variable to outputs of the asm statement that tweaks sr, so that gcc
doesn't insist on it being initialized. Move boot function template
closer to instantiation. Misc prettification.
* Split out the code that checks for CPUID 0x80000007 (powernow)
into another function.
* Cleaned up the code a bit.
More to come...
cpu0: AMD PowerNow! Technology.
cpu0: AMD features: TS FID VID
cpu0: available frequencies (Mhz): 798 864 931 997 1064 1197
cpu0: current frequency (Mhz): 1197
The frequency values are not correct (yet).
Thanks to Rhialto for testing and provide feedback.
These were written by William Allen Simpson and submitted in PR 21983,
and are added with minor adjustments and nits from christos@ and myself.
Approved by christos@ and groo@.
1) Don't put two pits in the same room.
2) Don't put bats and pits in the same room. (you will never hit the bat)
3) Don't start the player in a room with a pit or bad, if possible. Some
caves are so crowded the loop may go on forever, so we give up after 100
tries to put them in a safe location (as long as it's not with the wumpus).
4) Make the manpage reflect reality WRT the default number of rooms.
Bug #3 pointed out by salo.
games in a row. Also modify change made in rev 1.18 to work correctly.
take_action() returns 1 if the player dies, causing the game to allways
exit after any death, now the game correctly asks if the player would
like to play again.
that, but then fill out the control file with the "df" filename. Fix by
sending the remote file with the "df" filename, as suggested by Robert
Vargason on tech-userlevel. lpd -r now works properly with an if (input
filter)
The correct way to check for a zero record length is to check for it
without the LAST_FRAG marker in it, since it's legal to send a LAST_FRAG
marker with 0 bytes of data.
[This should be pulled up to 3.0]
- mmmmap: remove physmem check. it was wrong and superseded by the above.
- mmmmap: remove suser() check. there's no point to limit the operation
to root here, given that the almost same thing can be done by
normal read and write.
return -1. This causes filters to fail because we get a value for
child_pid we weren't expecting. Set SIGCHLD to SIG_DFL instead to clear
the reapchild setting when forking the child lpd.
otherwise generate an UVM trap or will access random memory. This is due to
the dereference of vp->v_specmountpoint that is really
vp->v_specinfo->si_mountpoint. The field v_specinfo is multiplexed with
other structs in the vun union in struct vnode like struct socket.
The patch adds a sanity check for accessing the specinfo fields by only
allowing VBLK nodes to be passed. In theory also VCHR could be valid since
its also a special node though mounting is only done on VBLK so be strict.
Ok'd by yamt.
lenght calculation phase instead of just growing the buffer like the older
patch did. I am leaving the bigger buffer too for now since it does not hurt.
md_view_labels() and a variable to mark its availability, so checking of
the available disklabels can be done before the root device is chosen.
Suggested by Martin Bertelsmann.
christos