core: map io.bfq.weight to 1..1000#18467
Conversation
Aaargh. See the comment in the code. Apparently the range is like that: $ sudo bash -c 'echo "default 1001" >/sys/fs/cgroup/user.slice/io.bfq.weight' bash: line 0: echo: write error: Numerical result out of range $ uname -r 5.11.0-0.rc4.129.fc34.x86_64
|
This pull request introduces 1 alert and fixes 2 when merging 1e6c897 into 436cde8 - view on LGTM.com new alerts:
fixed alerts:
|
bluca
left a comment
bluca
left a comment
There was a problem hiding this comment.
An... interesting interface. LGTM.
bluca
added
the
good-to-merge/waiting-for-ci 👍
|
bionic-i386: FAILED TESTS: TEST-30-ONCLOCKCHANGE TEST-55-UDEV-TAGS This looks unrelated to this PR, but maybe something changed recently to make those tests more flaky? |
mmh TEST-55 does sometimes fail, but I cannot recall ever seeing TEST-30 fail before? |
|
mmh looks like it was blocked for a while, but it is not currently? Only TEST-25 has a deny-list-ubuntu-ci |
|
still tempted to just drop bfq support. kernel people should figure out their api mess... |
|
lgtm though |
|
As you can see from linux-block ml, I've done my best to clear this interface mess, at no avail. I'm sorry for the community, and for the scheduler, whose quality and benefits have not to do with this problem.
… Il giorno 4 feb 2021, alle ore 16:37, Lennart Poettering ***@***.***> ha scritto:
still tempted to just drop bfq support. kernel people should figure out their api mess...
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or unsubscribe.
|
|
if kernel peeps can't figure out their mess I am not sure we should accept the bill for it though. |
keszybz
removed
the
good-to-merge/waiting-for-ci 👍
4 participants
Aaargh. See the comment in the code.
Apparently the range is like that:
$ sudo bash -c 'echo "default 1001" >/sys/fs/cgroup/user.slice/io.bfq.weight'
bash: line 0: echo: write error: Numerical result out of range
$ uname -r
5.11.0-0.rc4.129.fc34.x86_64
/cc @paolo-github