diff options
Diffstat (limited to 'sound/oss/sb_ess.c')
-rw-r--r-- | sound/oss/sb_ess.c | 28 |
1 files changed, 14 insertions, 14 deletions
diff --git a/sound/oss/sb_ess.c b/sound/oss/sb_ess.c index fae05fe3de43..180e95c87e3e 100644 --- a/sound/oss/sb_ess.c +++ b/sound/oss/sb_ess.c | |||
@@ -97,19 +97,19 @@ | |||
97 | * | 97 | * |
98 | * The documentation is an adventure: it's close but not fully accurate. I | 98 | * The documentation is an adventure: it's close but not fully accurate. I |
99 | * found out that after a reset some registers are *NOT* reset, though the | 99 | * found out that after a reset some registers are *NOT* reset, though the |
100 | * docs say the would be. Interresting ones are 0x7f, 0x7d and 0x7a. They are | 100 | * docs say the would be. Interesting ones are 0x7f, 0x7d and 0x7a. They are |
101 | * related to the Audio 2 channel. I also was suprised about the consequenses | 101 | * related to the Audio 2 channel. I also was surprised about the consequences |
102 | * of writing 0x00 to 0x7f (which should be done by reset): The ES1887 moves | 102 | * of writing 0x00 to 0x7f (which should be done by reset): The ES1887 moves |
103 | * into ES1888 mode. This means that it claims IRQ 11, which happens to be my | 103 | * into ES1888 mode. This means that it claims IRQ 11, which happens to be my |
104 | * ISDN adapter. Needless to say it no longer worked. I now understand why | 104 | * ISDN adapter. Needless to say it no longer worked. I now understand why |
105 | * after rebooting 0x7f already was 0x05, the value of my choice: the BIOS | 105 | * after rebooting 0x7f already was 0x05, the value of my choice: the BIOS |
106 | * did it. | 106 | * did it. |
107 | * | 107 | * |
108 | * Oh, and this is another trap: in ES1887 docs mixer register 0x70 is decribed | 108 | * Oh, and this is another trap: in ES1887 docs mixer register 0x70 is |
109 | * as if it's exactly the same as register 0xa1. This is *NOT* true. The | 109 | * described as if it's exactly the same as register 0xa1. This is *NOT* true. |
110 | * description of 0x70 in ES1869 docs is accurate however. | 110 | * The description of 0x70 in ES1869 docs is accurate however. |
111 | * Well, the assumption about ES1869 was wrong: register 0x70 is very much | 111 | * Well, the assumption about ES1869 was wrong: register 0x70 is very much |
112 | * like register 0xa1, except that bit 7 is allways 1, whatever you want | 112 | * like register 0xa1, except that bit 7 is always 1, whatever you want |
113 | * it to be. | 113 | * it to be. |
114 | * | 114 | * |
115 | * When using audio 2 mixer register 0x72 seems te be meaningless. Only 0xa2 | 115 | * When using audio 2 mixer register 0x72 seems te be meaningless. Only 0xa2 |
@@ -117,10 +117,10 @@ | |||
117 | * | 117 | * |
118 | * Software reset not being able to reset all registers is great! Especially | 118 | * Software reset not being able to reset all registers is great! Especially |
119 | * the fact that register 0x78 isn't reset is great when you wanna change back | 119 | * the fact that register 0x78 isn't reset is great when you wanna change back |
120 | * to single dma operation (simplex): audio 2 is still operation, and uses the | 120 | * to single dma operation (simplex): audio 2 is still operational, and uses |
121 | * same dma as audio 1: your ess changes into a funny echo machine. | 121 | * the same dma as audio 1: your ess changes into a funny echo machine. |
122 | * | 122 | * |
123 | * Received the new that ES1688 is detected as a ES1788. Did some thinking: | 123 | * Received the news that ES1688 is detected as a ES1788. Did some thinking: |
124 | * the ES1887 detection scheme suggests in step 2 to try if bit 3 of register | 124 | * the ES1887 detection scheme suggests in step 2 to try if bit 3 of register |
125 | * 0x64 can be changed. This is inaccurate, first I inverted the * check: "If | 125 | * 0x64 can be changed. This is inaccurate, first I inverted the * check: "If |
126 | * can be modified, it's a 1688", which lead to a correct detection | 126 | * can be modified, it's a 1688", which lead to a correct detection |
@@ -135,7 +135,7 @@ | |||
135 | * About recognition of ESS chips | 135 | * About recognition of ESS chips |
136 | * | 136 | * |
137 | * The distinction of ES688, ES1688, ES1788, ES1887 and ES1888 is described in | 137 | * The distinction of ES688, ES1688, ES1788, ES1887 and ES1888 is described in |
138 | * a (preliminary ??) datasheet on ES1887. It's aim is to identify ES1887, but | 138 | * a (preliminary ??) datasheet on ES1887. Its aim is to identify ES1887, but |
139 | * during detection the text claims that "this chip may be ..." when a step | 139 | * during detection the text claims that "this chip may be ..." when a step |
140 | * fails. This scheme is used to distinct between the above chips. | 140 | * fails. This scheme is used to distinct between the above chips. |
141 | * It appears however that some PnP chips like ES1868 are recognized as ES1788 | 141 | * It appears however that some PnP chips like ES1868 are recognized as ES1788 |
@@ -156,9 +156,9 @@ | |||
156 | * | 156 | * |
157 | * The existing ES1688 support didn't take care of the ES1688+ recording | 157 | * The existing ES1688 support didn't take care of the ES1688+ recording |
158 | * levels very well. Whenever a device was selected (recmask) for recording | 158 | * levels very well. Whenever a device was selected (recmask) for recording |
159 | * it's recording level was loud, and it couldn't be changed. The fact that | 159 | * its recording level was loud, and it couldn't be changed. The fact that |
160 | * internal register 0xb4 could take care of RECLEV, didn't work meaning until | 160 | * internal register 0xb4 could take care of RECLEV, didn't work meaning until |
161 | * it's value was restored every time the chip was reset; this reset the | 161 | * its value was restored every time the chip was reset; this reset the |
162 | * value of 0xb4 too. I guess that's what 4front also had (have?) trouble with. | 162 | * value of 0xb4 too. I guess that's what 4front also had (have?) trouble with. |
163 | * | 163 | * |
164 | * About ES1887 support: | 164 | * About ES1887 support: |
@@ -169,9 +169,9 @@ | |||
169 | * the latter case the recording volumes are 0. | 169 | * the latter case the recording volumes are 0. |
170 | * Now recording levels of inputs can be controlled, by changing the playback | 170 | * Now recording levels of inputs can be controlled, by changing the playback |
171 | * levels. Futhermore several devices can be recorded together (which is not | 171 | * levels. Futhermore several devices can be recorded together (which is not |
172 | * possible with the ES1688. | 172 | * possible with the ES1688). |
173 | * Besides the separate recording level control for each input, the common | 173 | * Besides the separate recording level control for each input, the common |
174 | * recordig level can also be controlled by RECLEV as described above. | 174 | * recording level can also be controlled by RECLEV as described above. |
175 | * | 175 | * |
176 | * Not only ES1887 have this recording mixer. I know the following from the | 176 | * Not only ES1887 have this recording mixer. I know the following from the |
177 | * documentation: | 177 | * documentation: |