What's left in libata-dev.git?

This is a summary of items and issues that remain outstanding for libata. Whenever a git branch is mentioned, it is referring to a branch at git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git

Current push just sent to Linus: http://marc.info/?l=linux-ide&m=119329974119307&w=2

Changes in 'upstream' branch pending for 2.6.24-rc, but not included in the above push:
Tejun Heo (2):
libata: relocate and fix post-command processing
libata: track SLEEP state and issue SRST to wake it up

Active libata branches:
ALL Superset branch for -mm testing
alpm Link power management (2.6.24-rc hopefully)
an Async notify (2.6.24-rc hopefully)
for-testing Interesting-but-not-ready stuff
master Vanilla linux-2.6.git clone
mv-ahci-pata Marvell 6141 PATA support (incomplete)
mv-ncq sata_mv NCQ support (incomplete)
new-eh sata_qstore, sata_sx4 new-EH conversions
(TEST ME! they have been in -mm for a while)
pmask proto_mask introduction
sii-lbt sata_sil large block transfer enablement
upstream pending for 2.6.24-rc
upstream-linus just pushed upstream for 2.6.24-rc

Patches and tasks in mbox queue, not yet in git:

1) Axboe: kill sg_last(), libata is last remaining user

2) Bart points out bugs in "Re: [git patches] libata update"

3) Tejun: libata: implement ata_wait_after_reset()

^^^ Tejun, what case does this solve? Still needed?

4) Alan: [PATCH] libata: Deal with ATA8-ACS proposed Trusted/Treacherous Computing features

5) Need a pass through bz.kernel.org

Please speak up if there is something missing.

Jeff

FWIW, current contents of ALL meta-branch:

upstream
an
alpm
for-testing
new-eh

i.e. that is what akpm will get on his next pull for -mm.

Not needed yet. It will be necessary to support a weird device (CF bridged over SATA) but it's a nice clean up regardless (instead of quoting ATADVDR in every reset functions). Any reason not to put it in?

tejun

<looks at it again> I suppose not :) I will put it in.

Jeff