Skip to content

selftests/bpf: add benchmark for ternary search tree map #231

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
wants to merge 2 commits into from
Closed

selftests/bpf: add benchmark for ternary search tree map #231

wants to merge 2 commits into from

Conversation

kernel-patches-bot
Copy link

Pull request for series with
subject: selftests/bpf: add benchmark for ternary search tree map
version: 1
url: https://patchwork.kernel.org/project/netdevbpf/list/?series=627818

Nobody and others added 2 commits March 31, 2022 05:12
adding ci files
8c07715
selftests/bpf: add benchmark for ternary search tree map
17b0b69 
Add a benchmark for ternary search tree map to compare lookup
performance and memory usage with hash table for string key.

To test the space efficiency of tst for string key with common prefix
(hierarchical key), the benchmark constructs a string key up to 256
bytes: up to 64 bytes base name and up to 191 bytes prefix. So the
space saving is about 2~3 fold and it can be confirmed from data of
tst-N-hk-dl section or tst-N-hk-sl section below.

It also tests the space saving when there are differences between string
length: the max length is 255 bytes and min length is 16. And there are
about 0.5~0.8 space saving in tst compared with htab as data of
tst-N-fk-dl section shows.

When the length of string key is the same and there are very few common
prefixes between these key, the space usage of tearnary will be greater
than hash table and it is about 3%~30% more space usage compared with
hash table as shown in tst-N-fk-sl section.

For the hierarchical key case, the lookup performance of tst is about
x2~x3 slower compared with hash table when the number of elements is
less than 1M. And the performance gap decreases when the number of
element increases.

The following are the detailed outputs of benchmark:

Notations:

* hk: hierarchical string key
* fk: flat string key (non-hierarchical string key)
* dl: the lengths of string keys are different
* sl: the lengths of string key are the same
* N: the number of string key

Details:

tst-1000-hk-dl            2.398 ± 0.000M/s (drops 0.000 ± 0.000M/s, mem 0.144 MiB)
tst-10000-hk-dl           2.458 ± 0.001M/s (drops 0.000 ± 0.000M/s, mem 1.479 MiB)
tst-100000-hk-dl          2.744 ± 0.006M/s (drops 0.000 ± 0.000M/s, mem 12.566 MiB)
tst-1000000-hk-dl         2.364 ± 0.007M/s (drops 0.000 ± 0.000M/s, mem 120.862 MiB)
tst-10000000-hk-dl        2.467 ± 0.032M/s (drops 0.000 ± 0.000M/s, mem 1204.017 MiB)

htab-1000-hk-dl           5.930 ± 0.006M/s (drops 0.000 ± 0.000M/s, mem 0.496 MiB)
htab-10000-hk-dl          5.070 ± 0.005M/s (drops 0.000 ± 0.000M/s, mem 4.960 MiB)
htab-100000-hk-dl         2.765 ± 0.022M/s (drops 0.000 ± 0.000M/s, mem 49.591 MiB)
htab-1000000-hk-dl        1.870 ± 0.032M/s (drops 0.000 ± 0.000M/s, mem 495.911 MiB)
htab-10000000-hk-dl       1.854 ± 0.104M/s (drops 0.000 ± 0.000M/s, mem 4959.095 MiB)

tst-1000-hk-sl            0.172 ± 0.384M/s (drops 0.000 ± 0.000M/s, mem 0.184 MiB)
tst-10000-hk-sl           1.831 ± 0.000M/s (drops 0.000 ± 0.000M/s, mem 1.821 MiB)
tst-100000-hk-sl          1.724 ± 0.002M/s (drops 0.000 ± 0.000M/s, mem 15.161 MiB)
tst-1000000-hk-sl         1.611 ± 0.005M/s (drops 0.000 ± 0.000M/s, mem 148.276 MiB)
tst-10000000-hk-sl        1.608 ± 0.007M/s (drops 0.000 ± 0.000M/s, mem 1480.046 MiB)

htab-1000-hk-sl           5.913 ± 0.001M/s (drops 0.000 ± 0.000M/s, mem 0.496 MiB)
htab-10000-hk-sl          5.067 ± 0.003M/s (drops 0.000 ± 0.000M/s, mem 4.960 MiB)
htab-100000-hk-sl         2.816 ± 0.015M/s (drops 0.000 ± 0.000M/s, mem 49.592 MiB)
htab-1000000-hk-sl        1.841 ± 0.026M/s (drops 0.000 ± 0.000M/s, mem 495.911 MiB)
htab-10000000-hk-sl       1.847 ± 0.102M/s (drops 0.000 ± 0.000M/s, mem 4959.107 MiB)

tst-1000-fk-dl            0.180 ± 0.402M/s (drops 0.000 ± 0.000M/s, mem 0.329 MiB)
tst-10000-fk-dl           2.546 ± 0.001M/s (drops 0.000 ± 0.000M/s, mem 3.211 MiB)
tst-100000-fk-dl          1.702 ± 0.003M/s (drops 0.000 ± 0.000M/s, mem 27.835 MiB)
tst-1000000-fk-dl         0.941 ± 0.006M/s (drops 0.000 ± 0.000M/s, mem 262.322 MiB)
tst-10000000-fk-dl        0.658 ± 0.039M/s (drops 0.000 ± 0.000M/s, mem 2618.858 MiB)

htab-1000-fk-dl           5.903 ± 0.002M/s (drops 0.000 ± 0.000M/s, mem 0.496 MiB)
htab-10000-fk-dl          5.057 ± 0.004M/s (drops 0.000 ± 0.000M/s, mem 4.960 MiB)
htab-100000-fk-dl         2.748 ± 0.019M/s (drops 0.000 ± 0.000M/s, mem 49.595 MiB)
htab-1000000-fk-dl        1.857 ± 0.029M/s (drops 0.000 ± 0.000M/s, mem 495.911 MiB)
htab-10000000-fk-dl       1.853 ± 0.103M/s (drops 0.000 ± 0.000M/s, mem 4959.107 MiB)

tst-1000-fk-sl            0.094 ± 0.210M/s (drops 0.000 ± 0.000M/s, mem 0.640 MiB)
tst-10000-fk-sl           1.656 ± 0.000M/s (drops 0.000 ± 0.000M/s, mem 6.251 MiB)
tst-100000-fk-sl          1.088 ± 0.005M/s (drops 0.000 ± 0.000M/s, mem 53.708 MiB)
tst-1000000-fk-sl         0.732 ± 0.020M/s (drops 0.000 ± 0.000M/s, mem 513.421 MiB)
tst-10000000-fk-sl        0.581 ± 0.031M/s (drops 0.000 ± 0.000M/s, mem 5139.383 MiB)

htab-1000-fk-sl           5.861 ± 0.004M/s (drops 0.000 ± 0.000M/s, mem 0.496 MiB)
htab-10000-fk-sl          5.077 ± 0.006M/s (drops 0.000 ± 0.000M/s, mem 4.959 MiB)
htab-100000-fk-sl         2.771 ± 0.020M/s (drops 0.000 ± 0.000M/s, mem 49.592 MiB)
htab-1000000-fk-sl        1.833 ± 0.029M/s (drops 0.000 ± 0.000M/s, mem 495.911 MiB)
htab-10000000-fk-sl       1.828 ± 0.099M/s (drops 0.000 ± 0.000M/s, mem 4959.107 MiB)

Signed-off-by: Hou Tao <houtao1@huawei.com>
@kernel-patches-bot
Copy link
Author

Master branch: d717e4c
series: https://patchwork.kernel.org/project/netdevbpf/list/?series=627818
version: 1

@kernel-patches-bot
Copy link
Author

At least one diff in series https://patchwork.kernel.org/project/netdevbpf/list/?series=627818 expired. Closing PR.

@kernel-patches-bot kernel-patches-bot deleted the series/627818=>bpf-next branch April 2, 2022 20:35
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jun 7, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
df99e69 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #224     tc_links_after:OK
  #225     tc_links_append:OK
  #226     tc_links_basic:OK
  #227     tc_links_before:OK
  #228     tc_links_both:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_first:OK
  #232     tc_links_invalid:OK
  #233     tc_links_last:OK
  #234     tc_links_prepend:OK
  #235     tc_links_replace:OK
  #236     tc_links_revision:OK
  Summary: 13/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jun 8, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
fdd0f47 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #224     tc_links_after:OK
  #225     tc_links_append:OK
  #226     tc_links_basic:OK
  #227     tc_links_before:OK
  #228     tc_links_both:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_first:OK
  #232     tc_links_invalid:OK
  #233     tc_links_last:OK
  #234     tc_links_prepend:OK
  #235     tc_links_replace:OK
  #236     tc_links_revision:OK
  Summary: 13/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 7, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
e886bcf 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 7, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
b937f9f 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 9, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
7bd5ef6 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 10, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
b5e9d96 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 10, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
5edb513 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 11, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
112c39b 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 14, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
3af5244 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 17, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
2582285 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
5befaac 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
1947966 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
f477552 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
ce0614b 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
4446bdb 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
d160133 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jul 19, 2023
@borkmann
selftests/bpf: Add mprog API tests for BPF tcx links
c6d479b 
Add a big batch of test coverage to assert all aspects of the tcx link API:

  # ./vmtest.sh -- ./test_progs -t tc_links
  [...]
  #225     tc_links_after:OK
  #226     tc_links_append:OK
  #227     tc_links_basic:OK
  #228     tc_links_before:OK
  #229     tc_links_chain_classic:OK
  #230     tc_links_dev_cleanup:OK
  #231     tc_links_invalid:OK
  #232     tc_links_prepend:OK
  #233     tc_links_replace:OK
  #234     tc_links_revision:OK
  Summary: 10/0 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/r/20230719140858.13224-9-daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request Jan 13, 2025
@melissawen @alexdeucher
drm/amd/display: fix divide error in DM plane scale calcs
5225fd2 
dm_get_plane_scale doesn't take into account plane scaled size equal to
zero, leading to a kernel oops due to division by zero. Fix by setting
out-scale size as zero when the dst size is zero, similar to what is
done by drm_calc_scale(). This issue started with the introduction of
cursor ovelay mode that uses this function to assess cursor mode changes
via dm_crtc_get_cursor_mode() before checking plane state.

[Dec17 17:14] Oops: divide error: 0000 [#1] PREEMPT SMP NOPTI
[  +0.000018] CPU: 5 PID: 1660 Comm: surface-DP-1 Not tainted 6.10.0+ #231
[  +0.000007] Hardware name: Valve Jupiter/Jupiter, BIOS F7A0131 01/30/2024
[  +0.000004] RIP: 0010:dm_get_plane_scale+0x3f/0x60 [amdgpu]
[  +0.000553] Code: 44 0f b7 41 3a 44 0f b7 49 3e 83 e0 0f 48 0f a3 c2 73 21 69 41 28 e8 03 00 00 31 d2 41 f7 f1 31 d2 89 06 69 41 2c e8 03 00 00 <41> f7 f0 89 07 e9 d7 d8 7e e9 44 89 c8 45 89 c1 41 89 c0 eb d4 66
[  +0.000005] RSP: 0018:ffffa8df0de6b8a0 EFLAGS: 00010246
[  +0.000006] RAX: 00000000000003e8 RBX: ffff9ac65c1f6e00 RCX: ffff9ac65d055500
[  +0.000003] RDX: 0000000000000000 RSI: ffffa8df0de6b8b0 RDI: ffffa8df0de6b8b4
[  +0.000004] RBP: ffff9ac64e7a5800 R08: 0000000000000000 R09: 0000000000000a00
[  +0.000003] R10: 00000000000000ff R11: 0000000000000054 R12: ffff9ac6d0700010
[  +0.000003] R13: ffff9ac65d054f00 R14: ffff9ac65d055500 R15: ffff9ac64e7a60a0
[  +0.000004] FS:  00007f869ea00640(0000) GS:ffff9ac970080000(0000) knlGS:0000000000000000
[  +0.000004] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0.000003] CR2: 000055ca701becd0 CR3: 000000010e7f2000 CR4: 0000000000350ef0
[  +0.000004] Call Trace:
[  +0.000007]  <TASK>
[  +0.000006]  ? __die_body.cold+0x19/0x27
[  +0.000009]  ? die+0x2e/0x50
[  +0.000007]  ? do_trap+0xca/0x110
[  +0.000007]  ? do_error_trap+0x6a/0x90
[  +0.000006]  ? dm_get_plane_scale+0x3f/0x60 [amdgpu]
[  +0.000504]  ? exc_divide_error+0x38/0x50
[  +0.000005]  ? dm_get_plane_scale+0x3f/0x60 [amdgpu]
[  +0.000488]  ? asm_exc_divide_error+0x1a/0x20
[  +0.000011]  ? dm_get_plane_scale+0x3f/0x60 [amdgpu]
[  +0.000593]  dm_crtc_get_cursor_mode+0x33f/0x430 [amdgpu]
[  +0.000562]  amdgpu_dm_atomic_check+0x2ef/0x1770 [amdgpu]
[  +0.000501]  drm_atomic_check_only+0x5e1/0xa30 [drm]
[  +0.000047]  drm_mode_atomic_ioctl+0x832/0xcb0 [drm]
[  +0.000050]  ? __pfx_drm_mode_atomic_ioctl+0x10/0x10 [drm]
[  +0.000047]  drm_ioctl_kernel+0xb3/0x100 [drm]
[  +0.000062]  drm_ioctl+0x27a/0x4f0 [drm]
[  +0.000049]  ? __pfx_drm_mode_atomic_ioctl+0x10/0x10 [drm]
[  +0.000055]  amdgpu_drm_ioctl+0x4e/0x90 [amdgpu]
[  +0.000360]  __x64_sys_ioctl+0x97/0xd0
[  +0.000010]  do_syscall_64+0x82/0x190
[  +0.000008]  ? __pfx_drm_mode_createblob_ioctl+0x10/0x10 [drm]
[  +0.000044]  ? srso_return_thunk+0x5/0x5f
[  +0.000006]  ? drm_ioctl_kernel+0xb3/0x100 [drm]
[  +0.000040]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? __check_object_size+0x50/0x220
[  +0.000007]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? drm_ioctl+0x2a4/0x4f0 [drm]
[  +0.000039]  ? __pfx_drm_mode_createblob_ioctl+0x10/0x10 [drm]
[  +0.000043]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? __pm_runtime_suspend+0x69/0xc0
[  +0.000006]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? amdgpu_drm_ioctl+0x71/0x90 [amdgpu]
[  +0.000366]  ? srso_return_thunk+0x5/0x5f
[  +0.000006]  ? syscall_exit_to_user_mode+0x77/0x210
[  +0.000007]  ? srso_return_thunk+0x5/0x5f
[  +0.000005]  ? do_syscall_64+0x8e/0x190
[  +0.000006]  ? srso_return_thunk+0x5/0x5f
[  +0.000006]  ? do_syscall_64+0x8e/0x190
[  +0.000006]  ? srso_return_thunk+0x5/0x5f
[  +0.000007]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  +0.000008] RIP: 0033:0x55bb7cd962bc
[  +0.000007] Code: 4c 89 6c 24 18 4c 89 64 24 20 4c 89 74 24 28 0f 57 c0 0f 11 44 24 30 89 c7 48 8d 54 24 08 b8 10 00 00 00 be bc 64 38 c0 0f 05 <49> 89 c7 48 83 3b 00 74 09 4c 89 c7 ff 15 62 64 99 00 48 83 7b 18
[  +0.000005] RSP: 002b:00007f869e9f4da0 EFLAGS: 00000217 ORIG_RAX: 0000000000000010
[  +0.000007] RAX: ffffffffffffffda RBX: 00007f869e9f4fb8 RCX: 000055bb7cd962bc
[  +0.000004] RDX: 00007f869e9f4da8 RSI: 00000000c03864bc RDI: 000000000000003b
[  +0.000003] RBP: 000055bb9ddcbcc0 R08: 00007f86541b9920 R09: 0000000000000009
[  +0.000004] R10: 0000000000000004 R11: 0000000000000217 R12: 00007f865406c6b0
[  +0.000003] R13: 00007f86541b5290 R14: 00007f865410b700 R15: 000055bb9ddcbc18
[  +0.000009]  </TASK>

Fixes: 1b04dcc ("drm/amd/display: Introduce overlay cursor mode")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3729
Reported-by: Fabio Scaccabarozzi <fsvm88@gmail.com>
Co-developed-by: Fabio Scaccabarozzi <fsvm88@gmail.com>
Signed-off-by: Fabio Scaccabarozzi <fsvm88@gmail.com>
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit ab75a0d)
Cc: stable@vger.kernel.org
kernel-patches-daemon-bpf-rc bot pushed a commit that referenced this pull request May 29, 2025
@rppt @akpm00
mm/cma: make detection of highmem_start more robust
07c9214 
Pratyush Yadav reports the following crash:

    ------------[ cut here ]------------
    kernel BUG at arch/x86/mm/physaddr.c:23!
    ception 0x06 IP 10:ffffffff812ebbf8 error 0 cr2 0xffff88903ffff000
    CPU: 0 UID: 0 PID: 0 Comm: swapper Not tainted 6.15.0-rc6+ #231 PREEMPT(undef)
    Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014
    RIP: 0010:__phys_addr+0x58/0x60
    Code: 01 48 89 c2 48 d3 ea 48 85 d2 75 05 e9 91 52 cf 00 0f 0b 48 3d ff ff ff 1f 77 0f 48 8b 05 20 54 55 01 48 01 d0 e9 78 52 cf 00 <0f> 0b 90 0f 1f 44 00 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90
    RSP: 0000:ffffffff82803dd8 EFLAGS: 00010006 ORIG_RAX: 0000000000000000
    RAX: 000000007fffffff RBX: 00000000ffffffff RCX: 0000000000000000
    RDX: 000000007fffffff RSI: 0000000280000000 RDI: ffffffffffffffff
    RBP: ffffffff82803e68 R08: 0000000000000000 R09: 0000000000000000
    R10: ffffffff83153180 R11: ffffffff82803e48 R12: ffffffff83c9aed0
    R13: 0000000000000000 R14: 0000001040000000 R15: 0000000000000000
    FS:  0000000000000000(0000) GS:0000000000000000(0000) knlGS:0000000000000000
    CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
    CR2: ffff88903ffff000 CR3: 0000000002838000 CR4: 00000000000000b0
    Call Trace:
     <TASK>
     ? __cma_declare_contiguous_nid+0x6e/0x340
     ? cma_declare_contiguous_nid+0x33/0x70
     ? dma_contiguous_reserve_area+0x2f/0x70
     ? setup_arch+0x6f1/0x870
     ? start_kernel+0x52/0x4b0
     ? x86_64_start_reservations+0x29/0x30
     ? x86_64_start_kernel+0x7c/0x80
     ? common_startup_64+0x13e/0x141

  The reason is that __cma_declare_contiguous_nid() does:

          highmem_start = __pa(high_memory - 1) + 1;

  If dma_contiguous_reserve_area() (or any other CMA declaration) is
  called before free_area_init(), high_memory is uninitialized. Without
  CONFIG_DEBUG_VIRTUAL, it will likely work but use the wrong value for
  highmem_start.

The issue occurs because commit e120d1b ("arch, mm: set high_memory
in free_area_init()") moved initialization of high_memory after the call
to dma_contiguous_reserve() -> __cma_declare_contiguous_nid() on several
architectures.

In the case CONFIG_HIGHMEM is enabled, some architectures that actually
support HIGHMEM (arm, powerpc and x86) have initialization of high_memory
before a possible call to __cma_declare_contiguous_nid() and some
initialized high_memory late anyway (arc, csky, microblase, mips, sparc,
xtensa) even before the commit e120d1b so they are fine with using
uninitialized value of high_memory.

And in the case CONFIG_HIGHMEM is disabled high_memory essentially becomes
the first address after memory end, so instead of relying on high_memory
to calculate highmem_start use memblock_end_of_DRAM() and eliminate the
dependency of CMA area creation on high_memory in majority of
configurations.

Link: https://lkml.kernel.org/r/20250519171805.1288393-1-rppt@kernel.org
Fixes: e120d1b ("arch, mm: set high_memory in free_area_init()")
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reported-by: Pratyush Yadav <ptyadav@amazon.de>
Tested-by: Pratyush Yadav <ptyadav@amazon.de>
Tested-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
bpf-next rfc V1
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@kernel-patches-bot