In D30218#680984, @jah wrote:Add basic fixes for nullfs and unionfs, conditionalize ZFS changes
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Feed Advanced Search
Advanced Search
Advanced Search
May 17 2021
May 17 2021
jah added a comment to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah updated the diff for D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
Add basic fixes for nullfs and unionfs, conditionalize ZFS changes
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
May 16 2021
May 16 2021
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
May 15 2021
May 15 2021
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
jah added inline comments to D30263: nullfs: busy the lower mount before calling VFS_* operations on it.
May 14 2021
May 14 2021
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Replace vfs_ref() with simple atomic load of v_mount
jah added inline comments to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
May 13 2021
May 13 2021
jah added inline comments to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah added inline comments to D30152: unionfs: reference the underlying FS' mount objects when using them.
jah added inline comments to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah updated the diff for D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
Fix manpage formatting, remove leftover include, fix incorrect deleted line due to fingerslip
jah retitled D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes from Follow mount point unbusying requirements in vfs_stdquotactl() to VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Chase D30218
jah added inline comments to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah updated the diff for D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
Allow the implementation to direct busy state changes
In D30152#678936, @markj wrote:Do you plan to revise this change after changing the VFS_QUOTACTL interface? I'm just not sure whether to wait until that's done before re-reading this.
May 12 2021
May 12 2021
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Use correct variable when passing through VFS_QUOTACTL
May 11 2021
May 11 2021
jah added inline comments to D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah requested review of D30218: VFS_QUOTACTL(9): allow implementation to indicate busy state changes.
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Follow the unbusy requirements for vfs_quotactl()
May 10 2021
May 10 2021
jah added inline comments to D30152: unionfs: reference the underlying FS' mount objects when using them.
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Style
May 7 2021
May 7 2021
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Use atomic_load_ptr to enforce single load of vp->v_mount
jah updated the diff for D30152: unionfs: reference the underlying FS' mount objects when using them.
Apply vfs_busy, minor refactoring
In D30152#677009, @kib wrote:In D30152#677007, @jah wrote:In D30152#676937, @jah wrote:A simpler approach would be to directly store the underlying mount objects in struct unionfs_mount, vfs_ref() them on mount and vfs_rel() them on unmount.
But that would have the effect of blocking forcible unmount of the underlying filesystems in vfs_mount_destroy(). Meanwhile, the underlying FS' vnodes would still be flushed before this happens so the underlying filesystems would be no more usable than they are with this approach.In D30152#676980, @kib wrote:Taking a reference on the mount point does not prevent unmount. For instance, calling VFS_QUOTACTL() requires busying mp, which, if successful, prevents unmount until unbusy. In fact, busying fs is currently enforced by VFS_QUOTACTL protocol since UFS (the only actual quotactl provider) in some cases unbusies and then busies the mp.
It certainly doesn't prevent the unmount operations (vflush, etc) from happening, but doesn't it prevent the unmount system call from returning? That's what I observed in testing, and is corroborated by reading vfs_mount_destroy().
Why 'preventing syscall from returning' matters? What matters is that unmount destroys the private mount data pointed to by mnt_data. This data is usually vital for VFS_XXX() methods, and destroying/freeing it while these methods operate causes undefined behavior, a crash in the best case.
In D30152#676937, @jah wrote:A simpler approach would be to directly store the underlying mount objects in struct unionfs_mount, vfs_ref() them on mount and vfs_rel() them on unmount.
But that would have the effect of blocking forcible unmount of the underlying filesystems in vfs_mount_destroy(). Meanwhile, the underlying FS' vnodes would still be flushed before this happens so the underlying filesystems would be no more usable than they are with this approach.
May 6 2021
May 6 2021
A simpler approach would be to directly store the underlying mount objects in struct unionfs_mount, vfs_ref() them on mount and vfs_rel() them on unmount.
But that would have the effect of blocking forcible unmount of the underlying filesystems in vfs_mount_destroy(). Meanwhile, the underlying FS' vnodes would still be flushed before this happens so the underlying filesystems would be no more usable than they are with this approach.
jah requested review of D30152: unionfs: reference the underlying FS' mount objects when using them.
Mar 31 2021
Mar 31 2021
jah committed rG8dc8feb53da0: Clean up a couple of MD warts in vm_fault_populate(): (authored by jah).
Clean up a couple of MD warts in vm_fault_populate():
Mar 27 2021
Mar 27 2021
Adjust wording in comment, rearrange return value assertion
jah added inline comments to D28717: vm_page_grab*: Consolidate common logic into vm_page_grab_release().
Mar 21 2021
Mar 21 2021
Mar 17 2021
Mar 17 2021
Mar 16 2021
Mar 16 2021
factor out PT page allocation/freeing
Mar 14 2021
Mar 14 2021
Mar 13 2021
Mar 13 2021
Fix la57 table accounting
--remove unnecessary include
--add la57 bootstrap accounting
--further cleanup to improve code reuse
Revert accidentally-committed PCID counter removal, incorporate resident_count
Mar 10 2021
Mar 10 2021
Mar 9 2021
Mar 9 2021
jah committed rGe4b8deb22227: amd64 pmap: convert to counter(9), add PV and pagetable page counts (authored by jah).
amd64 pmap: convert to counter(9), add PV and pagetable page counts
Mar 3 2021
Mar 3 2021
In D28923#650048, @kib wrote:I probably should have been more explicit, I proposed to drop the pt_page_count at all. But I hope that you will follow up with pmap_pt_page_alloc() patch so this does not matter much.
jah updated the diff for D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
- Revert "Remove PCID save count, factor out PT page allocation/freeing"
- Fix NULL check in pmap_large_map_getptp_unlocked()
Mar 1 2021
Mar 1 2021
jah updated the diff for D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
- Remove PCID save count, factor out PT page allocation/freeing
Feb 27 2021
Feb 27 2021
jah added inline comments to D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
Feb 26 2021
Feb 26 2021
jah added inline comments to D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
Feb 25 2021
Feb 25 2021
jah added inline comments to D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
jah requested review of D28923: amd64 pmap: convert to counter(9), add PV and pagetable page counts.
Feb 24 2021
Feb 24 2021
jah committed rG941747f63cfa: Fix divide-by-zero panic when ASLR is enabled and superpages disabled (authored by jah).
Fix divide-by-zero panic when ASLR is enabled and superpages disabled
jah committed rGfc6c898ec5f4: Fix divide-by-zero panic when ASLR is enabled and superpages disabled (authored by jah).
Fix divide-by-zero panic when ASLR is enabled and superpages disabled
Feb 15 2021
Feb 15 2021
jah committed rG41032835dc2d: Fix divide-by-zero panic when ASLR is enabled and superpages disabled (authored by jah).
Fix divide-by-zero panic when ASLR is enabled and superpages disabled
jah requested review of D28678: Fix divide-by-zero panic when ALSR is enabled and superpages disabled.
Feb 11 2021
Feb 11 2021
In D28592#640118, @mjg wrote:I have no comments about usefulness of adding these counters. I have to note the patch negatively affects scalability as it adds loads from the page form something which gets modified by other threads. If adding these stats, it would have to happen inside of vm_page_busy_sleep.
Jan 26 2021
Jan 26 2021
rctl(4): support throttling resource usage to 0
Jan 15 2021
Jan 15 2021
Jan 14 2021
Jan 14 2021
jah added inline comments to D28108: x86 busdma_bounce: do not make assumptions about alignment of malloc(9) results..
jah added inline comments to D28108: x86 busdma_bounce: do not make assumptions about alignment of malloc(9) results..
jah added inline comments to D28108: x86 busdma_bounce: do not make assumptions about alignment of malloc(9) results..
jah added inline comments to D28108: x86 busdma_bounce: do not make assumptions about alignment of malloc(9) results..
Jan 13 2021
Jan 13 2021
jah added inline comments to D28108: x86 busdma_bounce: do not make assumptions about alignment of malloc(9) results..
Jan 12 2021
Jan 12 2021
Jan 11 2021
Jan 11 2021
rctl(4): support throttling resource usage to 0
Jan 9 2021
Jan 9 2021
Dec 30 2020
Dec 30 2020
Oct 6 2020
Oct 6 2020
multimedia/libtuner: update to 1.1.5
Sep 12 2020
Sep 12 2020
amd64: prevent KCSan false positives on LAPIC mapping
Sep 8 2020
Sep 8 2020
jah added reviewers for D26354: amd64: prevent KCSan false positives on LAPIC mapping: kib, gbe, andrew.
Whitespace
Sep 1 2020
Sep 1 2020
Seemed like a cute driver and a fun thing to do with some spare time while I was traveling.
I've verified that spkrtest(8) cycles through the notes without error, but since I'm away from the one machine I have with an actual speaker, I haven't been able to test by ear.
Aug 14 2020
Aug 14 2020
kenv: avoid sleepable alloc for integer tunables