24a1cce34f
proc or any VM system structure will have to be rebuilt!!!
Much needed overhaul of the VM system. Included in this first round of
changes:
1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
haspage, and sync operations are supported. The haspage interface now
provides information about clusterability. All pager routines now take
struct vm_object's instead of "pagers".
2) Improved data structures. In the previous paradigm, there is constant
confusion caused by pagers being both a data structure ("allocate a
pager") and a collection of routines. The idea of a pager structure has
escentially been eliminated. Objects now have types, and this type is
used to index the appropriate pager. In most cases, items in the pager
structure were duplicated in the object data structure and thus were
unnecessary. In the few cases that remained, a un_pager structure union
was created in the object to contain these items.
3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
be removed. For instance, vm_object_enter(), vm_object_lookup(),
vm_object_remove(), and the associated object hash list were some of the
things that were removed.
4) simple_lock's removed. Discussion with several people reveals that the
SMP locking primitives used in the VM system aren't likely the mechanism
that we'll be adopting. Even if it were, the locking that was in the code
was very inadequate and would have to be mostly re-done anyway. The
locking in a uni-processor kernel was a no-op but went a long way toward
making the code difficult to read and debug.
5) Places that attempted to kludge-up the fact that we don't have kernel
thread support have been fixed to reflect the reality that we are really
dealing with processes, not threads. The VM system didn't have complete
thread support, so the comments and mis-named routines were just wrong.
We now use tsleep and wakeup directly in the lock routines, for instance.
6) Where appropriate, the pagers have been improved, especially in the
pager_alloc routines. Most of the pager_allocs have been rewritten and
are now faster and easier to maintain.
7) The pagedaemon pageout clustering algorithm has been rewritten and
now tries harder to output an even number of pages before and after
the requested page. This is sort of the reverse of the ideal pagein
algorithm and should provide better overall performance.
8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
have been removed. Some other unnecessary casts have also been removed.
9) Some almost useless debugging code removed.
10) Terminology of shadow objects vs. backing objects straightened out.
The fact that the vm_object data structure escentially had this
backwards really confused things. The use of "shadow" and "backing
object" throughout the code is now internally consistent and correct
in the Mach terminology.
11) Several minor bug fixes, including one in the vm daemon that caused
0 RSS objects to not get purged as intended.
12) A "default pager" has now been created which cleans up the transition
of objects to the "swap" type. The previous checks throughout the code
for swp->pg_data != NULL were really ugly. This change also provides
the rudiments for future backing of "anonymous" memory by something
other than the swap pager (via the vnode pager, for example), and it
allows the decision about which of these pagers to use to be made
dynamically (although will need some additional decision code to do
this, of course).
13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
object" code has been removed. MAP_COPY was undocumented and non-
standard. It was furthermore broken in several ways which caused its
behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
continue to work correctly, but via the slightly different semantics
of MAP_PRIVATE.
14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
threads design can be worked around in other ways. Both #12 and #13
were done to simplify the code and improve readability and maintain-
ability. (As were most all of these changes)
TODO:
1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
this will reduce the vnode pager to a mere fraction of its current size.
2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
information provided by the new haspage pager interface. This will
substantially reduce the overhead by eliminating a large number of
VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
improved to provide both a "behind" and "ahead" indication of
contiguousness.
3) Implement the extended features of pager_haspage in swap_pager_haspage().
It currently just says 0 pages ahead/behind.
4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
via a much more general mechanism that could also be used for disk
striping of regular filesystems.
5) Do something to improve the architecture of vm_object_collapse(). The
fact that it makes calls into the swap pager and knows too much about
how the swap pager operates really bothers me. It also doesn't allow
for collapsing of non-swap pager objects ("unnamed" objects backed by
other pagers).
|
||
|---|---|---|
| .. | ||
| procfs_ctl.c | ||
| procfs_fpregs.c | ||
| procfs_mem.c | ||
| procfs_note.c | ||
| procfs_regs.c | ||
| procfs_status.c | ||
| procfs_subr.c | ||
| procfs_vfsops.c | ||
| procfs_vnops.c | ||
| procfs.h | ||
| README | ||
saute procfs lyonnais procfs supports two levels of directory. the filesystem root directory contains a representation of the system process table. this consists of an entry for each active and zombie process, and an additional entry "curproc" which always represents the process making the lookup request. each of the sub-directories contains several files. these files are used to control and interrogate processes. the files implemented are: file - xxx. the exec'ed file. status - r/o. returns process status. ctl - w/o. sends a control message to the process. for example: echo hup > /proc/curproc/note will send a SIGHUP to the shell. whereas echo attach > /proc/1293/ctl would set up process 1293 for debugging. see below for more details. mem - r/w. virtual memory image of the process. parts of the address space are readable only if they exist in the target process. a more reasonable alternative might be to return zero pages instead of an error. comments? note - w/o. writing a string here sends the equivalent note to the process. [ not implemented. ] notepg - w/o. the same as note, but sends to all members of the process group. [ not implemented. ] regs - r/w. process register set. this can be read or written any time even if the process is not stopped. since the bsd kernel is single-processor, this implementation will get the "right" register values. a multi-proc kernel would need to do some synchronisation. this then looks like: % ls -li /proc total 0 9 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 0 17 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 1 89 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 10 25 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 2 2065 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 257 2481 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 309 265 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 32 3129 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 390 3209 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 400 3217 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 401 3273 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 408 393 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 48 409 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 50 465 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 57 481 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 59 537 dr-xr-xr-x 2 root kmem 0 Sep 21 15:06 66 545 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 67 657 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 81 665 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 82 673 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 83 681 dr-xr-xr-x 2 root wheel 0 Sep 21 15:06 84 3273 dr-xr-xr-x 2 jsp staff 0 Sep 21 15:06 curproc % ls -li /proc/curproc total 408 3341 --w------- 1 jsp staff 0 Sep 21 15:06 ctl 1554 -r-xr-xr-x 1 bin bin 90112 Mar 29 04:52 file 3339 -rw------- 1 jsp staff 118784 Sep 21 15:06 mem 3343 --w------- 1 jsp staff 0 Sep 21 15:06 note 3344 --w------- 1 jsp staff 0 Sep 21 15:06 notepg 3340 -rw------- 1 jsp staff 0 Sep 21 15:06 regs 3342 -r--r--r-- 1 jsp staff 0 Sep 21 15:06 status % df /proc/curproc /proc/curproc/file Filesystem 512-blocks Used Avail Capacity Mounted on proc 2 2 0 100% /proc /dev/wd0a 16186 13548 1018 93% / % cat /proc/curproc/status cat 446 439 400 81 12,0 ctty 748620684 270000 0 0 0 20000 nochan 11 20 20 20 0 21 117 the basic sequence of commands written to "ctl" would be attach - this stops the target process and arranges for the sending process to become the debug control process wait - wait for the target process to come to a steady state ready for debugging. step - single step, with no signal delivery. run - continue running, with no signal delivery, until next trap or breakpoint. <signame> - deliver signal <signame> and continue running. detach - continue execution of the target process and remove it from control by the debug process in a normal debugging environment, where the target is fork/exec'd by the debugger, the debugger should fork and the child should stop itself (with a self-inflicted SIGSTOP). the parent should do a "wait" then an "attach". as before, the child will hit a breakpoint on the first instruction in any newly exec'd image. $Id: README,v 3.1 1993/12/15 09:40:17 jsp Exp $