Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
tjh.dev
kernel
os
linux
[libata] sata_mv: add TODO list
Signed-off-by: Jeff Garzik <jeff@garzik.org>
+44
drivers/ata/sata_mv.c
+44
drivers/ata/sata_mv.c
···
21
21
*
22
22
*/
23
23
24
+
/*
25
+
sata_mv TODO list:
26
+
27
+
1) Needs a full errata audit for all chipsets. I implemented most
28
+
of the errata workarounds found in the Marvell vendor driver, but
29
+
I distinctly remember a couple workarounds (one related to PCI-X)
30
+
are still needed.
31
+
32
+
2) Convert to LibATA new EH. Required for hotplug, NCQ, and sane
33
+
probing/error handling in general. MUST HAVE.
34
+
35
+
3) Add hotplug support (easy, once new-EH support appears)
36
+
37
+
4) Add NCQ support (easy to intermediate, once new-EH support appears)
38
+
39
+
5) Investigate problems with PCI Message Signalled Interrupts (MSI).
40
+
41
+
6) Add port multiplier support (intermediate)
42
+
43
+
7) Test and verify 3.0 Gbps support
44
+
45
+
8) Develop a low-power-consumption strategy, and implement it.
46
+
47
+
9) [Experiment, low priority] See if ATAPI can be supported using
48
+
"unknown FIS" or "vendor-specific FIS" support, or something creative
49
+
like that.
50
+
51
+
10) [Experiment, low priority] Investigate interrupt coalescing.
52
+
Quite often, especially with PCI Message Signalled Interrupts (MSI),
53
+
the overhead reduced by interrupt mitigation is quite often not
54
+
worth the latency cost.
55
+
56
+
11) [Experiment, Marvell value added] Is it possible to use target
57
+
mode to cross-connect two Linux boxes with Marvell cards? If so,
58
+
creating LibATA target mode support would be very interesting.
59
+
60
+
Target mode, for those without docs, is the ability to directly
61
+
connect two SATA controllers.
62
+
63
+
13) Verify that 7042 is fully supported. I only have a 6042.
64
+
65
+
*/
66
+
67
+
24
68
#include <linux/kernel.h>
25
69
#include <linux/module.h>
26
70
#include <linux/pci.h>