{zfs,spl}-module-parameters.5: fix nonexistent parameters #12157
Conversation
Both were removed in 4fbdb10 ("remove kmem_cache module parameter KMC_EXPIRE_AGE") Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in git log -Soverflow_shift, and it wasn't exposed then, nor is it now zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work): > Although this did uncover a few low priority issues, this unfortuantely also causes ztest to ASSERT in many locations where the code is working correctly since it is designed to fail on IO errors. Developers can manually set this variable with the '-o' option to find and debug issues. Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
|
zio_decompress_fail_fraction still exists, although it is not exposed as a module parameter, it is used by zdb to simulate failures. Although there is a similar variable for decryption failure, that doesn't appear to be documented. |
|
It seems that I have to quote the messages For spl-m-p: Both were removed in 4fbdb10 ("remove kmem_cache module parameter KMC_EXPIRE_AGE") For zfs-m-p: zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work):
|
There was a problem hiding this comment.
Since zio_decompress_fail_fraction was added to the documentation it makes me wonder if the original intent was to make it available as a tunable option. However, since it clearly never was and is very developer focused I agree let's drop from from the user visible docs.
|
Before anyone requests a review from me (like with the other man page changes), I figured I'd take a look. The only question I have is about spl_kmem_cache_obj_per_slab. That one seems to still exist: 4fbdb10 says it removes spl_kmem_cache_obj_per_slab**_min** This comment is the limit of my expertise in this area. |
|
How fortunate that this doesn't touch |
Hah, yep, you're right. My bad! |
jwk404
added
Status: Accepted
zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in git log -Soverflow_shift, and it wasn't exposed then, nor is it now zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work): > Although this did uncover a few low priority issues, this unfortuantely also causes ztest to ASSERT in many locations where the code is working correctly since it is designed to fail on IO errors. Developers can manually set this variable with the '-o' option to find and debug issues. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes #12157
Both were removed in 4fbdb10 ("remove kmem_cache module parameter KMC_EXPIRE_AGE") Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in git log -Soverflow_shift, and it wasn't exposed then, nor is it now zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work): > Although this did uncover a few low priority issues, this unfortuantely also causes ztest to ASSERT in many locations where the code is working correctly since it is designed to fail on IO errors. Developers can manually set this variable with the '-o' option to find and debug issues. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
Both were removed in 4fbdb10 ("remove kmem_cache module parameter KMC_EXPIRE_AGE") Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in git log -Soverflow_shift, and it wasn't exposed then, nor is it now zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work): > Although this did uncover a few low priority issues, this unfortuantely also causes ztest to ASSERT in many locations where the code is working correctly since it is designed to fail on IO errors. Developers can manually set this variable with the '-o' option to find and debug issues. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
Both were removed in 4fbdb10 ("remove kmem_cache module parameter KMC_EXPIRE_AGE") Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
zfs_arc_overflow_shift was never a parameter: ca0bf58 ("Illumos 5497 - lock contention on arcs_mtx") is the only result in git log -Soverflow_shift, and it wasn't exposed then, nor is it now zfs_read_chunk_size was renamed to zfs_vnops_read_chunk_size in e53d678 ("Share zfs_fsync, zfs_read, zfs_write, et al between Linux and FreeBSD") zio_decompress_fail_fraction was never a parameter: it was added in c3bd3fb ("OpenZFS 9403 - assertion failed in arc_buf_destroy()") as a developer aid for setting in zdb, but it's a dangerous test tunable and has no place in public documentation, (not to mention that it obviously doesn't work): > Although this did uncover a few low priority issues, this unfortuantely also causes ztest to ASSERT in many locations where the code is working correctly since it is designed to fail on IO errors. Developers can manually set this variable with the '-o' option to find and debug issues. Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Richard Laager <rlaager@wiktel.com> Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Closes openzfs#12157
Motivation and Context
#12125 (comment), I hope for this to be pretty uncontroversial
Description
Commit messages outline history for all of these
How Has This Been Tested?
Looked at it, I guess? I made the selection through the very scientific process of getting all the names in the zfs-m-p page and contrasting that to a modinfo listing from zfs-2.0.3-6~bpo10+1, which yielded a list short enough for manual extraxion, like spl-m-p.
Types of changes
Checklist:
Signed-off-by.