kevin/linux
1
Fork 0
Code
6881647cce
linux/drivers/media/video/cx88/cx88-core.c

1096 lines
30 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/*
*
* device driver for Conexant 2388x based TV cards
* driver core
*
* (c) 2003 Gerd Knorr <kraxel@bytesex.org> [SuSE Labs]
*
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
* (c) 2005-2006 Mauro Carvalho Chehab <mchehab@infradead.org>
* - Multituner support
* - video_ioctl2 conversion
* - PAL/M fixes
*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include <linux/init.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/kmod.h>
#include <linux/sound.h>
#include <linux/interrupt.h>
#include <linux/pci.h>
#include <linux/delay.h>
[PATCH] v4l: 739: created make changelog to make easier to generate patches - Created make changelog to make easier to generate patches. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:17 -07:00
#include <linux/videodev2.h>
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
#include <linux/mutex.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#include "cx88.h"
V4L/DVB (3269): ioctls cleanups. - Now, all internal ioctls are at v4l2-common.h - removed unused ioctl at saa6752hs.h - all debug ioctl code moved to v4l2-common.c - removed duplicated stuff from other cards Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:31 -07:00
#include <media/v4l2-common.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
MODULE_DESCRIPTION("v4l2 driver module for cx2388x based TV cards");
MODULE_AUTHOR("Gerd Knorr <kraxel@bytesex.org> [SuSE Labs]");
MODULE_LICENSE("GPL");
/* ------------------------------------------------------------------ */
static unsigned int core_debug = 0;
module_param(core_debug,int,0644);
MODULE_PARM_DESC(core_debug,"enable debug messages [core]");
static unsigned int nicam = 0;
module_param(nicam,int,0644);
MODULE_PARM_DESC(nicam,"tv audio is nicam");
static unsigned int nocomb = 0;
module_param(nocomb,int,0644);
MODULE_PARM_DESC(nocomb,"disable comb filter");
#define dprintk(level,fmt, arg...) if (core_debug >= level) \
printk(KERN_DEBUG "%s: " fmt, core->name , ## arg)
static unsigned int cx88_devcount;
static LIST_HEAD(cx88_devlist);
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
static DEFINE_MUTEX(devlist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#define NO_SYNC_LINE (-1U)
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
/* @lpi: lines per IRQ, or 0 to not generate irqs. Note: IRQ to be
generated _after_ lpi lines are transferred. */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
static u32* cx88_risc_field(u32 *rp, struct scatterlist *sglist,
unsigned int offset, u32 sync_line,
unsigned int bpl, unsigned int padding,
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
unsigned int lines, unsigned int lpi)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct scatterlist *sg;
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
unsigned int line,todo,sol;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* sync instruction */
if (sync_line != NO_SYNC_LINE)
*(rp++) = cpu_to_le32(RISC_RESYNC | sync_line);
/* scan lines */
sg = sglist;
for (line = 0; line < lines; line++) {
while (offset && offset >= sg_dma_len(sg)) {
offset -= sg_dma_len(sg);
sg++;
}
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
if (lpi && line>0 && !(line % lpi))
sol = RISC_SOL | RISC_IRQ1 | RISC_CNT_INC;
else
sol = RISC_SOL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (bpl <= sg_dma_len(sg)-offset) {
/* fits into current chunk */
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
*(rp++)=cpu_to_le32(RISC_WRITE|sol|RISC_EOL|bpl);
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
*(rp++)=cpu_to_le32(sg_dma_address(sg)+offset);
offset+=bpl;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
} else {
V4L/DVB (4361): Cx88: add support for Norwood PCI TV Tuner (non-pro) This patch adds support for Norwood PCI TV Tuner (non-pro) Signed-off-by: Peter Naulls <peter@chocky.org> Signed-off-by: Michael Krufky <mkrufky@linuxtv.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-08-08 05:10:05 -07:00
/* scanline needs to be split */
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
todo = bpl;
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
*(rp++)=cpu_to_le32(RISC_WRITE|sol|
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
(sg_dma_len(sg)-offset));
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
*(rp++)=cpu_to_le32(sg_dma_address(sg)+offset);
todo -= (sg_dma_len(sg)-offset);
offset = 0;
sg++;
while (todo > sg_dma_len(sg)) {
[PATCH] v4l: 801: whitespaces cleanups - Whitespaces Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:45 -07:00
*(rp++)=cpu_to_le32(RISC_WRITE|
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
sg_dma_len(sg));
[PATCH] v4l: 801: whitespaces cleanups - Whitespaces Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:45 -07:00
*(rp++)=cpu_to_le32(sg_dma_address(sg));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
todo -= sg_dma_len(sg);
sg++;
}
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
*(rp++)=cpu_to_le32(RISC_WRITE|RISC_EOL|todo);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*(rp++)=cpu_to_le32(sg_dma_address(sg));
offset += todo;
}
offset += padding;
}
return rp;
}
int cx88_risc_buffer(struct pci_dev *pci, struct btcx_riscmem *risc,
struct scatterlist *sglist,
unsigned int top_offset, unsigned int bottom_offset,
unsigned int bpl, unsigned int padding, unsigned int lines)
{
u32 instructions,fields;
u32 *rp;
int rc;
fields = 0;
if (UNSET != top_offset)
fields++;
if (UNSET != bottom_offset)
fields++;
/* estimate risc mem: worst case is one write per page border +
V4L/DVB (3766): Correct buffer size calculations in cx88-core.c The computation in cx88_risc_buffer suffers from the mistake: a non-zero padding value can cause more page borders to be crossed, leading to big buffer over-runs. This patch changes the additive constant from 3 + 4 to 4 It also changees the constant in cx88_risc_databuffer from 3 + 4 to 2, because 2 dwords are the correct vaule. Signed-off-by: Duncan Sands <baldrick@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-04-11 06:18:57 -07:00
one write per scan line + syncs + jump (all 2 dwords). Padding
can cause next bpl to start close to a page border. First DMA
region may be smaller than PAGE_SIZE */
instructions = fields * (1 + ((bpl + padding) * lines) / PAGE_SIZE + lines);
instructions += 2;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if ((rc = btcx_riscmem_alloc(pci,risc,instructions*8)) < 0)
return rc;
/* write risc instructions */
rp = risc->cpu;
if (UNSET != top_offset)
rp = cx88_risc_field(rp, sglist, top_offset, 0,
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
bpl, padding, lines, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (UNSET != bottom_offset)
rp = cx88_risc_field(rp, sglist, bottom_offset, 0x200,
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
bpl, padding, lines, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* save pointer to jmp instruction address */
risc->jmp = rp;
V4L/DVB (3394): Bttv: correct bttv_risc_packed buffer size This patch fixes the strange crashes I was seeing after using bttv card, caused by a buffer overflow in bttv_risc_packed. The instruction buffer size calculation contains two errors: (a) a non-zero padding value can push the start of the next bpl section to just before a page border, leading to more scanline splits and thus additional instructions. (b) the first DMA region can be smaller than one page, so there can be a scanline split even if bpl*lines is smaller than PAGE_SIZE. For example, consider the case where offset is 0, bpl is 2, padding is 4094, lines is smaller than 2048, the first DMA region has size 1 and all others have size PAGE_SIZE, assumed to equal 4096. Then all bpl regions cross page borders and the number of instructions written is 2*lines+2, rather than lines+2 (the current estimate). With this patch the number of instructions for this example is estimated to be 2*lines+3. Also, the BUG_ON that was supposed to catch buffer overflows contained a thinko causing it fire only if the buffer was overrun by a factor of 16 or more, so it fixes the the BUG_ON's (using sizeof rather than "4"). Signed-off-by: Duncan Sands <baldrick@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-02-26 20:09:48 -07:00
BUG_ON((risc->jmp - risc->cpu + 2) * sizeof (*risc->cpu) > risc->size);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return 0;
}
int cx88_risc_databuffer(struct pci_dev *pci, struct btcx_riscmem *risc,
struct scatterlist *sglist, unsigned int bpl,
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
unsigned int lines, unsigned int lpi)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
u32 instructions;
u32 *rp;
int rc;
/* estimate risc mem: worst case is one write per page border +
V4L/DVB (3766): Correct buffer size calculations in cx88-core.c The computation in cx88_risc_buffer suffers from the mistake: a non-zero padding value can cause more page borders to be crossed, leading to big buffer over-runs. This patch changes the additive constant from 3 + 4 to 4 It also changees the constant in cx88_risc_databuffer from 3 + 4 to 2, because 2 dwords are the correct vaule. Signed-off-by: Duncan Sands <baldrick@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-04-11 06:18:57 -07:00
one write per scan line + syncs + jump (all 2 dwords). Here
there is no padding and no sync. First DMA region may be smaller
than PAGE_SIZE */
instructions = 1 + (bpl * lines) / PAGE_SIZE + lines;
instructions += 1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if ((rc = btcx_riscmem_alloc(pci,risc,instructions*8)) < 0)
return rc;
/* write risc instructions */
rp = risc->cpu;
V4L/DVB (6083): cx88-alsa: Rework buffer handling Rework the way the DMA buffer is handled and IRQs are generated. ALSA uses a ring-buffer of multiple periods. Each period is supposed to corrispond to one IRQ. The existing driver was generating one interrupt per ring-buffer, as opposed to per period. This meant that as soon as the IRQ was generated, the hardware was already starting to re-write the beginning of the buffer. Since the DMA happens on a per-line basis, there was only a narrow window to copy the data out before the buffer was overwritten. The cx88 core RISC program generator is modified so that it can set the IRQ and counter flags to count every X lines of DMA transfer. This way we can generate an interrupt every period instead of every full ring-buffer. Right now only period of one line are supported, but it should be possible to support longer periods. Note that a WRITE instruction generates an IRQ when it starts, not when the transfer is finished. Thus to generate an IRQ when line X is done, one must set the IRQ flag on the instruction that starts line X+1, not the one that ends line X. Change the line size so that there are four lines in the SRAM FIFO. If there are not four lines, the analog output from the cx88's internal DACs is full of clicks and pops. Try to handle FIFO sync errors. Sometimes the chip generates many of these errors before audio data starts. Up to 50 sync errors will be ignored and the counter reset. Have the IRQ handler save the RISC counter to the chip struct, and then have the pointer callback use this to calculate the pointer position. We could read the counter from the pointer callback, but sometimes the sync errors on start up cause the counter to go crazy. ALSA sees this and thinks there has been an overrun. The IRQ hander can avoid saving the counter position on sync errors. The chip "opened" flag wasn't necessary. ALSA won't try to open the same substream multiple times. Probably this code was cut&pasted from the bt87x driver, which has multiple sub-streams for one chip. Do error checking for the videobuf mapping functions. snd_card_cx88_runtime_free() is useless and can be deleted. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-23 21:06:34 -07:00
rp = cx88_risc_field(rp, sglist, 0, NO_SYNC_LINE, bpl, 0, lines, lpi);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* save pointer to jmp instruction address */
risc->jmp = rp;
V4L/DVB (3394): Bttv: correct bttv_risc_packed buffer size This patch fixes the strange crashes I was seeing after using bttv card, caused by a buffer overflow in bttv_risc_packed. The instruction buffer size calculation contains two errors: (a) a non-zero padding value can push the start of the next bpl section to just before a page border, leading to more scanline splits and thus additional instructions. (b) the first DMA region can be smaller than one page, so there can be a scanline split even if bpl*lines is smaller than PAGE_SIZE. For example, consider the case where offset is 0, bpl is 2, padding is 4094, lines is smaller than 2048, the first DMA region has size 1 and all others have size PAGE_SIZE, assumed to equal 4096. Then all bpl regions cross page borders and the number of instructions written is 2*lines+2, rather than lines+2 (the current estimate). With this patch the number of instructions for this example is estimated to be 2*lines+3. Also, the BUG_ON that was supposed to catch buffer overflows contained a thinko causing it fire only if the buffer was overrun by a factor of 16 or more, so it fixes the the BUG_ON's (using sizeof rather than "4"). Signed-off-by: Duncan Sands <baldrick@free.fr> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-02-26 20:09:48 -07:00
BUG_ON((risc->jmp - risc->cpu + 2) * sizeof (*risc->cpu) > risc->size);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return 0;
}
int cx88_risc_stopper(struct pci_dev *pci, struct btcx_riscmem *risc,
u32 reg, u32 mask, u32 value)
{
u32 *rp;
int rc;
if ((rc = btcx_riscmem_alloc(pci, risc, 4*16)) < 0)
return rc;
/* write risc instructions */
rp = risc->cpu;
*(rp++) = cpu_to_le32(RISC_WRITECR | RISC_IRQ2 | RISC_IMM);
*(rp++) = cpu_to_le32(reg);
*(rp++) = cpu_to_le32(value);
*(rp++) = cpu_to_le32(mask);
*(rp++) = cpu_to_le32(RISC_JUMP);
*(rp++) = cpu_to_le32(risc->dma);
return 0;
}
void
V4L/DVB (3516): Make video_buf more generic Video_buf were concerned to allow PCI devices to be used as video capture devices. This patch extends video_buf features by virtualizing pci-dependent functions and allowing other type of devices to use it. It is still DMA centric, although it may be used also by devices that emulates scatter/gather behavior or a DMA device Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-03-10 08:29:15 -07:00
cx88_free_buffer(struct videobuf_queue *q, struct cx88_buffer *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (6252): Adapt drivers to use the newer videobuf modules PCI-dependent videobuf_foo methods were renamed as videobuf_pci_foo. Also, videobuf_dmabuf is now part of videobuf-dma-sg private struct. So, to access it, a subroutine call is needed. This patch renames all occurences of those function calls to be consistent with the video-buf split. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org> http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981 Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
2007-08-23 12:37:49 -07:00
struct videobuf_dmabuf *dma=videobuf_to_dma(&buf->vb);
V4L/DVB (3485): BUG_ON() Conversion in drivers/video/media Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-03-13 09:17:11 -07:00
BUG_ON(in_interrupt());
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
videobuf_waiton(&buf->vb,0,0);
V4L/DVB (6252): Adapt drivers to use the newer videobuf modules PCI-dependent videobuf_foo methods were renamed as videobuf_pci_foo. Also, videobuf_dmabuf is now part of videobuf-dma-sg private struct. So, to access it, a subroutine call is needed. This patch renames all occurences of those function calls to be consistent with the video-buf split. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org> http://thread.gmane.org/gmane.comp.video.video4linux/34978/focus=34981 Reviewed-by: Ricardo Cerqueira <v4l@cerqueira.org>
2007-08-23 12:37:49 -07:00
videobuf_dma_unmap(q, dma);
videobuf_dma_free(dma);
V4L/DVB (3516): Make video_buf more generic Video_buf were concerned to allow PCI devices to be used as video capture devices. This patch extends video_buf features by virtualizing pci-dependent functions and allowing other type of devices to use it. It is still DMA centric, although it may be used also by devices that emulates scatter/gather behavior or a DMA device Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-03-10 08:29:15 -07:00
btcx_riscmem_free((struct pci_dev *)q->dev, &buf->risc);
V4L/DVB (6600): V4L: videobuf: don't chew up namespace STATE_.*, convert to VIDEOBUF_ s/STATE_NEEDS_INIT/VIDEOBUF_NEEDS_INIT/g s/STATE_PREPARED/VIDEOBUF_PREPARED/g s/STATE_QUEUED/VIDEOBUF_QUEUED/g s/STATE_ACTIVE/VIDEOBUF_ACTIVE/g s/STATE_DONE/VIDEOBUF_DONE/g s/STATE_ERROR/VIDEOBUF_ERROR/g s/STATE_IDLE/VIDEOBUF_IDLE/g Signed-off-by: Brandon Philips <bphilips@suse.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-11-06 16:02:36 -07:00
buf->vb.state = VIDEOBUF_NEEDS_INIT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
/* ------------------------------------------------------------------ */
/* our SRAM memory layout */
/* we are going to put all thr risc programs into host memory, so we
* can use the whole SDRAM for the DMA fifos. To simplify things, we
* use a static memory layout. That surely will waste memory in case
* we don't use all DMA channels at the same time (which will be the
* case most of the time). But that still gives us enougth FIFO space
* to be able to deal with insane long pci latencies ...
*
* FIFO space allocations:
* channel 21 (y video) - 10.0k
* channel 22 (u video) - 2.0k
* channel 23 (v video) - 2.0k
* channel 24 (vbi) - 4.0k
V4L/DVB (3293): Added digital support for cx88 (cx88-alsa) - This module is co-authored by Ricardo Cerqueira <v4l@cerqueira.org>. - Added digital audio support for cx88-based boards that have function 01 enabled. These boards can be identified by having PCI id 1471:8801 or 1471:8811. - Increased DMA buffer from 512 to 4096 seems to fix audio distortion. - Existing audio DMA uses conflict with cx88-alsa. Should be disabled when cx88-alsa module is compiled. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:44 -07:00
* channels 25+26 (audio) - 4.0k
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
* channel 28 (mpeg) - 4.0k
V4L/DVB (3293): Added digital support for cx88 (cx88-alsa) - This module is co-authored by Ricardo Cerqueira <v4l@cerqueira.org>. - Added digital audio support for cx88-based boards that have function 01 enabled. These boards can be identified by having PCI id 1471:8801 or 1471:8811. - Increased DMA buffer from 512 to 4096 seems to fix audio distortion. - Existing audio DMA uses conflict with cx88-alsa. Should be disabled when cx88-alsa module is compiled. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:44 -07:00
* TOTAL = 29.0k
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*
* Every channel has 160 bytes control data (64 bytes instruction
* queue and 6 CDT entries), which is close to 2k total.
*
* Address layout:
* 0x0000 - 0x03ff CMDs / reserved
* 0x0400 - 0x0bff instruction queues + CDs
* 0x0c00 - FIFOs
*/
struct sram_channel cx88_sram_channels[] = {
[SRAM_CH21] = {
.name = "video y / packed",
.cmds_start = 0x180040,
.ctrl_start = 0x180400,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x180400 + 64,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x180c00,
.fifo_size = 0x002800,
.ptr1_reg = MO_DMA21_PTR1,
.ptr2_reg = MO_DMA21_PTR2,
.cnt1_reg = MO_DMA21_CNT1,
.cnt2_reg = MO_DMA21_CNT2,
},
[SRAM_CH22] = {
.name = "video u",
.cmds_start = 0x180080,
.ctrl_start = 0x1804a0,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x1804a0 + 64,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x183400,
.fifo_size = 0x000800,
.ptr1_reg = MO_DMA22_PTR1,
.ptr2_reg = MO_DMA22_PTR2,
.cnt1_reg = MO_DMA22_CNT1,
.cnt2_reg = MO_DMA22_CNT2,
},
[SRAM_CH23] = {
.name = "video v",
.cmds_start = 0x1800c0,
.ctrl_start = 0x180540,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x180540 + 64,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x183c00,
.fifo_size = 0x000800,
.ptr1_reg = MO_DMA23_PTR1,
.ptr2_reg = MO_DMA23_PTR2,
.cnt1_reg = MO_DMA23_CNT1,
.cnt2_reg = MO_DMA23_CNT2,
},
[SRAM_CH24] = {
.name = "vbi",
.cmds_start = 0x180100,
.ctrl_start = 0x1805e0,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x1805e0 + 64,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x184400,
.fifo_size = 0x001000,
.ptr1_reg = MO_DMA24_PTR1,
.ptr2_reg = MO_DMA24_PTR2,
.cnt1_reg = MO_DMA24_CNT1,
.cnt2_reg = MO_DMA24_CNT2,
},
[SRAM_CH25] = {
.name = "audio from",
.cmds_start = 0x180140,
.ctrl_start = 0x180680,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x180680 + 64,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x185400,
V4L/DVB (3293): Added digital support for cx88 (cx88-alsa) - This module is co-authored by Ricardo Cerqueira <v4l@cerqueira.org>. - Added digital audio support for cx88-based boards that have function 01 enabled. These boards can be identified by having PCI id 1471:8801 or 1471:8811. - Increased DMA buffer from 512 to 4096 seems to fix audio distortion. - Existing audio DMA uses conflict with cx88-alsa. Should be disabled when cx88-alsa module is compiled. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:44 -07:00
.fifo_size = 0x001000,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.ptr1_reg = MO_DMA25_PTR1,
.ptr2_reg = MO_DMA25_PTR2,
.cnt1_reg = MO_DMA25_CNT1,
.cnt2_reg = MO_DMA25_CNT2,
},
[SRAM_CH26] = {
.name = "audio to",
.cmds_start = 0x180180,
.ctrl_start = 0x180720,
[PATCH] v4l: 800: whitespace cleanups - Whitespace Cleanups. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:37:43 -07:00
.cdt = 0x180680 + 64, /* same as audio IN */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_start = 0x185400, /* same as audio IN */
V4L/DVB (3293): Added digital support for cx88 (cx88-alsa) - This module is co-authored by Ricardo Cerqueira <v4l@cerqueira.org>. - Added digital audio support for cx88-based boards that have function 01 enabled. These boards can be identified by having PCI id 1471:8801 or 1471:8811. - Increased DMA buffer from 512 to 4096 seems to fix audio distortion. - Existing audio DMA uses conflict with cx88-alsa. Should be disabled when cx88-alsa module is compiled. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:44 -07:00
.fifo_size = 0x001000, /* same as audio IN */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.ptr1_reg = MO_DMA26_PTR1,
.ptr2_reg = MO_DMA26_PTR2,
.cnt1_reg = MO_DMA26_CNT1,
.cnt2_reg = MO_DMA26_CNT2,
},
[SRAM_CH28] = {
.name = "mpeg",
.cmds_start = 0x180200,
.ctrl_start = 0x1807C0,
.cdt = 0x1807C0 + 64,
V4L/DVB (3293): Added digital support for cx88 (cx88-alsa) - This module is co-authored by Ricardo Cerqueira <v4l@cerqueira.org>. - Added digital audio support for cx88-based boards that have function 01 enabled. These boards can be identified by having PCI id 1471:8801 or 1471:8811. - Increased DMA buffer from 512 to 4096 seems to fix audio distortion. - Existing audio DMA uses conflict with cx88-alsa. Should be disabled when cx88-alsa module is compiled. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
2006-01-09 10:32:44 -07:00
.fifo_start = 0x186400,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.fifo_size = 0x001000,
.ptr1_reg = MO_DMA28_PTR1,
.ptr2_reg = MO_DMA28_PTR2,
.cnt1_reg = MO_DMA28_CNT1,
.cnt2_reg = MO_DMA28_CNT2,
},
};
int cx88_sram_channel_setup(struct cx88_core *core,
struct sram_channel *ch,
unsigned int bpl, u32 risc)
{
unsigned int i,lines;
u32 cdt;
bpl = (bpl + 7) & ~7; /* alignment */
cdt = ch->cdt;
lines = ch->fifo_size / bpl;
if (lines > 6)
lines = 6;
BUG_ON(lines < 2);
/* write CDT */
for (i = 0; i < lines; i++)
cx_write(cdt + 16*i, ch->fifo_start + bpl*i);
/* write CMDS */
cx_write(ch->cmds_start + 0, risc);
cx_write(ch->cmds_start + 4, cdt);
cx_write(ch->cmds_start + 8, (lines*16) >> 3);
cx_write(ch->cmds_start + 12, ch->ctrl_start);
cx_write(ch->cmds_start + 16, 64 >> 2);
for (i = 20; i < 64; i += 4)
cx_write(ch->cmds_start + i, 0);
/* fill registers */
cx_write(ch->ptr1_reg, ch->fifo_start);
cx_write(ch->ptr2_reg, cdt);
cx_write(ch->cnt1_reg, (bpl >> 3) -1);
cx_write(ch->cnt2_reg, (lines*16) >> 3);
dprintk(2,"sram setup %s: bpl=%d lines=%d\n", ch->name, bpl, lines);
return 0;
}
/* ------------------------------------------------------------------ */
/* debug helper code */
[PATCH] v4l: 651: fix a number of sparse warnings - Fix a number of sparse warnings. Signed-off-by: Peter Hagervall <hager@cs.umu.se> Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:29 -07:00
static int cx88_risc_decode(u32 risc)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
static char *instr[16] = {
[ RISC_SYNC >> 28 ] = "sync",
[ RISC_WRITE >> 28 ] = "write",
[ RISC_WRITEC >> 28 ] = "writec",
[ RISC_READ >> 28 ] = "read",
[ RISC_READC >> 28 ] = "readc",
[ RISC_JUMP >> 28 ] = "jump",
[ RISC_SKIP >> 28 ] = "skip",
[ RISC_WRITERM >> 28 ] = "writerm",
[ RISC_WRITECM >> 28 ] = "writecm",
[ RISC_WRITECR >> 28 ] = "writecr",
};
static int incr[16] = {
[ RISC_WRITE >> 28 ] = 2,
[ RISC_JUMP >> 28 ] = 2,
[ RISC_WRITERM >> 28 ] = 3,
[ RISC_WRITECM >> 28 ] = 3,
[ RISC_WRITECR >> 28 ] = 4,
};
static char *bits[] = {
"12", "13", "14", "resync",
"cnt0", "cnt1", "18", "19",
"20", "21", "22", "23",
"irq1", "irq2", "eol", "sol",
};
int i;
printk("0x%08x [ %s", risc,
instr[risc >> 28] ? instr[risc >> 28] : "INVALID");
for (i = ARRAY_SIZE(bits)-1; i >= 0; i--)
if (risc & (1 << (i + 12)))
printk(" %s",bits[i]);
printk(" count=%d ]\n", risc & 0xfff);
return incr[risc >> 28] ? incr[risc >> 28] : 1;
}
void cx88_sram_channel_dump(struct cx88_core *core,
struct sram_channel *ch)
{
static char *name[] = {
"initial risc",
"cdt base",
"cdt size",
"iq base",
"iq size",
"risc pc",
"iq wr ptr",
"iq rd ptr",
"cdt current",
"pci target",
"line / byte",
};
u32 risc;
unsigned int i,j,n;
printk("%s: %s - dma channel status dump\n",
core->name,ch->name);
for (i = 0; i < ARRAY_SIZE(name); i++)
printk("%s: cmds: %-12s: 0x%08x\n",
core->name,name[i],
cx_read(ch->cmds_start + 4*i));
V4L/DVB (6082): cx88: Improve risc instruction printing in sram channel dump When the risc instrunctions from the CMDS were printed, instruction arguments weren't taken into account. This changes output like: cx88[0]: risc0: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc1: 0x0cac2800 [ INVALID sol eol 23 21 19 18 13 count=2048 ] cx88[0]: risc2: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc3: 0x0cac2c00 [ INVALID sol eol 23 21 19 18 13 count=3072 ] cx88[0]: risc0: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc1: 0x0cac2800 [ arg #1 ] cx88[0]: risc2: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc3: 0x0cac2c00 [ arg #1 ] Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-21 04:19:16 -07:00
for (n = 1, i = 0; i < 4; i++) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
risc = cx_read(ch->cmds_start + 4 * (i+11));
printk("%s: risc%d: ", core->name, i);
V4L/DVB (6082): cx88: Improve risc instruction printing in sram channel dump When the risc instrunctions from the CMDS were printed, instruction arguments weren't taken into account. This changes output like: cx88[0]: risc0: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc1: 0x0cac2800 [ INVALID sol eol 23 21 19 18 13 count=2048 ] cx88[0]: risc2: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc3: 0x0cac2c00 [ INVALID sol eol 23 21 19 18 13 count=3072 ] cx88[0]: risc0: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc1: 0x0cac2800 [ arg #1 ] cx88[0]: risc2: 0x1d010400 [ write sol eol irq1 cnt0 count=1024 ] cx88[0]: risc3: 0x0cac2c00 [ arg #1 ] Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-21 04:19:16 -07:00
if (--n)
printk("0x%08x [ arg #%d ]\n", risc, n);
else
n = cx88_risc_decode(risc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
for (i = 0; i < 16; i += n) {
risc = cx_read(ch->ctrl_start + 4 * i);
printk("%s: iq %x: ", core->name, i);
n = cx88_risc_decode(risc);
for (j = 1; j < n; j++) {
risc = cx_read(ch->ctrl_start + 4 * (i+j));
printk("%s: iq %x: 0x%08x [ arg #%d ]\n",
core->name, i+j, risc, j);
}
}
printk("%s: fifo: 0x%08x -> 0x%x\n",
core->name, ch->fifo_start, ch->fifo_start+ch->fifo_size);
printk("%s: ctrl: 0x%08x -> 0x%x\n",
core->name, ch->ctrl_start, ch->ctrl_start+6*16);
printk("%s: ptr1_reg: 0x%08x\n",
core->name,cx_read(ch->ptr1_reg));
printk("%s: ptr2_reg: 0x%08x\n",
core->name,cx_read(ch->ptr2_reg));
printk("%s: cnt1_reg: 0x%08x\n",
core->name,cx_read(ch->cnt1_reg));
printk("%s: cnt2_reg: 0x%08x\n",
core->name,cx_read(ch->cnt2_reg));
}
[PATCH] make lots of things static Another large rollup of various patches from Adrian which make things static where they were needlessly exported. Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-05-01 08:59:29 -07:00
static char *cx88_pci_irqs[32] = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
"vid", "aud", "ts", "vip", "hst", "5", "6", "tm1",
"src_dma", "dst_dma", "risc_rd_err", "risc_wr_err",
"brdg_err", "src_dma_err", "dst_dma_err", "ipb_dma_err",
"i2c", "i2c_rack", "ir_smp", "gpio0", "gpio1"
};
void cx88_print_irqbits(char *name, char *tag, char **strings,
V4L/DVB (5480): Fix cx88_print_irqbits calls to use ARRAY_SIZE cx88_print_irqbits were expecting a string pointer with 32 bytes. Better to pass the string size and use ARRAY_SIZE on its calls. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-03-29 04:47:04 -07:00
int len, u32 bits, u32 mask)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
unsigned int i;
printk(KERN_DEBUG "%s: %s [0x%x]", name, tag, bits);
V4L/DVB (5480): Fix cx88_print_irqbits calls to use ARRAY_SIZE cx88_print_irqbits were expecting a string pointer with 32 bytes. Better to pass the string size and use ARRAY_SIZE on its calls. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-03-29 04:47:04 -07:00
for (i = 0; i < len; i++) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (!(bits & (1 << i)))
continue;
if (strings[i])
printk(" %s", strings[i]);
else
printk(" %d", i);
if (!(mask & (1 << i)))
continue;
printk("*");
}
printk("\n");
}
/* ------------------------------------------------------------------ */
int cx88_core_irq(struct cx88_core *core, u32 status)
{
int handled = 0;
V4L/DVB (6064): cx88: Add symbolic names for the PCI interrupt bits Used for the PCI_INTMSK and PCI_INT_STAT registers. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-18 02:57:55 -07:00
if (status & PCI_INT_IR_SMPINT) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
cx88_ir_irq(core);
handled++;
}
if (!handled)
cx88_print_irqbits(core->name, "irq pci",
V4L/DVB (5480): Fix cx88_print_irqbits calls to use ARRAY_SIZE cx88_print_irqbits were expecting a string pointer with 32 bytes. Better to pass the string size and use ARRAY_SIZE on its calls. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-03-29 04:47:04 -07:00
cx88_pci_irqs, ARRAY_SIZE(cx88_pci_irqs),
status, core->pci_irqmask);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return handled;
}
void cx88_wakeup(struct cx88_core *core,
struct cx88_dmaqueue *q, u32 count)
{
struct cx88_buffer *buf;
int bc;
for (bc = 0;; bc++) {
if (list_empty(&q->active))
break;
buf = list_entry(q->active.next,
struct cx88_buffer, vb.queue);
/* count comes from the hw and is is 16bit wide --
* this trick handles wrap-arounds correctly for
* up to 32767 buffers in flight... */
if ((s16) (count - buf->count) < 0)
break;
do_gettimeofday(&buf->vb.ts);
dprintk(2,"[%p/%d] wakeup reg=%d buf=%d\n",buf,buf->vb.i,
count, buf->count);
V4L/DVB (6600): V4L: videobuf: don't chew up namespace STATE_.*, convert to VIDEOBUF_ s/STATE_NEEDS_INIT/VIDEOBUF_NEEDS_INIT/g s/STATE_PREPARED/VIDEOBUF_PREPARED/g s/STATE_QUEUED/VIDEOBUF_QUEUED/g s/STATE_ACTIVE/VIDEOBUF_ACTIVE/g s/STATE_DONE/VIDEOBUF_DONE/g s/STATE_ERROR/VIDEOBUF_ERROR/g s/STATE_IDLE/VIDEOBUF_IDLE/g Signed-off-by: Brandon Philips <bphilips@suse.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-11-06 16:02:36 -07:00
buf->vb.state = VIDEOBUF_DONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
list_del(&buf->vb.queue);
wake_up(&buf->vb.done);
}
if (list_empty(&q->active)) {
del_timer(&q->timeout);
} else {
mod_timer(&q->timeout, jiffies+BUFFER_TIMEOUT);
}
if (bc != 1)
printk("%s: %d buffers handled (should be 1)\n",__FUNCTION__,bc);
}
void cx88_shutdown(struct cx88_core *core)
{
/* disable RISC controller + IRQs */
cx_write(MO_DEV_CNTRL2, 0);
/* stop dma transfers */
cx_write(MO_VID_DMACNTRL, 0x0);
cx_write(MO_AUD_DMACNTRL, 0x0);
cx_write(MO_TS_DMACNTRL, 0x0);
cx_write(MO_VIP_DMACNTRL, 0x0);
cx_write(MO_GPHST_DMACNTRL, 0x0);
/* stop interrupts */
cx_write(MO_PCI_INTMSK, 0x0);
cx_write(MO_VID_INTMSK, 0x0);
cx_write(MO_AUD_INTMSK, 0x0);
cx_write(MO_TS_INTMSK, 0x0);
cx_write(MO_VIP_INTMSK, 0x0);
cx_write(MO_GPHST_INTMSK, 0x0);
/* stop capturing */
cx_write(VID_CAPTURE_CONTROL, 0);
}
int cx88_reset(struct cx88_core *core)
{
dprintk(1,"%s\n",__FUNCTION__);
cx88_shutdown(core);
/* clear irq status */
cx_write(MO_VID_INTSTAT, 0xFFFFFFFF); // Clear PIV int
cx_write(MO_PCI_INTSTAT, 0xFFFFFFFF); // Clear PCI int
cx_write(MO_INT1_STAT, 0xFFFFFFFF); // Clear RISC int
/* wait a bit */
msleep(100);
/* init sram */
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH21], 720*4, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH22], 128, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH23], 128, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH24], 128, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH25], 128, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH26], 128, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH28], 188*4, 0);
/* misc init ... */
cx_write(MO_INPUT_FORMAT, ((1 << 13) | // agc enable
(1 << 12) | // agc gain
(1 << 11) | // adaptibe agc
(0 << 10) | // chroma agc
(0 << 9) | // ckillen
(7)));
/* setup image format */
cx_andor(MO_COLOR_CTRL, 0x4000, 0x4000);
/* setup FIFO Threshholds */
cx_write(MO_PDMA_STHRSH, 0x0807);
cx_write(MO_PDMA_DTHRSH, 0x0807);
/* fixes flashing of image */
cx_write(MO_AGC_SYNC_TIP1, 0x0380000F);
cx_write(MO_AGC_BACK_VBI, 0x00E00555);
cx_write(MO_VID_INTSTAT, 0xFFFFFFFF); // Clear PIV int
cx_write(MO_PCI_INTSTAT, 0xFFFFFFFF); // Clear PCI int
cx_write(MO_INT1_STAT, 0xFFFFFFFF); // Clear RISC int
/* Reset on-board parts */
cx_write(MO_SRST_IO, 0);
msleep(10);
cx_write(MO_SRST_IO, 1);
return 0;
}
/* ------------------------------------------------------------------ */
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_swidth(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
return (norm & (V4L2_STD_MN & ~V4L2_STD_PAL_Nc)) ? 754 : 922;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_hdelay(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
return (norm & (V4L2_STD_MN & ~V4L2_STD_PAL_Nc)) ? 135 : 186;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_vdelay(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
return (norm & V4L2_STD_625_50) ? 0x24 : 0x18;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_fsc8(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (norm & V4L2_STD_PAL_M)
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
return 28604892; // 3.575611 MHz
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (norm & (V4L2_STD_PAL_Nc))
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
return 28656448; // 3.582056 MHz
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (norm & V4L2_STD_NTSC) // All NTSC/M and variants
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
return 28636360; // 3.57954545 MHz +/- 10 Hz
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
/* SECAM have also different sub carrier for chroma,
but step_db and step_dr, at cx88_set_tvnorm already handles that.
The same FSC applies to PAL/BGDKIH, PAL/60, NTSC/4.43 and PAL/N
*/
return 35468950; // 4.43361875 MHz +/- 5 Hz
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_htotal(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[PATCH] v4l: PAL-M support fix for CX88 chipsets This patch fixes PAL-M chroma subcarrier frequency (FSC) to its correct value of 3.5756115 MHz and adjusts horizontal total samples for PAL-M, according with formula Line Draw Time / (4*FSC). Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Cc: <video4linux-list@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-23 22:04:50 -07:00
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
unsigned int fsc4=norm_fsc8(norm)/2;
[PATCH] v4l: PAL-M support fix for CX88 chipsets This patch fixes PAL-M chroma subcarrier frequency (FSC) to its correct value of 3.5756115 MHz and adjusts horizontal total samples for PAL-M, according with formula Line Draw Time / (4*FSC). Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Cc: <video4linux-list@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-23 22:04:50 -07:00
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
/* returns 4*FSC / vtotal / frames per seconds */
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
return (norm & V4L2_STD_625_50) ?
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
((fsc4+312)/625+12)/25 :
((fsc4+262)/525*1001+15000)/30000;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
static unsigned int inline norm_vbipack(v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
return (norm & V4L2_STD_625_50) ? 511 : 400;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
int cx88_set_scale(struct cx88_core *core, unsigned int width, unsigned int height,
enum v4l2_field field)
{
unsigned int swidth = norm_swidth(core->tvnorm);
unsigned int sheight = norm_maxh(core->tvnorm);
u32 value;
dprintk(1,"set_scale: %dx%d [%s%s,%s]\n", width, height,
V4L2_FIELD_HAS_TOP(field) ? "T" : "",
V4L2_FIELD_HAS_BOTTOM(field) ? "B" : "",
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
v4l2_norm_to_name(core->tvnorm));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (!V4L2_FIELD_HAS_BOTH(field))
height *= 2;
// recalc H delay and scale registers
value = (width * norm_hdelay(core->tvnorm)) / swidth;
value &= 0x3fe;
cx_write(MO_HDELAY_EVEN, value);
cx_write(MO_HDELAY_ODD, value);
V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-12-17 19:30:47 -07:00
dprintk(1,"set_scale: hdelay 0x%04x (width %d)\n", value,swidth);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
value = (swidth * 4096 / width) - 4096;
cx_write(MO_HSCALE_EVEN, value);
cx_write(MO_HSCALE_ODD, value);
dprintk(1,"set_scale: hscale 0x%04x\n", value);
cx_write(MO_HACTIVE_EVEN, width);
cx_write(MO_HACTIVE_ODD, width);
dprintk(1,"set_scale: hactive 0x%04x\n", width);
// recalc V scale Register (delay is constant)
cx_write(MO_VDELAY_EVEN, norm_vdelay(core->tvnorm));
cx_write(MO_VDELAY_ODD, norm_vdelay(core->tvnorm));
dprintk(1,"set_scale: vdelay 0x%04x\n", norm_vdelay(core->tvnorm));
value = (0x10000 - (sheight * 512 / height - 512)) & 0x1fff;
cx_write(MO_VSCALE_EVEN, value);
cx_write(MO_VSCALE_ODD, value);
dprintk(1,"set_scale: vscale 0x%04x\n", value);
cx_write(MO_VACTIVE_EVEN, sheight);
cx_write(MO_VACTIVE_ODD, sheight);
dprintk(1,"set_scale: vactive 0x%04x\n", sheight);
// setup filters
value = 0;
value |= (1 << 19); // CFILT (default)
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (core->tvnorm & V4L2_STD_SECAM) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
value |= (1 << 15);
value |= (1 << 16);
}
V4L/DVB (6021): cx88: Copy board information into card state The cx88 driver state stored the ID of the board type in core->board. Every time the driver need to get some information about the board configuration, it uses the board number as an index into board configuration array. This patch changes it so that the board number is in core->boardnr, and core->board is a copy of the board configuration information. This allows access to board information without the extra indirection. e.g. cx88_boards[core->board].mpeg becomes core->board.mpeg. This has a number of advantages: - The code is simpler to write. - It compiles to be smaller and faster, without needing the extra array lookup to get at the board information. - The cx88_boards array no longer needs to be exported to all cx88 modules. - The boards array can be made const - It should be possible to avoid keeping the (large) cx88_boards array around after the module is loaded. - If module parameters or eeprom info override some board configuration setting, it's not necessary to modify the boards array, which would affect all boards of the same type. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-15 10:41:57 -07:00
if (INPUT(core->input).type == CX88_VMUX_SVIDEO)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
value |= (1 << 13) | (1 << 5);
if (V4L2_FIELD_INTERLACED == field)
value |= (1 << 3); // VINT (interlaced vertical scaling)
if (width < 385)
value |= (1 << 0); // 3-tap interpolation
if (width < 193)
value |= (1 << 1); // 5-tap interpolation
if (nocomb)
value |= (3 << 5); // disable comb filter
cx_write(MO_FILTER_EVEN, value);
cx_write(MO_FILTER_ODD, value);
dprintk(1,"set_scale: filter 0x%04x\n", value);
return 0;
}
static const u32 xtal = 28636363;
static int set_pll(struct cx88_core *core, int prescale, u32 ofreq)
{
static u32 pre[] = { 0, 0, 0, 3, 2, 1 };
u64 pll;
u32 reg;
int i;
if (prescale < 2)
prescale = 2;
if (prescale > 5)
prescale = 5;
pll = ofreq * 8 * prescale * (u64)(1 << 20);
do_div(pll,xtal);
reg = (pll & 0x3ffffff) | (pre[prescale] << 26);
if (((reg >> 20) & 0x3f) < 14) {
printk("%s/0: pll out of range\n",core->name);
return -1;
}
dprintk(1,"set_pll: MO_PLL_REG 0x%08x [old=0x%08x,freq=%d]\n",
reg, cx_read(MO_PLL_REG), ofreq);
cx_write(MO_PLL_REG, reg);
for (i = 0; i < 100; i++) {
reg = cx_read(MO_DEVICE_STATUS);
if (reg & (1<<2)) {
dprintk(1,"pll locked [pre=%d,ofreq=%d]\n",
prescale,ofreq);
return 0;
}
dprintk(1,"pll not locked yet, waiting ...\n");
msleep(10);
}
dprintk(1,"pll NOT locked [pre=%d,ofreq=%d]\n",prescale,ofreq);
return -1;
}
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
int cx88_start_audio_dma(struct cx88_core *core)
{
V4L/DVB (3266): Fix NICAM buzz on analog sound Apparently, having the number of lines fixed at 4 reduces (or even kills) the buzz found in NICAM stereo with analog sound. Signed-off-by: Marcin Rudowski <mar_rud@poczta.onet.pl> Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-02-06 04:15:14 -07:00
/* constant 128 made buzz in analog Nicam-stereo for bigger fifo_size */
int bpl = cx88_sram_channels[SRAM_CH25].fifo_size/4;
V4L/DVB (4504): Enable audio DMA restart on channel change even when cx88-alsa is compiled cx88-alsa can be compiled in (in distro built kernels, for example), but not used. In those cases, the audio DMA restart is needed for proper sound on NICAM channels. This patch enables the DMA restart even with ALSA, but the functions now check if cx88-alsa is really active. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-08-17 14:40:28 -07:00
/* If downstream RISC is enabled, bail out; ALSA is managing DMA */
if (cx_read(MO_AUD_DMACNTRL) & 0x10)
return 0;
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
/* setup fifo + format */
V4L/DVB (3266): Fix NICAM buzz on analog sound Apparently, having the number of lines fixed at 4 reduces (or even kills) the buzz found in NICAM stereo with analog sound. Signed-off-by: Marcin Rudowski <mar_rud@poczta.onet.pl> Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-02-06 04:15:14 -07:00
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH25], bpl, 0);
cx88_sram_channel_setup(core, &cx88_sram_channels[SRAM_CH26], bpl, 0);
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
V4L/DVB (3266): Fix NICAM buzz on analog sound Apparently, having the number of lines fixed at 4 reduces (or even kills) the buzz found in NICAM stereo with analog sound. Signed-off-by: Marcin Rudowski <mar_rud@poczta.onet.pl> Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-02-06 04:15:14 -07:00
cx_write(MO_AUDD_LNGTH, bpl); /* fifo bpl size */
cx_write(MO_AUDR_LNGTH, bpl); /* fifo bpl size */
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
/* start dma */
cx_write(MO_AUD_DMACNTRL, 0x0003); /* Up and Down fifo enable */
V4L/DVB (4504): Enable audio DMA restart on channel change even when cx88-alsa is compiled cx88-alsa can be compiled in (in distro built kernels, for example), but not used. In those cases, the audio DMA restart is needed for proper sound on NICAM channels. This patch enables the DMA restart even with ALSA, but the functions now check if cx88-alsa is really active. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-08-17 14:40:28 -07:00
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
return 0;
}
int cx88_stop_audio_dma(struct cx88_core *core)
{
V4L/DVB (4504): Enable audio DMA restart on channel change even when cx88-alsa is compiled cx88-alsa can be compiled in (in distro built kernels, for example), but not used. In those cases, the audio DMA restart is needed for proper sound on NICAM channels. This patch enables the DMA restart even with ALSA, but the functions now check if cx88-alsa is really active. Signed-off-by: Ricardo Cerqueira <v4l@cerqueira.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-08-17 14:40:28 -07:00
/* If downstream RISC is enabled, bail out; ALSA is managing DMA */
if (cx_read(MO_AUD_DMACNTRL) & 0x10)
return 0;
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
/* stop dma */
cx_write(MO_AUD_DMACNTRL, 0x0000);
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
static int set_tvaudio(struct cx88_core *core)
{
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
v4l2_std_id norm = core->tvnorm;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (6021): cx88: Copy board information into card state The cx88 driver state stored the ID of the board type in core->board. Every time the driver need to get some information about the board configuration, it uses the board number as an index into board configuration array. This patch changes it so that the board number is in core->boardnr, and core->board is a copy of the board configuration information. This allows access to board information without the extra indirection. e.g. cx88_boards[core->board].mpeg becomes core->board.mpeg. This has a number of advantages: - The code is simpler to write. - It compiles to be smaller and faster, without needing the extra array lookup to get at the board information. - The cx88_boards array no longer needs to be exported to all cx88 modules. - The boards array can be made const - It should be possible to avoid keeping the (large) cx88_boards array around after the module is loaded. - If module parameters or eeprom info override some board configuration setting, it's not necessary to modify the boards array, which would affect all boards of the same type. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-15 10:41:57 -07:00
if (CX88_VMUX_TELEVISION != INPUT(core->input).type)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return 0;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (V4L2_STD_PAL_BG & norm) {
[PATCH] v4l: 648: some clean up in cx88 tvaudio c - Some clean up in cx88-tvaudio.c Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:27 -07:00
core->tvaudio = WW_BG;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (V4L2_STD_PAL_DK & norm) {
[PATCH] v4l: 648: some clean up in cx88 tvaudio c - Some clean up in cx88-tvaudio.c Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:27 -07:00
core->tvaudio = WW_DK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (V4L2_STD_PAL_I & norm) {
[PATCH] v4l: 648: some clean up in cx88 tvaudio c - Some clean up in cx88-tvaudio.c Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:27 -07:00
core->tvaudio = WW_I;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (V4L2_STD_SECAM_L & norm) {
[PATCH] v4l: 648: some clean up in cx88 tvaudio c - Some clean up in cx88-tvaudio.c Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:27 -07:00
core->tvaudio = WW_L;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (V4L2_STD_SECAM_DK & norm) {
[PATCH] v4l: 648: some clean up in cx88 tvaudio c - Some clean up in cx88-tvaudio.c Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-08 22:36:27 -07:00
core->tvaudio = WW_DK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if ((V4L2_STD_NTSC_M & norm) ||
(V4L2_STD_PAL_M & norm)) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
core->tvaudio = WW_BTSC;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (V4L2_STD_NTSC_M_JP & norm) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
core->tvaudio = WW_EIAJ;
} else {
printk("%s/0: tvaudio support needs work for this tv norm [%s], sorry\n",
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
core->name, v4l2_norm_to_name(core->tvnorm));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
core->tvaudio = 0;
return 0;
}
cx_andor(MO_AFECFG_IO, 0x1f, 0x0);
cx88_set_tvaudio(core);
[PATCH] v4l: CX88 updates and card additions - Remove $Id CVS logs for V4L files - add ioctl indirection via cx88_ioctl_hook and cx88_ioctl_translator to cx88-blackbird.c. - declare the indirection hooks from cx88-blackbird.c. - dcprintk macro which uses core instead of dev->core on cx88-video.c. - replace dev->core occurances with core on cx88-video.c. - CodingStyle fixes. - MaxInput replaced by a define. - cx8801 structures moved from cx88.h. - The output_mode needs to be set for the Hauppauge Nova-T DVB-T for versions after 2.6.12. - Corrected GPIO values for cx88 cards #28 & #31 for s-video and composite. - Updated DViCO FusionHDTV5 Gold & added DVB support. - Fixed DViCO FusionHDTV 3 Gold-Q GPIO. - Some clean up in cx88-tvaudio.c - replaced hex values when writing to AUD_CTL to EN_xx for better reading. - Allow select by hand between Mono, Lang1, Lang2 and Stereo for BTSC. - Support for stereo NICAM and BTSC improved. - Broken stereo check removed. - Added support for remote control to Cinergy DVBT-1400. - local var renamed from rc5 to a better name (ircode). - LGDT330X QAM lock bug fixes. - Some reorg: move some bits to struct cx88_core, factor out common ioctl's to cx88_do_ioctl. - Get rid of '//' comments, replace them with #if 0 and /**/. - Minor clean-ups: remove dcprintk and replace all instances of "dev->core" with "core". - Added some registers to control PCI controller at CX2388x chips. - New tuner standby API. - Small mpeg fixes and cleanups for blackbird. - fix mpeg packet size & count - add VIDIOC_QUERYCAP ioctl for the mpeg stream - return more information in struct v4l2_format - fix default window height - small cleanups Signed-off-by: Uli Luckas <luckas@musoft.de> Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Nickolay V. Shmyrev <nshmyrev@yandex.ru> Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Patrick Boettcher <patrick.boettcher@desy.de> Signed-off-by: Catalin Climov <catalin@climov.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-09 13:03:41 -07:00
/* cx88_set_stereo(dev,V4L2_TUNER_MODE_STEREO); */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
/*
This should be needed only on cx88-alsa. It seems that some cx88 chips have
bugs and does require DMA enabled for it to work.
*/
cx88_start_audio_dma(core);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return 0;
}
[PATCH] V4L: Enables audio DMA setting on cx88 chips, even when dma not in use - Enabled audio DMA transfer code even when DMA not in use to solve a problem on some broken cx88 chips. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-12-01 01:51:34 -07:00
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
int cx88_set_tvnorm(struct cx88_core *core, v4l2_std_id norm)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
u32 fsc8;
u32 adc_clock;
u32 vdec_clock;
u32 step_db,step_dr;
u64 tmp64;
u32 bdelay,agcdelay,htotal;
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
u32 cxiformat, cxoformat;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
core->tvnorm = norm;
fsc8 = norm_fsc8(norm);
adc_clock = xtal;
vdec_clock = fsc8;
step_db = fsc8;
step_dr = fsc8;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
if (norm & V4L2_STD_NTSC_M_JP) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatNTSCJapan;
cxoformat = 0x181f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_NTSC_443) {
V4L/DVB (5098): Added support for V4L2_STD_NTSC_443 Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:20 -07:00
cxiformat = VideoFormatNTSC443;
cxoformat = 0x181f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_PAL_M) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatPALM;
cxoformat = 0x1c1f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_PAL_N) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatPALN;
cxoformat = 0x1c1f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_PAL_Nc) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatPALNC;
cxoformat = 0x1c1f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_PAL_60) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatPAL60;
cxoformat = 0x181f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_NTSC) {
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat = VideoFormatNTSC;
cxoformat = 0x181f0008;
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
} else if (norm & V4L2_STD_SECAM) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
step_db = 4250000 * 8;
step_dr = 4406250 * 8;
V4L/DVB (5098): Added support for V4L2_STD_NTSC_443 Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:20 -07:00
cxiformat = VideoFormatSECAM;
cxoformat = 0x181f0008;
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
} else { /* PAL */
cxiformat = VideoFormatPAL;
cxoformat = 0x181f0008;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
dprintk(1,"set_tvnorm: \"%s\" fsc8=%d adc=%d vdec=%d db/dr=%d/%d\n",
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
v4l2_norm_to_name(core->tvnorm), fsc8, adc_clock, vdec_clock,
step_db, step_dr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
set_pll(core,2,vdec_clock);
dprintk(1,"set_tvnorm: MO_INPUT_FORMAT 0x%08x [old=0x%08x]\n",
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxiformat, cx_read(MO_INPUT_FORMAT) & 0x0f);
cx_andor(MO_INPUT_FORMAT, 0xf, cxiformat);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
// FIXME: as-is from DScaler
dprintk(1,"set_tvnorm: MO_OUTPUT_FORMAT 0x%08x [old=0x%08x]\n",
V4L/DVB (5097): Convert cx8800 driver to video_ioctl2 handler video_ioctl2 handler provides V4L2 API parsing. Using it makes the driver simpler, and isolates API parsing. This allows future reusage of driver controls using other ways, like sysfs and/or procfs and increases isolation of driver-specific handling from the generic common ioctl processing. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:17 -07:00
cxoformat, cx_read(MO_OUTPUT_FORMAT));
cx_write(MO_OUTPUT_FORMAT, cxoformat);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
// MO_SCONV_REG = adc clock / video dec clock * 2^17
tmp64 = adc_clock * (u64)(1 << 17);
do_div(tmp64, vdec_clock);
dprintk(1,"set_tvnorm: MO_SCONV_REG 0x%08x [old=0x%08x]\n",
(u32)tmp64, cx_read(MO_SCONV_REG));
cx_write(MO_SCONV_REG, (u32)tmp64);
// MO_SUB_STEP = 8 * fsc / video dec clock * 2^22
tmp64 = step_db * (u64)(1 << 22);
do_div(tmp64, vdec_clock);
dprintk(1,"set_tvnorm: MO_SUB_STEP 0x%08x [old=0x%08x]\n",
(u32)tmp64, cx_read(MO_SUB_STEP));
cx_write(MO_SUB_STEP, (u32)tmp64);
// MO_SUB_STEP_DR = 8 * 4406250 / video dec clock * 2^22
tmp64 = step_dr * (u64)(1 << 22);
do_div(tmp64, vdec_clock);
dprintk(1,"set_tvnorm: MO_SUB_STEP_DR 0x%08x [old=0x%08x]\n",
(u32)tmp64, cx_read(MO_SUB_STEP_DR));
cx_write(MO_SUB_STEP_DR, (u32)tmp64);
// bdelay + agcdelay
bdelay = vdec_clock * 65 / 20000000 + 21;
agcdelay = vdec_clock * 68 / 20000000 + 15;
dprintk(1,"set_tvnorm: MO_AGC_BURST 0x%08x [old=0x%08x,bdelay=%d,agcdelay=%d]\n",
(bdelay << 8) | agcdelay, cx_read(MO_AGC_BURST), bdelay, agcdelay);
cx_write(MO_AGC_BURST, (bdelay << 8) | agcdelay);
// htotal
tmp64 = norm_htotal(norm) * (u64)vdec_clock;
do_div(tmp64, fsc8);
V4L/DVB (4683): Norm_notchfilter is used on just one point and argument is bogus Removed norm_notchfilter function. All the code is bound to 4 x FSC, so, any other filter won't work fine. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-09-29 08:39:36 -07:00
htotal = (u32)tmp64 | (HLNotchFilter4xFsc << 11);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
dprintk(1,"set_tvnorm: MO_HTOTAL 0x%08x [old=0x%08x,htotal=%d]\n",
htotal, cx_read(MO_HTOTAL), (u32)tmp64);
cx_write(MO_HTOTAL, htotal);
V4L/DVB (4002): Adjust VBI offset to match value reported in VBI format The actual offset of the start of VBI data is incorrectly documented for both the cx2388x and bt8x8. For the cx2388x, it appears to be about 220 to 224 + VBI_V_DEL*2. The driver has been incorrectly reporting an offset of 244. This patch adjusts VBI_V_DEL to 10, so that the offset of 244 is correct. This is the same offset as the bttv driver, and expected by some software. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-05-23 19:54:44 -07:00
// vbi stuff, set vbi offset to 10 (for 20 Clk*2 pixels), this makes
// the effective vbi offset ~244 samples, the same as the Bt8x8
cx_write(MO_VBI_PACKET, (10<<11) | norm_vbipack(norm));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
// this is needed as well to set all tvnorm parameter
cx88_set_scale(core, 320, 240, V4L2_FIELD_INTERLACED);
// audio
set_tvaudio(core);
// tell i2c chips
V4L/DVB (5102): make videodev to auto-generate standards v4l2_tvnorm were meant to describe video standards and its names to V4L2 API. However, this were doing by some static structures at the driver. This patch changes the internals in a way that, at the driver, only a v4l2_tvnorm (a 64 bit integer) should be filled, with all supported tvnorms. videodev will dynamically generate the proper API array based on supported standards. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-01-20 09:58:33 -07:00
cx88_call_i2c_clients(core,VIDIOC_S_STD,&norm);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
// done
return 0;
}
/* ------------------------------------------------------------------ */
struct video_device *cx88_vdev_init(struct cx88_core *core,
struct pci_dev *pci,
struct video_device *template,
char *type)
{
struct video_device *vfd;
vfd = video_device_alloc();
if (NULL == vfd)
return NULL;
*vfd = *template;
vfd->minor = -1;
vfd->dev = &pci->dev;
vfd->release = video_device_release;
snprintf(vfd->name, sizeof(vfd->name), "%s %s (%s)",
V4L/DVB (6021): cx88: Copy board information into card state The cx88 driver state stored the ID of the board type in core->board. Every time the driver need to get some information about the board configuration, it uses the board number as an index into board configuration array. This patch changes it so that the board number is in core->boardnr, and core->board is a copy of the board configuration information. This allows access to board information without the extra indirection. e.g. cx88_boards[core->board].mpeg becomes core->board.mpeg. This has a number of advantages: - The code is simpler to write. - It compiles to be smaller and faster, without needing the extra array lookup to get at the board information. - The cx88_boards array no longer needs to be exported to all cx88 modules. - The boards array can be made const - It should be possible to avoid keeping the (large) cx88_boards array around after the module is loaded. - If module parameters or eeprom info override some board configuration setting, it's not necessary to modify the boards array, which would affect all boards of the same type. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-15 10:41:57 -07:00
core->name, type, core->board.name);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return vfd;
}
struct cx88_core* cx88_core_get(struct pci_dev *pci)
{
struct cx88_core *core;
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
mutex_lock(&devlist);
V4L/DVB (6312): cx88: Replace list_for_each+list_entry with list_for_each_entry Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-10-10 01:37:40 -07:00
list_for_each_entry(core, &cx88_devlist, devlist) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (pci->bus->number != core->pci_bus)
continue;
if (PCI_SLOT(pci->devfn) != core->pci_slot)
continue;
V4L/DVB (6022): cx88: Move card core creation from cx88-core.c to cx88-cards.c A lot of code in cx88-cards.c was only used by cx88-core.c when the core state is first allocated and initialized. Moving that task to cx88-cards makes the driver simpler and the files more self contained. - Module parameters tuner, radio, card, and latency move to cx88-cards.c - cx88_boards is made static - cx88_subids is made static and const - cx88_bcount is eliminated - cx88_idcount is eliminated - cx88_card_list() is made static - cx88_card_setup_pre_i2c() is made static - cx88_card_setup() is made static - cx88_pci_quirks() is moved from cx88-core to cx88-cards The function argument "char *name" is made const too - get_ressources() is moved from cx88-core to cx88-cards, and renamed to cx88_get_resources() - The code to allocate and initialize the core state struct and the chip is moved out of cx88-core.c:cx88_get_core() and into a new function in cx88-cards.c, cx88_core_create(). This makes both functions simpler. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-15 10:41:58 -07:00
if (0 != cx88_get_resources(core, pci)) {
mutex_unlock(&devlist);
return NULL;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
atomic_inc(&core->refcount);
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
mutex_unlock(&devlist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return core;
}
[PATCH] v4l: CX88 cards update This patch adds support for various CX88 cards and allows specifying card addresses. Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: cybercide@f2s.com <cybercide@f2s.com> Signed-off-by: Catalin Climov <catalin@climov.com> Signed-off-by: Nickolay V Shmyrev <nshmyrev@yandex.ru> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-06-23 22:05:03 -07:00
V4L/DVB (6022): cx88: Move card core creation from cx88-core.c to cx88-cards.c A lot of code in cx88-cards.c was only used by cx88-core.c when the core state is first allocated and initialized. Moving that task to cx88-cards makes the driver simpler and the files more self contained. - Module parameters tuner, radio, card, and latency move to cx88-cards.c - cx88_boards is made static - cx88_subids is made static and const - cx88_bcount is eliminated - cx88_idcount is eliminated - cx88_card_list() is made static - cx88_card_setup_pre_i2c() is made static - cx88_card_setup() is made static - cx88_pci_quirks() is moved from cx88-core to cx88-cards The function argument "char *name" is made const too - get_ressources() is moved from cx88-core to cx88-cards, and renamed to cx88_get_resources() - The code to allocate and initialize the core state struct and the chip is moved out of cx88-core.c:cx88_get_core() and into a new function in cx88-cards.c, cx88_core_create(). This makes both functions simpler. Signed-off-by: Trent Piepho <xyzzy@speakeasy.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2007-08-15 10:41:58 -07:00
core = cx88_core_create(pci, cx88_devcount);
if (NULL != core) {
cx88_devcount++;
list_add_tail(&core->devlist, &cx88_devlist);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
mutex_unlock(&devlist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return core;
}
void cx88_core_put(struct cx88_core *core, struct pci_dev *pci)
{
release_mem_region(pci_resource_start(pci,0),
pci_resource_len(pci,0));
if (!atomic_dec_and_test(&core->refcount))
return;
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
mutex_lock(&devlist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
cx88_ir_fini(core);
if (0 == core->i2c_rc)
i2c: Discard the i2c algo del_bus wrappers They are all only calling i2c_del_adapter, so we may as well do it directly. Signed-off-by: Jean Delvare <khali@linux-fr.org>
2006-12-10 13:21:33 -07:00
i2c_del_adapter(&core->i2c_adap);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
list_del(&core->devlist);
iounmap(core->lmmio);
cx88_devcount--;
V4L/DVB (3380): Semaphore to mutex conversion on drivers/media - Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
2006-01-15 02:52:23 -07:00
mutex_unlock(&devlist);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
kfree(core);
}
/* ------------------------------------------------------------------ */
EXPORT_SYMBOL(cx88_print_irqbits);
EXPORT_SYMBOL(cx88_core_irq);
EXPORT_SYMBOL(cx88_wakeup);
EXPORT_SYMBOL(cx88_reset);
EXPORT_SYMBOL(cx88_shutdown);
EXPORT_SYMBOL(cx88_risc_buffer);
EXPORT_SYMBOL(cx88_risc_databuffer);
EXPORT_SYMBOL(cx88_risc_stopper);
EXPORT_SYMBOL(cx88_free_buffer);
EXPORT_SYMBOL(cx88_sram_channels);
EXPORT_SYMBOL(cx88_sram_channel_setup);
EXPORT_SYMBOL(cx88_sram_channel_dump);
EXPORT_SYMBOL(cx88_set_tvnorm);
EXPORT_SYMBOL(cx88_set_scale);
EXPORT_SYMBOL(cx88_vdev_init);
EXPORT_SYMBOL(cx88_core_get);
EXPORT_SYMBOL(cx88_core_put);
V4L/DVB (6247): Fix bug #8689: Fixes IR stop/start during suspend/resume IR workqueue should be disabled during suspend. This avoids some troubles, like the one reported on bug #8689: "The Hauppauge HVR 1100 ir-remote control does not work after resume from suspend to ram or disk." This patch disables IR before suspending, re-enabling it after resume. Thanks to Peter Poklop <Peter.Poklop@gmx.at> for reporting it and helping with the fix. Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org> Reviewed-by: Peter Poklop <Peter.Poklop@gmx.at>
2007-10-01 04:51:39 -07:00
EXPORT_SYMBOL(cx88_ir_start);
EXPORT_SYMBOL(cx88_ir_stop);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/*
* Local variables:
* c-basic-offset: 8
* End:
[PATCH] v4l: CX88 updates and card additions - Remove $Id CVS logs for V4L files - add ioctl indirection via cx88_ioctl_hook and cx88_ioctl_translator to cx88-blackbird.c. - declare the indirection hooks from cx88-blackbird.c. - dcprintk macro which uses core instead of dev->core on cx88-video.c. - replace dev->core occurances with core on cx88-video.c. - CodingStyle fixes. - MaxInput replaced by a define. - cx8801 structures moved from cx88.h. - The output_mode needs to be set for the Hauppauge Nova-T DVB-T for versions after 2.6.12. - Corrected GPIO values for cx88 cards #28 & #31 for s-video and composite. - Updated DViCO FusionHDTV5 Gold & added DVB support. - Fixed DViCO FusionHDTV 3 Gold-Q GPIO. - Some clean up in cx88-tvaudio.c - replaced hex values when writing to AUD_CTL to EN_xx for better reading. - Allow select by hand between Mono, Lang1, Lang2 and Stereo for BTSC. - Support for stereo NICAM and BTSC improved. - Broken stereo check removed. - Added support for remote control to Cinergy DVBT-1400. - local var renamed from rc5 to a better name (ircode). - LGDT330X QAM lock bug fixes. - Some reorg: move some bits to struct cx88_core, factor out common ioctl's to cx88_do_ioctl. - Get rid of '//' comments, replace them with #if 0 and /**/. - Minor clean-ups: remove dcprintk and replace all instances of "dev->core" with "core". - Added some registers to control PCI controller at CX2388x chips. - New tuner standby API. - Small mpeg fixes and cleanups for blackbird. - fix mpeg packet size & count - add VIDIOC_QUERYCAP ioctl for the mpeg stream - return more information in struct v4l2_format - fix default window height - small cleanups Signed-off-by: Uli Luckas <luckas@musoft.de> Signed-off-by: Torsten Seeboth <Torsten.Seeboth@t-online.de> Signed-off-by: Nickolay V. Shmyrev <nshmyrev@yandex.ru> Signed-off-by: Michael Krufky <mkrufky@m1k.net> Signed-off-by: Patrick Boettcher <patrick.boettcher@desy.de> Signed-off-by: Catalin Climov <catalin@climov.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-09 13:03:41 -07:00
* kate: eol "unix"; indent-width 3; remove-trailing-space on; replace-trailing-space-save on; tab-width 8; replace-tabs off; space-indent off; mixed-indent off
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*/
Copy Permalink