ToDo List Kernel (2005.12.01 19:00:00)

We currently migrate to MediaWiki from our old installation, but not all content has been migrated yet. Take a look at the Wiki Team page for instructions how to help or browse through our new wiki at wiki.linux-vserver.org to find the information already migrated.

Stuff ToDo for the Kernel

• {005} check tagxid on ro filesystems/mounts
• {025} make VXCs required for CAPs

work in progress (mask capabilities)

• {010} add vfsmount tagxid=xx

(tag)xid=<xid> mounts implemented, but untested

• {002} unwire uid hashes

maybe other hashes are more important

• {025} do some decent profiling
• {020} check inode xid assignment and iattr scope

somewhat excercised with testfs.sh

• {010} CPU virtualization
• {020} vserver child reaper (initpid)

• {020} alternate solution for closing_wait (HZ)

maybe remove the entire patch? obsoleted!

• {015} accounting and limits for IPC stuff

shm, msq and locks for now (not all enforced)

• {015} look into acct_update_integrals()

some kind of per context integrals maybe?

• {020} separate token bucket per cpu

• {005} virtualize /proc/locks (get_locks_status)

done, not really tested

• {005} replace hash locks by rw locks

• {010} add slab cache for vxi/nxi
• {010} check that SOCK_USER_SOCKET is used at all
• {010} fix the inode sizes inode->i_size = 64;
• {010} fix the sys32_sysinfo() for parisc32
• {010} revalidate checks in syscall commands
• {010} allow proc_dointvec_bset read without CAP_SYS_MODULE

• {015} look into atomic_inc_return() for resources

i.e. inc, then check, on over limit dec again ...

• {010} add proper socket limits (maybe splitup?)
• {015} remove/change vx_info_proc_limit() clamping
• {010} check set_?x_info() vs. init_?x_info()

• {020} check /proc/sys/fs/overflow{u,g}id and tagging
• {015} check dqhash and quota support on ext3
• {020} add semaphore accounting and limits
• {010} check __user annotation for vc_* commands
• {020} add xid information to kernel log messages

and process stack traces

• {030} switch to seq_printf() for procfs
• {020} check NBIPV4 max (maybe option?)
• {010} check out vm_stat_account()

The numbers before the items are related to the priority