8e06f5f ("selftests/bpf: Tests execution support for test_loader.c")
Reported-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/bpf/f4c4aee644425842ee6aa8edf1da68f0a8260e7c.camel@gmail.com/T/
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-3-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Florian Westphal found a bug in test_loader.c processing of __retval
tag. Because of this bug the function test_loader.c:do_prog_test_run()
never executed and all __retval test tags were ignored. This hid an
issue with progs/refcounted_kptr.c tests.
When __retval tag bug is fixed and refcounted_kptr.c tests are run
kernel reports various issues and eventually hangs. Shortest reproducer
is the following command run a few times:
$ for i in $(seq 1 4); do (./test_progs --allow=refcounted_kptr &); done
Commenting out __retval tags for these tests until this issue is resolved.
Reported-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/bpf/f4c4aee644425842ee6aa8edf1da68f0a8260e7c.camel@gmail.com/T/
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20230420232317.2181776-2-eddyz87@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
The recent support for inline annotations in control flow graphs
generated by bpftool introduced the usage of the "__fallthrough" macro
in a switch/case block in btf_dumper.c. This change went through the
bpf-next tree, but resulted in a merge conflict in linux-next, because
this macro has been renamed "fallthrough" (no underscores) in the
meantime.
To address the conflict, we temporarily switch to a simple comment
instead of a macro.
Related: commit f7a858bffc ("tools: Rename __fallthrough to fallthrough")
Fixes: 9fd496848b ("bpftool: Support inline annotations when dumping the CFG of a program")
Reported-by: Sven Schnelle <svens@linux.ibm.com>
Reported-by: Thomas Richter <tmricht@linux.ibm.com>
Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/all/yt9dttxlwal7.fsf@linux.ibm.com/
Link: https://lore.kernel.org/bpf/20230412123636.2358949-1-tmricht@linux.ibm.com/
Link: https://lore.kernel.org/bpf/20230420003333.90901-1-quentin@isovalent.com
10 ms is a lot of time to spend busy-waiting. Sleeping is clearly
allowed here, because we have just returned from ice_sq_send_cmd(),
which takes a mutex.
On kernels with HZ=100, this msleep may be twice as long, but I don't
think it matters.
I did not actually observe any retries happening here.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
The 'buf_cpy'-related code in ice_sq_send_cmd_retry() looks broken.
'buf' is nowhere copied into 'buf_cpy'.
The reason this does not cause problems is that all commands for which
'is_cmd_for_retry' is true go with a NULL buf.
Let's remove 'buf_cpy'. Add a WARN_ON in case the assumption no longer
holds in the future.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
The driver polls for ice_sq_done() with a 100 µs period for up to 1 s
and it uses udelay to do that.
Let's use usleep_range instead. We know sleeping is allowed here,
because we're holding a mutex (cq->sq_lock). To preserve the total
max waiting time, measure the timeout in jiffies.
ICE_CTL_Q_SQ_CMD_TIMEOUT is used also in ice_release_res(), but there
the polling period is 1 ms (i.e. 10 times longer). Since the timeout was
expressed in terms of the number of loops, the total timeout in this
function is 10 s. I do not know if this is intentional. This patch keeps
it.
The patch lowers the CPU usage of the ice-gnss-<dev_name> kernel thread
on my system from ~8 % to less than 1 %.
I received a report of high CPU usage with ptp4l where the busy-waiting
in ice_sq_send_cmd dominated the profile. This patch has been tested in
that usecase too and it made a huge improvement there.
Tested-by: Brent Rowsell <browsell@redhat.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
sq_cmd_timeout is initialized to ICE_CTL_Q_SQ_CMD_TIMEOUT and never
changed, so just use the constant directly.
Suggested-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Double the GNSS data polling interval from 10 ms to 20 ms.
According to Karol Kolacinski from the Intel team, they have been
planning to make this change.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
The ice-gnss-<dev_name> kernel thread, which reads data from the u-blox
GNSS module, keep a CPU core almost 100% busy. The main reason is that
it busy-waits for data to become available.
A simple improvement would be to replace the "mdelay(10);" in
ice_gnss_read() with sleeping. A better fix is to not do any waiting
directly in the function and just requeue this delayed work as needed.
The advantage is that canceling the work from ice_gnss_exit() becomes
immediate, rather than taking up to ~2.5 seconds (ICE_MAX_UBX_READ_TRIES
* 10 ms).
This lowers the CPU usage of the ice-gnss-<dev_name> thread on my system
from ~90 % to ~8 %.
I am not sure if the larger 0.1 s pause after inserting data into the
gnss subsystem is really necessary, but I'm keeping that as it was.
Of course, ideally the driver would not have to poll at all, but I don't
know if the E810 can watch for GNSS data availability over the i2c bus
by itself and notify the driver.
Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Reviewed-by: Arkadiusz Kubalewski <arkadiusz.kubalewski@intel.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Tested-by: Sunitha Mekala <sunithax.d.mekala@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
ethtool uses `ETHTOOL_GRXRINGS` to compute how many queues are supported
by RSS. The driver should return the smaller of either:
- The maximum number of RSS queues the device supports, OR
- The number of RX queues configured
Prior to this change, running `ethtool -X $iface default` fails if the
number of queues configured is larger than the number supported by RSS,
even though changing the queue count correctly resets the flowhash to
use all supported queues.
Other drivers (for example, i40e) will succeed but the flow hash will
reset to support the maximum number of queues supported by RSS, even if
that amount is smaller than the configured amount.
Prior to this change:
$ sudo ethtool -L eth1 combined 20
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 20 RX ring(s):
0: 0 1 2 3 4 5 6 7
8: 8 9 10 11 12 13 14 15
16: 0 1 2 3 4 5 6 7
24: 8 9 10 11 12 13 14 15
32: 0 1 2 3 4 5 6 7
...
You can see that the flowhash was correctly set to use the maximum
number of queues supported by the driver (16).
However, asking the NIC to reset to "default" fails:
$ sudo ethtool -X eth1 default
Cannot set RX flow hash configuration: Invalid argument
After this change, the flowhash can be reset to default which will use
all of the available RSS queues (16) or the configured queue count,
whichever is smaller.
Starting with eth1 which has 10 queues and a flowhash distributing to
all 10 queues:
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 10 RX ring(s):
0: 0 1 2 3 4 5 6 7
8: 8 9 0 1 2 3 4 5
16: 6 7 8 9 0 1 2 3
...
Increasing the queue count to 48 resets the flowhash to distribute to 16
queues, as it did before this patch:
$ sudo ethtool -L eth1 combined 48
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
0: 0 1 2 3 4 5 6 7
8: 8 9 10 11 12 13 14 15
16: 0 1 2 3 4 5 6 7
...
Due to the other bugfix in this series, the flowhash can be set to use
queues 0-5:
$ sudo ethtool -X eth1 equal 5
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
0: 0 1 2 3 4 0 1 2
8: 3 4 0 1 2 3 4 0
16: 1 2 3 4 0 1 2 3
...
Due to this bugfix, the flowhash can be reset to default and use 16
queues:
$ sudo ethtool -X eth1 default
$ sudo ethtool -x eth1
RX flow hash indirection table for eth1 with 16 RX ring(s):
0: 0 1 2 3 4 5 6 7
8: 8 9 10 11 12 13 14 15
16: 0 1 2 3 4 5 6 7
...
Fixes: 91cd94bfe4 ("ixgbe: add basic support for setting and getting nfc controls")
Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
