freebsd-nq/sys
Søren Schmidt 3427651043 Ten'th update to the new ATA/ATAPI driver:
It been awhile since the last major update, as a benefit there
are some cool things in this one (and new bugs probably :) )...

The ATA driver has grown "real" timeout support for all devices.
This means that it should be possible to get in contact with
(especially) lost ATAPI devices. It also means that the ATA
driver is now usable on notebooks as it will DTRT on resume.

An experimental hack at utilizing the Promise66's at UDMA66 is
in there, but I cant test it. If someone feels like sending
me one, give me a ping.

The ATAPI DMA enableling scheme has been changed, also better DMA
support for the Aladdin chipset has been implemented for ATAPI
devices. Note that the Aladdin apparently only can do DMA reads
on ATAPI devices, and the Promise cant do ATAPI DMA at all.
I have seen problems on some ATAPI devices that should be able
to run in DMA mode, so if you encounter problems with hanging
atapi devices during the probe, or during access, disable DMA
in atapi-all.c, and let me know. It might be nessesary to do this
via a "white list" for known good devices...

The ATAPI CDROM driver can now use eject/close without hanging and
the bug that caused reading beyond the end of a CD has been fixed.
Media change is also handled proberly. DVD drives are identified
and are usable as CDROM devices at least, I dont have the HW to
test this further, see above :).

The ATAPI tape driver has gotten some support for using the DSC
method for not blocking the IDE channel during read/write when
the device has full buffers. It knows about the OnStream DI-30
device, support is not completed yet, but it can function as a
primitive backup medium, without filemarks, and without bad media
handeling. This is because the OnStream device doesn't handle this
(like everybody else) in HW. It also now supports getting/setting
the record position on devices that supports it.

Some rather major cleanups and rearrangements as well (cvs -b diff
is your freind). I'm closing in on declaring this for beta code,
most of the infrastruture is in place by now.

As usual USE AT YOUR OWN RISK!!, this is still alpha level code.
This driver can hose your disk real bad if anything goes wrong, but
now you have been warned :)

But please tell me how it works for you!

Enjoy!

-Søren
1999-09-21 19:50:40 +00:00
..
alpha Fix bug in pipe code relating to writes of mmap'd but illegal address 1999-09-20 19:08:48 +00:00
amd64 Print out flags value 1999-09-21 10:51:47 +00:00
boot netboot removed until somebody fixes it so it compiles 1999-09-09 01:28:26 +00:00
cam Remove unneeded disk-related includes. 1999-09-12 08:23:21 +00:00
coda Seperate the export check in VFS_FHTOVP, exports are now checked via 1999-09-11 00:46:08 +00:00
compat This is what was "fdfix2.patch," a fix for fd sharing. It's pretty 1999-09-19 17:00:25 +00:00
compile
conf Add md driver to LINT 1999-09-21 11:15:58 +00:00
contrib Use devstat_end_transaction_buf() rather than Use devstat_end_transaction() 1999-09-18 21:30:27 +00:00
ddb $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
dev Ten'th update to the new ATA/ATAPI driver: 1999-09-21 19:50:40 +00:00
fs More removals of vnode->v_lastr, replaced by preexisting seqcount 1999-09-20 23:27:58 +00:00
geom Use devstat_end_transaction_buf() rather than Use devstat_end_transaction() 1999-09-18 21:30:27 +00:00
gnu More removals of vnode->v_lastr, replaced by preexisting seqcount 1999-09-20 23:27:58 +00:00
i4b Finish making LINT compile. Disable references to pnp in files that 1999-09-03 20:38:47 +00:00
i386 Add md driver to LINT 1999-09-21 11:15:58 +00:00
isa Two more devstat_end_transaction() -> devstat_end_transaction_buf(). 1999-09-19 12:43:34 +00:00
isofs/cd9660 More removals of vnode->v_lastr, replaced by preexisting seqcount 1999-09-20 23:27:58 +00:00
kern Kill some spammage that seems to have gotten in through diffs from marcel's 1999-09-21 03:47:42 +00:00
libkern $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
miscfs Fix handling of a device EOF that occurs in the middle of a block. The 1999-09-20 23:17:47 +00:00
modules Change the name of the installed KLM to contain a leading "if_". 1999-09-20 06:11:48 +00:00
msdosfs More removals of vnode->v_lastr, replaced by preexisting seqcount 1999-09-20 23:27:58 +00:00
net Don't call if_up() here, just set IFF_UP. 1999-09-14 01:17:30 +00:00
netatalk Change suser_xxx() to suser() where it applies. 1999-04-27 12:21:16 +00:00
netatm Fixed printf format errors (don't assume that ntohl() returns u_long; 1999-08-29 10:28:10 +00:00
netinet Restore previous version of FindLinkIn(). 1999-09-21 14:44:32 +00:00
netipx Get rid of the old XNS checksum code and implement it the IPX way. 1999-08-28 18:21:55 +00:00
netkey Fix a printf(3) formatter to match its variable. 1999-08-17 22:11:02 +00:00
netnatm Examine all occurrences of sprintf(), strcat(), and str[n]cpy() 1998-12-04 22:54:57 +00:00
netns $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
nfs Add comment to clarify a commit rpc optimization already being performed. 1999-09-20 19:10:28 +00:00
nfsclient Add comment to clarify a commit rpc optimization already being performed. 1999-09-20 19:10:28 +00:00
nfsserver Asynchronized client-side nfs_commit. NFS commit operations were 1999-09-17 05:57:57 +00:00
ntfs Seperate the export check in VFS_FHTOVP, exports are now checked via 1999-09-11 00:46:08 +00:00
pc98 Set si_iosize_max rather than d_maxio. 1999-09-20 19:58:38 +00:00
pccard $Id$ -> $FreeBSD$ 1999-09-06 11:23:05 +00:00
pci Close PR #13665. I managed to figure out the problem, no thanks to the 1999-09-20 20:26:14 +00:00
posix4 $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
powerpc Fix bug in pipe code relating to writes of mmap'd but illegal address 1999-09-20 19:08:48 +00:00
rpc $Id$ -> $FreeBSD$ 1999-08-27 23:45:13 +00:00
svr4 This is what was "fdfix2.patch," a fix for fd sharing. It's pretty 1999-09-19 17:00:25 +00:00
sys green and I committed crossing fixes for this. He won. 1999-09-21 11:12:08 +00:00
tools $Id$ -> $FreeBSD$ (some mangled and/or hidden ones) 1999-08-28 02:21:15 +00:00
ufs More removals of vnode->v_lastr, replaced by preexisting seqcount 1999-09-20 23:27:58 +00:00
vm cleanup madvise code, add a few more sanity checks. 1999-09-21 05:00:48 +00:00
Makefile $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00