diff options
Diffstat (limited to 'drivers/net/via-rhine.c')
-rw-r--r-- | drivers/net/via-rhine.c | 110 |
1 files changed, 3 insertions, 107 deletions
diff --git a/drivers/net/via-rhine.c b/drivers/net/via-rhine.c index 98b6f3207d3d..e48e76ce73a0 100644 --- a/drivers/net/via-rhine.c +++ b/drivers/net/via-rhine.c | |||
@@ -25,117 +25,13 @@ | |||
25 | version. He may or may not be interested in bug reports on this | 25 | version. He may or may not be interested in bug reports on this |
26 | code. You can find his versions at: | 26 | code. You can find his versions at: |
27 | http://www.scyld.com/network/via-rhine.html | 27 | http://www.scyld.com/network/via-rhine.html |
28 | 28 | [link no longer provides useful info -jgarzik] | |
29 | |||
30 | Linux kernel version history: | ||
31 | |||
32 | LK1.1.0: | ||
33 | - Jeff Garzik: softnet 'n stuff | ||
34 | |||
35 | LK1.1.1: | ||
36 | - Justin Guyett: softnet and locking fixes | ||
37 | - Jeff Garzik: use PCI interface | ||
38 | |||
39 | LK1.1.2: | ||
40 | - Urban Widmark: minor cleanups, merges from Becker 1.03a/1.04 versions | ||
41 | |||
42 | LK1.1.3: | ||
43 | - Urban Widmark: use PCI DMA interface (with thanks to the eepro100.c | ||
44 | code) update "Theory of Operation" with | ||
45 | softnet/locking changes | ||
46 | - Dave Miller: PCI DMA and endian fixups | ||
47 | - Jeff Garzik: MOD_xxx race fixes, updated PCI resource allocation | ||
48 | |||
49 | LK1.1.4: | ||
50 | - Urban Widmark: fix gcc 2.95.2 problem and | ||
51 | remove writel's to fixed address 0x7c | ||
52 | |||
53 | LK1.1.5: | ||
54 | - Urban Widmark: mdio locking, bounce buffer changes | ||
55 | merges from Beckers 1.05 version | ||
56 | added netif_running_on/off support | ||
57 | |||
58 | LK1.1.6: | ||
59 | - Urban Widmark: merges from Beckers 1.08b version (VT6102 + mdio) | ||
60 | set netif_running_on/off on startup, del_timer_sync | ||
61 | |||
62 | LK1.1.7: | ||
63 | - Manfred Spraul: added reset into tx_timeout | ||
64 | |||
65 | LK1.1.9: | ||
66 | - Urban Widmark: merges from Beckers 1.10 version | ||
67 | (media selection + eeprom reload) | ||
68 | - David Vrabel: merges from D-Link "1.11" version | ||
69 | (disable WOL and PME on startup) | ||
70 | |||
71 | LK1.1.10: | ||
72 | - Manfred Spraul: use "singlecopy" for unaligned buffers | ||
73 | don't allocate bounce buffers for !ReqTxAlign cards | ||
74 | |||
75 | LK1.1.11: | ||
76 | - David Woodhouse: Set dev->base_addr before the first time we call | ||
77 | wait_for_reset(). It's a lot happier that way. | ||
78 | Free np->tx_bufs only if we actually allocated it. | ||
79 | |||
80 | LK1.1.12: | ||
81 | - Martin Eriksson: Allow Memory-Mapped IO to be enabled. | ||
82 | |||
83 | LK1.1.13 (jgarzik): | ||
84 | - Add ethtool support | ||
85 | - Replace some MII-related magic numbers with constants | ||
86 | |||
87 | LK1.1.14 (Ivan G.): | ||
88 | - fixes comments for Rhine-III | ||
89 | - removes W_MAX_TIMEOUT (unused) | ||
90 | - adds HasDavicomPhy for Rhine-I (basis: linuxfet driver; my card | ||
91 | is R-I and has Davicom chip, flag is referenced in kernel driver) | ||
92 | - sends chip_id as a parameter to wait_for_reset since np is not | ||
93 | initialized on first call | ||
94 | - changes mmio "else if (chip_id==VT6102)" to "else" so it will work | ||
95 | for Rhine-III's (documentation says same bit is correct) | ||
96 | - transmit frame queue message is off by one - fixed | ||
97 | - adds IntrNormalSummary to "Something Wicked" exclusion list | ||
98 | so normal interrupts will not trigger the message (src: Donald Becker) | ||
99 | (Roger Luethi) | ||
100 | - show confused chip where to continue after Tx error | ||
101 | - location of collision counter is chip specific | ||
102 | - allow selecting backoff algorithm (module parameter) | ||
103 | |||
104 | LK1.1.15 (jgarzik): | ||
105 | - Use new MII lib helper generic_mii_ioctl | ||
106 | |||
107 | LK1.1.16 (Roger Luethi) | ||
108 | - Etherleak fix | ||
109 | - Handle Tx buffer underrun | ||
110 | - Fix bugs in full duplex handling | ||
111 | - New reset code uses "force reset" cmd on Rhine-II | ||
112 | - Various clean ups | ||
113 | |||
114 | LK1.1.17 (Roger Luethi) | ||
115 | - Fix race in via_rhine_start_tx() | ||
116 | - On errors, wait for Tx engine to turn off before scavenging | ||
117 | - Handle Tx descriptor write-back race on Rhine-II | ||
118 | - Force flushing for PCI posted writes | ||
119 | - More reset code changes | ||
120 | |||
121 | LK1.1.18 (Roger Luethi) | ||
122 | - No filtering multicast in promisc mode (Edward Peng) | ||
123 | - Fix for Rhine-I Tx timeouts | ||
124 | |||
125 | LK1.1.19 (Roger Luethi) | ||
126 | - Increase Tx threshold for unspecified errors | ||
127 | |||
128 | LK1.2.0-2.6 (Roger Luethi) | ||
129 | - Massive clean-up | ||
130 | - Rewrite PHY, media handling (remove options, full_duplex, backoff) | ||
131 | - Fix Tx engine race for good | ||
132 | - Craig Brind: Zero padded aligned buffers for short packets. | ||
133 | 29 | ||
134 | */ | 30 | */ |
135 | 31 | ||
136 | #define DRV_NAME "via-rhine" | 32 | #define DRV_NAME "via-rhine" |
137 | #define DRV_VERSION "1.2.0-2.6" | 33 | #define DRV_VERSION "1.4.0" |
138 | #define DRV_RELDATE "June-10-2004" | 34 | #define DRV_RELDATE "June-27-2006" |
139 | 35 | ||
140 | 36 | ||
141 | /* A few user-configurable values. | 37 | /* A few user-configurable values. |